• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Ratko TA, Belinson SE, Brown HM, et al. Hematopoietic Stem-Cell Transplantation in the Pediatric Population [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Feb. (Comparative Effectiveness Reviews, No. 48.)

Cover of Hematopoietic Stem-Cell Transplantation in the Pediatric Population

Hematopoietic Stem-Cell Transplantation in the Pediatric Population [Internet].

Show details

Systematic Reviews

Table 36 lists the indications to be addressed as part of the systematic reviews of this report.

Table 36. Pediatric HSCT indications to be addressed with systematic review.

Table 36

Pediatric HSCT indications to be addressed with systematic review.

Systematic Reviews: Malignant, Nonhematopoietic Disease

Ewing's Sarcoma Family of Tumors Systematic Review

Background and Indication

The Ewing's sarcoma family of tumors (ESFT) is the second most common primary malignant bone tumor in children, adolescents and young adults. ESFTs include Ewing tumor of bone (classic Ewing sarcoma and primitive neuroectodermal tumor or PNET) and extraosseous Ewing (i.e., Ewing sarcoma in a site other than bone). The incidence of ESFT is approximately 3 cases per 1,000,000 persons per year. The incidence in the U.S. population is one per 1,000,000 in the population.352 The median age of patients is 15 years, and more than 50 percent of patients are adolescents. Primary sites of bone disease include lower extremity (41 percent), pelvis (26 percent), chest wall (16 percent), upper extremity (9 percent), spine (6 percent) and skull (2 percent).352 Primary sites of extraosseous Ewing's are trunk (32 percent), extremity (26 percent), head and neck (18 percent), retroperitoneum (16 percent) and other sites (9 percent).352 Approximately 25 percent of patients will have metastatic disease at diagnosis.352

Certain adverse prognostic factors place some patients with ESFT into a high-risk category: relapsed or resistant disease, primary tumor site in the axial skeleton, including pelvis, large tumor volume, and the presence of metastatic disease (patients with isolated lung metastases are considered to have better prognosis than patients with metastases to bone and/or bone marrow). Treatment of ESFT includes systemic chemotherapy in conjunction with either surgery or radiation or both for local tumor control.

Overall survival rates for localized ESFT have dramatically improved over the last 30 years, however, the prognosis for patients with high-risk tumors treated with conventional chemotherapy, radiation and surgery remain poor, with long-term survival rates for patients with metastatic disease less than 35 percent.352 Patients with lung-only metastases have been reported to have 4-year EFS of approximately 40 percent, whereas patients with bone/bone marrow metastases have 4-year EFS of approximately 28 percent and with combined lung and bone/bone marrow metastases 4-year EFS of approximately 14 percent. Relapsed ESFT treated with conventional-dose chemotherapy, radiation and surgery has been reported to have a 2-year event free survival of less than 10 percent.

Chemotherapy for patients with ESFT initially was based on four drugs: doxorubicin, cyclophosphamide, vincristine, and dactinomycin. More recently, treatment has included ifosfamide, with or without etoposide. Dose-intensive chemotherapy regimens as well as HSCT have been investigated in patients with high-risk ESFT in an effort to improve survival.

Evidence Summary

The overall grade of strength of evidence for overall survival and the use of single and tandem HSCT for the treatment of high-risk Ewing's Sarcoma Family of Tumors (ESFT) is shown in Table 37.

Table 37. Overall grade of strength of evidence for overall survival and the use of single and tandem HSCT for the treatment of high-risk Ewing's Sarcoma Family of Tumors (ESFT).

Table 37

Overall grade of strength of evidence for overall survival and the use of single and tandem HSCT for the treatment of high-risk Ewing's Sarcoma Family of Tumors (ESFT).

Single HSCT

The literature using dose-intensive chemotherapeutic regimens or HSCT consists of case series with small numbers of patients and case reports without direct comparisons between conventional or dose-intensive chemotherapy and HSCT. The evidence compiled for this review includes, for HSCT, 24 case series353-376 (including two Phase II studies) and six case reports.377-382 The comparator is conventional chemotherapy and includes seven case series (including one Phase II study).116, 376, 383-387 No information on quality of life (QOL) was provided and data on adverse events were sparse and based on small numbers of patients.

The evidence suggests that treatment-related mortality is higher in the patients that underwent HSCT compared to the chemotherapy comparators. The rate of secondary malignancies appeared lower in some reports of dose-intensive chemotherapy compared to HSCT and similar in one report of dose-intensive chemotherapy compared to HSCT.

Tandem Autologous-Autologous HSCT

The literature using tandem HSCT consists of case series with small numbers of patients and a case report.355, 380 A direct comparison between tandem HSCT and single HSCT is reported in one case series.354 The evidence compiled for this review includes, for tandem HSCT, two case series and one case report. The comparator is single HSCT and includes 24 case series and six case reports. Data on transplant-related mortality and infectious complications were sparse; data on other adverse effects were not reported.

Results

Table 38 arrays the study selection criteria for ESFT.

Table 38. Study selection criteria for ESFT.

Table 38

Study selection criteria for ESFT.

Table 39 shows the study design and population. Seventeen studies were based in Europe,353, 354, 356, 360, 362, 363, 366, 368, 369, 372, 374-376, 378, 380, 385, 386 seven in Asia,357, 358, 370, 371, 379, 382 373 and 12 in the U.S. and Canada.355, 359, 361, 364, 365, 367, 377, 381, 383, 384, 387, 388 The total number of patients for which data was abstracted from the 36 studies was 751 (468 HSCT and 283 chemotherapy). Twenty-eight studies included patients who underwent a single autologous or allogeneic HSCT.353, 354, 356-371, 377-379, 381, 382 372-376 Three studies reported outcomes for tandem autologous-autologous HSCT.354, 355, 380

Table 39. ESFT study characteristics and population.

Table 39

ESFT study characteristics and population.

Seven studies included in this analysis involved patients who underwent conventional chemotherapy.383-388 376 The patients who underwent conventional therapy were used as the comparators to the single HSCT population and the single HSCT population was used as the comparator to tandem HSCT population.

Table 40 shows the outcomes that were reported across studies.

Table 40. ESFT outcomes reported.

Table 40

ESFT outcomes reported.

Overall Survival

Data on overall survival were reported or generated in 20 HSCT studies353, 355-358, 360, 362-366, 368-371, 373-376, 389 and four comparator studies (Table 41).385-388 No direct comparisons can be made from the published data as there are no comparative studies.

Table 41. Overall survival for treatment (single HSCT and tandem auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

Table 41

Overall survival for treatment (single HSCT and tandem auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

None of the studies evaluated quality of life. Data on treatment-related mortality was reported in 14 HSCT studies353, 355 356, 363-365, 367-371, 374, 375 376 and three comparative studies.385, 387 388 (Table 42). Eleven HSCT353, 355, 356, 359, 360, 364, 370, 372, 374, 375 376 and two comparator studies385, 388 reported serious infectious complications. Six HSCT studies353, 365, 374, 375, 377 376 and four comparator studies383, 384, 386, 388 reported a secondary malignancy. Seven HSCT studies356, 359, 361, 381 372, 374, 375 and one comparator study385 reported other long-term complications involving severe organ dysfunction.

Table 42. Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups.

Table 42

Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups.

Ongoing Studies

Two ongoing Phase III trials will include an HSCT arm in the treatment of patients with high-risk ESFT:

  • A study in localized and disseminated Ewing Sarcoma (EWING 2008; NCT00987636) will include a randomized trial arm for high-risk Ewing's (localized and unfavorable histological response or tumor volume greater than 200 mL) examining whether HSCT compared with standard chemotherapy improves EFS. Patients with pulmonary metastases will be randomized to HSCT versus standard chemotherapy and whole lung irradiation. Very high-risk patients (with primary disseminated disease) will be randomized to HSCT versus standard chemotherapy. Estimated enrollment is 1,383 with an estimated study completion date of March 2018.
  • A randomized trial is comparing chemotherapy with or without peripheral stem-cell transplantation, radiation, and/or surgery (EURO-EWING 99; NCT00020566). Primary outcome measures include EFS and OS. Estimated enrollment is 1,200 with an estimated primary completion date of December 2011.

Conclusion

Low strength evidence on overall survival suggests no benefit with single HSCT compared to conventional therapy for the treatment of high-risk ESFT.

The body of evidence on overall survival with tandem HSCT compared to single HSCT for the treatment of high-risk ESFT and overall survival is insufficient to draw conclusions.

Wilms Tumor Systematic Review

Background and Setting

Wilms tumor is the fifth most common pediatric malignancy and the most common type of renal tumor in children. The incidence of Wilms tumor is approximately 0.8 cases per 100,000 persons, with approximately 500 new cases diagnosed each year in the U.S., 6 percent involving both kidneys.390 Most cases occur sporadically, whereas some are hereditary or associated with certain syndromes. Wilms tumor is diagnosed at a mean age of 3.5 years, and is unusual after the age of 6.391Overall survival rates for Wilms tumor are approximately 90 percent with first-line therapy consisting of surgery, chemotherapy and in some cases radiation therapy (to the abdomen and/or lungs).390 However, approximately 15 percent of patients with favorable (nonanaplastic) histology and 50 percent of patients with anaplastic histology experience tumor recurrence.381 Recurrent Wilms tumor is a heterogeneous disease and treatment is generally based upon patient risk stratification. For patients with favorable prognostic features, standard-dose chemotherapy may be curative.

Patients with relapsed disease and adverse prognostic factors are considered as a high-risk relapse category. Adverse prognostic factors include initial advanced tumor stage, anaplastic histology, early recurrence (less than 6 months after diagnosis), recurrence in multiple organs or in a previously irradiated field, and initial chemotherapy consisting of vincristine, actinomycin D, and doxorubicin (versus vincristine and actinomycin D alone). Since the identification of this high-risk group of patients with relapsed disease and the poor outcome after initial treatment with chemotherapy consisting of vincristine, actinomycin D, and doxorubicin (VAD) and radiation therapy, investigation now focuses on the activity of ifosfamide, etoposide, and platinum analogs as single agents or in combination, and in more intensive doses. Other intensive dose strategies include the use of myeloablative chemotherapeutic regimens and HSCT.

Evidence Summary

The overall grade of strength of comparative study evidence for overall survival and the use of HSCT for the treatment of high-risk relapsed Wilms tumor is shown in Table 43.

Table 43. Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk relapsed Wilms tumor.

Table 43

Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk relapsed Wilms tumor.

The literature using dose-intensive chemotherapeutic regimens consists of case series with small numbers of patients, without direct comparisons between conventional intensive chemotherapy and HSCT.

The evidence compiled for this review includes 13 case series364, 392-403 and seven case reports.378, 404-409 The comparator is conventional chemotherapy. Although direct comparisons are difficult to make between dose-intensive chemotherapy and HSCT in high-risk relapsed Wilms, based on the current systematic review, there does not appear to be a difference in progression-free or overall survival between the two groups. No information on quality of life was provided and data on adverse events was sparse and therefore insufficient to make conclusions regarding adverse effects and quality of life.

Results

Thirty-eight articles were retrieved for full-text screening. Twenty reports were included in this review, and the remaining 18 articles were excluded. Table 44 arrays the criteria that were used to select studies for this section.

Table 44. Wilms tumor study selection criteria.

Table 44

Wilms tumor study selection criteria.

Table 45 shows the study designs and population. Of the included publications, 13 were case series364, 392-403 and seven were case reports.378, 404-407 Nine studies were based in Europe,378, 392-394, 397, 398, 400, 404, 405 one in Asia,401 two in South America,399, 410 and eight in the U.S.395, 396, 402, 403, 406-409

Table 45. Wilms tumor study characteristics and population.

Table 45

Wilms tumor study characteristics and population.

The total number of patients for which data was abstracted from the twenty studies was 202: 114 patients received HSCT, whereas 88 patients received chemotherapy.

Fifteen studies included patients who underwent HSCT,378, 392-400, 404, 406, 407 two studies contained data for patients treated either with HSCT or conventional therapy,401, 410 one study contained a report of double sequential high-dose chemotherapy with HSCT,405 and two studies included in this analysis contained only patients that underwent conventional chemotherapy.402, 403 The patients who underwent conventional therapy were used as the comparators to the HSCT population. No studies were identified using tandem autologous HSCT. Patients from these 20 studies received HSCT or conventional chemotherapy for relapsed (first or subsequent), progressive disease, or metastatic disease and one study included patients in first complete remission with bilateral disease (stage V).

Table 46 shows the outcomes that were reported across studies.

Table 46. Wilms tumor outcomes reported.

Table 46

Wilms tumor outcomes reported.

Overall Survival

Data on overall survival were reported in fifteen studies (Table 47).378, 392, 394-398, 400-403, 405-407, 410 No direct comparisons can be made from the published data as there are no comparative studies.

Table 47. Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

Table 47

Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

None of the studies evaluated quality of life. Data on treatment-related mortality was reported in 10 studies (Table 48).378, 392, 394, 395, 397, 398, 400, 402, 403, 406 Two studies reported a case of serious infection leading to death394, 403 and one study reported no serious infectious complications.407 One study reported a secondary malignancy.403 One study reported a case of mild veno-occlusive disease.408 There were no reports of other long-term complications.

Table 48. Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/-radiation) groups.

Table 48

Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/-radiation) groups.

Ongoing Studies

One Phase II trial is ongoing studying chemotherapy followed by surgery and radiation, with or without HSCT in patients with relapsed or refractory Wilms tumor or clear cell sarcoma of the kidney. The study design is interventional and uses one of three regimens (one of which includes HSCT) depending upon patient risk stratification. Primary outcome measures include unified treatment strategy, improvement of current survival rates, efficacy and toxicity and prognostic variables. Estimated enrollment is 75 (50 for HSCT and 25 for each of the non-HSCT regimens). Estimated final data collection date is November 2008 (NCT00025103).

Conclusion

Low strength evidence on overall survival suggests no benefit with single HSCT compared to conventional therapy for the treatment of high-risk relapsed Wilms tumor.

Rhabdomyosarcoma Systematic Review

Background and Setting

The incidence of rhabdomyosarcoma is 4 to 7 cases per 1 million children age 15 or younger;411 approximately 350 new cases are diagnosed each year in the United States.412 The majority of children have an initial presentation of nonmetastatic disease. In this setting conventional treatments have produced at least a 60-70 percent chance of cure.411 Metastatic rhabdomyosarcoma in comparison is generally a lethal disease, with less than 20 percent of patients being cured from their disease.411 Despite the development of new chemotherapy options, the prognosis of these patients remains generally poor.

Some centers have used HDC with HSCT in the setting of high-risk rhabdomyosarcoma. High-risk rhabdomyosarcoma includes primary metastatic or stage III or greater disease and relapsed or refractory disease. Patients with relapsed or refractory disease experience 5-year survival of approximately 30 percent.413 In most series, numbers remain small as the majority of rhabdomyosarcoma cases are cured with conventional treatment; no randomized controlled trials exist.

Data are generally from case series, save two comparative studies414, 415 with patients who received high-dose chemotherapy and HSCT; case reports are also available. While comparative, the study by McDowell and colleagues415 is treated here as two single arms. The focus was to treat a subgroup of high-risk patients with sequential HDC and HSCT and compare them to standard high-risk patients receiving standard chemotherapy. This stratification makes this patient population treated with HSCT not comparable to other treated groups, as they are of generally higher risk than is found in other studies. Prognostic factors identified in prior research were used in identifying those with the poorest prognosis.366, 416, 417 This study provides outcome data for the stratified high-risk rhabdomyosarcoma group, and tested the hypothesis that the highest risk patients may benefit from sequential HDC and stem-cell rescue. Patients traditionally viewed as high-risk, may not have uniform survival outcomes, and may be further stratified based on prognostic factors. Evidence was evaluated in three groups: studies confined to patients with metastatic disease, studies of mixed tumor stage, and “other” (congenital alveolar, cranial parameningeal disease with metastases, and allogeneic transplantation for metastatic disease).

Evidence Summary

The overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk rhabdomyosarcoma is shown in Table 49.

Table 49. Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk rhabdomyosarcoma.

Table 49

Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk rhabdomyosarcoma.

The evidence compiled for this review includes two comparative studies,414, 418 one study comprising two single arms,415 15 case series (nine on HSCT357, 363, 365, 419-424 and six on the comparator conventional chemotherapy387, 413, 416, 425-427) and eight case reports on HSCT.428-435 Two case reports on allogeneic transplantation were also included.420, 436 The total number of patients abstracted from the 26 studies was 887: 340 patients received HSCT, whereas 547 patients received conventional chemotherapy. Patients with embryonal tumors have a better prognosis than those with alveolar histology. Prognostic factors such as age at diagnosis and location of the metastatic disease may help stratify high-risk patients into two groups, those of standard risk and those of poor risk. Treatment with conventional chemotherapy offers three-year survival of about 39 percent.416 Treatment with HSCT does not appear to alter the survival for patients with metastatic rhabdomyosarcoma above what is already achieved with conventional chemotherapy.

The effects of HSCT on survival for pediatric patients with high-risk rhabdomyosarcoma of mixed tumor stage and those with congenital alveolar rhabdomyosarcoma, cranial parameningeal rhabdomyosarcoma with metastasis or the use of allogeneic transplantation for metastatic rhabdomyosarcoma is uncertain. No information on quality of life (QOL) was provided, and data on adverse events was sparse and therefore insufficient to make conclusions regarding adverse effects and quality of life. Two ongoing trials focused on treatment for malignant solid tumors are enrolling children with rhabdomyosarcoma. One is focused on the toxicity of killer IG-like receptor mismatched cord blood, and the other is investigating a tumor lysate-pulsed dendritic cell vaccine for immune augmentation after stem-cell transplantation. Future research aimed to further stratify high-risk pediatric patients with nonmetastatic disease will be important as the field moves towards more targeted therapies.

Results

Sixty articles were retrieved for full-text screening, including articles identified from the bibliography of identified articles and articles containing patients with rhabdomyosarcoma identified in another disease search. Twenty-six reports were included in this review, and the remaining 34 articles were excluded. The total number of patients abstracted from the 26 studies was 887: 346 patients received HSCT, whereas 547 patients received conventional chemotherapy.

Table 50 shows the criteria that were used to select studies for this section.

Table 50. Rhabdomyosarcoma study selection criteria.

Table 50

Rhabdomyosarcoma study selection criteria.

Table 51 shows the study design and population. Of the included publications, two were comparative studies (McDowell et al.415 was abstracted as two single arms); one study was comprised of two single arms. There were 15 case series (nine on HSCT357, 363, 365, 419-424and six on the comparator conventional chemotherapy387, 413, 416, 425-427) and seven case reports on HSCT.428-435 Two case reports on allo-transplantation were also included.420, 436 Eight studies were based in Europe,363, 414, 415, 419, 420, 428, 429, 436 eight in Asia,357, 421, 422, 430-434 one in the Middle East,423 and nine in North America.365, 387, 416, 418, 424-427, 435

Table 51. Rhabdomyosarcoma study characteristics and population.

Table 51

Rhabdomyosarcoma study characteristics and population.

All patients across 18 treatment studies received autologous HSCT as consolidation of primary treatments. Patients in three studies received allogeneic HSCT as consolidation of primary treatments. All patients were considered to have high-risk disease prior to transplant.

For the comparison of tandem to single HSCT, no studies were identified in the search.

All studies were specific to the pediatric age group, with age primarily reported as age at diagnosis; 15 studies reported either mean age or only had one patient. Mean age at diagnosis was approximately 8 years with a range of birth to 17 years. Median or categorical age at diagnosis, reported by 15 studies, was 8 years with a range of 3 to 13.1 years. Across all studies patients were approximately split equally by gender. Studies included patients with diverse histology, approximately 40-50 percent of the patients of alveolar histology, save two studies419, 424 where 63 percent were of alveolar histology. The majority of the remaining patients had embryonal tumors with a small proportion diagnosed with a tumor not otherwise specified or unknown. Induction regimens varied across and within study (i.e., different chemotherapeutic agents and different (cumulative) dosages). The induction regimen consisted of multiple cycles of chemotherapy with or without radiation and/or surgery.

Conditioning regimens also varied across and within studies. The most common regimens included the following agents: melphalan, thiotepa, busulfan, cyclophosphamide, carboplatin and etoposide, either alone or in combination; MEC (melphalan, VP16, and carboplatin) is a common backbone used alone or in combination with radiation therapy or additional drugs. Treatment periods ranged from 1989 to 2005.

Table 52 shows the pediatric outcomes that were reported across the 26 included studies.

Table 52. Rhabdomyosarcoma outcomes reported.

Table 52

Rhabdomyosarcoma outcomes reported.

Overall Survival

Data on overall survival were reported in all but two studies420, 422 (Table 52). Survival data is presented (Table 53). Individual studies varied in their method for calculating overall survival. In general studies of patients with metastatic disease used time since diagnosis, where studies with patients of mixed tumor stage used time from treatment. Similar trends were observed in the 1-, 3-, and 5-year OS across studies. While not direct, comparisons with adequate numbers of participants can be made from both the McDowell415 and Carli414 studies.

Table 53. Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

Table 53

Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups.

The study published by McDowell and colleagues415 stratified patients with metastatic rhabdomyosarcoma into two groups, poor risk and standard risk. Poor-risk patients were identified as those 10 years of age or older with bone or bone marrow involvement.415 These patients were given sequential HDC and HSCT, while the standard-risk patients (younger than 10 years of age and not bone or bone marrow involvement) were treated with conventional chemotherapy. Patients in the standard risk group had 3 year EFS and OS of 54.92 percent and 62.14 percent, respectively, comparable to rates in other studies. While those in the poor-risk group had 3 year EFS and OS of 16.17 percent and 23.17 percent, respectively, statistically worse than those in the standard-risk group in this study and no improvement on prior studies.

Carli et al.414 published results from the European Collaborative MMT4-91. Fifty-two patients in complete remission after induction were given HDC and stem-cell rescue. Outcomes were then compared to 44 patients also in complete remission after induction, but went onto receive conventional chemotherapy. No differences in OS were observed.

The data from additional case series and case reports appear consistent with these findings.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

None of the studies evaluated quality of life, and serious adverse events were reported by fifteen studies (Table 52). Data on treatment-related mortality was reported in twelve studies (Table 54).357, 363, 365, 387, 414, 415, 419, 423-425, 427, 434 McDowell reported two cases of treatment-related mortality in the comparator group and there were seven serious adverse events in the treatment group with five resulting in death; however it is unclear how many occurred in 100 days of treatment.415 Toxic death from sepsis was reported in the treatment group in two studies.414, 420 Bisogno et al.419 reported seven of 55 evaluable patients experienced serious infectious complications while Sandler and colleagues427 reported 40 percent of patients experiencing serious infection with seven leading to death. One study reported a secondary malignancy, myelodysplastic syndrome related to alkylating agents.435 No treatment related mortality was observed in 11 studies.363, 421, 422, 424, 429-433, 435, 436 Two studies416, 426 did not report on adverse events. There were no reports of secondary malignancies, serious hemorrhagic events, irreversible veno-occlusive disease or other long term complications.

