• 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.

National Institutes of Health (US). Office for Medical Applications of Research. NIH Consensus Statements [Internet]. Bethesda (MD): National Institutes of Health (US); 1977-2002.

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of NIH Consensus Statements

NIH Consensus Statements [Internet].

Show details

79Adjuvant Therapy for Patients With Colon and Rectum Cancer

National Institutes of Health Consensus Development Conference Statement April 16-18, 1990

The National Institutes of Health Consensus Development Conference on Adjuvant Therapy for Patients With Colon and Rectum Cancer brought together surgeons, medical oncologists, radiation oncologists, gastroenterologists, other health care providers, and the public to address the issues regarding adjuvant therapy for colon and rectum cancer. Following 1 1/2 days of presentations by experts and discussion by the audience, a consensus panel weighed the evidence and prepared a consensus statement.

Among their findings, the panel recommended that patients with Stage III colon cancer should receive adjuvant therapy with 5-fluorouracil (5-FU) and levamisole. Specific adjuvant therapy is not recommended for Stage II colon cancer patients outside of clinical trials. For rectal cancer, the panel recommended that adjuvant therapy combining chemotherapy and radiation therapy improves local control and survival for Stage II and III patients. The most effective combination at present appears to be 5-FU, methyl-CCNU, and high-dose pelvic irradiation. However, the use of methyl-CCNU outside of clinical trials is discouraged because of documented toxicities. The panel concluded that patients with Stage I colon and rectal cancers are at low risk of recurrence and do not warrant adjuvant therapy. The panel also recommended that the American Joint Committee on Cancer system for classifying stages of colon and rectal cancer, known as the TNM system, become the standard measurement used in clinical trials and in clinical practice.

The full text of the consensus panel's statement follows.

Introduction

Colorectal cancer is a major public health problem in the United States. The annual incidence of colorectal cancer is more than 150,000, with approximately 110,000 new cases of colon cancer (tumors above the peritoneal reflection or more than 12 cm proximal to the anal verge) and 45,000 new cases of rectal cancer each year. Over the past 30 years, the population-adjusted incidence has remained constant at approximately 47 cases per 100,000, and thus the number of cases has increased due to population growth. These tumors are predominantly of a single histologic type, adenocarcinoma. The average age at presentation is 60 to 65 years. About 75 percent of the individuals with these cancers will have a primary surgical resection with the hope of complete tumor eradication. Recently, mortality from colorectal cancer has decreased overall, more for rectal than for colon tumors. Despite the high resectability rate and a general improvement in therapy, nearly half of all patients with colorectal cancer still die from metastatic tumor.

Adjuvant therapy is administered in addition to surgical treatment of the primary colorectal cancer with the intent to improve outcome. Adjuvant therapy options include chemotherapy, radiation therapy, and immunotherapy that are administered before or after curative intent surgery (i.e., surgery with negative microscopic margins). Over the past three decades, many clinical studies have failed to demonstrate benefits from adjuvant therapy. Claims of efficacy have been viewed with skepticism. Recently, new data from several studies have demonstrated delays in tumor recurrence and increases in survival for specific groups of patients.

Five general principles underlie adjuvant therapy:

  1. There may be occult, viable tumor cells in circulation (intravascular, intralymphatic, or intraperitoneal) and/or established, microscopic foci of tumor cells locally, at distant sites, or both.
  2. Therapy is most effective when tumor burden is minimal and cell kinetics are optimal.
  3. Agents with proven effectiveness against the tumor are available.
  4. Cytotoxic therapy shows a dose-response relationship and, therefore, must be administered in maximally tolerated doses, and the duration of therapy must be sufficient to eradicate all tumor cells.
  5. The risk-to-benefit ratio for therapy must be favorable for individuals who may remain asymptomatic for their natural life expectancy after tumor resection.

Numerous staging systems, including those of Dukes', Kirklin, Astler-Coller, the Gastrointestinal Tumor Study Group (GITSG), and the Tumor/Node/Metastasis (TNM) system jointly agreed to by the International Union Against Cancer (UICC) and American Joint Committee on Cancer (AJCC), have been used to classify colorectal cancers. Although the modified Astler-Coller Dukes' system (MAC) is probably the most commonly used in the United States, the current TNM staging system should be utilized to accurately translate other systems into a standard format and will be used throughout this consensus statement (see tables 1 and 2).