Table 54. Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups.

Table 54

Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups.

Ongoing Research

Twenty children age 21 or younger were to be enrolled in a Phase I study examining the toxicity of killer IG-like receptor mismatched umbilical cord blood for pediatric patients with malignant solid tumors. This study is ongoing and no longer recruiting, and no results have been published.

There are no trials specifically looking at HSCT outcomes in patients with rhabdomyosarcoma; however, ongoing trials are investigating support networks for transplant recipients (NCT00782145), prevention of fungal infection (NCT00079222) and genetic susceptibility (NCT00949052) to secondary malignancy among stem-cell recipients.

One ongoing open-label nonrandomized study, at the University of Michigan Cancer Center, is investigating a tumor lysate-pulsed dendritic cell vaccine for immune augmentation after stem-cell transplantation for pediatric patients with high-risk solid tumors (NCT00405327). This study is ongoing and no longer recruiting patients, and final data collection for the primary outcome is scheduled for June 2012.

Conclusion

Moderate strength evidence on overall survival suggests no benefit with single HSCT compared to conventional therapy for the treatment of high-risk metastatic rhabdomyosarcoma.

The body of evidence on overall survival with single HSCT compared to conventional therapy for the treatment of high-risk rhabdomyosarcoma of mixed tumor type is insufficient to draw conclusions

The body of evidence on overall survival with single HSCT compared to conventional therapy for the treatment of congenital alveolar rhabdomyosarcoma, cranial parameningeal rhabdomyosarcoma with metastasis, or the use of allogeneic transplantation for metastatic rhabdomyosarcoma was insufficient to draw conclusions.

Retinoblastoma Systematic Review

Background and Setting

Retinoblastoma is the most common primary intraocular tumor in children, with an incidence of 1 in 15,000 births,437 and accounts for 4 percent of all childhood cancers. Majority of children present with intraocular disease where conventional treatments have produced at least a 90 percent chance of cure.438 Patients with trilateral retinoblastoma have an initial diagnosis of intraocular disease, with the subsequent development of a primary intra-cranial primitive neuro-ectodermal tumor and have traditionally had extremely poor prognosis and are included in this review. Extraocular or metastatic retinoblastoma in comparison to intraocular disease is generally lethal specifically when the disease has reached the central nervous system. Despite the development of new chemotherapy options, the prognosis of these patients is generally poor. Some centers have used HDC with HSCT in the setting of extraocular disease. Data from case series and case reports are available. Numbers remain small, as extraocular and trilateral retinoblastoma are rare conditions; no randomized controlled trials exist. Evidence was evaluated in three groups; studies confined to patients with CNS involvement, those with patients without CNS disease and patients with trilateral retinoblastoma.

Evidence Summary

The overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of metastatic retinoblastoma is shown in Table 55.

Table 55. Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of metastatic retinoblastoma.

Table 55

Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of metastatic retinoblastoma.

The evidence compiled for this review includes five case reports439-443 on HSCT and 15 case series (eight on HSCT438, 444-450 and five on the comparator conventional chemotherapy451-455 and two retrospective reviews with data on both HSCT and conventional chemotherapy456, 457). The total number of patients abstracted from the 20 studies was 267: 91 patients in 15 studies received HSCT, whereas 176 patients in seven studies received conventional chemotherapy.

Prognostic factors are not well defined except that patients with metastatic disease to the CNS have shorter survival than those with metastatic disease to other areas. Treatment with HSCT does not appear to alter the survival for patients with metastatic retinoblastoma to the CNS. These patients continue to have very poor prognosis. Treatment with HSCT may alter the 5-year survival for patients with metastatic retinoblastoma to sites other than the CNS, but these effects are uncertain. Treatment with HSCT may alter the 5-year survival for patients with trilateral retinoblastoma, but these effects are uncertain. Additional research with more patients is needed to confirm these findings. No information on quality of life was provided and data on adverse events was sparse and therefore insufficient to make conclusions regarding adverse effects and quality of life. One Phase III multicenter study of multimodal therapy (induction, HDC, and HSCT and/or radiotherapy) for young children with extraocular retinoblastoma is ongoing.

Results

Forty-one articles were retrieved for full-text screening. Twenty reports were included in this review, and the remaining 21 articles were excluded. The total number of patients abstracted from the twenty studies was 267: 91 patients in 15 studies received HSCT, whereas 176 patients in seven studies received conventional chemotherapy.

Table 56 shows the criteria that were used to select retinoblastoma studies.

Table 56. Retinoblastoma study selection criteria.

Table 56

Retinoblastoma study selection criteria.

Table 57 shows the study design and population. Of the included publications, five were case reports on HSCT and 15 were case series (eight on HSCT438, 444-450 and five on the comparator conventional chemotherapy451-455 and two retrospective reviews with data on both HSCT and conventional chemotherapy456, 457). Five studies were based in Europe,441, 445, 447, 454, 456 three in Asia,439, 446, 452 three in South America,451, 453, 455 and nine in North America.438, 440, 442-444, 448-450, 457

Table 57. Retinoblastoma study characteristics and population.

Table 57

Retinoblastoma study characteristics and population.

All patients across the 15 treatment studies received HSCT as consolidation of primary treatments. Other than the patients with trilateral retinoblastoma442, 444 all patients had metastatic disease prior to transplant. For the comparison of tandem HSCT to single HSCT; no studies were identified in the search.

All studies were specific to the pediatric age group, with age primarily reported as age at diagnosis; 14 studies reported either mean age or only had one patient. Mean age at diagnosis was 21.8 months with a range of 4 months to 51.8 months. Median age, reported by 13 studies, was 26.3 months with a range of 1 week to 145 months. Patients were approximately split equally by gender. Induction regimens varied across and within study (i.e., different chemotherapeutic agents and different (cumulative) dosages). The induction regimen consisted of multiple cycles of chemotherapy with or without radiation, following primary enucleation.

Conditioning regimens also varied across and within studies. The most common regimens included the following agents; cyclophosphamide, thiotepa, etoposide, carboplatin and etoposide either alone or in combination, ICE (ifosfamide, carboplatin, and etoposide) is a common backbone used alone or in combination with radiation therapy or additional drugs. Treatment periods ranged from 1982 to 2007.

Table 58 shows the outcomes that were reported across studies.

Table 58. Retinoblastoma outcomes reported.

Table 58

Retinoblastoma outcomes reported.

Overall Survival

Data on overall survival were reported in all 20 studies (Table 58). Survival data are presented stratified by if patients were identified as having metastatic spread to the CNS, then by year (Table 59). A study of trilateral retinoblastoma was also separated into its own category. Ten studies presented data for patients with CNS involvement442, 443, 447, 449, 451, 453-457 and the same ten studies plus nine more 438-441, 445, 446, 448, 450 presented data on patients without CNS involvement. One study presented data exclusively on trilateral retinoblastoma.444 The individual studies either did not define overall survival or used different starting points for this variable (i.e., either years from diagnosis or years from first transplant). No direct comparisons can be made from the published data as there are no comparative studies.

Table 59. Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups: Retinoblastoma.

Table 59

Overall survival for treatment (single auto HSCT) and comparison (conventional chemotherapy +/- radiation) groups: Retinoblastoma.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

No studies evaluated quality of life, and adverse effects were only reported by intervention studies. Data on treatment-related mortality was reported in two intervention studies (Table 60). Two patients died from septicemia and multi-organ failure during induction therapy.444, 449 Two studies reported cases of serious infection, both attributed to Candida albicans.447, 448 One comparator study451 reported three secondary malignancies (two osteogenic sarcoma, and one nonlymphocytic leukemia) and one intervention study450 reported three secondary malignancies (osteosarcoma, two occurring in irradiated fields). There were no reports of serious hemorrhagic events, irreversible veno-occlusive disease or other long-term complications among patients treated with HSCT or conventional chemotherapy.

Table 60. Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups: Retinoblastoma.

Table 60

Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups: Retinoblastoma.

Ongoing Studies

A Phase III multicenter study of multimodal therapy (induction, HDC, and HSCT and/or radiotherapy) for young children with extraocular retinoblastoma was identified (NCT00554788). This trial estimates it will enroll 60 children ages 10 years of age and younger and will be complete in February 2014. Event-free survival is the primary outcome measure.

Twenty children ages 21 or younger were to be enrolled in a Phase I study examining the toxicity of killer IG-like receptor mismatched umbilical cord blood for pediatric patients with malignant solid tumors. This study is ongoing and no longer recruiting, and no results have been published.

Conclusion

Low strength evidence on overall survival suggests no benefit with single HSCT compared to conventional therapy for the treatment of extraocular retinoblastoma with CNS involvement.

The body of evidence on overall survival with single HSCT compared to conventional therapy for the treatment of extraocular retinoblastoma without CNS involvement was insufficient to draw conclusions.

The body of evidence on overall survival with single HSCT compared to conventional therapy for the treatment of trilateral retinoblastoma without CNS involvement was insufficient to draw conclusions.

Neuroblastoma Systematic Review

Background and Setting

Neuroblastoma is the most common extracranial solid tumor of childhood, and accounts for 8 to 10 percent of all childhood cancers and for approximately 15 percent of cancer deaths in children.103 At least 40 percent of all children with neuroblastoma are designated as high-risk patients.103, 104 Despite the development of new treatment options, the prognosis of patients with high-risk neuroblastoma is generally poor; more than half of patients experience disease recurrence and long-term survival with current treatments is about 30 percent.104

Many centers have used HDC with HSCT in the setting of high-risk or recurrent disease.103, 106 Results from randomized controlled trials (RCTs) comparing HDC/HSCT with conventional therapy have shown higher survival rates with HSCT, although higher levels of adverse effects have been reported and overall rates are unsatisfactory.105, 107, 108 Sequential tandem HSCT has been developed to improve further the outcome of patients with high-risk neuroblastoma.

Evidence Summary

The overall grade of strength of evidence for overall survival in pediatric patients with high-risk neuroblastoma is shown in Table 61.

Table 61. Overall grade of strength of evidence for overall survival: Neuroblastoma.

Table 61

Overall grade of strength of evidence for overall survival: Neuroblastoma.

The evidence compiled for this review includes six observational studies on HSCT, and three RCTs reporting outcomes data on single HSCT. The total number of patients included in the nine studies was 4,044: 682 patients received tandem HSCT, whereas 3,362 patients received single HSCT.

Tandem HSCT results in no significant differences in survival rates than single HSCT. In addition, no significant differences in secondary malignant disease and treatment-related mortality between treatment groups were identified. No information on QOL was provided and data on adverse effects are very limited; no definitive conclusions can be made regarding adverse effects and quality of life.

The ongoing randomized trial by the Children's Oncology Group will address whether tandem HSCT is superior to single HSCT in patients with high-risk neuroblastoma.

Results

Eighteen reports describing nine unique studies were included in this review. Data from the European Group for Blood and Marrow Transplantation (EBMT) registry on outcomes for single and tandem HSCT have been reported in two publications.113, 458 George et al. have reported outcomes of tandem HSCT across four U.S. centers in seven publications.459-465 Two further studies have been reported in multiple publications; two reports by Sung et al. on tandem HSCT466, 467 and two reports of the RCT by Matthay et al. on single HSCT.107, 111 The report with the largest sample size and longest followup period from each of the above series was included in the primary analysis for this review. The total number of patients included in the nine studies was 4,044: 682 patients received tandem HSCT, whereas 3,362 patients received single HSCT.

Table 62 shows the criteria that were used to select studies for this section.

Table 62. Study selection criteria: Neuroblastoma.

Table 62

Study selection criteria: Neuroblastoma.

Table 63 shows the study design and population. Of the included publications, six were observational studies (three provided comparisons of tandem vs. single HSCT466, 468, 469; three of tandem HSCT459, 462), and three were RCTs reporting outcomes data on single HSCT.105, 107, 108, 111 Five were multicenter studies (two reporting on outcomes for tandem HSCT and three trials on single HSCT). Three studies were based in Europe,105, 108, 113 three in Asia,466, 468, 469 and three in North America.107, 459, 462The EBMT data represents the largest cohort of patients recruited over 28 years (1978–2006).113

Table 63. Study characteristics and population: Neuroblastoma.

Table 63

Study characteristics and population: Neuroblastoma.

All patients across eight (of nine) studies received HSCT as consolidation of primary treatments, Eighty percent of patients in the EBMT cohort received HSCT as consolidation therapy; relapse was the indication in another 10 percent while the status prior to HSCT was not specified in a further 10 percent of patients.113 The vast majority of patients across studies presented with stage IV disease at diagnosis (range: 81 to 100 percent For the EBMT data, the stage was reported only in 53 percent of the cohort but there was a high prevalence for advanced disease with stage IV in more than 90 percent of the reported cases.113

Eight studies were specific to the pediatric age group; the EBMT cohort consisted of 2 percent (of 3,421) patients over 18 years of age. Eight studies reported the age of the participants at diagnosis; Sung et al. (2007) reported age at both diagnosis and HSCT.466 The median age was reported in six studies on tandem HSCT; the remaining three trials on single HSCT reported only the number of cases above and below one year of age. The majority of patients (86 to 97 percent across all studies were over 12 months of age at diagnosis.

All studies used different induction regimens (i.e., different chemotherapeutic agents and different (cumulative) dosages). The induction regimen across studies consisted of multiple cycles (1-10) of chemotherapy followed by surgery for resection of the primary tumor. The timing of surgery varied during induction and took place at diagnosis or after 2 to 7 cycles of chemotherapy. Tumor-field radiotherapy was used in patients with residual tumor and/or metastatic disease in at least six (of nine) studies: Sung et al. employed radiotherapy in the early study period (diagnosis by December 2003).466 There was no postoperative radiotherapy in Pritchard et al.; in this latter study, 41 percent of patients randomized to the single HSCT arm received nine or more cycles of induction chemotherapy.108

Various conditioning regimens were used across studies. The primary conditioning regimen consisted of carboplatin, etoposide and melphalan. Total body radiation was used as part of the treatment regimen in six studies.107, 113, 459, 462, 466, 469 In at least four studies, there were also differences in treatment that patients received within the study itself (for example, in external radiotherapy, immunotherapy, and retinoic acid).

Peripheral blood stem cells were used as the sole source of support in six studies,105, 459, 462, 466, 468, 469 and bone marrow in two studies;107, 108 the EBMT cohort used peripheral stem cells (56 percent), bone marrow (41 percent) and a combination of both (3 percent) as a source of support after HDC.113 The median follow-up durations from first transplant across three studies comparing tandem and single HSCT were 2.3 years, 9 years, and 5 years, respectively.113, 468, 469

Table 64 shows the outcomes that were reported across nine studies. Of note, the study by Hobbie et al.462 was a subgroup analysis of George et al.459 reporting on the long-term adverse events of tandem HSCT for high-risk disease. For purposes of data analysis and synthesis, these two reports were considered as unique studies; George et al.459 reported on overall survival (OS), event-free survival (EFS), treatment-related mortality and secondary malignancies, while Hobbie et al.462 reported on other adverse effects of HSCT.

Table 64. Outcomes reported: Neuroblastoma.

Table 64

Outcomes reported: Neuroblastoma.

Overall Survival

Data on OS were reported in seven (of nine) primary studies (Table 64). Six studies presented 3- and/or 5-year rates and the study by George et al.459 also presented 7-year rates (Table 65). No significant differences in either the 3-year or 5-year OS between treatment groups were identified in the two comparative studies (Table 65).113, 468 Multivariate analysis of EBMT data showed significantly better OS rates in patients younger than 2 years of age at diagnosis (Hazard Ratio [HR], 1.6; 95 percent; Confidence Interval [CI], 1.4-1.9; p<0.0001).113 It should be noted that the individual studies either did not define OS or used different starting points for this variable (i.e., either years from diagnosis or years from first transplant).

Table 65. Overall survival for treatment (tandem HSCT) and comparison (single HSCT) groups: Neuroblastoma.

Table 65

Overall survival for treatment (tandem HSCT) and comparison (single HSCT) groups: Neuroblastoma.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

None of the studies evaluated quality of life (Table 64). Data on treatment-related mortality were reported in six studies (Table 66). There were 20 (of 197) cases in the tandem group and 36 (of 373) cases in the single HSCT group. Secondary malignancies were reported in five studies (Table 66). There were three (of 212) cases in the tandem group (one synovial cell sarcoma, one myelodysplasia with clonal trisomy 8, and one thyroid cancer); two cases were reported in the George et al.459 study. The case of thyroid cancer was reported in the 2010 study by Sung et al.469, and occurred in a patient receiving only the first HSCT. Three (of 408) cases of secondary malignancies were reported in the single HSCT group (two acute myeloblastic leukemias and one follicular carcinoma of the thyroid).

Table 66. Adverse effects for treatment (tandem HSCT) and comparison (single HSCT) groups: Neuroblastoma.

Table 66

Adverse effects for treatment (tandem HSCT) and comparison (single HSCT) groups: Neuroblastoma.

Infectious complications were reported in four studies (Table 66). Sepsis was more prevalent in the single HSCT group (n=219) compared to the tandem group (n=126) (26 vs. 2 percent). All infectious complications were attributed to sepsis in the single HSCT group. Further serious infections in the tandem group included two cases of viral pneumonia and three cases of Epstein-Barr virus and cytomegalovirus, all resulting in toxicity-related deaths. Other reported serious adverse effects included one case of pulmonary hemorrhage in the tandem group and three cases of bleeding in the single HSCT group.

The frequency of veno-occlusive disease was reported across four studies (Table 66).108, 111, 459, 466 There were nine (of 126) cases in the tandem group and two (of 30) cases in the single HSCT group. Only one study (n=13) by Hobbie et al.462 reported further long-term complications including developmental delays (i.e., hearing loss, 92 percent), cataracts (54 percent), and growth hormone deficiency (54 percent) following tandem HSCT.

Ongoing Research

In North America, the Children's Oncology Group is studying, in a randomized fashion, whether tandem HDC/HSCT is superior to a single HDC/HSCT in patients with high-risk neuroblastoma up to 30 years of age. This is an international trial (U.S., Canada, Australia, New Zealand) being undertaken across 142 centers and is currently recruiting patients with an expected enrollment of 495 patients. The primary outcomes of interest include 3-year EFR, response after induction therapy, and incidence rate of local recurrence. The projected completion of accrual is spring 2012 (NCT00567567).a

Conclusion

The body of evidence on overall survival with tandem HSCT compared to single HSCT for the treatment of high-risk neuroblastoma was insufficient to draw conclusions.

Germ-Cell Tumors Systematic Review

Background and Setting

Germ cell tumors (GCT) are rare in children younger than 15 years, accounting for approximately 3 percent of cancer cases in this age group.115 Childhood GCT can be divided into gonadal (ovarian and testicular) and extragonadal (e.g., mediastinal or retroperitoneal) neoplasms.118 Gonadal GCT (particularly testicular GCT) are much more common among adolescents aged 15 to 19 years, representing approximately 14 percent of cancer diagnoses in this age group.115 GCTs are highly sensitive to chemotherapy. Cisplatin-based combination chemotherapy, followed by appropriate surgical resection of residual disease, is curative in 80 percent of patients; however, about 20-30 percent of patients may develop recurrent disease.114, 118, 119 HDC with HSCT has been explored primarily in adults with relapsed testicular GCT through observational studies.115, 118, 119, 470

Reports from salvage treatment strategies used in adult recurrent GCT include larger numbers of patients, but the differences between children and adults regarding the location of the primary GCT site, pattern of relapse, and the biology of childhood disease may limit the applicability of adult salvage approaches to children. Sequential tandem HSCT has been developed to improve further the outcome for children with relapsed GCT.

Evidence Summary

The overall grade of strength of evidence for overall survival in pediatric patients with tandem HSCT compared to single HSCT for the treatment of relapsed germ cell tumors is shown in Table 67. The evidence compiled for this review includes four observational studies.114, 119, 120, 470 The total number of pediatric patients included in the four studies was 71: 29 patients received tandem HSCT, whereas 42 patients received single HSCT. Tandem HSCT results in no significant differences in survival rates than single HSCT. No information on QOL was provided, and data on adverse effects are very limited; no definitive conclusions can be made regarding adverse effects and QOL. Results to date are based on small observational studies that have focused on adult patients with gonadal disease. Tandem HSCT may be particularly beneficial in patients with more advanced testicular cancer at diagnosis and greater likelihood of exhibiting cisplatin resistance when compared to single HSCT. However, the reports have great variability in patient selection, prior treatments, the choice of the conditioning regimen and variability of doses within the same regimen. Furthermore, many reports have either combined the data from single and tandem transplants or the numbers are very small.

Table 67. Overall grade of strength of evidence for overall survival: Germ cell tumor.

Table 67

Overall grade of strength of evidence for overall survival: Germ cell tumor.

Randomized (prospective) trials focused on young children and adolescents will be needed to determine if tandem HSCT transplants is superior to single HSCT utilizing an optimal conditioning regimen.

Results

Seventeen articles were retrieved for full-text screening. Four reports were included in this review, and the remaining 13 articles were excluded. The total number of pediatric patients included in the four studies was 71 (of 539): 29 patients received tandem HSCT, whereas 42 patients received single HSCT. Table 68 shows the study selection criteria.

Table 68. Germ cell tumor study selection criteria.

Table 68

Germ cell tumor study selection criteria.

Table 69 shows the study design and population. All four publications were observational studies. Tandem transplants were performed in two (50 percent) studies. Only one study reported outcomes data of tandem versus single HSCT.119 Two were multicenter studies (Center for International Blood and Marrow Transplant Research [CIBMTR] cohort by Lazarus et al.119 and a European Group for Blood and Marrow Transplantation [EBMT] cohort by De Giorgi et al.114 and two were U.S. single-center studies.120, 470

Table 69. Germ cell tumor study characteristics and population.

Table 69

Germ cell tumor study characteristics and population.

Only one small study by De Giorgi et al.114 was specific to the pediatric age group; approximately 10 percent of all patients across the remaining three studies were in the pediatric age range (Einhorn, 2007: n=17 [of 184];470 Lazarus, 2007: n=32 [of 300];119 Agarwal, 2009: n=4 [of 37]120 ). The corresponding authors for the three studies were approached for outcomes data (and if available, patient characteristics) specific to the pediatric age groups.b Data on study variables and outcome events for the pediatric age range (11-20 years) for Lazarus et al.119 were obtained from the CIBMTR.471c

All study patients received HSCT as salvage treatment for relapsed disease. The majority of patients (65-100 percent) across three studies had advanced testicular cancer; the EBMT cohort consisted of pediatric patients with extragonadal GCT.114 Most patients received a cisplatin-based chemotherapy regimen initially and surgery for residual disease when appropriate. Various conditioning regimens were used across studies. The primary conditioning regimen consisted of carboplatin and etoposide. Peripheral blood stem cells were used as either the sole or primary source of support in all studies.

For the CIBMTR cohort, the tandem and single HSCT groups were comparable for median age, testicular versus abdominal origin, number of chemotherapy regimens prior to HSCT, and year of HSCT (over 50 percent of transplants were performed between 1996 and 1998).471 The interval from diagnosis to first HSCT for the CIBMTR cohort was 12 (range: 2-34) months for the tandem group and 9 (range: 3-17) months for the single HSCT group. Eighty-three percent and 65 percent of patients had residual cancer at time of HSCT, respectively.471 There were observed differences in the intensity of the transplant preparative regimen between the two study groups; 58 percent of the tandem group received a regimen containing 3 or more chemotherapeutic agents in contrast to 95 percent in the single HSCT group.471 In addition, in comparison to the single HSCT group, the tandem group had a greater likelihood of cisplatin-resistance at time of transplantation (58 percent vs. 10 percent), and was more likely to receive blood (83 percent vs. 60 percent) rather than marrow as the stem cell source.471 Median followup in the CIBMTR cohort was 56 (range: 45-74) months for the tandem group and 59 (range: 13-124) months for the single HSCT group, respectively.471

The Einhorn et al.470 tandem series exhibited more favorable prognostic features compared to the CIBMTR tandem cohort. No patients in this series received more than two chemotherapeutic agents as part of their transplant preparative regimen.470 Seventy-eight percent of patients exhibited platinum sensitivity and all patients received peripheral-blood stem cells.470 Median followup in the Einhorn series was comparable to the CIBMTR cohort (48 [range: 14-118] months).470

Table 70 shows the pediatric outcomes that were reported across the four studies.

Table 70. Germ cell tumor outcomes reported.

Table 70

Germ cell tumor outcomes reported.

Overall Survival

Data on OS were reported in all four studies (Table 70). Data were available to compute three-year rates across all studies, and five-year rates for three studies (Table 71). Similar trends were observed between treatment groups in the one-, three-, and five-year OS across studies (Table 71). For the CIBMTR cohort, five-year survival probability was 36 percent (95 percent confidence interval (CI), 10-69 percent) in the tandem group compared to 49 percent (24-68 percent) in the single HSCT group.471 OS was defined across three studies as the interval between salvage chemotherapy or transplant and death from any cause.

Table 71. Overall survival for tandem HSCT and comparison (single HSCT) groups: Germ cell tumor.

Table 71

Overall survival for tandem HSCT and comparison (single HSCT) groups: Germ cell tumor.

Event-free Survival

Information on event-free survival can be found in Appendix D.

Adverse Effects

None of the studies evaluated quality of life (Table 70). Data on treatment-related mortality was available from three studies (Table 72).114, 120, 471 There was no reported cases of treatment-related mortality in the two single HSCT series (N=22). For the CIBMTR cohort, cumulative incidence of treatment-related mortality was 10 percent (2-27 percent) at 5 years for the single HSCT group (n=20); none of the 12 patients in the tandem group had treatment-related mortality (Table 72). Relapse/progression incidence, on the other hand, was 64 percent (30–85 percent) for the tandem group up to five years after transplant compared to 41 percent (20–62 percent) for the single HSCT group.471 Other adverse events were reported in only two single HSCT studies. There were no secondary malignancies (Table 72). Veno-occlusive disease occurred in two (of 18) patients in the EBMT cohort by De Giorgi et al.114

Table 72. Adverse effects for tandem HSCT and comparison (single HSCT) groups: Germ cell tumor.

Table 72

Adverse effects for tandem HSCT and comparison (single HSCT) groups: Germ cell tumor.

Ongoing Research

Two U.S. nonrandomized studies are underway on tandem transplants. The first is a two-center (M.D. Anderson Cancer Center; Fred Hutchinson Cancer Research Center) Phase II study being undertaken to evaluate if bevacizumab, when given in combination with two cycles of HDC, can help to control GCTs in patients aged 12 to 65 years. The study is currently recruiting patients with an estimated enrollment of 25 participants. The primary outcome of interest is 2-year EFS. The estimated final data collection date for this trial is June 2014. (NCT00936936).

The second study (Phase I/II) is being undertaken at the Children's Memorial Hospital in Chicago to assess the feasibility and toxicity of tandem rescue with peripheral blood cells following HDC as consolidation in pediatric patients with high risk solid tumors, including relapsed GCT. The study is currently recruiting patients with an estimated enrollment of 12 participants. The estimated final data collection date is September 2012 (NCT00179816).

Conclusion

The body of evidence on overall survival with tandem HSCT compared to single HSCT for the treatment of relapsed pediatric germ cell tumors was insufficient to draw conclusions.

Central Nervous System/Embryonal Tumors Systematic Review

Background and Setting

Classification of brain tumors is based on both histopathologic characteristics of the tumor and location in the brain.122 Central nervous system (CNS) embryonal tumors are the most common malignant brain tumor in childhood. Embryonal tumors of the CNS primarily include medulloblastoma (MB), supratentorial primitive neuroectodermal tumor (PNET), and atypical teratoid/rhabdoid tumor (AT/RT).122 MBs account for 20 percent of all childhood CNS tumors.123, 124 The other types of embryonal tumors are rare by comparison.122

PNETs are a heterogeneous group of highly malignant neoplasms comprising 3 to 5 percent of all childhood brain tumors, most commonly located in the cerebral cortex and pineal region.123, 125 AT/RT, on the other hand, comprise approximately 2-3 percent of these tumors with a peak incidence in children less than three years of age, and is associated with characteristic genetic abnormalities.123, 125, 126 The prognosis for these tumors is worse than for MB, despite identical therapies.122, 123, 125

Recurrence of all forms of CNS embryonal tumors is not uncommon, usually occurring within 18 months of treatment; however, recurrent tumors may develop many years after initial treatment.122 The treatment of these tumors continues to evolve especially in children less than three years of age because of the concern of the deleterious effects of craniospinal radiation on the immature nervous system. Therapeutic approaches have attempted to delay and sometimes avoid the use of radiation, and have included trials investigating different chemotherapy regimens to improve outcome.122 Many centers have used HDC with HSCT to improve further the outcome for children with CNS embryonal tumors.

Evidence Summary

The overall grade of strength of evidence for overall survival with tandem HSCT compared to single HSCT for the treatment of CNS embryonal tumors is shown in Table 73.

Table 73. Overall grade of strength of evidence for overall survival: CNS embryonal tumors.

Table 73

Overall grade of strength of evidence for overall survival: CNS embryonal tumors.

The evidence compiled for this review includes ten observational studies133, 472-480 and two randomized clinical trials (RCT).481, 482 Nine studies reported outcomes for HSCT,133, 473-477, 481 479, 480 and three studies (including two RCTs) were multi-institutional treatment protocols on CNS embryonal tumors.478, 481, 482 For HSCT studies, 15 patients received tandem transplant, whereas 132 patients received single HSCT.

Based on the currently available evidence, it is not possible to clarify the role of HSCT (single or tandem procedure), as studies are limited individually by low numbers of patients enrolled and collectively by inconsistencies in the patients' ages. The prognosis and treatment varies depending upon the age of the patient and type of embryonal tumor. Most studies to date have focused on children with newly diagnosed medulloblastoma. Comparison between studies, moreover, remains challenging, given the heterogeneity of these tumors and the varied therapies used across centers.

Results

Twelve reports were included in this review. Table 74 shows the criteria that were used to select studies for this section. For HSCT studies, 15 patients received tandem transplant (MB, n=13; PNET, n=1; AT/RT, n=1), whereas 132 patients received single HSCT (MB, n=61; PNET, n=52; AT/RT, n=19).

Table 74. Study selection criteria: CNS embryonal tumors.

Table 74

Study selection criteria: CNS embryonal tumors.

Table 75 shows the study design and population. Ten publications were observational studies133, 472-480 and two were randomized clinical trials (RCTs).481, 482 Nine studies reported outcomes for HSCT, and three studies (including two RCTs) were multi-institutional treatment protocols on CNS embryonal tumors.478, 481, 482 Of the nine HSCT studies, tandem transplants were performed in three studies, one of which reported comparative data of tandem vs. single HSCT.133 Sixty percent of these patients were considered as average-risk (i.e., Chang stage M0 having no metastasis), and 40 percent as high-risk (i.e., Chang stage M1-M4 having metastasis).

Table 75. Study characteristics and population: CNS embryonal tumors.

Table 75

Study characteristics and population: CNS embryonal tumors.

All patients across the nine transplant studies received HSCT as consolidation of primary treatments. All studies used different induction regimens (i.e., different chemotherapeutic agents and different (cumulative) dosages). The induction regimen across studies primarily consisted of five cycles of chemotherapy followed by consolidation phase. Various conditioning regimens were used across studies. The conditioning regimen primarily consisted of carboplatin, thiotepa, etoposide, busulfan and/or melphalan. Approximately 30 percent of patients (29-37 percent) also received radiation therapy across these studies. Peripheral blood stem cells were used as the sole source of support in five studies (two on single HSCT and three on tandem HSCT); combination of peripheral blood and bone marrow was used across the remaining four studies.

Data on conventional care were based on results from three multi-institutional treatment protocols on CNS embryonal tumors (one on multiple tumor types which consisted of MB, PNET and AT/RT; and two on MB).478, 481, 482 The study by Geyer et al.481 was a RCT from the U.S. Children's Cancer Group (COG) of two multi-agent chemotherapy regimens (with deferred radiotherapy) for children younger than 3 years of age with various malignant brain tumors in a large cohort of patients. Maintenance therapy for all patients in the COG protocol comprised of eight cycles of vincristine, carboplatin and cyclophosphamide; over 40 percent of patients received radiotherapy.481 Two studies reported on outcomes for MB patients; the RCT by Packer et al.482 reported on outcomes with radiotherapy and adjuvant chemotherapy for children three years and older with nonmetastatic disease, and the European multicenter study by Taylor et al.478 reported on outcomes for ages three years and older with metastasis.

Table 76 shows the outcomes that were reported across the 12 studies.

Table 76. Outcomes reported: CNS embryonal tumors.

Table 76

Outcomes reported: CNS embryonal tumors.

Overall Survival

Data on OS were reported in ten (of 12) studies (Table 76). For comparisons between tandem vs. single HSCT, data were available to compute 2-year rates for two studies, 3-year rates for two studies, and 5-year rates for four studies (Table 77). For Sung et al.466 (n=14), 2-year survival probability was 82 percent (95 percent confidence interval (CI), 59-100 percent) in the tandem group (MB, n=10; PNET, n=1) compared to 67 percent (13-100 percent) in the single HSCT group (MB, n=1; PNET, n=2). The AT/RT patient reported in Gidwani et al.472 has remained disease free for two years following tandem HSCT. OS was defined across studies as the interval between diagnosis to death or last followup.

Table 77. Overall survival for tandem HSCT and comparison (single HSCT) groups: CNS embryonal tumors.

Table 77

Overall survival for tandem HSCT and comparison (single HSCT) groups: CNS embryonal tumors.

For the conventional-care group of studies, data were available to compute 3-year rates for one study,478 and 5-year rates for three studies (Table 78).478, 481, 482 There were no comparative studies between single HSCT vs. conventional care. For Geyer et al.481 on multiple tumor types, five-year survival probability overall was 43 percent (3 percent) for children under three years of age; for MB, PNET and AT/RT, the corresponding rates were 43 percent (5 percent), 31 percent (7 percent), and 29 percent (9 percent), respectively. Similar rates were observed for MB patients with metastatic disease in the multicenter study by Taylor et al.478 Packer et al.482 reported higher survival rates in their cohort of MB patients without metastasis.

Table 78. Overall survival for single HSCT and comparison (conventional care) groups: CNS embryonal tumors.

Table 78

Overall survival for single HSCT and comparison (conventional care) groups: CNS embryonal tumors.

Adverse Effects

Only one HSCT study on MB patients evaluated quality of life (Table 76).474 Dhall et al.474 reported that mean intellectual functioning and QOL for children less than three years of age surviving without radiotherapy (n=4 [of 21]) was within the average range at both followup periods of testing (using the Parent Form of the Child Health Questionnaire [which is a 50-item QOL measure]). Data on treatment-related mortality were reported in 11 studies (Table 79).133, 474-479, 481-483 There was one (of 15) case (7 percent) in the tandem group, nine (of 132) cases (8 percent) in the single HSCT group, and 71 (of 663) cases (11 percent) in the conventional care group.

Table 79. Adverse effects for treatment (tandem HSCT) and comparison (single HSCT) groups: CNS embryonal tumors.

Table 79

Adverse effects for treatment (tandem HSCT) and comparison (single HSCT) groups: CNS embryonal tumors.

Secondary malignancies were reported in three single HSCT studies476, 477, 483and three studies on conventional care (Table 79 and Table 80).478, 481, 482 Three (of 69) cases (4 percent) of secondary malignancies were reported in the single HSCT group, and 12 (663) cases (2 percent) in the conventional care group. Other adverse events across studies are reported in Table 79 and Table 80, respectively.

Table 80. Adverse effects for treatment (single HSCT) and comparison (conventional care) groups: CNS embryonal tumors.

Table 80

Adverse effects for treatment (single HSCT) and comparison (conventional care) groups: CNS embryonal tumors.

Ongoing Research

In North America, the Children's Hospital of Los Angeles is leading a Phase III trial (“Head Start III”) studying combination chemotherapy with or without etoposide followed by single HSCT in treating patients (10 years or younger) with newly diagnosed brain tumors including MB, PNET, and AT/RT. This is an international trial (U.S., Canada, Australia, New Zealand, Switzerland) being undertaken across 37 centers and is currently recruiting patients with an expected enrollment of 120 patients. The primary outcomes of interest include time to tumor progression, disease recurrence or death of any cause, EFS at 2 years and toxicity. The projected completion of accrual is December 2010 (NCT00392886).

The St. Jude Children's Research Hospital is leading a Phase III trial studying two different regimens of radiation therapy when given together with chemotherapy and HSCT (1 to 3 procedures) to see how this regimen works in treating patients (3 years to 21 years) with newly diagnosed MB, PNET, or AT/RT. This is an international trial (U.S., Canada, Australia) being undertaken across nine centers and is currently recruiting patients with an expected enrollment of 342 patients. The primary outcomes of interest include the relationship of protein expression in tumors and PFS up to seven years of followup. The projected completion of accrual is April 2011 (NCT00085202).

In addition to the above studies, there are two trials underway by the Children's Oncology Group (COG). The first trial is open for children aged 3 years or younger at diagnosis with newly diagnosed PNET or high-risk medulloblastoma (NCT00336024). The second trial is a Phase III study for patients under 21 years of age with AT/RT. Both studies are using multi-agent chemotherapy, radiation, and high-dose chemotherapy with hematopoietic stem-cell rescue (NCT00653068).”

Conclusion

The body of evidence on overall survival with tandem HSCT compared to single HSCT for the treatment of CNS embryonal tumors was insufficient to draw conclusions.

The body of evidence on overall survival with single HSCT compared to conventional therapy for the treatment of CNS embryonal tumors was insufficient to draw conclusions.

Glial Tumor Systematic Review

Background and Setting

Glial tumors comprise a heterogeneous group of neoplasms that are the largest single group of primary brain tumors in children and adolescents and contribute significant morbidity and mortality.484 The World Health Organization (WHO) classifies glial tumors into four major categories: astrocytic, ependymal, oligodendroglial or mixed gliomas, and choroid plexus tumors. According to SEER data, pediatric age-adjusted incidence rate of primary CNS glial tumors per 100,000 persons was:

  • Astrocytoma (excluding pilocytic), 0.411
  • Glioblastoma, 0.138
  • Ependymoma/anaplastic ependymoma, 0.226
  • Choroid plexus tumor, 0.025
  • Oligodendroglioma, 0.083
  • The age-adjusted mortality rate of brain and other nervous system tumors was 0.65 per 100,000 persons.

Data on glial tumors are primarily from case series, save one comparative study with an historic cohort485 with patients who received high-dose chemotherapy and HSCT. Case reports were also available. Differences in patient selection, accrual of small numbers of patients with patient data not stratified by tumor type, and differences in conditioning regimens make differences in overall survival between HSCT and conventional chemotherapy difficult to interpret. Although randomized evidence for gross total resection is lacking, retrospective analysis reaffirms the value of surgical resection in prolonging survival.

A greater than 90 percent surgical resection of newly diagnosed malignant gliomas, both anaplastic astrocytoma and glioblastoma multiforme, in childhood and adolescence confers a statistically significant survival advantage when followed by local field irradiation and conventional chemotherapy, or autologous stem-cell rescue.485 Evidence was evaluated in five groups: anaplastic astrocytoma and glioblastoma multiforme (astrocytic tumors), choroid plexus tumor, ependymoma, and other glial tumor patients. Data for other glial tumors was presented but separate analysis by type was not possible. Patients were classified into newly diagnosed or recurrent/progressive disease due to a poorer overall survival for recurrent/progressive patients.

High-Grade Glioma: Anaplastic Astrocytoma (AA)/Glioblastoma Multiforme (GBM)

The prognosis for patients diagnosed with high-grade glioma is poor. The median survival is less than 1 year, the majority die within two years despite some exceptional survivors.486 Patients with grade II astrocytoma may survive for 5 or more years while patients with AA often die within 2 or 3 years and frequently show progression to GBM with survival times substantially less than 2 years.487

Choroid Plexus Carcinomas

Choroid plexus carcinomas are rare typically occurring among children under 12 years of age with the greatest prevalence among children less than 2 years of age. 488 Choroid plexus tumors account for 1-4 percent of all childhood brain tumors with 25 percent of these patients developing progressive disease. 488 The role of surgery is well established in these tumors. Total resection of the tumor is often limited by tumor vascularity, large tumor size, and the tumor's tendency to invade the brain.488 The added benefits of radiation and chemotherapy on overall survival after total resection are unclear.488

Ependymomas

Ependymomas are significantly more prevalent in infants and young children, than in adults, and account for 6-10 percent of brain tumors in children.489 Sixty percent of ependymal tumors in children are infratentorial with 40 percent supratentorial. With conventional therapy the estimated 5-year OS and PFS are 50-64 percent and 23-45 percent, respectively.490 Factors significant in the prognosis of patients are extent of tumor resection and age.490 Patients with gross total resection have higher survival rates compared to incompletely resected gliomas (67-80 percent and 22-47 percent 5-year OS, respectively), and younger children tend to have a worse prognosis (more aggressive biological behavior of the tumor, avoidance of irradiation, and unacceptable neurotoxicity).490

Evidence Summary

The overall grade of strength of evidence for overall survival with HSCT for the treatment of high-risk glial tumors is shown in Table 81. The evidence compiled for this review includes one comparative cohort study of HSCT versus conventional therapy, one noncomparative cohort study, four randomized clinical trials, three Phase II trials, and 30 case series. The total number of patients abstracted was 1012: 215 patients received HSCT and 797 received conventional therapy.

Table 81. Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk glial tumors.

Table 81

Overall grade of strength of evidence for overall survival and the use of HSCT for the treatment of high-risk glial tumors.

Results

Thirty-eight publications comprising 40 studies were included in this review. The total number of patients abstracted from the 39 studies was 1,012: 215 patients received HSCT, whereas 797 patients received conventional chemotherapy. Study selection criteria are shown in Table 82.

Table 82. Study selection criteria: Glial tumors.

Table 82

Study selection criteria: Glial tumors.

Table 83 shows the study design and population. Of the included publications for HSCT, sixteen were case series,367, 488, 492-504 two were cohort studies,485, 490 and one was a Phase II trial. 505 One study, by Finlay et al.485, compared myeloablative chemotherapy with autologous stem-cell transplant to conventional therapy and used a group of historic controls from a trial by the children's cancer group (CCG-945). All studies were published after 1995 with treatment periods ranging from 1986-2005. Four studies were conducted in France488, 493, 494, 497, two in Italy495, 502, one in the U.K.504, and twelve in the U.S.132, 367, 485, 490, 496, 498-501, 503, 506A total of 215 patients were treated with autologous bone marrow transplantation; Two-hundred and two of these were given single autologous stem-cell transplants while 11 patients received tandem or sequential transplant.367, 504 Due to the small number of patients in each tumor histology (AA 4, GBM 6, BSG 2, EPD 2, CPC 1), and similar survival outcomes to single autologous transplant, these patients were not analyzed separately. Stem cell source varied by study. Six studies treated patients with peripheral blood stem cells,367, 488, 490, 500, 502, 503 ten studies treated patients with bone marrow transplant, 132, 485, 493-496, 499, 501, 504, 506and two studies used multiple sources.497, 498 Eight studies investigated patients with newly diagnosed high-risk disease488, 490, 494, 495, 498, 500, 503, 504 and the remaining studies contained patients who had recurrent or relapsed disease.492

Table 83. Study characteristics and population: Glial tumors.

Table 83

Study characteristics and population: Glial tumors.

Conventional therapy included 14 case series,132, 488, 507-519 one cohort study,485 two Phase II trials,505, 520 and four clinical trials.489, 491, 521, 522 Case reports were excluded as comparators. All studies were published after 1995 with treatment periods ranging from 1986-2005. Four studies were conducted in France488, 508, 510, 511, two in Germany520, 522, Italy509, Taiwan516, Turkey507, and the U.K.512, and ten in the U.S.485, 489, 491, 505, 513-515, 517, 518, 521 A total of 797 patients were treated with conventional chemotherapy; the vast majority of these patients were given chemotherapy alone (N=458; 57 percent) in 16 studies. Patients received combination radiotherapy in seven studies (N=213; 27 percent), and in four studies patients received radiotherapy alone (N=126; 16 percent). Twelve studies investigated newly diagnosed high-risk patients (60 percent) and the remaining studies contained patients who had recurrent or relapsed disease. Previous treatments included a mix of excision, chemotherapy, and radiotherapy.