Patients with Stage II (T3-4, N0, M0 or MAC B2-3) and Stage III (T1-4, N1-3, M0 or MAC C1-3) cancers have limited disease that may be resected with curative intent but remain at high enough risk for tumor recurrence that they may benefit from additional treatment. Adjuvant therapy, as previously defined, is not applicable to metastatic disease (Stage IV) patients. Individuals with Stage I disease (T1,2 N0 or MAC A or B1) colorectal cancer have a 5-year survival with surgery alone of greater than or equal to 80 to 90 percent. With such a high probability of cure after surgery alone, the risk:benefit ratio for most presently available adjuvant therapy favors no further treatment for this group. Additional factors that may identify subsets within these early-stage patients most at risk for tumor recurrence are being sought, but currently TNM (or MAC) staging remains the only proven prognostic method. Thus, the approximate number of new patients in the United States each year in the conventionally defined high-risk group is 55,000 colon cancer patients and 20,000 rectal cancer patients. Roughly half of these individuals will die from recurrent colorectal cancer. The goal of adjuvant therapy is to reduce the expected 35,000 to 40,000 deaths per year.

Although the history of adjuvant therapy for colorectal cancer has spanned 30 years, only in the past 5 to 8 years have a number of clinical trials yielded reproducible positive results. To evaluate this recent information and resolve issues regarding adjuvant therapy for patients with colon and rectal cancer, the National Cancer Institute and the Office of Medical Applications of Research of the National Institutes of Health convened a Consensus Development Conference on April 16-18, 1990. After a day and a half of presentations by experts and discussion by the audience, a consensus panel drawn from specialists and generalists from the medical profession and related scientific disciplines, clinical investigators, and public representatives considered the evidence and agreed on answers to the following key questions:

  1. Who is at risk for recurrence after colon and rectal cancer resection?
  2. Is there effective adjuvant therapy for patients with colon cancer?
  3. Is there effective adjuvant therapy for patients with rectal cancer?
  4. What are the directions for future research?

Who Is at Risk for Recurrence After Colon and Rectal Cancer Resection?

Surgery remains the critical modality in the treatment of colon and rectal cancer. Negative surgical margins, including radial margins in the rectum and retroperitoneal margins for selected sites within the colon, must be as wide as anatomically possible. Classic oncologic principles must be followed in removal of the primary tumor, the associated vascular arcade, and draining lymph nodes. No-touch techniques are not essential. The goal of any colorectal resection must be to achieve negative margins.

Clarifying the role of adjuvant therapy for colon and rectal cancer and maximizing benefit of adjuvant regimens require identifying those individuals most likely to develop recurrent disease. As a first step, patients should undergo evaluation of the remainder of the large bowel for synchronous lesions. The presence of inflammatory bowel disease, familial adenomatous polyposis, hereditary nonpolyposis colorectal cancer, or more subtle familial associations should be assessed. An abdominal Computerized Tomographic (CT) scan for liver metastases and a preoperative Carcino-Embryonic Antigen (CEA) level should be obtained. In patients presenting with emergent perforation or obstruction, or those explored without complete preoperative assessment, these studies should be done postoperatively. Pelvic CT and Magnetic Resonance Imaging (MRI) are neither specific nor sensitive enough to alter the surgical approach to a rectal lesion but may add useful preoperative data. Transrectal ultrasound appears promising in defining depth of penetration preoperatively but needs to be studied further.

At laparotomy, a complete surgical exploration is mandatory. Colon lesions should be resected with contiguous and regional lymph nodes. Lesions adherent to contiguous organs should be resected in continuity rather than dissected free. Adequate radial margins must be obtained to minimize local recurrence in rectal lesions.

The surgical pathologist should specify the gross and microscopic extent of all surgical margins, the depth of gross and microscopic penetration, the number of nodes removed, the number involved, and whether the apical node (highest level) is positive. The disease should be defined by TNM stage. Characteristics such as venous or lymphatic invasion, perineural invasion, histologic subtype, and grade should be stated.