All studies were specific to the pediatric age group, with age primarily reported as age at diagnosis or transplant; only one study520 lacked information about participant's median age. Three studies495, 499, 517 did not provide information on patient age. Median ages ranged from under one year of age to 18 years of age. Five studies reported no patient gender132, 492, 504, 510, 520, but among the remaining studies with five or more patients gender was distributed equally. Studies included patients of diverse histology:

Nineteen HSCT studies (N=215):

  • Astrocytoma 31 (14.4 percent)
  • Choroid plexus tumor 4 (1.9 percent)
  • Ependymoma 70 (32.6 percent)
  • High-grade glioma 2 (0.93 percent)
  • Glioblastoma multiforme 56 (26.0 percent)
  • Oligodendroglioma 4 (1.9 percent)
  • Glioma [unspecified] 1 (0.47 percent)
  • Pontine glioma [unspecified] 40 (18.6 percent)
  • Ganglioma 1 (0.47 percent)

Twenty-one conventional treatment studies (N=797):

  • Astroblastoma 1 (0.1 percent)
  • Astrocytoma 109 (13.7 percent)
  • Brainstem glioma 54 (6.8 percent)
  • Choroid plexus tumor 69 (8.6 percent)
  • Ependymoma 435 (54.6 percent)
  • Ganglioma 1 (0.1 percent)
  • Glioblastoma multiforme 80 (10 percent)
  • Glioma [unspecified] 14 (1.8 percent)
  • Oligodendroglioma 2 (0.3 percent)
  • Other 9 (1.1 percent)
  • Pontine glioma [unspecified] 0 (0 percent)

Induction regimens varied across and within studies (i.e., different chemotherapeutic agents and different (cumulative) dosages) and consisted of multiple cycles of chemotherapy and/or radiation and/or surgery. Conditioning regimens also varied. The most common regimens included thiotepa, etoposide, carboplatin, cyclophosphamide (with or without mesna), busulfan and carmustine (either alone or in combination with radiation therapy or additional drugs). Table 84 shows the pediatric outcomes that were reported across the 39 included studies.

Table 84. Outcomes reported: Glial tumors.

Table 84

Outcomes reported: Glial tumors.

Overall Survival

Data on overall survival were reported in all but three studies,505, 515, 518 and calculated from the raw data from the additional studies (Table 84). Survival data are presented by five histologic categories (Table 85). Individual studies varied in their method for calculating overall survival.

Table 85. Overall survival for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups: Glial tumors.

Table 85

Overall survival for single auto HSCT and comparison (conventional chemotherapy +/- radiation) groups: Glial tumors.

Fourteen studies examined overall survival for astrocytic gliomas.123, 132, 367, 485, 491, 492, 495, 500-502, 509, 510, 517, 518 Survival data were reported for 20 patients with astrocytic tumors who underwent autologous transplant and 106 conventional therapy patients. Fourteen studies examined overall survival for glioblastoma multiforme.123, 132, 367, 485, 491, 492, 495, 498, 500-502, 504, 509, 518 Survival data was reported for 45 patients with glioblastoma multiforme who underwent autologous transplant and 92 conventional therapy patients. Of the noncomparative studies reporting yearly OS, none had HSCT treatment. OS at 5 years ranged from 0 percent in recurrent patients to 25 percent for newly diagnosed patients. One study grouped OS of newly diagnosed AA and GBM patients and stratified by noninfant or infant status. These patients had a 5-year OS of 36±13 for noninfants and 25±15 for infants.

Finlay et al.485, compared a historic chemotherapy cohort (CCG-945) to astrocytoma and GBM patients receiving HSCT. This study provided 5-year recurrent HSCT and conventional therapy OS estimates of 40 percent and 4 percent for astrocytoma and 12 percent and 0 percent for glioblastoma multiforme respectively. The OS was statistically significantly better for HSCT compared to chemotherapy at p=0.010 and retained this significance when stratified by tumor histology. The authors also found evidence that degree of surgical debulking impacted survival. OS estimates stratified by treatment and degree of debulking minimized the treatment effect and yielding a nonsignificant survival difference of p=0.39 due to the poor prognosis of patients with bulky tumor in both treatment types. However, when the authors looked at HSCT versus chemotherapy treatment among only surgically debulked patients the HSCT patients had a better survival (p=0.017).

Three noncomparative studies reported yearly GBM OS with a 5-year OS estimate for autologous transplant of newly diagnosed GBM of 0-22 percent and newly diagnosed conventional therapy of 22 percent.498, 502, 521 Data comparing HSCT to conventional therapy was provided by Finlay et al.485 for recurrent/progressive GBM and AA showed an increase in survival for HSCT. No comparison was made for newly diagnosed AA due to a lack of HSCT studies. Data for newly diagnosed GBM seems to show a similarly poor prognosis for both HSCT and conventional therapy patients.

Seventeen studies examined overall survival for ependymoma132, 367, 489, 490, 495, 497, 501, 504, 507-509, 511-514, 516, 520 Survival data was reported for 71 patients with ependymal tumors who underwent transplant and 442 conventional therapy patients. No studies were comparative between HSCT and conventional therapy. Five studies reported overall survival with a 5 year overall survival estimates for autologous transplant of recurrent tumor of 10 percent and newly diagnosed of 38 percent.367, 490, 495, 497, 504 Conventional therapy did not include recurrent disease and found estimates of 35.2 percent to 64 percent for newly diagnosed anaplastic ependymoma with newly diagnosed nonanaplastic, mixed or unspecified ependymoma estimates of 52-74 percent.489, 508, 511, 513, 514, 516, 520 One study stratified by metastatic/nonmetastatic disease and obtained a 5-year OS of 33 percent and 59 percent respectively.512 For patients with newly diagnosed ependymoma, patients treated with HSCT appear to have inferior overall survival when compared to those treated with conventional therapy.

Sixty-four patients with choroid plexus carcinoma were in three conventional therapy studies and four patients with HSCT across three studies reported survival.488, 499, 504, 513, 522 All HSCT patients died between five and 25 months. A conventional therapy study of 29 patients had survival of 35 percent at last followup (median 25 months, range 3-85 months). Two studies reported 5-year OS of 21.5 and 36 percent.

Event-free Survival

Data on event-free survival can be found in Appendix D.

Adverse Effects

Nine HSCT studies reported adverse events in a patient population of 138 patients composed of 13 anaplastic astrocytoma, three anaplastic glioma, 49 ependymoma, one ganglioma, 30 glioblastoma multiforme, one oligodendroglioma, and 41 pontine tumors (Table 86).485, 490, 493, 497, 498, 500, 501, 503, 506 The conventional therapy studies reported adverse events for 113 ependymoma patients, 30 anaplastic astrocytoma patients, 40 glioblastoma multiforme patients, 18 high-grade glioma patients, seven pontine tumor patients and 15 choroid plexus tumor patients.508, 512, 513, 521 Overall, the level of adverse event reporting for both HSCT and conventional therapy may be underreported. Many studies included tumor types not relevant to this report in their design, and the authors in most instances did not give data on a tumor group or per patient basis when discussing adverse events.

Table 86. Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) group: Glial tumors.

Table 86

Adverse effects for single auto HSCT and comparison (conventional chemotherapy +/- radiation) group: Glial tumors.

Ongoing Research

Six trials were identified with currently unpublished results (Table 87). One trial was completed, two were ongoing, and three were recruiting participants. Anaplastic astrocytoma was investigated in four studies, brainstem glioma in one study, choroid plexus carcinoma in two studies, ependymoma in three studies, and glioblastoma multiforme in four studies. The estimated total enrollment of these trials is 363 participants, but with the exception of two studies, nonpediatric patients will also be enrolled. All studies include overall survival or event-free survival as outcomes relevant to this report.

Table 87. Ongoing trials: Glial tumors.

Table 87

Ongoing trials: Glial tumors.

Conclusions

Low strength evidence on overall survival suggests a benefit with single HSCT compared to conventional therapy for the treatment of:

  • High-risk recurrent or progressive anaplastic astrocytoma
  • High-risk recurrent glioblastoma multiforme.

Low strength evidence on overall survival suggests a harm due to higher treatment-related mortality with single HSCT compared to conventional chemotherapy for the treatment of nonanaplastic mixed or unspecified ependymoma.

The body of evidence on overall survival with single HSCT compared to conventional therapy was insufficient to draw conclusions for treatment of:

  • High-risk newly diagnosed anaplastic astrocytoma
  • High-risk newly diagnosed glioblastoma multiforme
  • Newly diagnosed anaplastic, nonanaplastic, mixed, or unspecified ependymoma
  • Recurrent ependymoma
  • Choroid plexus carcinoma
  • Other gliomas.

Systematic Reviews: Nonmalignant Disease

Inherited Metabolic Diseases Systematic Review

Background and Setting

Inherited metabolic diseases (IMD), also known as inborn errors of metabolism, are rare genetic diseases of biochemistry. IMDs are caused by defects of enzymes which result in the accumulation of substrates in tissues and organs. As substrates accumulate, progressive damage to the skeletal structure, connective tissues, organs, and in more severe disorders, the central nervous system occurs. Symptoms and the severity range widely among the IMDs. Many of the diseases are characterized by a rapid deterioration and have a life expectancy of a few years, while some of the IMDs have a slower course and patients may live into adulthood. While each condition is rare, the collection of these diseases has caused significant morbidity and mortality. Estimates of cumulative incidence for IMDs range from 1 in 1500 to 1 in 5,000 live births.270, 523, 524

In this report, IMDs will be discussed in three sections: 1) diseases with rapid progression of symptoms and life expectancies of 10 years or less, 2) diseases with slower progression of symptoms and life expectancies of more than 10 years, and 3) diseases with two different forms, one form which has a rapid progression of symptoms and one form which has a slow progression of symptoms. For diseases that have a rapid progression of symptoms, the expected outcome following HSCT is prolonged life expectancy. For diseases with a slow progression of symptoms, the expected outcomes following HSCT are improvements in neurocognitive and neurodevelopmental functioning.

The diseases with rapid progression of symptoms that were systematically reviewed include: Wolman disease, Gaucher disease Type II, Niemann-Pick Type A, mucolipidosis II (I-cell disease), cystinosis, and infantile free sialic acid disease. The diseases with slow progression of symptoms that were systematically reviewed are mucopolysaccharidosis II (Hunter's disease), mucopolysaccharidosis III (Sanfilippo disease), mucopolysaccharidosis IV (Morquio syndrome), Fabry's disease, Gaucher disease Type III, aspartylglucosaminuria, β-mannosidosis, mucolipidosis III, mucolipidosis IV, Niemann-Pick Type C, glycogen storage disease Type 2 (Pompe disease), Salla disease, and adrenomyeloneuropathy. Diseases with forms that progress rapidly and forms that progress slowly and that were systematically reviewed are Farber's disease, GM1 gangliosidosis, Tay-Sachs disease, Sandhoff's disease, ceroid lipofuscinosis, and galactosialidosis.

Evidence Summary

Diseases With Rapid Progression

The overall grade of strength of evidence for overall survival with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression is shown in Table 88.

Table 88. Overall grade of strength of evidence for overall survival with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression.

Table 88

Overall grade of strength of evidence for overall survival with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression.

Diseases With Slow Progression

The overall grade of strength of evidence for stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with slow progression is shown in Table 89.

Table 89. Overall grade of strength of evidence for stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with slow progression.

Table 89

Overall grade of strength of evidence for stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with slow progression.

Diseases With Forms That Progress Rapidly and Slowly

The overall grade of strength of evidence for overall survival and stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression and slow progression forms is shown in Table 90.

Table 90. Overall grade of strength of evidence for overall survival and stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression and slow progression form.

Table 90

Overall grade of strength of evidence for overall survival and stabilization of neurocognitive and neurodevelopmental symptoms with the use of HSCT for the treatment of inherited metabolic diseases with rapid progression and slow progression form.

Results

Table 91 shows the criteria that were used to select studies for this section.

Table 91. Study selection criteria: Inherited metabolic diseases.

Table 91

Study selection criteria: Inherited metabolic diseases.

Diseases With Rapid Progression

Wolman Disease

Wolman disease is a rare autosomal recessive disorder characterized by a deficiency of lysosomal acid lipase which causes an accumulation of cholesterol esters and triglycerides in the spleen, liver, adrenal glands, bone marrow, small intestines, and lymph nodes.261 Fewer than 80 cases have been identified. Symptoms appear immediately, within the first week of life, and include failure to thrive, jaundice, anemia, relentless vomiting, abdominal distention, steatorrhea, and hepatosplenomegaly. Because of the failure to absorb nutrients, severe malnutrition occurs and life expectancy is less than 6 months.526 Several patients with Wolman disease have undergone HSCT (Table 92).527-530

Table 92. Study characteristics and population for Wolman disease.

Table 92

Study characteristics and population for Wolman disease.

Refer to Appendix E Table E1 for details of neurocognitive and neurodevelopmental outcomes. In summary, two patients died of treatment-related mortality, one at 2.5 months post-transplant and one at 8 months post-transplant and one died from the natural progression of the disease.527 Three (of 4) patients who survived HSCT are long-term survivors, with followup from 4 to 11 years. They are highly functional in language skills, and social and behavioral skills. One attends regular school and two attend special schools.529, 531

The case report of the patient with Wolman disease who underwent HSCT reported growth in height, weight and head circumference.529 Of the three surviving Wolman disease patients in the case series, one showed improvement in motor skills and another is reported to have average gross motor skills and below average fine motor skills531

Evidence for this rapidly progressing disease which has a life expectancy of 6 months, consists of two case reports and two case series. A total of seven patients with Wolman disease have undergone HSCT. Two died from the procedure and 1 died from disease progression. Four have survived and have been followed for 0.3 to 11 years, with normal or near normal functioning. For three patients who have survived long-term followup from 4 to 11 years, HSCT altered the course of Wolman disease.

Gaucher Disease Type II

Gaucher disease is caused by a deficiency in the enzyme glucocerebrosidase, which leads to an accumulation of glucosylceramide in the spleen, liver, lungs, bone marrow, and sometimes the brain.261 There are three types of Gaucher disease. Gaucher Type I is discussed in the Narrative Review section of this report. Gaucher Type III is discussed in this Systematic Review section under diseases with slow progression.

Type II is the acute neuronopathic form, exhibiting hepatosplenomegaly as early as three months of age. There is severe central nervous system involvement and death occurs within two years of life. There is no effective treatment for Type II because of the rapid progression of symptoms and neurological involvement. No HSCT and Gaucher Type II studies were found in the literature.

Niemann-Pick Disease Type A

Niemann-Pick disease is characterized by the accumulation of lipids in the spleen, liver, lungs, bone marrow, and the brain. There are three types of this disease. Type A occurs most frequently in the Ashkenazi Jewish population (1 in 40,000), while the frequency of Type A and B in the general population is estimated to be 1 in 250,000.532 Type B is discussed in the Narrative Review section of this report. Type C is discussed under the heading “Other Lipidoses” within this Systematic Review.

Type A is the most severe form, occurring in infants and characterized by jaundice, an enlarged liver, and brain damage, with life expectancy of 3 years.261 Reports of HSCT on 3 Type A patients have been found in the literature (Table 93).

Table 93. Study characteristics and population for Niemann-Pick Type A.

Table 93

Study characteristics and population for Niemann-Pick Type A.

Refer to Appendix E Table E1 for details of neurocognitive and neurodevelopmental outcomes. In summary, a case report of a patient with Niemann-Pick Type A who underwent HSCT at 3 months of age, showed initial normal neurocognitive and neurodevelopmental progress, followed by brain atrophy at 0.6 years post-transplant, and the onset of seizure disorders and developmental delays by 1.7 years post-transplant.533 At the time of the report, the patient was alive at 2.7 years' followup. Two patients receiving HSCT continued to decline neurocognitively and neurodevelopmentally, and died 2 years post-transplant, from natural progression of disease.534 Autopsy of one patient showed very little enzyme present in target tissues of brain and liver.534

Evidence for Niemann-Pick Type A, which has a life expectancy of 3 years, consists of 1 case report and 1 case series.533, 534 HSCT did not prevent neurocognitive and neurodevelopmental decline in these patients. Based on these reports, HSCT does not show a benefit for Niemann-Pick Type A.

Mucolipidosis II (I-cell Disease)

Mucolipidosis II is an autosomal recessive disorder caused by a defective enzyme, N-acetylglucosamine-1-phosphotransferase, which is instrumental in the transport of enzymes. This defect causes a deficiency of lysosomal enzymes in fibroblasts, and an excess of lysosomal enzymes in tissues and extracellular fluids.535 This is a rare, panethnic disorder with an estimated frequency of 1 in 640,000 live births.524

The skeletal system is most severely affected. Death from progressive psychomotor retardation, pneumonia, or congestive heart failure usually occurs in early childhood. Symptom management of this disease includes antibiotics for respiratory infections and nutritional supplements.

There are reports of four cases of mucolipidosis II undergoing HSCT (Table 94).536-538 One mucolipidosis II patient was included in the retrospective study of 81 patients in the Japan Marrow Donor Program. The patient failed to engraft and no further information on that case could be separated from the aggregate data in that study.525

Table 94. Study characteristics and population for mucolipidosis II.

Table 94

Study characteristics and population for mucolipidosis II.

Refer to Appendix E Table E1 for details of neurocognitive and neurodevelopmental outcomes. In summary, one patient in a case series of combined diseases, did not have neurocognitive or neurodevelopmental followup, only adverse events reported.536 The patient experienced infectious complications, grade 2 skin aGVHD, and skin cGVHD. The patient was alive at last followup, which had a median of 2 years for the case series.536 One patient with delayed language skills continued to develop neurocognitively after HSCT, although abilities remained below real age. The patient's gross motor skills remained at level of a 1.5 year-old and fine motor skills were slowly developing through 5-years of followup.537 At the time of the report, the patient was alive at 5 years' followup. One patient with severe psychomotor retardation prior to HSCT gained developmental milestones of a 4- to 8-month old. The patient had no change in joint contractures and skeletal symptoms and died of disease progression at 5.6 years post-transplant.538

There was mention in one of the discussion sections537 of a personal communication with another physician who reportedly used HSCT to treat a patient with mucolipidosis II and that at 1 year post-transplant, the patient showed improvement in development and growth retardation. To our knowledge, this case has not been published.

Evidence for this disease which has a life expectancy of less than 1 decade, consists of three case reports. One patient died at 5.6 years post-transplant of disease progression, one is alive at 2 years' followup but with unknown neurological status, and the other patient was reported as showing progress neurocognitively, although below real age levels, and attends a special school. Based on these three patients with differing outcomes, there is uncertainty as to the benefit of HSCT for I-cell disease.

Cystinosis

Cystinosis is a rare autosomal recessive disease caused by a defect in cystinosin, which is needed to transport cystine out of lysosomes, which then results in the accumulation of cystine crystals in most major organs of the body.539 The incidence is estimated at 1 in 100,000-200,000, although the incidence in French Canadians may be higher.540 There are three types of cystinosis: classic nephropathic cystinosis, a rare adolescent form, and a mild adult-onset form.

Symptoms in the classic form present in the first year of life. Progressive renal damage and end stage renal failure is the usual cause of death, commonly within the first decade of life.539 The adolescent form of the disease is milder with a slower progression to renal failure. The adult form is benign, with no renal involvement.540 Renal transplant, oral cysteamine therapy, cysteamine eyedrops, and dialysis have prolonged survival into adulthood for patients with the nephropathic form.539 No studies of HSCT to treat cystinosis were found in the literature.

Infantile Sialic Acid Storage Disease

Infantile free sialic acid storage disease (ISSD) is a rare autosomal disorder caused by the accumulation of free sialic acid in lysosomes, due to a defect in the lysosomal membrane transport system.541 More than 27 ISSD cases have been reported. Dysmyelination of the brain occurs in ISSD. Symptoms present at birth and life expectancy is about a year, with cause of death commonly from respiratory infections.542 Disease management is symptom specific. No studies of HSCT to treat ISSD were found in the literature.

Diseases With Slow Progression

Hunter Syndrome (Mucopolysaccharidosis Type II)

Hunter Syndrome is a rare X-linked recessive disorder caused by a deficiency of the enzyme iduronate sulfatase, needed to degrade heparin sulfate and dermatan sulfate. The disease is panethnic, with an estimated incidence in Europe between 1 in 110,000–300,000; a higher incidence of 1 in 34,000 has been noted in the Jewish population living in Israel.543

There are two clinical forms of the disease, severe and attenuated. Onset of symptoms in the severe form occur at age 2 to 4 years. Survival can be expected into the second decade of life. Cause of death is usually heart disease, from valvular, myocardial, and ischemic factors.263 In the attenuated form, symptoms begin later in life, with minimal to no CNS involvement. Survival can extend into the fifth to sixth decade of life.263

Treatment is symptom specific: developmental, occupational, and physical therapy; shunting for hydrocephalus; tonsillectomy and adenoidectomy; positive pressure ventilation; carpal tunnel release; cardiac valve replacement; inguinal hernia repair; and hip replacement.544 HSCT has been attempted in MPS II patients, with both the severe (n=8) and attenuated forms (n=10), in attempts to slow or stop the progression of the disease (Table 95). An enzyme replacement therapy, Elaprase®, was approved by the FDA in 2006 for treatment of MPS II, following clinical trials which proved efficacy in patients with the attenuated form of the disease, aged 5-31 years.

Table 95. Study characteristics and population for mucopolysaccharidosis II (Hunter disease).

Table 95

Study characteristics and population for mucopolysaccharidosis II (Hunter disease).

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, among 32 patients undergoing HSCT, seven died of the treatment. In eight MPS II patients with the severe form, five showed decreases in neurocognitive scores545-547 and one showed stable scores.548 In 10 MPS II patients with the attenuated form, there are neurocognitive test scores for 6 patients. Four showed stable scores and two showed slight decreases in their neurocognitive scores.538, 545, 549, 550 Among the three case series and two case reports that did not specify if patients had the severe or attenuated form, there was neurocognitive information on three patients549: two patients showed neurocognitive decline and one patient was stable and attends a special school.

Of 32 MPS II patients undergoing HSCT, there was followup neurodevelopmental information for 19 patients. Improvements in joint stiffness were reported in 14 of the 19 patients,538, 545, 548, 551-553 and one patient showed improvement in both fine and gross motor skills.548

The two clinical trials of enzyme-replacement therapy for MPS II patients reported pre- and post-treatment measurements for 6-minute walk tests.548, 554, 555 The 1-year followup in the Phase II/III trial (N=96) showed improvements in distance walked by the ERT weekly group (p=0.01) and the enzyme-replacement therapy every other week group (p=0.07) compared to the placebo group. The open label extension (n=12) reported that 8 patients improved and 4 experienced no change in walk test results after 1 year of followup.

Evidence for the attenuated form of this disease with a life expectancy into adulthood, consists of three case reports and three case series. HSCT showed stabilization of cognitive skills in four of six patients. Though the numbers are small, HSCT may benefit MPS II patients with the attenuated form.

Evidence for the severe form of this disease with life expectancy into the second decade of life, consists of three case reports and one case series. Neurocognitive decline continued in seven of eight patients. Though the numbers are small, HSCT does not appear to benefit MPS II patients with the severe form.

Sanfilippo Syndrome (Mucopolysaccharidosis Type III)

Sanfilippo Syndrome is an autosomal recessive disorder, with an incidence of 1 in 70,000 births.560 There are four types of Sanfilippo Syndrome, differentiated by the specific enzyme deficiency needed to break down heparan sulfate (Type A: heparan sulfate sulfatase, Type B: N-acetyl-o-glucosaminidase, Type C: Acetyl CoA: o-glucosaminide N-acetyltransferase, and Type D: N-acetyl-o-glucosamine-6-sulfate sulfatase).

Type A is the most severe form. Unlike most mucopolysaccharidoses, Sanfilippo disease has milder somatic symptoms, but severe progressive CNS involvement.263 Initial clinical symptoms occur slowly from 1-6 years of age. Mental deterioration is progressive and severe by ages 6 to 10 years.560 Life expectancy is from 12-20 years, with cause of death primarily caused by cardiopulmonary arrest due to airway obstruction and/or pulmonary infection.263 Symptom management of this disease includes anticonvulsants and sedative medications to improve sleep quality. HSCT has been attempted in several MPS III patients (Table 96).553, 561-564

Table 96. Study characteristics and population for mucopolysaccharidosis III (Sanfilippo disease).

Table 96

Study characteristics and population for mucopolysaccharidosis III (Sanfilippo disease).

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, one patient died 5 months post-transplant of pneumonia.561 Of nine MPS III patients undergoing HSCT, there is neurocognitive followup information on six patients. There was a continuing deterioration in six patients553, 563, 564 and no significant improvement reported in one patient.562

There is neurodevelopmental information for three of the nine MPS III patients undergoing HSCT.563, 564 One patient experienced a slow and continuous decline in skeletal and muscular symptoms and was wheelchair-bound by 7.4 years after the transplant. This patient experienced the same physical deterioration as his untreated sibling.563 Twins experienced less neurodevelopmental decline compared to untreated brothers who were wheelchair-bound by the time they reached the age of the twins.564

Evidence for this disease with a life expectancy into the second decade consists of two case reports and two case series.553, 561, 562, 564 HSCT did not alter the neurocognitive decline but may have had some effect on the neurodevelopmental decline in two patients. Although the numbers are small, HSCT does not appear to benefit MPS III.

Morquio Syndrome (Mucopolysaccharidosis Type IV)

Morquio Syndrome is an autosomal recessive disorder with an estimated incidence of 1 in 200,000 births.565 There are two types, differentiated by which enzyme needed to degrade keratin sulfate is deficient (Type A: N-acetylgalactosamine 6-sulfatase, and Type B: β-galactosidase). Type A is the more severe form. Onset of symptoms occurs around 2 years of age. In most cases, normal intelligence is preserved.263 Life expectancy can extend into the third or fourth decade of life with the more severe form, while those with the milder form have been reported to live decades longer.264 Common causes of death include myelopathy, restrictive chest wall movement, and valvular heart disease.565 Spinal fusion to stabilize the upper cervical spine and prevent irreversible spinal cord injury can be a life-saving treatment for MPS IV patients. HSCT has been attempted on two MPS IV patients (Table 97).558, 566

Table 97. Study characteristics and population for mucopolysaccharidosis IV (Morquio syndrome).

Table 97

Study characteristics and population for mucopolysaccharidosis IV (Morquio syndrome).

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, one patient was 15 years old at the time of transplant and a pretransplant MRI showed no pathological findings in the brain or spinal cord. The patient had mild bone deformities at the time of transplant, and there was no followup for this patient.558 From echocardiograph, aortic stenosis and left ventricular dilatation were detected in one patient prior to HSCT. There was no change in cardiac symptoms after HSCT.566

Evidence for this disease with a life expectancy that varies from adolescence into adulthood, is based on two cases of HSCT found in the literature. The reports did not provide any post transplant neurocognitive or neurodevelopmental followup data.

Fabry Disease

Fabry disease is an X-linked recessive disorder characterized by decreased activity of α-galactosidase A. The prevalence is estimated at 1/40,000-60,000 males.526 The onset of symptoms and the severity of the disease vary widely. Males may exhibit symptoms in childhood or adolescence, or remain asymptomatic into adulthood. Female carriers may be asymptomatic or have symptoms as severe as affected males.526 Pain episodes, called Fabry pain crises, consist of burning, tingling, and numbness in the hands and feet, and can last several hours to days.261 Decline in kidney function in early adulthood is the main cause of premature death in Fabry disease. Cardiovascular disease is also a cause of premature death, with hypertension, mitral valve prolapse, or congestive heart failure occurring.261

Renal transplantation and long-term hemodialysis have prolonged life in Fabry's disease patients, and enzyme-replacement therapy using recombinant alpha-galatosidase has been shown to be safe and effective.261

Gaucher Disease Type III

Gaucher disease is caused by a deficiency in the enzyme glucocerebrosidase, which leads to an accumulation of glucosylceramide in the spleen, liver, lungs, bone marrow, and sometimes the brain.261 There are three types of Gaucher disease. Gaucher Type I is discussed in the Narrative Review section. Gaucher Type II is discussed in the Systematic Review under diseases with rapid progression.

Gaucher Type III is the subacute neuronopathic form, usually beginning later in childhood or adolescence, with loss of muscle coordination and cognitive deterioration progressing more slowly than in Type II.261 Gaucher Type III patients may live into adulthood. Enzyme replacement therapy can be used to alleviate severe visceral symptoms, but is not effective in altering the neurologic progression of the disease.261 Combinations of enzyme replacement therapy using recombinant imiglucerase or velaglucerase, substrate reduction therapy using miglustat, and HSCT have been attempted in Type III Gaucher patients (Table 98).

Table 98. Study characteristics and population for Gaucher Type III.

Table 98

Study characteristics and population for Gaucher Type III.

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, among eight patients undergoing HSCT (two case reports and one case series of 6 patients), five showed stable neurocognitive scores.561, 569 All eight patients showed improved growth, although skeletal symptoms persisted.561, 568, 569 A case series that included two patients that had HSCT followed by enzyme-replacement therapy, report only followup data.567 Both patients have borderline mental retardation at last followup, but the mental status prior to HSCT and enzyme-replacement therapy is not specified.

Of 23 Gaucher Type III patients treated with enzyme-replacement therapy, neurocognitive followup is available on nine patients. Seven of the nine patients showed stable neurocognitive function,574, 575 one deteriorated clinically,574 and one who was showing improvement following enzyme-replacement therapy, deteriorated when therapy was discontinued.572 Enzyme-replacement therapy improves growth, but cannot change skeletal deformities. In the enzyme-replacement therapy case series of 11 patients for which grading severity of marrow involvement was provided, one worsened, five remained constant, and five experienced complete improvement.571 In a 2-year randomized controlled trial of substrate reduction therapy (miglustat) with enzyme-replacement therapy (imiglucerase; n=21) compared to enzyme-replacement therapy alone (n=9), there was no significant difference between study groups using several neurocognitive measurements.570

Evidence for HSCT for the treatment of Gaucher Type III which has a life expectancy extending into adulthood, consists of two case reports and two case series. In one case series, HSCT was followed by enzyme-replacement therapy. Among the patients who were treated with HSCT only, five of eight had stable neurocognitive scores at last followup. Among patients treated with enzyme-replacement therapy only, seven of nine had stable neurocognitive scores at last followup. Patients undergoing HSCT and patients treated with enzyme-replacement therapy have shown improved growth, although skeletal symptoms persist. HSCT appears to have a similar benefit compared to enzyme-replacement therapy.

Aspartylglucosaminuria

Aspartylglucosaminuria is a rare autosomal recessive disease characterized by a deficiency in the enzyme aspartylglucosaminidase, leading to an accumulation of glycoproteins in the liver, spleen, and thyroid. There is a higher prevalence of this disease in Finland, where the carrier frequency is estimated to be 1 in 36576 and the estimated incidence of the disease is 1 in 35,000. The estimated incidence outside of Finland is 1 in 2,000,000 births.

In the first year of life, recurrent infections, diarrhea, and hernia may occur. During adolescence, the intellectual disabilities worsen. The central nervous system is affected. Survival to mid-adulthood is expected, with most deaths attributed to pneumonia or other pulmonary complications.576 Anticonvulsant medications have been used to control seizures. HSCT has been attempted as a potential treatment of this disease (Table 99).

Table 99. Study characteristics and population for aspartylglucosaminuria.

Table 99

Study characteristics and population for aspartylglucosaminuria.

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, there were no reports of treatment-related mortality in the 10 patients undergoing HSCT. Of 10 patients with aspartylglucosaminuria undergoing HSCT, there is neurocognitive followup on nine. Two patients have improved concentration and cooperation.579 Two patients have stabilized developmentally at 5 years of age (real ages 15 and 11 years), and can speak in sentences and understand words in two languages.577 Five patients had, on average, lower developmental ages compared to 12 untreated patients, but direct comparisons may not be appropriate because the severity of disease differs widely in this disease. Two of the five transplanted patients were more severely retarded than any of the nontransplanted patients, potentially skewing the average age differential higher in the transplanted group.578

Evidence for aspartylglucosaminuria which has a life expectancy into mid-adulthood, consists of one case report and three case series, with a total of 10 transplants. Neurocognitive and neurodevelopmental measurements did not show clear improvements following HSCT. Small numbers in studies, and differences in severity of disease make interpretations of results difficult.

ß-Mannosidosis

ß-mannosidosis is a rare autosomal recessive disorder caused by a deficiency in the enzyme ß-mannosidase, resulting in the accumulation of oligosaccarides in lysosomes. Twenty cases have been identified worldwide, but the incidence may be higher because people with milder symptoms may never be diagnosed.

The onset of symptoms varies from infancy to adolescence, and the severity of symptoms varies from relatively mild to moderately severe.581 Mental retardation is present in all individuals with this disease. There is no cure for ß-mannosidosis and treatment is symptom-specific.

No reports of HSCT for ß-mannosidosis patients have been found.

Mucolipidosis III (Pseudo-Hurler Polydystrophy)

Mucolipidosis III is a rare autosomal recessive disorder caused by a deficiency of the enzyme, N-acetylglucosamine-1-phosphotransferase. A defect of this enzyme affects the function of all lysosomal enzymes, which in turn causes the accumulation of a variety of substrates.535

Symptoms present between the ages of 4 to 5 years and include joint stiffness and short stature. Survival to adulthood is expected. There is no cure and treatment is symptom-specific, and may include: low-impact physical therapy for stiff joints, myringotomy tube placement for recurrent otitis media, tendon release for carpal tunnel syndrome, bilateral hip replacement for older adolescents with milder disease, and monthly bisphosphonate pamidronate IV for bone pain associated with osteoporosis.

No reports of HSCT for mucolipidosis III have been found.

Mucolipidosis IV

Mucolipidosis IV is a rare autosomal recessive disorder caused by a defect in the protein mucolipin-1, which is needed in the transport of lipids and proteins. This defect results in the build-up of lipids and proteins in lysosomes, affecting the development and maintenance of the brain and retinas.582 An estimated 1 in 40,000 have mucolipidosis IV, with 70 percent having Ashkenazi Jewish ancestry.

There is a severe and more common form called typical mucolipidosis IV (about 95 percent) and a milder form called atypical mucolipidosis IV. In the severe form, mental and motor developmental delays occur within the first year of life. Most are unable to walk independently. Those with the milder form have less severe psychomotor and ophthalmic symptoms, and may be ambulatory. Life expectancy extends to adulthood, though a shorter life span is expected.582

Treatment is symptom-specific and may include: physical therapy for spasticity and ataxia, antiepileptic drugs, topical lubricating eyedrops, artificial tears, gels, or ointments for ocular irritation, and surgery for strabismus.

There are no reports of HSCT attempted in patients with mucolipidosis IV.

Niemann-Pick Disease C

Niemann-Pick disease is characterized by the accumulation of lipids in the spleen, liver, lungs, bone marrow, and the brain. There are three types of this disease. Type A is discussed in the “Sphingolipidoses” section of this Systematic Review and Type B is discussed in the Narrative Review section. The incidence of Type C is estimated to be 1 in 150,000 and is most common in Nova Scotia among those of French-Acadian descent.523

Prolonged neonatal jaundice may occur, with no other symptoms until 1-2 years later or potentially until teen or adult years, when the disease develops a slow, progressive neurodegenerative course.261 Death may occur in the late second or third decade of life, commonly from aspiration pneumonia. Management of this disease is symptom-specific for seizures, dystonia, and cataplexy, and may include chest physical therapy with aggressive bronchodilation and antibiotics for recurrent infections and seizure management. A randomized controlled study using substrate reduction therapy versus standard care has been conducted, and there are two case reports of HSCT to treat this disease (Table 100).

Table 100. Study characteristics and population for Niemann-Pick Type C.

Table 100

Study characteristics and population for Niemann-Pick Type C.

Refer to Appendix E Table E2 for details of neurocognitive and neurodevelopmental outcomes. In summary, results from one case report of a patient with Niemann-Pick Type C undergoing HSCT showed that the transplant did not stop a progressive decline in developmental age, and an MRI confirmed brain atrophy. The patient became bedridden during the conditioning phase of the treatment. She never recovered developmentally following the transplant.584 The second case of HSCT showed a resolution of lung disease in the patient, and normal neurocognitive and neurodevelopmental progress, except for delayed speech.583 An abstract referenced in the most recent report of HSCT583 describes the resolution of lung disease in a Niemann-Pick Type C transplanted patient at 2 months post-transplant, but the patient died 3 months post-transplant of an adenovirus pulmonary infection.

Results from the randomized, controlled trial comparing substrate reduction therapy to routine symptom management and the retrospective cohort of substrate reduction therapy for Niemann-Pick Type C combined data for pediatric and adult patients.586, 588 The randomized, controlled trial did not find a significant difference in the mini-mental status examination (p=0.165) but found significantly improved ambulatory indexes in the treated group588 and the cohort study reported majority stable or improved scores in ambulation.586 The open-label extension study, which focused on pediatric patients, reported that eight of ten patients were stable in ambulation.585

Evidence for HSCT and Niemann-Pick Type C which has a life expectancy into the second to third decade, consists of two case reports. HSCT for one patient was not successful in stopping the neurocognitive and neurodevelopmental decline. One HSCT patient is developing normally at 1.7 years post-transplant. Based on two case reports, it is unclear if HSCT provides a benefit in the treatment of Niemann-Pick Type C.

Glycogen Storage Disease Type 2 (Pompe Disease)

Pompe disease is an autosomal recessive disorder caused by a deficiency in acid maltase, which results in the accumulation of lysosomal glycogen in tissues and cells. Cardiac, skeletal, and smooth muscle cells are the most seriously affected.589 The incidence is estimated at 1 in 40,000 live births. Age of onset and severity of symptoms varies among patients.

In infantile-onset Pompe disease, symptoms begin within the first few months of life and life expectancy is less than one year, with cause of death usually from cardiorespiratory failure or respiratory infection. The juvenile and adult-onset forms of the disease have either no or less severe cardiac involvement. Life expectancy ranges from early childhood to late adulthood, depending on the rate of disease progression. Respiratory failure is the most common cause of death.589 Several clinical trials of enzyme-replacement therapy in patients with infantile-onset Pompe disease have shown promising cardiac responses and variable skeletal responses to the treatment.590, 591

There have been no reports of HSCT in the treatment of Pompe disease.

Salla Disease

Salla disease is a type of sialic acid storage disease, which is a rare autosomal disorder caused by the accumulation of free sialic acid in lysosomes, due to a defect in the lysosomal membrane transport system.541 Salla disease is autosomal recessive. One hundred twenty Salla disease cases have been reported. Patients appear normal at birth, then develop psychomotor delay and ataxia during infancy, as dysmyelination of the brain occurs. Life expectancy is slightly reduced.541 Disease management is symptom-specific.

There are no reports of HSCT used to treat Salla disease.

Adrenomyeloneuropathy

Adrenomyeloneuropathy is a variant of the X-linked recessive disorder, adrenoleukodystrophy, which is discussed in the Narrative Review section of this report. These disorders are caused by the accumulation of very long chain fatty acids in the brain and adrenal cortex, due to a deficiency in the enzyme that breaks down fatty acids.592 About 40 percent of males with adrenoleukodystrophy develop adrenomyeloneuropathy, which presents in their late twenties as a chronic disorder of the spinal cord and peripheral nerves.593 The severity of symptoms varies greatly, even within one family. Depending on the severity of symptoms, life expectancy can reach late adulthood, though ambulation with a cane or walker may be necessary. HSCT has been shown to prevent the progression of symptoms in adrenoleukodystrophy if performed prior to the development of neurological symptoms.

A single case of HSCT for a 39-year-old male with adrenomyeloneuropathy was found in the literature.561 No pediatric cases treated with HSCT have been reported.

Diseases With Forms That Progress Rapidly and Slowly

Farber Disease

Farber disease is an autosomal recessive disorder characterized by a deficiency in ceramidase, resulting in the accumulation of ceramide in various tissues, the central nervous system, and most notably the joints. Fifty cases of this disease have been reported in the literature.594

Symptoms can begin in the first few weeks of life.261 Nodules forming on the vocal cords cause hoarseness and breathing difficulties, which sometimes require the insertion of a breathing tube. Life expectancy in Type 1, the more severe form which has central nervous system involvement, is 2 years of age with progressive neurological deterioration as cause of death. Patients with the milder form, Type 2/3 with either no or mild central nervous system symptoms, can live to their teenage years with chronic respiratory failure as the most common cause of death.595

Physical therapy or surgery may provide relief of contractures, and surgery to remove nodules, granulomas, and possibly enlarged lymph nodes may be recommended. Hematopoietic stem-cell transplantation has been attempted in two patients with Type 1 Farber and in five patients with Type 2/3 Farber (Table 101).

Table 101. Study characteristics and population for Farber's disease.

Table 101

Study characteristics and population for Farber's disease.

Refer to Appendix E, Table E3 for details of neurocognitive and neurodevelopmental outcomes. In summary, no treatment-related mortality was reported in the seven patients with Farber disease undergoing HSCT. There is neurocognitive followup on the two patients with Type 1 Farber disease with CNS involvement. In one patient at the time of transplant, her developmental age was equivalent to her real age. After 1.4 years followup, at age 2.1 years, her developmental age had deteriorated to 0.6 years.598 The second Type I patient had mental regression prior to the transplant, which worsened following the transplant. This patient died 6 months post-transplant of disease progression.553 No neurocognitive followup was provided for the five Farber disease patients who had Type 2 disease, which has little or no CNS involvement.

The five patients reported in the case series on Farber Type 2/3 had nodule and joint inflammation. HSCT was successful in reducing the number of subcutaneous nodules and reducing the number of joints with limited range of motion in five of five patients.596, 597

Evidence for Type 1 Farber disease with CNS involvement and a life expectancy of 2 years, consists of one case report and one case series. HSCT did not stop the neurocognitive deterioration in these patients. Evidence for Type 2 Farber disease without CNS involvement and a life expectancy extending into the second decade, consists of two case series. In all five patients with Farber Type 2/3 undergoing HSCT, both the number of subcutaneous nodules and the number of joints with limited range of motion were reduced. Based on these five patients, HSCT appears to improve the quality of life of patients with Farber Type 2/3.

GM1 Gangliosidosis

GM1 gangliosidosis is an autosomal recessive disorder caused by a deficiency in ß-galactosidase. There are three subtypes, classified by age at presentation: infantile (type 1), juvenile (type 2), and adult (type 3). Estimated incidence is 1 in 100,000-200,000 live births.599 The infantile form, which can present as early as six months, is characterized by overall developmental retardation and generalized seizures. Survival is 2-4 years, with death most commonly due to aspiration pneumonia. Symptoms in the juvenile form begin around 1 year and are primarily neurological. Progression of this form of the disease is slow, and survival through the fourth decade of life is possible. The adult form is a slowly progressive disease characterized by spasticity, ataxia, dysarthria, and loss of cognitive function.600

Research in the areas of enzyme replacement therapy and gene therapy for this disease are ongoing, but have not advanced to human trials.599 A case report describes the use of HSCT to treat a patient with the juvenile form of the disease (Table 102).

Table 102. Study characteristics and population for GM1 gangliosidosis.

Table 102

Study characteristics and population for GM1 gangliosidosis.

Refer to Appendix E, Table E3 for details of neurocognitive and neurodevelopmental outcomes. In summary, a case report of a patient with GM1 gangliosidosis juvenile form describes a slow deterioration in neurocognitive and neurodevelopmental measurements.601

There have been no reports of HSCT for the infantile form of GM1 gangliosidosis, which has a life expectancy of 2 to 4 years. Evidence for the juvenile form of GM1 gangliosidosis, which has a life expectancy extending into the second through fourth decade, consists of 1 case report. Based on this case report, HSCT did not alter the course of the disease.

Tay-Sachs Disease

Tay-Sachs disease is an autosomal recessive disorder caused by a deficiency in the isoenzyme hexosaminidase A, resulting in the accumulation of GM2 ganglioside in the brain. The Ashkenazi Jewish population is most at risk, with a carrier rate estimated at 1 in 30.600 There are infantile-, juvenile-, and adult-onset forms of the disease. In the infantile form, patients have no hexosaminidase A enzyme and in the juvenile and adult forms, patients have low levels of hexosaminidase A enzyme. The infantile form is the most severe, and other than a marked startle reaction to noise, infants appear normal until about 6 months of age when developmental delays begin. Life expectancy is 4 to 5 years, with aspiration or bronchopneumonia the most common causes of death.261 The juvenile and adult forms are rare and symptoms are less severe.

Anticonvulsant medication to control seizures, proper hydration to keep airways open, and feeding tubes to provide nutritional supplements have been recommended. HSCT, substrate reduction therapy, and a combination of both, have been attempted on several Tay-Sachs disease patients (Table 103).

Table 103. Study characteristics and population for Tay-Sachs disease.

Table 103

Study characteristics and population for Tay-Sachs disease.

Refer to Appendix E, Table E3 for details of neurocognitive and neurodevelopmental outcomes. Two case series of HSCT to treat several different diseases included one patient with Tay-Sachs in each series (disease form not specified). One patient died at 4.6 years post-transplant of a possible infection.556 The other patient had psychomotor retardation at the time of transplant and further regressed to a vegetative state at 1.7 years' followup.553

The case report602 was of a patient with the juvenile form of Tay-Sachs. In the case report, brain MRI, EEG, and neuropsychological tests showed neurological deterioration at 1.5 years post-transplant. At that time, substrate reduction therapy was initiated, but was not successful in stopping the deterioration. Neurodevelopmental followup in this case report showed motor skills deteriorating by 0.5 years post-transplant in this patient; her deterioration was comparable to her untreated sister's.