In assessing risk of recurrence after a potentially curative resection of a colorectal adenocarcinoma, we considered the usefulness of several possible prognostic factors in defining subgroups of patients at low, intermediate, and high risk for recurrence. These included stage of disease, site of disease, marker status, histologic features, host response parameters, and cellular and molecular characteristics.

Pathologic stage is the most important determinant of risk of recurrence and survival probability after primary surgical resection of colon or rectal cancer. The degree of penetration of the primary lesion, the presence or absence of lymph node involvement, and the number of involved nodes are all significant independent risk factors. Several recent retrospective reviews and prospective trials of surgery alone for Stage I and II colon cancer or Stage I rectal cancer indicate that prognosis is favorable. In contrast, the prognosis for patients with more advanced stages is poor. Local recurrence after rectal cancer resection also correlates with TNM staging.

There are clear differences in natural history and patterns of failure between colon and rectal cancer that mandate testing distinct adjuvant strategies for lesions in the two sites. Therefore, separate consideration of adjuvant approaches for colon and rectal cancers is appropriate. The impact of site within the colon or within the rectum on recurrence risk and survival is less significant and will not be addressed in this consensus statement.

Elevation of a preoperative CEA (>5 ng/ml) is an indicator of increased risk for recurrence. CEA elevations correlate with stage and histology. A normal preoperative CEA does not obviate the need for adjuvant therapy in node-positive patients. However, an elevated CEA may determine a high-risk subset of patients with Stage II colon cancer appropriate for study of adjuvant therapy compared with an observation-only control.

Pathologic features such as grade, colloid or signet ring histology, venous or lymphatic invasion, perineural invasion, and perforation have been reported to be significant in univariate analysis. However, in multivariate models where stage is included, many or all of these characteristics lose independent prognostic significance for recurrence or survival.

Pathologic findings suggesting host response such as mononuclear cell infiltration of the primary or reactive regional lymphadenopathy are not reproducible prognostic features. These findings may be of greater interest in the context of current or future adjuvant vaccine trials or studies of Tumor Infiltrating Lymphocytes (TIL)/infiltrating mononuclear cells as adjuvant therapy.

It is premature to use certain cellular or molecular characteristics as standard determinants of recurrence risk or survival in colorectal cancer. Ploidy status and S phase fraction have not consistently correlated with overall recurrence and survival. With refinement and standardization of the methodology, these biologic features may help to determine Stage II subsets at high risk for recurrence.

At a molecular level, the Ki-ras-2 allele is frequently mutated in colon and rectal cancers, but there is thus far no clear correlation between the presence of this mutation and tumor phenotype or clinical course. Deletions of p53 on chromosome 17p and "DCC" on chromosome 18q have recently been documented and may be critical to mechanisms of tumor progression. The effects of transforming oncogenes and potential suppressor genes are highly pleiotropic. Even in the most well-defined systems of viral oncogenesis, hundreds of gene sequences change in expression, all with the potential of contributing to the final tumor phenotype. Data from several investigators suggest similar complexity of changes in gene expression and/or structure in colon and rectal cancer. Data from colorectal tumor cell lines well characterized as poorly, moderately, or well differentiated suggest that production and interaction of growth factors correlate with the degree of differentiation. The clinical significance of these findings remains to be elucidated.

Pursuit of these and future laboratory observations represents critical opportunities for companion studies to current and future clinical trials of adjuvant therapy in colon and rectal cancer. Strategies must be implemented for obtaining optimal tissue and prioritizing assignment of available specimens to studies defining the biological and clinical significance of these cellular and molecular characteristics in relation to etiology, mechanisms of progression, and clinical behavior of colorectal cancer.

Is There Effective Adjuvant Therapy for Patients With Colon Cancer?