Among the two patients with the juvenile form who were treated with substrate reduction therapy,603 one who had mild cognitive impairment pretreatment experienced an acute psychotic event at 1.3 years post-treatment, and one who had severe cognitive impairment pretreatment had increased spasticity and seizures post-treatment. The 2 Tay-Sachs disease patients with the juvenile form of the disease who were treated with substrate reduction therapy, continued to have neurodevelopmental decline following the treatment.

Evidence for the juvenile form of Tay-Sachs disease which has a life expectancy of 15 years, consists of one case report. The patient continued to show neurocognitive and neurodevelopmental decline similar to what was experienced in the untreated sibling. Based on this case report, HSCT does not show a benefit in the treatment of the juvenile form of Tay-Sachs disease.

Ceroid Lipofuscinosis

Neuronal ceroid lipofuscinoses are autosomal recessive disorders which are the most common class of neurodegenerative diseases in children.600 A defect in the enzyme that degrades fatty acylated proteins causes the storage of autofluorescent lipopigments in lysosomes.604 Worldwide incidence of this disease is estimated at 1 in 20,000-100,000, but the incidence is higher in Finland.600

Depending on which gene is affected, symptoms may begin during early infancy, late infancy, or during juvenile years. Symptoms develop by the end of age 1 in the early infantile form with life expectancy from 6 to 13 years. In the late infantile form, symptoms begin from 2 to 4 years of age, with a life expectancy extending from 6 to 40 years. In the juvenile form, symptoms begin between 5 to 10 years of age with a life expectancy from teens to thirties.600

There is no cure for these disorders and treatment is symptom-specific: antiepileptic drugs and benzodiazepines for seizures, anxiety, and spasticity, gastric tubes for swallowing problems, and antidepressants and antipsychotic agents for patients with the juvenile form. HSCT has been performed in several patients with the early infantile form of the disease (Table 104).

Table 104. Study characteristics and population for ceroid lipofucinosis.

Table 104

Study characteristics and population for ceroid lipofucinosis.

Refer to Appendix E, Table E3 for details of neurocognitive and neurodevelopmental outcomes. Neurocognitive decline continued in three of three patients with ceroid lipofuscinosis with the infantile form undergoing HSCT, as measured by cerebral cortical atrophy and periventricular white matter hyperintensity. HSCT did not prevent the neurodevelopmental decline in the three patients with infantile ceroid lipofuscinosis. By followup of 2 to 4 years, all three were hypotonic and spastic.

Evidence for this disease which has a life expectancy of 6-13 years, consists of one case series of three patients. The procedure was unable to stop the neurocognitive and neurodevelopmental decline in all three patients. Based on this case series, HSCT does not show a benefit of HSCT for the treatment of infantile ceroid lipofuscinosis.

Galactosialidosis

Galactosialidosis is a rare autosomal recessive condition in which there is a deficiency of two lysosomal enzymes, neuraminidase and β-galactosidase. This enzyme deficiency causes the accumulation of oligosaccharides in many tissues such as the liver, bone marrow, and brain.576 There are three forms which differ by age of onset of symptoms and symptom severity. One-hundred cases have been reported, with 60 percent of the juvenile/adult forms in patients of Japanese descent.606

In the early infantile form, fluid accumulation begins before birth. Life expectancy does not extend beyond late infancy, with kidney failure or cardiomegaly as common causes of death. Symptoms in the late infantile form of the disease are similar to those in the early infantile form, though less severe and the onset is later in the first year of life. Life expectancy can extend into the second decade of life, depending on severity of symptoms. The juvenile/adult form of the disease is least severe, with symptoms first occurring usually in the teen years. There is no cure for galactosialidosis and treatment is symptom specific.

A retrospective study of 81 patients in the Japan Marrow Donor Program who underwent unrelated bone marrow transplantations for immunodeficiency and metabolic diseases reported a single case of galactosialidosis within its study population.525 The form of galactosialidosis was not specified in the report. Outcomes were cumulative overall and event-free survival, and cumulative acute and chronic graft-versus-host disease. Engraftment occurred in the galactosialidosis case, but no other information on that case could be separated from the aggregate data.

Sandhoff's Disease

Sandhoff's disease is caused by a deficiency in both hexosaminidase A and B, resulting in the accumulation of GM2 ganglioside in lysosomes. Symptoms are similar to those in Tay-Sachs disease, presenting at about 6 months of age. Life expectancy is 3 years of age.600 Symptom management includes anticonvulsant medication to control seizures, and proper hydration and nutrition to keep airways open.

A case of a patient with Sandhoff's disease undergoing HSCT is reported in the literature, but the form of the disease is not specified (Table 105). There is also a single arm study reporting the use of substrate reduction therapy in 3 patients with Sandhoff's disease (juvenile form).

Table 105. Study characteristics and population for Sandhoff's disease.

Table 105

Study characteristics and population for Sandhoff's disease.

Refer to Appendix E, Table E3 for details of neurocognitive and neurodevelopmental outcomes. In summary, there is no neurocognitive or neurodevelopmental information in the patient with Sandhoff's disease (form unspecified) who underwent HSCT. The three patients with Sandhoff's disease who were treated with substrate reduction therapy experienced stable neurocognitive scores, but neurodevelopmental decline occurred.603 One became wheelchair dependent by 1.8 years post-treatment, and two had gait disturbance.

Evidence for Sandhoff's disease consists of one case report. The report did not specify if the patient had the infantile form or the juvenile form of the disease. No neurocognitive or neurodevelopmental followup information on the single Sandhoff's disease patient was provided; no conclusions on effectiveness can be made.

Adverse Effects

Table 106 summarizes the adverse effects reported in patients undergoing HSCT for inherited metabolic disorders.

Table 106. Adverse effects for treatment (HSCT) in IMD patients.

Table 106

Adverse effects for treatment (HSCT) in IMD patients.

Ongoing Research

“Stem Cell Transplantation for Inborn Errors of Metabolism,” a study sponsored by the Masonic Cancer Center of the University of Minnesota, is ongoing and no longer recruiting. The study is comparing patients treated by bone marrow, peripheral blood, or umbilical cord blood transplantation after March 2001 with historical controls. Outcomes to be measured include: survival, change in neuropsychometric function, rate of donor cell engraftment, rate of graft-versus-host disease, and toxicity of HSCT therapy. Patients with the following diseases were eligible to participate in the study: adrenoleukodystrophy, metachromatic leukodystrophy, globoid cell leukodystrophy, Gaucher disease, fucosidosis, Wolman's disease, Niemann-Pick disease, Batten disease, GM1 gangliosidosis, Tay-Sachs disease, and Sandhoff disease. The study began in January 1995 and the estimated study completion date was June 2010.

Conclusions

Rapidly Progressive Diseases
  • High strength evidence on overall survival suggests a benefit with single HSCT compared to conventional management for Wolman's disease.
  • Low strength evidence on overall survival suggests no benefit with single HSCT compared to symptom management or disease natural history for Niemann-Pick Type A.
  • The body of evidence on overall survival with single HSCT compared to symptom management is insufficient to draw conclusions for mucolipidosis II (I-cell disease), Gaucher disease type II, cystinosis and infantile free sialic acid disease.
Slowly Progressive Diseases
  • Low strength evidence on neurodevelopmental outcomes suggests a benefit with single HSCT compared to enzyme replacement therapy for the attenuated and severe forms of MPS II (Hunter's disease).
  • Low strength evidence on neurocognitive outcomes suggests a benefit with single HSCT compared to enzyme replacement therapy for the attenuated form of MPS II (Hunter's disease).
  • Low strength evidence on neurocognitive outcomes suggests no benefit with single HSCT compared to enzyme replacement therapy for the severe form of MPS II (Hunter's disease).
  • Low strength evidence on neurocognitive outcomes suggests no benefit with single HSCT compared to enzyme replacement therapy for Gaucher Type III.
  • Low strength evidence on neurocognitive or neurodevelopmental outcomes suggests no benefit with single HSCT compared to symptom management, substrate reduction therapy or disease natural history for MPS III (Sanfilippo).
  • The body of evidence on neurocognitive or neurodevelopmental outcomes with single HSCT compared to symptom management and/or disease natural history is insufficient to draw conclusions for Niemann-Pick type C, MPS IV (Morquio syndrome), aspartylglucosaminuria, Fabry's disease, β-mannosidosis, mucolipidosis III or IV, glycogen storage disease type II (Pompe disease), Salla disease, and adrenomyeloneuropathy.
Disease With Both Rapidly and Slowly Progressive Forms
  • High strength evidence on number of subcutaneous nodules and number of joints with limited range of motion suggests a benefit with single HSCT compared to symptom management or disease natural history for Farber's disease Type 2/3.
  • Low strength evidence on neurocognitive outcomes suggests no benefit with single HSCT compared to symptom management or disease natural history for infantile ceroid lipofuscinosis.
  • The body of evidence on overall survival and/or neurocognitive and neurodevelopmental outcomes with single HSCT compared to symptom management and or disease natural history is insufficient to draw conclusions for galactosialidosis (type unspecified) and Sandhoff disease (type unspecified), Farber's disease type I, infantile and juvenile forms of GM1, infantile and juvenile forms of Tay-Sachs, infantile GM1 gangliosidosis, and juvenile ceroid lipofuscinosis.

Autoimmune Diseases Systematic Review

Type 1 Diabetes Mellitus

Background and Setting

Type 1 diabetes mellitus (DM1) is a T-cell mediated autoimmune disease characterized by selective, relentless and irreversible destruction of insulin-producing pancreatic beta-cells.607 DM1 is the most common autoimmune disorder in childhood, with an estimated incidence of 15,000 newly diagnosed cases in the U.S. annually based on 2002-2003 data.608 The disease typically is clinically diagnosed after approximately 60 to 80 percent of beta-cell mass has been destroyed.609 At this stage of disease, exogenous insulin treatment is required to maintain glucose homeostasis and survival. While DM1 comprises 5-10 percent of all diabetic causes, it is ultimately associated with a high frequency of vascular-related complications, including heart disease, stroke, blindness, and renal disease, with highly compromised quality of life and life expectancy.610

According to the U.S. Centers for Disease Control and Prevention, diabetes was the seventh leading cause of death listed on U.S. death certificates in 2006. Intensive insulin therapy (IIT) represents the gold standard treatment for DM1, to maintain tight control of blood glucose levels, as reflected by levels of HbA1C. IIT is delivered by multiple daily injections or by continuous subcutaneous infusion. Both methods have been shown to decrease the risk of diabetic retinopathy, nephropathy, and neuropathy by 39 to 90 percent and reduce their rate of progression by 39 to 60 percent when compared to standard insulin therapy with 1 to 2 injections daily.611 However, IIT is complicated by lack of patient acceptance and compliance, cannot fully prevent diabetic complications, and is associated with increased risk of severe hypoglycemia compared to standard therapy.

While DM1 does not typically develop into a fulminant, life-threatening form, it is a relentlessly progressive disorder despite IIT. The natural history may be transiently altered, but not halted, by coadministration of IIT and immune modulating therapies that include cyclosporine, azathioprine, prednisone, etanercept, and antithymocyte globulin (ATG).607 These approaches may induce a slower decline or some initial improvement in C-peptide levels, which directly reflect beta-cell mass and endogenous insulin production. However, the majority of patients continue to require increasing amounts of exogenous insulin. Furthermore, the toxic effects of immune suppressants, concerns about potential risks associated with immune suppression, and the need for continuous treatment in an otherwise healthy young population limit the use of these agents in conjunction with IIT.

For these reasons, based on a theory of possible reconstitution of immune tolerance after “immunologic reset,” nonmyeloablative autologous HSCT has been investigated as a way to effect an intense, but brief, immune suppression and preserve islet cell mass in children with newly diagnosed DM1. It is hypothesized that early intervention with HSCT will prevent the development of DM1-associated complications, improve quality of life, and ultimately increase life expectancy in this population. The effects of HSCT on insulin use and C-peptide levels will be compared to those parameters in children treated with IIT, in the context of adverse events associated with HSCT and IIT.

Evidence Summary

The overall grade of the strength of evidence for insulin independence and the use of HSCT for the treatment of autoimmune type I juvenile diabetes mellitus is shown in Table 107.

Table 107. Overall grade of strength of evidence for insulin independence and the use of HSCT for the treatment of autoimmune Type I diabetes mellitus.

Table 107

Overall grade of strength of evidence for insulin independence and the use of HSCT for the treatment of autoimmune Type I diabetes mellitus.

Evidence compiled for this review includes one prospective Phase I/II study of autologous HSCT (n=18 pediatric patients) that reported pre- and post-HSCT data on C-peptide levels and daily insulin use. Comparator data were obtained from the IIT control arms of two studies (total n=35) in newly diagnosed pediatric DM1 patients.

In the HSCT study, among 18 pediatric patients, the majority (89 percent) became free from insulin, either continuously (63 percent) or transiently (37 percent). Insulin independence was maintained for 7 to 52 months at total followup that ranged from 9 to 56 months. Among the 6 patients who resumed insulin, daily doses were lower than prior to HSCT. There was no treatment-related mortality in the HSCT study.

Results

The electronic literature search identified 15 citations relevant to HSCT and DM1, from which seven were retrieved for full-text screening, including those found in examination of the bibliographies of retrieved articles. A total of three reports were included in this review.612-614

Table 108 shows the criteria that were used to select studies for this section.

Table 108. Study selection criteria: Type I DM.

Table 108

Study selection criteria: Type I DM.

Table 109 shows the characteristics of one Phase I/II study of HSCT,612 and the IIT control arms of two randomized trials that compared IIT with IIT plus an immunosuppressant agent.613, 614 All three studies included pediatric patients with DM1 who had been clinically diagnosed within 6 weeks prior to study entry.

Table 109. Type 1 juvenile diabetes mellitus study characteristics and population.

Table 109

Type 1 juvenile diabetes mellitus study characteristics and population.

In the HSCT study, peripheral blood hematopoietic stem cells were mobilized with cyclophosphamide (2 g/m2) and granulocyte colony-stimulating factor (10 μg/kg daily).612 Patients were conditioned with a nonmyeloablative regimen comprising cyclophosphamide (50 mg/kg daily for 4 days) and rabbit antithymocyte globulin (0.5 mg/kg daily for 1 day, then 1 mg/kg daily for 4 days) prior to stem cell infusion. The IIT studies utilized 3 to 4 injections of short- or intermediate-acting insulin, with blood glucose levels monitored and maintained as near to normal as possible.613, 614

Table 110 shows the outcomes that were reported across the studies included in this report.

Table 110. Outcomes reported: Type I DM.

Table 110

Outcomes reported: Type I DM.

Insulin Requirements

Daily pretransplant insulin use ranged from 0.13 to 0.59 IU/kg in the HSCT study.612 Insulin was suspended in 16 of 18 (89 percent) pediatric patients following HSCT.612 Among the 16 who became insulin-independent, 10 were reported continuously free for an average of 31 months (range: 14-52 months) at followup times that ranged from 9 to 56 months. Patients who ultimately resumed insulin remained free from its use for about 15 months (range 7 to 47 months), at followup times that ranged from 9 to 58 months. However, daily insulin doses after exogenous treatment was resumed were relatively small, ranging from 0.1 to 0.3 IU/kg, compared to premobilization doses that ranged from 0.13 to 0.44 IU/kg, maintaining good glucose control.

In one IIT study, daily insulin use averaged 0.91 ± 0.28 IU/kg at study entry, with no significant change at 12 or 24 months (0.61 ± 0.28 and 0.70 ± 0.24 IU/kg, respectively).613 In the second IIT study, average daily insulin use at 6 months was reported to have increased by 23 percent from that at baseline (p<0.05) but the dose was not specified.614 No patients became insulin independent in either study.

Adverse Events

No treatment-related mortality was reported in the HSCT study.612 One post-conditioning case of bilateral pneumonia was reported that responded quickly to intravenous broad-spectrum antibiotics. With long-term followup, six cases of oligospermia were reported, and one case of leukopenia. The majority of adverse effects in the HSCT study were mild and included nausea, vomiting, fever, and alopecia.

No severe adverse effects were reported with IIT in either study.613, 614

Ongoing Research

According to the website ClinicalTrials.gov, five clinical studies are recruiting pediatric patients, as shown in Table 111. None of these originates in the U.S. Of the ongoing trials, only one offers a comparison between autologous mesenchymal stem cells and placebo (NCT01157403).

Table 111. Ongoing clinical trials of HSCT in DM1.

Table 111

Ongoing clinical trials of HSCT in DM1.

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits or harms with single autologous HSCT in patients with newly diagnosed type I juvenile diabetes.

Moderate strength evidence suggests that an extended interval of insulin independence can be achieved with single autologous HSCT in patients with newly diagnosed type I juvenile diabetes.

Systemic Lupus Erythematosus

Background and Setting

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease that is associated with inflammation and eventual organ damage.607 It may involve any organ system, with a wide range of disease severity. The exact cause of SLE is unknown. Diagnosis of SLE is the same regardless of age at onset, and is based on a combination of laboratory and clinical criteria. SLE is likely if four of the 11 revised American College of Rheumatology (ACR) criteria are present in a patient simultaneously or over time.615

SLE is rare in childhood, with an estimated incidence of 10 to 20 per 100,000 children, with some variation depending on ethnicity. Juvenile-onset SLE (prior to age 18 years) accounts for 15 to 20 percent of cases,616 which in general have a more severe presentation, faster development of organ damage, and a higher disease burden over a lifetime. For all age groups, 5-year survival rates have improved with advances in management of organ damage and complications, from 59 to 93 percent in the 1980s to 94 to 100 percent by the late 1990s.617 Patients aged younger than 24 years have the highest rate of SLE-related all-cause mortality, about 8-fold greater than the average for all SLE cases.618

The clinical course of SLE is marked by the alternation of periods of active disease and quiescence. However, children and adolescents with SLE enter adult life with considerable morbidity, secondary to sequelae of disease activity, side effects of medications, and comorbid conditions. The most common symptoms of SLE include fever, rash, fatigue, weight loss, arthritis, and renal disease.619 Lupus nephritis is one of the main clinical presentations of pediatric SLE, and it determines the course of illness as the major threat to long-term survival. Other major manifestations include neuropsychiatric, cardiac, and lung.

SLE has no known cure. Depending on severity, it is often treated with high-dose corticosteroids and immune suppressants, which are responsible for much of the permanent organ damage observed in these patients. Other treatments include hydroxychloroquine, cyclophosphamide, cyclosporine A, mycophenolate mofetil, azathioprine, nonsteroidal anti-inflammatory drugs (NSAIDs), rituximab, and abatacept.607 Only three agents have received U.S. Food and Drug Administration marketing approval for SLE: corticosteroids, hydroxychloroquine, and aspirin.

Autologous HSCT has been used to treat a small number of pediatric SLE cases, all of which have been severe, life-threatening, and refractory to nearly all drug therapies, with a dismal prognosis. Accordingly, this systematic review will present only results from HSCT reports, with the comparison being usual care.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic lupus erythematosus is shown in Table 112.

Table 112. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic lupus erythematosus.

Table 112

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic lupus erythematosus.

Overall, 12 of 17 (71 percent) SLE patients treated with autologous HSCT entered a state of complete drug-free remission, for periods that ranged from about 4 months620 to 66 months.621 The former reflects the followup time at preparation of the paper. In the largest series (n=9), patients experienced complete drug-free remission for a median 24 months, and a range of 12 to 66 months.621

Three studies reported SLE Disease Activity Index (DAI) score changes pre- and post-HSCT.621-624 Patients who underwent autologous HSCT and experienced a complete drug-free remission had substantial reduction in their SLEDAI scores. In one study, two of four patients succumbed to treatment-related mortality, one at 63 days from multiorgan failure, the other on day 15 due to multiple causes.623

Results

A total of seven reports were included in this review. Table 113 shows the criteria that were used to select studies for this section.

Table 113. Study selection criteria: SLE.

Table 113

Study selection criteria: SLE.

Table 114 shows the characteristics of studies of autologous HSCT in 17 patients (16 female) aged 13-21 years with SLE. All had severe, life-threatening SLE that was refractory to most first- and second-line drugs, variously including corticosteroids, pulsed cyclophosphamide, 6-mercaptopurine, azathioprine, plasmapheresis, and hydroxychloroquine. Although rituximab and abatacept have been studied in adults with SLE, we did not identify any studies of those agents in the pediatric setting. Outcomes reported are included in Table 115.

Table 114. Systemic lupus erythematosus study characteristics and population.

Table 114

Systemic lupus erythematosus study characteristics and population.

Table 115. Outcomes reported: SLE.

Table 115

Outcomes reported: SLE.

Five studies used only peripheral blood stem cells.620-622, 625, 626 One used bone marrow cells;624 and, one used both sources.623 Conditioning regimens typically included cyclophosphamide plus ATG; two studies included total-body irradiation,624, 626 and one used modified BEAM regimens with ATG.623

Complete Drug-Free Remission

Table 116 shows the proportions of patients with severe, refractory SLE who entered a state of complete drug-free remission following intense immune suppression and autologous HSCT. Overall, 12 of 17 (71 percent) entered a state of complete drug-free remission, for periods that ranged from about 4 months620 to 66 months.621 The former reflects the followup time at preparation of the paper. In the largest series, patients experienced complete drug-free remission for a median 24 months, and a range of 12 to 66 months.621

Table 116. Complete drug-free remission in patients with SLE undergoing autologous HSCT.

Table 116

Complete drug-free remission in patients with SLE undergoing autologous HSCT.

Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) Score

Three studies reported SLEDAI score changes pre- and post-HSCT,622-624 as shown in Table 117. In studies that reported this outcome, patients who underwent autologous HSCT and experienced a complete drug-free remission had substantial reduction in their SLEDAI scores.

Table 117. Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) Score.

Table 117

Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) Score.

Mortality and Other Serious Adverse Events Associated With Autologous HSCT

As shown in Table 115, six studies reported information on treatment-related mortality among autologous HSCT recipients.620-624, 626 In one study, 2 of 4 patients succumbed to treatment-related mortality, one at 63 days from multiorgan failure, the other on day 15 due to multiple causes.623 Other than those mentioned, all other adverse effects of the autologous HSCT conditioning regimens were reported by the authors to be mild to moderate and without clinical sequelae.

Ongoing Research

According to the Web site ClinicalTrials.gov, two clinical studies are recruiting pediatric patients, as shown in Table 118.

Table 118. Ongoing clinical trials of HSCT in SLE.

Table 118

Ongoing clinical trials of HSCT in SLE.

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits or harms with single autologous HSCT for the treatment of severe, refractory SLE in children.

Moderate strength evidence suggests that an extended drug-free clinical remission can be achieved with single autologous HSCT for the treatment of severe, refractory SLE in children.

Juvenile Idiopathic Arthritis (JIA)

Background and Setting

Juvenile idiopathic arthritis (JIA) is the most common chronic rheumatic condition in children, with a prevalence of between 16 and 150 per 100,000.607 It is defined as persistent arthritis in one or more joints in a child or adolescent less than 16 years old, after excluding other causes. According to the current International League of Association of Rheumatologists (ILAR) the term JIA refers to seven different disease categories: systemic, persistent oligoarticular, extended oligoarticular, polyarticular rheumatoid factor (RF) negative, polyarticular RF positive, enthesitis-related arthritis, and psoriatic.627

While the cause of JIA is not defined, evidence suggests altered immune system function, particularly T-cell regulation, has a major role in the pathogenesis of joint damage and disease progression.628, 629 JIA subtypes vary by the number of joints involved, and by age of onset. The most common form is the early onset (before age 6 years) oligo- or mono-articular JIA, with 1 to 4 asymmetrical joints affected, a high frequency of positive antinuclear antibodies (ANA), and high risk (30 percent) of chronic uveitis. These forms have generally good prognosis, and may be managed with intra-articular steroids and physiotherapy.627

At the other end of the spectrum, systemic-onset JIA is distinct from other JIA subtypes, with a systemic inflammatory component. While about 50 percent of cases will have a waxing and waning course, with favorable long-term prognosis, the other 50 percent will have an unremitting course with polyarthritis; prolongation of active systemic illness past 6 months is a particularly bad prognostic sign.607 Despite current treatment that includes methotrexate, corticosteroids, biological response modifiers (blocking agents of TNF-alpha, IL-1, IL-6) or blockers of costimulatory immune cell-surface molecules (e.g., CD28 or CD20),and other immune suppressants, most children with systemic polyarticular JIA do not achieve long-term clinical remission.607, 627 More than one-third will have ongoing active disease into adulthood, with sequelae secondary to chronic inflammation. Those who do not respond to early use of antirheumatic agents will experience considerable morbidity from joint damage, osteoporosis, growth retardation, psychosocial morbidity, reduced quality of life and educational and employment disadvantage.627

Autologous HSCT has been used to treat a small number of pediatric JIA cases, all of which have been severe, progressive and refractory to standard drug therapies. Accordingly, this systematic review will present only results from HSCT reports, compared to usual care and the disease course.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory juvenile idiopathic arthritis is shown in Table 119.

Table 119. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory juvenile idiopathic arthritis.

Table 119

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory juvenile idiopathic arthritis.

Overall the evidence signals that autologous HSCT following chemotherapy-induced immune suppression may be associated with prolonged resolution of JIA into a drug-free, much-improved state. Among all cases reported, 21 of 43 (56 percent) achieved extended drug-free remission for 3 to 60 months. In the largest series, drug-free remission was reported in 53 percent, with a median duration of 29 months. There were four cases of treatment-related mortality, with no other reports of long-term benefits and harms.

Results

A total of four reports were included in this review.630-633 Table 120 shows the criteria that were used to select studies for this section.

Table 120. Study selection criteria: JIA.

Table 120

Study selection criteria: JIA.

Table 121 presents the study characteristics and populations of 4 reports on the use of autologous HSCT treatment in patients with JIA. All patients had severe, refractory illness, mostly systemic (79 percent) but also polyarticular (21 percent). The largest experience is from the registry of the EBMT, which reports on 34 of 41 total patients who had received autologous HSCT at nine European centers.630 Four cases were treated in Japan631, 632 and five were from Italy.633 Patients ranged in age from 3 to 21 years, with 49 percent females and 51 percent males.

Table 121. Juvenile idiopathic arthritis study characteristics and population.

Table 121

Juvenile idiopathic arthritis study characteristics and population.

Among 43 cases, 26 (60 percent) used bone marrow stem cells, 12 (28 percent) used peripheral-blood stem cells, and the stem-cell source was not identified for five.633 Various conditioning regimens were reported, typically involving cyclophosphamide and ATG.

Table 122 shows outcomes that were reported in the four studies in this systematic review. Complete drug-free remission is the key outcome, which is generally defined as disappearance of signs and symptoms of JIA and cessation of antirheumatic agents.

Table 122. Outcomes reported: JIA.

Table 122

Outcomes reported: JIA.

Complete Drug-free Remission

Overall, 56 percent of patients in these reports achieved a complete drug-free remission following autologous HSCT (Table 123). In the largest study, a complete drug-free response was achieved in 53 percent of cases, for an average duration of about 2.5 years, although some patients maintained this response for 60 months.630 In the Rabusin series, four of five (80 percent) patients achieved a complete drug-free response at 3 months' followup, with a relapse in one at 6 months, and ultimately relapse in the other three at 8, 12, and 18 months.633 One patient in the Kishimoto study had disease flares at 11 and 23 months followup, but became medication-free and asymptomatic at 39 months followup.631

Table 123. Complete drug-free remission in patients with JIA undergoing autologous HSCT.

Table 123

Complete drug-free remission in patients with JIA undergoing autologous HSCT.

Overall Survival

One study reported Kaplan-Meier overall survival of about 79 percent at 5 years' followup.630

Treatment-related Mortality and Other Adverse Events

Overall, treatment-related mortality was reported in 4 of 43 (9 percent) compiled cases in this review. In the EBMT experience, treatment-related mortality was reported in 3 of 34 cases (9 percent), attributed to macrophage-activation syndrome.630 In another study, one patient who had an uncontrollable disease flare after autologous HSCT underwent a subsequent cord-blood allogeneic HSCT, but developed a CMV infection and died 48 days after allogeneic HSCT.631 No treatment-related mortality was reported in the other two studies.632, 633

Ongoing Research

According to ClinicalTrials.gov, there are no clinical trials of HSCT actively recruiting patients with JIA.

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits or harms with single autologous HSCT for the treatment of severe, refractory JIA in children.

Moderate strength evidence suggests that an extended drug-free clinical remission can be achieved with single autologous HSCT for the treatment of severe, refractory JIA in children.

Systemic Sclerosis

Background and Setting

Systemic sclerosis is a generalized, highly heterogeneous autoimmune disorder characterized by diffuse, disabling skin thickening combined with fibrotic changes in many organs, in particular the heart and lung, ultimately resulting in end-stage failure.634

Systemic sclerosis is a rare disease. It is diagnosed in approximately 67 male patients and 265 female patients per 100,000 people each year in the U.S. Systemic sclerosis usually appears in women aged 30 to 40 years, and it occurs in slightly older men. In approximately 85 percent of cases, systemic sclerosis develops in individuals aged 20 to 60 years. Cases also are observed in children and in the elderly population. Immunologic mechanisms and heredity (certain HLA subtypes) play a role in etiology. Systemic sclerosis-like syndromes can result from exposure to vinyl chloride, bleomycin, pentazocine, epoxy and aromatic hydrocarbons, contaminated rapeseed oil, or L-tryptophan.

Systemic sclerosis pathophysiology involves vascular damage and activation of fibroblasts; collagen and other extracellular proteins in various tissues are overproduced. The disease varies in severity and progression, ranging from generalized skin thickening with rapidly progressive and often fatal visceral involvement, primarily end-stage organ failure.635-637 The course depends on the type of systemic sclerosis but is often unpredictable. Typically, progression is slow. Overall, 5- and 10-year survival is about 20 to 80 percent and 15 to 65 percent, respectively, according to the major organ affected at diagnosis.638-640 Patients with diffuse skin disease eventually develop visceral complications, which are the usual causes of death. Prognosis is poor if cardiac, pulmonary, or renal manifestations are present early. Heart failure may be intractable. Acute renal insufficiency, if untreated, progresses rapidly and causes death within months.

Results from Phase II open studies suggest intravenous pulse cyclophosphamide therapy may improve skin score and pulmonary function.641-643 However, no treatment has been shown definitively to halt disease progression. Autologous HSCT has been used to treat a small number of pediatric systemic sclerosis cases, all of which have been severe, progressive and refractory. Accordingly, this systematic review will present only results from HSCT reports, compared to usual care and the disease course.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic sclerosis is shown in Table 124.

Table 124. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic sclerosis.

Table 124

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory systemic sclerosis.

Systemic sclerosis is very rare in children, with one registry report comprising the sole source of published evidence on the use of autologous HSCT in this setting. There are no proven therapies for advanced, progressive systemic sclerosis with visceral involvement, which has a dismal prognosis. In this context, that four of five patients (80 percent) entered a state of complete clinical remission, with the other one in partial remission, signals that autologous HSCT following chemotherapy-induced immune suppression may be associated with these results.

Results

One report was included in this review. Table 125 shows the criteria that were used to select studies for this section.

Table 125. Study selection criteria: Systemic sclerosis.

Table 125

Study selection criteria: Systemic sclerosis.

Table 126 shows the study and patient characteristics of one report on the use of autologous HSCT in pediatric systemic sclerosis patients.644 This registry report provides details on 5 children age 9 to 17 years, all of whom had lung disease at inclusion and systemic sclerosis (4 diffuse, 1 limited). Disease had been diagnosed within 62 months (range: 26-85 months) before HSCT. They all received the same mobilization regimen comprising cyclophosphamide and G-CSF plus cell selection before transplantation with either CD34+ selection alone (n=3) or CD34+/4+/8+ (n=2). Different conditioning regimens were used, comprising cyclophosphamide plus antiCD52 (CAMPATH 1) (n=3), cyclophosphamide plus TBI plus ATG (n=1), or cyclophosphamide alone (n=1).

Table 126. Systemic sclerosis study characteristics and population.

Table 126

Systemic sclerosis study characteristics and population.

Only scant details on pediatric patients are available in the registry report. All five children were reported alive after a median duration of about 38 months (range: 14-68 months). Four (80 percent) entered complete remission, with one partial remission. Disease ultimately progressed in one patient, and one relapsed about 9 months after experiencing a complete remission. One 19-year-old patient, not included in the group of five reported above, succumbed to diffuse alveolar hemorrhage 11 days post-transplant.

Ongoing Research

According to ClinicalTrials.gov, one Phase I/II study in China is actively recruiting individuals with systemic sclerosis to undergo HSCT with allogeneic mesenchymal stem cells (NCT00962923). A pilot study of total body irradiation in combination with cyclophosphamide, antithymocyte globulin, and autologous CD34-selected peripheral blood stem cell transplantation in children with refractory autoimmune disorders is active but not recruiting (NCT00010335).

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits or harms with single autologous HSCT for the treatment of severe, refractory systemic sclerosis in children.

Moderate strength evidence suggests that an extended drug-free clinical remission can be achieved with single autologous HSCT for the treatment of severe, refractory systemic sclerosis in children.

Multiple Sclerosis

Background and Setting

Multiple sclerosis (MS) is a CNS demyelinating disease with an autoimmune etiology.645 It is characterized by the presence of inflammatory, demyelinating lesions scattered throughout the CNS at different sites and at different times. Clinical features secondary to the CNS lesions include loss of sensation, muscle weakness, visual loss, incoordination, cognitive impairment, fatigue, pain, and bladder and bowel disturbance.645

Pediatric MS is diagnosed after two clinical episodes of CNS demyelination that are separated by at least 30 days.646 In adults, three of the following four features are required: nine or more white matter lesions or one gadolinium-enhancing lesion; three or more periventricular lesions; one juxtacortical lesion; one infratentorial lesion. These criteria may be applied to identify pediatric cases, but have not been clinically validated in this population.647

The worldwide prevalence of pediatric MS is unknown. Data from individual countries or MS centers reported prevalence rates of MS in childhood ranging from 3.1 to 4.4 percent of all MS cases.648-650 A Canadian study estimated the incidence of initial pediatric demyelinating events (including MS, neuromyelitis optica, acute disseminated encephalomyelitis, complete transverse myelitis, and recurrent optic neuritis) as 0.9 per 100,000 individuals.651

The natural history of MS is extremely variable, with a waxing and waning pattern that may gradually worsen over time.645 Four broad categories are recognized: relapsing remitting MS (RRMS), which accounts for about 85 percent of cases; secondary progressive MS (SPMS), which represents a progression of RRMS with accumulating irreversible neurological deficit and disability; primary progressive MS (PPMS), which is characterized by progressive disease from onset, and accounts for 10 to 15 percent of MS cases; and, progressive relapsing MS (PRMS), defined as progressive disease from onset with superimposed relapses.

Malignant MS is a poorly defined subset of MS that comprises a heterogeneous group of demyelinating disorders that is applied only to cases that succumb within 5 years of onset, accounting for less than 5 percent of all MS subjects.652

The therapeutic approach to MS has evolved over the past two decades. Four first-line disease modifying therapies have received U.S. Food and Drug Administration approval for use in adults with RRMS: glatiramer acetate, intramuscular and subcutaneous interferon-β1a, and subcutaneous interferon-β1b.647 Evidence supporting their use in children is very limited.

Most treatment decisions in children with MS are based in part on results achieved in adults. Corticosteroids are used to treat acute, symptomatic relapses, but are associated with serious adverse effects in children.645 Second-line agents in children have included cyclophosphamide, mitoxantrone, mycophenolate mofetil, daclizumab, rituximab or natalizumab, primarily described in retrospective case series and reports with limited follow-up.647

There is no consensus on how to treat malignant MS. Approaches have included plasmapheresis, aggressive immunosuppression with mitoxantrone, cladribine, and cyclophosphamide, with no documented effect in this setting. Given the extremely poor prognosis of malignant MS, and the lack of effective treatment, autologous HSCT has been used to treat a small number of pediatric malignant MS cases. Accordingly, this systematic review will present only results from HSCT reports compared to usual care and the disease course.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory malignant multiple sclerosis is shown in Table 127.

Table 127. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory malignant multiple sclerosis.

Table 127

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory malignant multiple sclerosis.

Malignant MS is very rare in children, with three reports comprising the sole identified source of published evidence on the use of autologous HSCT in this setting. There are no proven therapies for malignant MS, which has a dismal prognosis. In this context, that five of five patients (100 percent) entered a state of clinical remission, with no relapses at followup, signals that autologous HSCT following chemotherapy-induced immune suppression may be associated with these results.

Results

A total of three reports were included in this review. Table 128 shows the criteria that were used to select studies for this section.

Table 128. Study selection criteria: MS.

Table 128

Study selection criteria: MS.

Table 129 shows the study and patient characteristics of three reports (total N=5) on the use of autologous HSCT in pediatric patients.653-655 They ranged in age from 9 to 18 years, with two females and three males.

Table 129. Multiple sclerosis study characteristics and population.

Table 129

Multiple sclerosis study characteristics and population.

In the first report, peripheral blood stem cells were mobilized using cyclophosphamide and G-CSF.653 One patient underwent conditioning using a BEAM regimen (BCNU, etoposide, cytosine-arabinoside, melphalan), the other (a young male) was conditioned using cyclophosphamide. Total body irradiation was not used in either case. ATG and methylprednisolone were administered 1 and 2 days after stem-cell infusion. In the second report, peripheral blood stem cells were mobilized using cyclophosphamide and G-CSF, and the patient was conditioned with busulfan and ATG.654 In the third report, peripheral blood stem cells were mobilized using cyclophosphamide and G-CSF, and conditioned using a BEAM regimen plus cyclophosphamide plus ATG.655

In the Fagius study, both patients experienced stabilization of their disease course.653 One was alive at 35 months followup, with expanded disability status scale (EDSS) score reduced from 4.0 at HSCT to 0 at last followup. The second patient in the Fagius study was alive at 28 months, with EDSS score reduced from 8.0 at HSCT to 1.0 at followup. Adverse effects of autologous HSCT were not reported individually, but were as expected with the conditioning regimen, comprising fever, mucositis, and alopecia.

Kimiskidis and colleagues reported their patient showed rapid neurological improvement that was sustained and gradually increased following HSCT.654 His EDSS score dropped from 5.0 at HSCT to 1.5 at 66 months' followup. The patient experienced no relapses since HSCT, with no immunomodulatory therapies, and at publication had graduated from college.

Mancardi and colleagues reported results for two pediatric cases.655 The first patient was alive at 29 months' followup, with EDSS score reduced from 7.5 to 4.0, the ability to walk 500 meters unaided, independent in activities of daily living, and no relapses. HSCT-related adverse effects were not reported. The second patient was alive at 14 months' followup, with dramatically improved neurological condition and EDSS score reduced from 9.0 to 4.5; information was not provided about the patient's mobility or capability to perform activities of daily living. This patient experienced fever for two weeks post-HSCT, but no pathogen was identified.

In total, five pediatric patients with malignant MS have been reported alive following intense immune suppression and autologous HSCT. All had durable remission of severe, disabling disease and life-threatening disease for 14 to 66 months followup, improved EDSS scores, and improvement in neurological function. Where reported, mobility improved, along with the ability to perform activities of daily living. No patient relapsed in the followup periods reported.

Ongoing Research

According to ClinicalTrials.gov, there are no active clinical studies involving HSCT recruiting patients with any type of multiple sclerosis.

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits or harms with single autologous HSCT for the treatment of severe, refractory MS in children.

Moderate strength evidence suggests that an extended drug-free clinical remission can be achieved with single autologous HSCT for the treatment of severe, refractory MS in children.

Crohn's Disease

Background and Setting

Crohn's disease is an idiopathic, chronic inflammatory disease of the gastrointestinal tract that primarily affects the small intestine and colon.

There is wide discrepancy in the prevalence and incidence estimates of Crohn's disease in North America.656 Prevalence has been estimated between about 44 to nearly 200 per 100,000 persons, perhaps representing the effect of environmental or genetic factors in its development. Similarly, incidence estimates vary considerably, from about 3.1 to about 5 cases per 100,000 person-years; the incidence in children is estimated at about 5 per 100,000. With about 300 million people in the United States, approximately 9,000 to 44,000 cases are diagnosed with Crohn's disease annually.

The natural history of Crohn's disease is characterized by recurring flares with periods of inactive disease and remission.657 The waxing and waning nature dictates that patients require medication for a large period of their life, primarily to maintain remission but also to control flares. About 50 percent of cases will remain in a state of remission or mild intermittent disease, but about 5 percent will have severe, drug-refractory disease.656 Surgery is required in up to 80 percent of cases at some point.657

While Crohn's disease-related mortality is relatively low, the range and severity of symptoms varies from mild to disabling. The most common symptoms of Crohn's disease are abdominal pain, often in the lower right area, and diarrhea. Rectal bleeding, weight loss, arthritis, skin problems, and fever may also occur. Bleeding may be serious and persistent, leading to anemia. Children with Crohn's disease may suffer delayed development and stunted growth. The most common complication is blockage of the intestine. Nutritional complications are common in Crohn's disease, with deficiencies of proteins, calories, and vitamins. These deficiencies may be caused by inadequate dietary intake, intestinal loss of protein, or poor absorption, also referred to as malabsorption. Other complications associated with Crohn's disease include arthritis, skin problems, inflammation in the eyes or mouth, kidney stones, gallstones, or other diseases of the liver and biliary system. Some of these problems resolve during treatment for disease in the digestive system, but some must be treated separately.

Current therapy for Crohn's disease consists of corticosteroids, immunomodulators and biological therapy blocking TNF-alpha (e.g., infliximab).657 Corticosteroids efficiently suppress inflammation, but have not been shown definitively to alter the natural course of Crohn's disease. Immunomodulators and biologicals such as azathioprine, 6-mercaptopurine, methotrexate, etanercept, and infliximab can induce and maintain remission, but their overall effect on the long-term course of Crohn's disease and the ultimate need for surgery are not definitively established.657 Their postoperative role also is not defined. In general, the optimal timing of therapies relative to disease course is not clear.

Autologous HSCT has been used to treat a small number of pediatric Crohn's disease cases, all of which have been severe, progressive, disabling, and refractory to nearly all drug therapies. Accordingly, this systematic review will present only results from HSCT reports, with the comparison considered usual care and the disease course.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Crohn's Disease is shown in Table 130.

Table 130. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Crohn's disease.

Table 130

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Crohn's disease.

The evidence signals that autologous HSCT following chemotherapy-induced immune suppression may be associated with prolonged resolution of severe, refractory, disabling Crohn's disease into a drug-free, much-improved state, 3 to 6 months post-HSCT.

Results

Two reports were included in this review. Table 131 shows the criteria that were used to select studies for this section.

Table 131. Crohn's disease study selection criteria.

Table 131

Crohn's disease study selection criteria.

One pilot study658 and one long-term follow-up study659 (Table 132) reported results on the use of autologous HSCT utilizing a conditioning regimen of high-dose cyclophosphamide plus equine or rabbit ATG and T-cell depleted CD-34+ enriched peripheral blood stem cells mobilized with cyclophosphamide and G-CSF. Mesna, methylprednisolone, and G-CSF were started in conjunction with the conditioning regimen.

Table 132. Crohn's disease study characteristics and population.

Table 132

Crohn's disease study characteristics and population.

Patients were 5 males, ages 15-21 years, and 2 females, ages 18 and 21 years. Pretransplant CDAI scores averaged 288 ± 37 (range: 101-337), with mean Karnofsky performance score (KPS) of 48 ± 10 (range: 40-60). All were highly symptomatic, completely disabled, with a clinical history and histologic evidence of Crohn's disease, and had failed treatment with corticosteroids, mesalamine, metronidazole, azathioprine, 6-mercaptopurine, and infliximab. Failure was defined as a Crohn's Disease Activity Index (CDAI) of 250-400 despite those therapies. All immunosuppressive and disease-modifying agents were discontinued at stem-cell mobilization, except systemic corticosteroids, which were tapered over 2 to 6 months. The key outcome was clinical remission, defined as CDAI≤150, and freedom from immune suppressant therapy following taper and discontinuation of corticosteroids post-HSCT. Adverse effects of autologous HSCT were also reported.

In the Oyama pilot study, all patients were alive at mean followup of 24 ± 15 months (range: 7-36 months).658 They had a rapid and dramatic post-HSCT improvement, discontinued all immunosuppressive therapies, regained normal appetite and oral intake, with cessation of diarrhea and abdominal pain. The mean CDAI score improved by 77 percent, declining from 288 ± 37 (range: 250-337), to 66 ± 13 (range 51-78). The mean Karnofsky Performance Status improved by 92 percent, from 48 ± 10 (range: 40-60) to 92 ± 10 (range: 80-100). Adverse HSCT-related events were not individually documented, but the procedure was reported as well tolerated. One patient (not identified) developed Mallory-Weiss syndrome that responded to intravenous fluids. One patient (unidentified) relapsed at 15 months after achieving remission at 6 months. Among the total patient population (pediatric cases not reported separately), after a median follow-up of 18 months (range 7-37 months) 11 of 12 remained in drug-free clinical remission.