Adjuvant therapy for colon cancer should be considered as separate from that for rectal cancer. Patterns of failure differ significantly with local pelvic failure common in rectal cancer because of narrow radial margins defined by the anatomic limits of dissection. Colon cancer, because far fewer limits exist in the extent of surgical resection and margins obtainable, tends to fail in the peritoneal cavity, the liver, or in distant sites with only a small component of isolated local failure. Thus, while a local modality such as radiotherapy may have an important role in the rectum, it is limited to special situations in the colon. Systemic chemotherapy or immunotherapy is the dominant adjuvant modality in colon cancer at this time.

In our identification of effective adjuvant therapies for patients with colon and rectal cancers we have relied exclusively on data derived from prospectively randomized trials. Comparisons using historical controls are subject to bias especially considering the changing prognosis stage for stage of these patients over time. Several aspects contribute to the usefulness of trial results in making consensus recommendations for patient care. First and foremost, an adequate number of patients must be accrued to detect treatment effects that are considered to be worthwhile but realistic. A clinical trial that fails to provide a statistically significant result might be too small to provide sufficient evidence either for or against a specific therapy. Estimates of the magnitude of treatment effects with confidence intervals to reflect the statistical uncertainty of the observed trial results are recommended. The level of statistical significance (p-value) alone is a poor measure of the magnitude of a treatment effect as it is influenced by the sample size of the data being evaluated.

In our deliberations, we attached credibility to high-quality studies, that is, studies with a low percentage of randomized patients excluded from analysis, a large sample size, good treatment compliance, and adequate duration of followup. We gave relatively low credibility to retrospective subset analyses in the absence of confirmatory studies. We regarded overall survival as the primary end point for evaluating treatment effectiveness. In addition, disease-free survival was an important end point. Analyses that weigh the benefits of delaying symptoms of recurrent disease against toxic effects of adjuvant therapy are important for assessing overall treatment gains considering quality of life.

Because individual trials may not be large enough to provide convincing evidence of small but worthwhile treatment effects, meta-analyses have been performed to increase the precision for estimating such treatment effects. These evaluations of data from randomized clinical trials are also useful for putting individual trial results within the context of other available data. Although there are potential methodologic problems with such analyses that relate to the selection and variability of the studies included, these analyses can provide a global summary of treatment effects. For the recommendations made by this panel, we relied on results from individual, well-conducted trials and used the meta-analyses to provide confirmatory evidence.

In the 1960's fluorodeoxyuridine (FUDR) and 5-fluorouracil (5-FU) were given as adjuvant agents for varying periods postoperatively with a wide range of dosage schedules. Activity was demonstrable but not convincing statistically in individual trials. A recent meta-analysis suggested some benefit with the use of 5-FU.

Levamisole alone was first tested in the adjuvant setting in a small group of nonrandomized colorectal cancer patients reported by Verhaegen. Despite the failure of this agent to produce a response either alone or with chemotherapy in metastatic disease, an improvement in survival was noted in this series.

The North Central Cancer Treatment Group (NCCTG) and Mayo Clinic, encouraged by Verhaegen's findings, designed a clinical trial that tested the combination of 5-FU and levamisole in the adjuvant therapy of resectable colorectal cancer patients with Stage II and Stage III lesions. Quality control of data acquisition and analysis was excellent. The data are mature at a median followup of 7 years. The 5-FU plus levamisole combination significantly reduced recurrence. Subset analysis for Stage III patients treated with 5-FU and levamisole showed a significant improvement in overall survival. These results were considered sufficiently encouraging to recommend a confirmatory trial.

An intergroup study was designed with the same methodology with two exceptions: patients with Stage II disease were randomized to observation or to 5-FU plus levamisole and those with Stage III disease were randomized to the three arms of observation, levamisole alone, and 5-FU plus levamisole; only patients with colon carcinoma were studied.

This trial was well performed. Median followup is 3 years. Levamisole alone produced no significant effect. A dramatic effect of 5-FU and levamisole relative to surgery alone on disease-free survival and overall survival was demonstrated for Stage III patients. Therapy with 5-FU and levamisole reduced the risk of cancer recurrence by 41 percent and the overall death rate by 33 percent. Overall survival percentages at 3 1/2 years were estimated to be 71 percent versus 55 percent for the control group. Analysis failed to demonstrate an outcome difference in Stage II patients, but 1 to 2 years of additional observation is needed. Side effects from this combination were well tolerated and primarily associated with 5-FU.

There are potential pitfalls in interpreting these results. The mechanism of action of levamisole is not understood. The positive results seen in the NCCTG/Mayo Clinic study were based on a relatively small number of patients and a subset analysis. The short followup is a concern for the interpretation of results from the intergroup study. At this time, however, these studies are the most compelling investigations demonstrating a statistically significant improvement in disease-free survival and overall survival in the adjuvant setting for Stage III patients. Longer followup will hopefully strengthen the conclusion that adjuvant 5-FU/levamisole for Stage III patients should be the control arm for ongoing clinical trials and should be offered to similarly staged patients off trial unless medical or psychosocial contraindications exist.

If subsequent analysis of 5-FU/levamisole demonstrates significant benefit for Stage II patients, identical recommendations for trial or nontrial patients would pertain. However, the high survival rates observed in node-negative patients without adjuvant therapy would require a close examination of risk and benefit if this subset of patients were to be treated outside of clinical trials. It is noteworthy to observe the high response rates in metastatic colon cancer with the use of 5-FU and leucovorin. This combination, with or without levamisole, is under current clinical trial investigation.

The National Surgical Adjuvant Breast and Bowel Project (NSABP) protocol CO1 demonstrated a significant, although modest, improvement in disease-free survival at a median followup of 7 years using a combination of methyl-chloroethyl-cyclohexyl-nitrosourea (methyl-CCNU), vincristine, and 5-FU. This effect seems to be diminishing with time. A major concern, demonstrated in other studies as well, is the leukemogenic effect and renal toxicity of methyl-CCNU.

The GITSG protocol 6175 failed to demonstrate any benefit for the adjuvant use of methyl-CCNU/5-FU, methanol extracted residue of BCG (MER) or methyl-CCNU/5-FU plus MER for similarly staged node-positive and node-negative colon cancer patients.

Recognition that the liver is the most common site of tumor recurrence has led to multiple controlled and uncontrolled studies to evaluate portal vein infusion as adjuvant therapy in high-risk patients. One study combined systemic infusion with hepatic irradiation. All of these studies sought to demonstrate an improvement in regional control of hepatic metastases. Although some have reported significant improvement in disease-free survival and overall survival (presumed to be secondary to systemic effect), none, with the exception of the original study by Taylor et al., have succeeded in reducing liver metastases. This form of regional adjuvant therapy remains investigational. In view of the improved overall survival results, other approaches for using short duration intraoperative or immediate postoperative (perioperative) chemotherapy might be investigated.

Whenever a new therapy is found to be effective in a clinical trial, it is always difficult to determine how to advise other patient groups not precisely identical to the clinical study group. Although the available clinical trial data demonstrate benefit for patients who receive their therapy within 6 weeks of surgery, there is no evidence that patients who receive therapy after that period benefit significantly. Because there is no appropriate way to design a prospective trial that intentionally delays therapy beyond 6 weeks, this question cannot be answered by existing data or by subsequent clinical trial. The decision to begin adjuvant chemotherapy in appropriate patient groups later than 6 weeks after surgery must be left to individual decisions between the patient and doctor.

Is There Effective Adjuvant Therapy for Patients With Rectal Cancer?

Oncologic principles of surgical removal of the primary and draining lymph nodes apply to rectal cancer as well as to colon cancer. In general, two standard surgical approaches, abdominoperineal resection and low anterior resection, have been used in these trials. In contrast to colon cancer, there is a significant risk of symptomatic local-regional failure as the only or first site of recurrence in patients with curatively resected rectal cancer. In node-negative patients, local recurrence may be as low as 5 to 10 percent for Stage I tumors but approaches 25 to 30 percent for Stage II tumors. In patients with Stage III tumors, local recurrence rates increase to 50 percent or more for more invasive tumors. The principal reason for local recurrence in resected rectal cancer appears to be related to the anatomic constraints in obtaining wide radial margins, even though the proximal and distal margins are adequate. Thus, in addition to the standard treatment outcomes of overall and disease-free survival, local-regional control should be considered in patients with rectal cancer. Clearly, this is not only an academic issue because local recurrence is associated with significant morbidity and often not effectively palliated by local treatment (surgery and/or radiation therapy) or by systemic chemotherapy at the time of recurrence.