In the second report,659 all 3 patients not previously reported in the Oyama paper658 were alive at 60 months follow-up, in an immune suppressant drug-free state but still with active Crohn's disease. All had undergone colectomy with or without ileostomy at 18 to 44 months following HSCT.

Ongoing Research

According to ClinicalTrials.gov, two Phase I clinical trials in the U.S. are recruiting pediatric patients with severe Crohn's disease (CDAI>250) for autologous HSCT (NCT00692939, NCT00278577).

Conclusion

The overall body of evidence is insufficient to draw conclusions on long-term benefits and harms with single autologous HSCT in children with severe, refractory, disabling CD.

Moderate strength evidence suggests that an extended clinical remission, free from immune suppressant therapy following taper and discontinuation of corticosteroids 3-6 months post-HSCT, can be achieved with single autologous HSCT in children with severe, refractory, disabling CD.

Miscellaneous Nonhematologic Autoimmune Diseases

Background and Setting
Myasthenia Gravis

Myasthenia gravis is an autoimmune disease characterized by failure of neuromuscular transmission secondary to destruction of acetylcholine receptors at the neuromuscular junction synapse by anti-acetylcholine antibodies.660 The estimated incidence of Myasthenia gravis is about 1 per 30,000.661 It typically presents in adulthood, but has been diagnosed in children as young as one year of age. Myasthenia gravis affects women more than men (67 percent of cases), with a peak onset in the 20s. Spontaneous remissions occur in about 25 percent of patients, but rarely last more than two years and do not typically recur.

Myasthenia gravis is controlled in most cases by the use of acetylcholinesterase inhibitors, but more severe, progressive disease is treated with immunomodulating approaches, including IVIG, corticosteroids, azathioprine, thymectomy, and plasmapheresis.661 High-dose cyclophosphamide has been used to treat severe MG, with good initial response in 90 percent of cases, although 80 percent have recurrence and require continual immunosuppression by 5 years following treatment.662 High-dose chemotherapy with allogeneic HSCT has been reported in one case of severe, refractory disease.

Calcinosis Cutis

Calcinosis cutis is a term used to describe a group of disorders in which calcium deposits form in the skin, first described by Virchow in 1855. It occurs in four major types according to etiology: dystrophic, metastatic, iatrogenic, and idiopathic. It may be associated with autoimmune diseases such as dermatomyositis, systemic lupus, systemic sclerosis, and others.663, 664 The incidence and prevalence of calcinosis cutis in the U.S. is unknown.

Damage caused by calcium deposits may be localized or systemic. Lesions may become painful, limit mobility of an adjacent joint, or compress adjacent neural structures. Ulceration and secondary infection may occur. Vascular calcification may cause ischemia and necrosis of the affected organ. Medical therapy is limited and of unproven benefit. Intralesional corticosteroids, probenecid, colchicine, magnesium or aluminum antacids, sodium etidronate and diphophosphonates, myoinositol hexaphosphonate, warfarin, diltiazem, sodium thiosulfate may be effective. Pediatric use of most of these agents is unapproved.

High-dose immunoablation with allogeneic HSCT has been reported in one pediatric patient with diffuse, severe refractory calcinosis cutis.

Overlap Syndrome

An overlap syndrome is an autoimmune disease of connective tissue in which the patient presents with symptoms of two or more diseases. As many as 25 percent of all patients with connective tissue disease show signs of an overlap syndrome. Examples of overlap syndromes include mixed connective tissue disease and scleromyositis, but the exact diagnosis depends from which diseases the patient shows symptoms. In overlap syndromes, features of systemic lupus, systemic sclerosis, polymyositis, dermatomyositis, rheumatoid arthritis and Sjögren's syndrome are found often.665

The prevalence of mixed connective tissue disease is not known precisely, falling somewhere between that of systemic sclerosis and polymyositis and systemic lupus. It is found more often in females than males (8:1 ratio), and occurs in children. Morbidity is greater in children than adults, with higher prevalence of myocarditis, glomerulonephritis, thrombocytopenia, seizures, and aseptic meningitis.665

Mixed connective tissue disease is viewed as incurable, with variable prognosis. The presentation ranges from mild self-limited disease, to major organ involvement that requires aggressive treatment. No controlled clinical trials have been performed to evaluate therapy in mixed connective tissue disease. Treatment strategies generally involve conventional therapies that are used for other autoimmune diseases such as systemic lupus, systemic sclerosis, and polymyositis. Given the heterogeneous clinical course of mixed connective tissue disease, therapy is individualized according to specific organ involvement and the severity of underlying disease activity. Agents include corticosteroids, antimalarials, methotrexate, cytotoxics (most often cyclophosphamide), and vasodilators, with varying degrees of success.666

Nonmyeloablative allogeneic HSCT has been reported in one severe, refractory pediatric case.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory myasthenia gravis, overlap syndrome, or diffuse calcinosis cutis is shown in Table 133.

Table 133. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory myasthenia gravis, overlap syndrome, or diffuse calcinosis cutis.

Table 133

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory myasthenia gravis, overlap syndrome, or diffuse calcinosis cutis.

One case report each showed prolonged resolution of myasthenia gravis or overlap syndrome into a drug-free, much-improved state following chemotherapy-induced immunosuppression with allogeneic HSCT. Similarly, immunosuppression and autologous HSCT was followed by complete resolution of disabling diffuse calcinosis cutis in one patient.

Results

A total of three reports for these miscellaneous diseases were included in this review. Table 134 shows the criteria that were used to select studies for this section.

Table 134. Study selection criteria: MG, OS, CC.

Table 134

Study selection criteria: MG, OS, CC.

As shown in Table 135, three case reports are available, one each on the use of allogeneic HSCT to treat myasthenia gravis667 or overlap syndrome,668 and one on autologous HSCT to treat calcinosis cutis.669

Table 135. Miscellaneous nonhematologic autoimmune disease study characteristics and population.

Table 135

Miscellaneous nonhematologic autoimmune disease study characteristics and population.

The myasthenia gravis case report outlined the outcome of reduced-intensity, matched-sibling, peripheral blood allogeneic HSCT using busulfan, fludarabine, and alemtuzumab.667 The patient was severely affected 17-year-old male who had failed prior treatment with pyridostigmine, IVIG, corticosteroids, thymectomy, azathioprine, mycophenolate mofetil, plasmapheresis, rituximab, and high-dose cyclophosphamide.

The patient with overlap syndrome was a 15-year-old female with pulmonary vasculitis, severe Cushing's syndrome, stunted growth, profound adrenal steroid dependency, and iatrogenic liver toxicity secondary to failed treatment with methotrexate and cyclophosphamide.668 She underwent reduced-intensity allogeneic HSCT using fludarabine, cyclophosphamide, and total body irradiation, followed by infusion of HLA-matched bone marrow stem cells.

The third case report involved a 16-year-old female who had diffuse, severely disabling calcinosis cutis with arthritis, myalgia, anemia, recurrent pulmonary hemorrhage, CNS abnormalities, and painful skin ulcers.669 Her condition did not adequately respond to corticosteroids, cyclophosphamide, azathioprine, methotrexate, hydroxychloroquine, and thalidomide. She developed pulmonary hypertension and ischemic digital necrosis, at which time she was referred for high-dose immunosuppression and autologous HSCT. Peripheral blood stem cells were mobilized using cyclophosphamide and G-CSF, and reinfused following a conditioning regimen comprising BCNU, etoposide, cytarabine, and melphalan.

The patient with myasthenia gravis achieved T- and B-cell immune reconstitution within 7 months post-HSCT, and during the next 12 months was weaned off pyridostigmine, developed normal muscle strength and lost 60 pounds of weight.667 He experienced mucositis that required total parenteral nutrition and patient-controlled analgesia for 11 days, 1 episode of gram-positive bacteremia that resolved with vancomycin, and CMV reactivation that resolved with ganciclovir. His oropharyngeal muscles and speech normalized, although he still had ophthalmoplegia. At 40 months post-HSCT, despite the presence of elevated acetylcholine receptor antibody levels, he was free of all myasthenia gravis treatments, was able to play basketball, and was reported as completely independent.

The patient with overlap syndrome achieved greater than 90 percent donor chimerism at 12 months post-HSCT.668 She was weaned off methylprednisolone, IVIG, and asthma medications over the next year, her cushingoid features resolved, she grew approximately 7 inches over 3 years' followup and became a full-time student in a regular classroom. She had no evidence of clinical graft-versus-host disease or systemic infection over 36 months' followup, but continued to have occasional periods of fatigue and mild Gottron-like rash, which was reported to decrease in frequency at followup.

The patient with calcinosis cutis engrafted promptly, with no significant HSCT-related complications reported.669 She regained mobility and ability to perform unaided activities of daily living, such as sitting, standing, and walking. At 6 weeks post-HSCT, the subcutaneous calcinosis nodules began to liquefy and calcium salts extruded through her skin. Deep calcinosis plaques disappeared, all skin ulcers healed completely, and her pulmonary blood pressure normalized. At 24 months' followup, the patient was free from clinical and laboratory evidence of disease activity.

Ongoing Research

According to ClinicalTrials.gov, one Phase I study is recruiting patients with severe, refractory myasthenia gravis for autologous HSCT (NCT00424489). A Phase I study is recruiting patients with severe, refractory systemic vasculitis and overlap syndrome for autologous HSCT (NCT00278512). No studies are recruiting for HSCT in patients with calcinosis cutis.

Conclusion

The overall body of evidence is insufficient to draw conclusions on benefits and harms with allogeneic HSCT to treat severe, refractory MG or OS, and autologous HSCT for the treatment of severe, refractory CC is insufficient to draw conclusions.

The overall body of evidence is insufficient to demonstrate that an extended drug-free remission can be achieved with allogeneic HSCT to treat severe, refractory MG or OS, and autologous HSCT for the treatment of severe, refractory CC.

Hematologic Autoimmune Diseases

Background and Setting
Evans Syndrome

Evans syndrome is an uncommon autoimmune disease characterized by simultaneous or sequential development of autoimmune thrombocytopenia and autoimmune hemolytic anemia, with some patients also being neutropenic.670-672 While the etiology is unknown, evidence suggests this disease is secondary to a more generalized immune dysregulation, with several clinical and laboratory features in common with systemic lupus and autoimmune lymphoproliferative syndrome.671, 673

The exact frequency of Evans syndrome is unknown. Familial occurrence is rare. It has a chronic, relapsing course, with substantial morbidity and mortality. In a 1997 survey of North American pediatric hematologists, the median reported age at diagnosis was about 8 years (range: 0.2–27 years).674 This late presentation age may indicate the disease was undiagnosed until the second presentation of cytopenia, which was usually months to years after the first presentation. Evans syndrome in adults has been anecdotally reported. No randomized trials have been conducted in patients with Evans syndrome, and the evidence for treatment is based on case reports, case series, and retrospective studies.675 Corticosteroids, IVIG, danazol, cyclosporine, azathioprine, cyclophosphamide, vincristine, rituximab, alemtuzumab, and splenectomy have been used, but response to therapy varies even within the same individual.

Autoimmune Hemolytic Anemia

Autoimmune hemolytic anemia occurs when an individual develops anti-self antibodies that destroy red blood cells. The incidence of autoimmune hemolytic anemia has been reported in the range of 1 per 50,000 to 75,000, rising with age, mostly as secondary rather than idiopathic disease.676 In children, the onset of autoimmune hemolytic anemia is more likely to be sudden and severe compared to that in adults. It has a relatively good prognosis in most cases, with good response to corticosteroids, and often not requiring splenectomy. It can develop into a refractory state that does not respond well to steroids, IVIG, azathioprine, cyclophosphamide, plasmapheresis, or splenectomy.

Autoimmune Thrombocytopenia

Chronic autoimmune thrombocytopenia is a disorder of diminished platelet count, secondary to the development of anti-self antibodies directed against platelet surface glycoproteins, resulting in splenic platelet destruction.677 Acute idiopathic thrombocytopenia purpura has an annual incidence in the U.S. of about 1.6 per 10,000, but it is estimated that chronic idiopathic thrombocytopenia purpura develops in 7 to 28 percent of children who have acute disease. Chronic, refractory autoimmune thrombocytopenia has been reported to have a mortality rate of 4 to 16 percent, largely attributed to bleeding or infection 678. It may respond to corticosteroids and IVIG, but can become refractory and nonresponsive to immunosuppressants that include cyclophosphamide, azathioprine, vinblastine, mycophenolate mofetil, and rituximab.

Given the poor response among a proportion of patients with severe Evans syndrome, autoimmune hemolytic anemia, and autoimmune thrombocytopenia to immunosuppressant therapies, with attendant serious adverse effects, HSCT has been investigated in a small number of children with severe, refractory Evans syndrome, autoimmune hemolytic anemia, and autoimmune thrombocytopenia.

Evidence Summary

The overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia is shown in Table 136.

Table 136. Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia.

Table 136

Overall grade of strength of evidence for drug-free clinical remission and the use of HSCT for the treatment of severe, refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia.

One case report each showed resolution of Evans syndrome or autoimmune hemolytic anemia into a drug-free, much-improved state following chemotherapy-induced immune suppression with allogeneic HSCT. Similarly, one case report showed resolution of severe, refractory autoimmune hemolytic anemia following autologous HSCT. The single case report of autologous HSCT for autoimmune thrombocytopenia showed no response to the procedure.

Results

A total of seven publications comprising eight studies (six case reports and two case series) on autoimmune hematologic conditions were included in this review. Table 137 shows the criteria that were used to select studies for this section.

Table 137. Study selection criteria: Refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia.

Table 137

Study selection criteria: Refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia.

Evans Syndrome

Four studies (three case reports and one case series) described results of allogeneic HSCT in eight patients with severe, refractory Evans syndrome (Table 138).

Table 138. Evans syndrome, autoimmune hemolytic anemia, and autoimmune thrombocytopenia study characteristics and population.

Table 138

Evans syndrome, autoimmune hemolytic anemia, and autoimmune thrombocytopenia study characteristics and population.

The case series described 5 cases of Evans syndrome reported to the European Group for Blood and Marrow Transplantation (EBMT) registry between 1984 and 2007679. After receiving unspecified standard therapy, patients (100% male, age range 2-21) were referred for allogeneic HSCT (3 bone marrow, 1 peripheral blood, and 1 cord blood) with various combination conditioning regimens including cyclophosphamide, fludarabine, busulfan, thiotepa, ATG, TBI and received immunosuppressive cyclosporine A with either methotrexate or mycophenolate mofetil therapy for GVHD. Three patients were alive at 36, 85 and 113 months following allogeneic HSCT; one was dead from disease at 59 months and one was dead from interstitial pneumonitis at 6 months following allogeneic HSCT.679 All five patients were reported to have aGVHD (grade 1 n=1, grade 2 n=1, grade 3 n=1, NR n=2) and cGVHD (extensive: n=1, limited: n=1, NR: n=3), but no other HSCT-related adverse events were reported.

In the Connor et al. report, the female subject presented at age 6 months with Evans syndrome.680 Over the next several years, she was treated with corticosteroids, IVIG, cyclosporine A, mycophenolate mofetil, rituximab alone and with other drugs, and underwent a splenectomy, without experiencing durable remission. Her condition worsened, she developed pulmonary hypertension with dilated right ventricle and tricuspid regurgitation. She was referred for an unrelated, single-antigen mismatched allogeneic HSCT, using a conditioning regimen of alemtuzumab, fludarabine, and melphalan, with cyclosporine A and mycophenolate mofetil as graft-versus-host disease prophylaxis. The patient developed full donor chimerism, was weaned off all immunosuppressants, developed normalized pulmonary pressures, and exhibited normal right ventricular size and function at 10 months following allogeneic HSCT.680 No HSCT-related adverse events were reported.

A second case report describes the results achieved with an unrelated cord blood HSCT in a 7-year old boy with severe, refractory Evans syndrome who failed a previous double autologous HSCT.681 His disease was poorly responsive to previous therapy, including corticosteroids, IVIG, cyclosporine A, mycophenolate mofetil, vincristine, vinblastine, cyclophosphamide, rituximab, and danazol. He became pancytopenic, with disfiguring hypercorticism and polyneuropathy, and underwent double high-dose chemotherapy with autologous HSCT, with temporary improvement. After suffering a massive intracranial bleed, he received a 7/10 HLA-matched female cord blood HSCT, with a conditioning regimen comprising busulfan, ATG, thiotepa and etoposide. The patient developed graft-versus-host disease with grade II skin, liver, and mucosa involvement that resolved after a short course of prednisone and cyclosporine A.681 At 1.5 years' followup, he was reported with 100 percent donor chimerism, normal blood counts, in good clinical condition, free of graft-versus-host disease and the need for immunosuppressant drugs.

The third case involved a nearly 5-year-old boy who presented with Evans syndrome at age 5 months.682 His disease responded poorly to courses of therapy with corticosteroids, IVIG, 6-mercaptopurine, danazol, cyclosporine, azathioprine, vincristine, and anti-D, with increasingly worse mucosal bleeding and intracranial hemorrhage. He was referred for a matched-sibling cord blood allogeneic HSCT, with a myeloablative conditioning regimen consisting of total body irradiation, followed by cyclophosphamide, and cyclosporine-A graft-versus-host disease prophylaxis. The patient engrafted with adequate absolute neutrophil count by day 16 post-HSCT.682 However, he developed graft-versus-host disease with severe pulmonary insufficiency that resolved promptly following high-dose corticosteroid treatment. He ultimately experienced fulminant hepatic failure of unknown origin 286 days after transplant, and died on day 289.

Autoimmune Hemolytic Anemia

One case series and two case reports describe results on the use of HSCT to treat severe, refractory autoimmune hemolytic anemia (Table 138). The first report involves a boy, 8 years of age, who had severe AIHA that was refractory to prednisone, IVIG, cyclophosphamide, plasmapheresis, and splenectomy.683 As a consequence of life-threatening anemia, he underwent initial high-dose immunosuppressive chemotherapy with cyclophosphamide plus ATG followed by infusion of peripheral blood stem cells that had been mobilized using G-CSF. Because his disease did not respond to the initial HSCT procedure, he was treated again with a high-dose BEAM regimen and infusion of autologous stem cells. The patient was considered to be in hematological remission at 35 days following autologous HSCT, with no infectious complications.683 Although he relapsed at 7 months; this resolved with a course of corticosteroids and he was reported in complete hematological remission at 20 months' followup.

The second autoimmune hemolytic anemia case was that of a 12-year-old male whose disease had failed to respond to prednisone, azathioprine, cyclosporine A, cyclophosphamide, and splenectomy.684 As his condition worsened, he underwent high-dose immunosuppression using thoraco-abdominal irradiation, cyclophosphamide, and CAMPATH-1G followed by infusion of autologous peripheral blood stem cells mobilized using cyclophosphamide. He relapsed 7 weeks after HSCT, and underwent allogeneic HSCT with HLA-compatible unrelated donor bone marrow stem cells following conditioning using busulfan, thiotepa, and fludarabine. Graft-versus-host disease prophylaxis comprised cyclosporine A, methotrexate, and ALG. The patient experienced an uncomplicated post-HSCT course, donor cell engraftment, restoration of normal immune system function, beneficial effects on body growth and skeletal deformities, with normal hemoglobin levels at 18 months after allogeneic HSCT without the need for immunosuppressant therapy.684

The case series reported on 7 cases of AIHA reported to the European Group for Blood and Marrow Transplantation (EBMT) registry between 1984 and 2007.679 After receiving unspecified standard therapy, patients (100% male, age range 2-21) were referred for allogeneic HSCT with various combination conditioning regimens including cyclophosphamide, fludarabine, busulfan, thiotepa, ATG, TBI and received immunosuppressive cyclosporine A with either methotrexate or mycophenolate mofetil therapy for GVHD. Of the 7 patients treated, 4 were alive at 3.9, 86, 112, 124 months, respectively. Three patients died during the study, one from hepatic VOD at 0.7 months of followup, one from infectious complications at followup of 1.4 months, and one died from disease progression at 5.2 months followup.679

Autoimmune Thrombocytopenia

One pediatric case of autoimmune thrombocytopenia was reported in a nonrandomized Phase I/II study that included patients who had severe autoimmune thrombocytopenia that had not responded to prednisone, IVIG, azathioprine, danazol, plasmapheresis, interferon, or splenectomy.685 This 17-year-old male underwent high-dose immunosuppression using cyclophosphamide followed by infusion of peripheral blood stem cells that were mobilized by G-CSF treatment (Table 138). The patient did not respond to autologous HSCT.685

Ongoing Research

According to ClinicalTrials.gov, no clinical trials of HSCT are recruiting patients with severe, refractory Evans syndrome, autoimmune hemolytic anemia, or autoimmune thrombocytopenia.

Conclusion

The overall body of evidence is insufficient to draw conclusions about the comparative benefits or harms of single autologous or allogeneic HSCT compared to conventional therapy or disease natural history pediatric patients with severe, refractory ES, AIHA, or AITP.

The overall body of evidence is insufficient to conclude that an extended drug-free clinical remission can be achieved with allogeneic HSCT for severe, refractory ES or AIHA and autologous HSCT for severe, refractory AIHA.

The projected date was confirmed as personal communication to Hussein Noorani by Dr. Julie Park, Study Chair of the Children's Oncology Group, October 15, 2010.

Data from Einhorn et al. (2007) were provided as personal communication to Hussein Noorani by Dr. Lawrence Einhorn, August 11 and September 1, 2010, respectively; data on outcome events from Agarwal et al. (2009) was provided as personal communication to Hussein Noorani by Dr. Rajni Agarwal, August 10, 2010.

The data presented here are preliminary and were obtained from the Statistical Center of the Center for International Blood and Marrow Transplant Research. The analysis has not been reviewed or approved by the Advisory or Scientific Committees of the CIBMTR.

Footnotes

a

The projected date was confirmed as personal communication to Hussein Noorani by Dr. Julie Park, Study Chair of the Children's Oncology Group, October 15, 2010.

b

Data from Einhorn et al. (2007) were provided as personal communication to Hussein Noorani by Dr. Lawrence Einhorn, August 11 and September 1, 2010, respectively; data on outcome events from Agarwal et al. (2009) was provided as personal communication to Hussein Noorani by Dr. Rajni Agarwal, August 10, 2010.

c

The data presented here are preliminary and were obtained from the Statistical Center of the Center for International Blood and Marrow Transplant Research. The analysis has not been reviewed or approved by the Advisory or Scientific Committees of the CIBMTR.

PubReader format: click here to try

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (3.6M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...