Patients with locally invasive rectal cancer are at high risk for systemic relapse similar to patients with locally advanced colon cancer. The systemic risk for node-negative rectal cancer patients also appears comparable to similarly staged colon cancer patients.

By definition, "effective" adjuvant therapy of surgically resected rectal cancer should address both the local and systemic relapse potential of the disease. Over the past 10 to 15 years, the use of adjuvant radiation therapy has been evaluated in several randomized studies in both the preoperative and postoperative setting. An important variable in the studies of preoperative irradiation has been the total dose ranging from a single 5 Gray (Gy) fraction within 24 hours before surgery to high-dose fractionated (35-50 Gy in 4-7 weeks) irradiation over several weeks before surgery. Overall, there appears to be a significant decrease in local recurrence rates in patients receiving moderate to high dose (>35 Gy) preoperative irradiation (10 to 20 percent versus 25 to 50 percent in controls), usually without any impact on survival.

More recently, randomized trials in patients with resected rectal cancer have assessed the impact of postoperative irradiation. This approach allows for accurate pathological staging and inclusion of only high-risk patients in the study. An improvement in local control is found in most studies where postoperative irradiation (40-50 Gy) is compared to surgery alone. However, similar to preoperative irradiation, no significant benefits in terms of disease-free or overall survival are evident.

In light of a minimal, if any, impact of adjuvant radiation therapy alone on survival, it is evident that systemic treatment is necessary to improve the prognosis of surgically resected rectal cancer patients. Several large, well-designed randomized trials employing systemic therapy in at least one study arm were updated at the consensus conference. The GITSG 7175 and 7180 studies, the NSABP R-01 study, and the NCCTG-Mayo 794751 study were reviewed in detail and form the principal basis for our recommendations.

The major evidence in favor of the use of chemotherapy for systemic disease control comes from these trials. All relevant trials demonstrate an advantage for chemotherapy using 5-FU and methyl-CCNU with or without radiotherapy for overall survival, disease-free survival, or both. The second GITSG study (7180) compared the effects of radiation therapy with 5-FU or 5-FU/methyl-CCNU. Preliminary analysis documented no differences in recurrence rates, suggesting that methyl-CCNU may be unnecessary. Unfortunately, early termination of this trial makes a definitive answer difficult. In the first GITSG trial (7175), the combination of 5-FU and methyl-CCNU with radiotherapy continues to show significant improvement in local control and survival with 10-year followup compared with surgery alone and somewhat better than either modality on its own. In the NCCTG study with a median followup of 6 years, the same combination is shown to be superior to radiotherapy alone (using DFS as the end point, overall survival is only marginally significant). In both studies, combined modality treatment yields local failure rates about one-half of single modality therapy. In the NSABP study, a combination of 5-FU, methyl-CCNU, and vincristine is significantly better than surgery alone but local failure remains high (23 percent). The one trial that does not fit this pattern is a European Organization for Research on Treatment of Cancer (EORTC) study (40741), where the addition of only 5-FU during preoperative radiotherapy had an apparently adverse effect on survival.

Overall, these results are strongly suggestive that the best current adjuvant therapy for rectal cancer involves postoperative treatment with both chemotherapy and radiotherapy. However, the absolute necessity for radiotherapy for survival benefits is by no means convincingly demonstrated at this time, and we must await the completion of current trials for clarification of this issue. Likewise, the presence of methyl-CCNU in most of the successful chemotherapy regimens is problematic in view of its demonstrated leukemogenesis and nephrotoxicity. It should be noted that the NCCTG study that documents benefits from combined modality therapy utilized only two doses of methyl-CCNU compared with seven doses in the NSABP and GITSG studies. Thus, if methyl-CCNU is found to be essential it may be possible to deliver fewer doses. We are optimistic about the likely effectiveness of combined modality therapy, but we recognize the need to study the effectiveness of combinations that do not include methyl-CCNU, such as 5-FU and leucovorin or related combinations in concert with radiotherapy for local control.

As with the colon carcinoma trials, the majority of the adjuvant treatments in rectal carcinoma were initiated within 6 weeks of surgery. It is not known whether therapy initiated after that point would be beneficial. The decision to begin adjuvant chemotherapy later than 6 weeks after surgery must be decided individually between the patient and doctor.

The administration of adjuvant therapy must be balanced against the possible deleterious side effects that might negate any beneficial effect. Chronic radiation effects, although infrequent, can be severe, including radiation enteritis and bowel obstruction, because tumoricidal doses (45-55 Gy) of radiation are necessary. Clearly, careful attention to the technical details of radiation therapy, including CT-assisted, contrast-enhanced treatment planning to exclude small bowel and the use of multiple radiation fields, can reduce acute and chronic radiation side effects. Additionally, surgical attention should be directed to preventing small intestine loops from entering the pelvis, using either natural structures or synthetic mesh. Chemotherapy should be administered by oncologists skilled in the administration of these toxic agents. The use of methyl-CCNU with its toxic effects remains problematic, and research should be directed toward finding effective drug combinations that do not use this agent.

What Are the Directions for Future Research?

The history of adjuvant treatment for colon cancer is confusing in part because of conflicting data from small, poorly controlled trials with inadequate followup. This experience makes it likely that major advances will require large, prospective randomized trials with well-balanced populations, carefully stratified by known prognostic factors of significance. There has also been little supportive biochemical or immunologic data generated from recent trials to support or reject the biological rationale on which the trial was based. Future trials hopefully should build in laboratory studies to confirm or reject the principal underlying hypothesis.

A review of existing clinical trials served to emphasize the need to continue efforts to improve the quality of trial data and the need to apply appropriate quality-assurance measures. For example, future trials will need to use such measures to document adherence to preestablished surgical resection margins, staging, and compliance with adjuvant therapy.

Although several adjuvant 5-FU containing combinations now show disease-free and overall survival benefits for patients with colon cancer, a wide variety of doses, schedules, routes of administration, and duration of therapy has been used. It is impossible to conclude what dose, schedule, and duration of 5-FU is optimal. Future trials should standardize drug dose and schedule, and link dose, dose intensity, and duration to some biological or biochemical end point. Likewise, the potential mechanism of levamisole effect is presently undefined. Levamisole drug dose and schedule have been selected empirically, and the extent of immunomodulation achieved or even what is desirable has not been established. Although lack of rationale for the activity of the combination should not preclude acceptance of its demonstrated benefits, efforts to build subsequent clinical trials on the provocative results of the 5-FU/levamisole trial are hampered by the absence of data on the immunomodulatory effect, if any, or on the mechanism of action of the combination. Future clinical trials should include sufficient laboratory investigation to address these questions of mechanism of action and biological rationale.

The highest priority for future adjuvant clinical trials in colon cancer remains the search for optimal regimens that improve disease-free survival and overall survival. At present, the 5-FU/levamisole regimen is the standard to which new therapies should be compared. Issues that remain at high priority are the study of other modulators of 5-FU, which include leucovorin, phosphonacetyl-l-aspartate (PALA), and others. Crucial for the understanding of these trials will be the biochemical determination of successful modulation of 5-FU effect to be done in concert with the ongoing clinical trial. Other combinations active in advanced disease, such as 5-FU/interferon have merit for study provided they are built into prospective trial design.

Adjuvant therapy is most conclusively established for patients with Stage III disease. However, other groups of patients, such as those with Stage II disease, possess 5-year disease-free and overall survival outcomes of 80 percent or less and might therefore be legitimate candidates for inclusion into subsequent prospective clinical trials with an untreated arm. Patients with early disease such as Stage I patients have an excellent 5-year survival (>80 to 90 percent) with surgery alone and are not likely to benefit from adjuvant chemotherapy unless the development of newer prognostic markers allows one to select poor prognostic subsets. Newer prognostic markers such as DNA content, proliferative activity, surface glycoprotein, gastrin receptor, oncogenes/tumor suppressor genes, and allelic deletions may allow the refinement of prognostic groups further than presently possible with stage alone.

Although local relapse is not a common problem in colon cancer, there are certain groups (i.e., TNM T4,N1,N2) that have significant local failure rates and should be included in separate clinical trials testing radiation-containing combined modality therapies.

Two recent clinical trials suggest a disease-free and a survival benefit from the direct portal administration of chemotherapy, although the magnitude of the effect is less than that seen in the systemic therapy trials with 5-FU/levamisole. Furthermore, neither trial reduced the frequency of relapse in the liver. Subsequent clinical trials will require the direct comparison of effective systemic adjuvant chemotherapy with a similar regimen given by regional infusion before intraportal drug administration can be a recommended alternative to systemically administered adjuvant chemotherapy.

As with colon carcinoma, the prospective clinical trial will be the major tool for defining future effective therapy in rectal carcinoma. Future trials must define the relative merits of combination chemotherapy, including 5-FU/levamisole; 5-FU, methyl-CCNU, and vincristine; and also introduce newer combinations of promise, including 5-FU/leucovorin and other biochemical techniques for 5-FU modulation.

Current evidence indicates that combined modality therapy can significantly reduce the incidence of symptomatic local recurrence in the disease, and the next series of clinical trials must define the proper dose, sequence, and integration of these modalities. The impact on local recurrence, disease-free survival, and overall survival must be measured against any increased toxicity inherent in the combined modality approach.

Adjuvant therapy for colon and rectal carcinoma potentially affects 35,000 to 40,000 patients each year. These therapies, while bringing improved disease-free survival and overall survival, do entail longer and more complex treatments, and there is presently little information to answer questions regarding either physiologic or financial burdens experienced by patients undergoing adjuvant therapy either within or outside the clinical trials setting. Resources need to be allocated to incorporate these important questions into the design of future colorectal adjuvant therapy trials. In addition, differences in disease incidence and outcome observed in various ethnic and socioeconomic groups suggest a need to perform adequate adjuvant trials to address these questions.

The cost to the patient should be considered as an integral part of adjunctive therapy. Ability to pay should never be a factor in a patient's decision to receive treatment. New protocols should always consider cost to the patient. Cost is not only considered in dollars. Cost can also mean quality of life. Decisions affecting the quality of a patient's life require an understanding between the patient and his or her doctor. Basic qualities for all people should be hope, comfort, and freedom from pain.

Conclusions and Recommendations

In answer to the question "Who is at risk for recurrence after colon and rectal cancer resection?" the consensus panel recommends the following:

  • The TNM system based on a complete pathological description can effectively describe risk groups for recurrence and should be used in clinical trials research and clinical practice.
  • Patients with colon and rectal cancer should be studied separately when defining adjuvant strategies.
  • Patients with Stage III colon cancer or Stage II/III rectal cancer are at high risk for recurrence and warrant adjuvant therapy.
  • Anatomic or biologic features may define subsets of patients with Stage II colon cancer at intermediate risk of recurrence sufficient to merit testing of adjuvant treatment compared to observation only controls. These features include:
    • T4 N0 M0
    • T3 N0 M0 plus one or more of the following:
      • Pre-op CEA > 5 ng/ml.
      • Aneuploid DNA content.
      • High S phase.
      • Colloid, signet ring, or poorly differentiated histology.
      • 17p or 18q deletion.
  • Patients with Stage I lesions are at low risk of recurrence and should not receive adjuvant treatment.
  • Correlation of laboratory observations and clinical data must be pursued to define the biological and clinical significance of these cellular and molecular characteristics.

In answer to the question "Is there effective adjuvant therapy for patients with colon cancer?" we conclude that:

  • Optimal adjuvant therapy for Stage II and III colon cancer has not yet been devised. Continued clinical trials in this disease are essential to discover more active adjuvant therapies.
  • Based on current clinical trial data, Stage III patients unable to enter a clinical trial should be offered adjuvant 5-FU and levamisole as administered in the intergroup trial unless medical or psycho-social contraindications exist.
  • The panel cannot recommend any specific adjuvant therapy at this time for Stage II patients outside of clinical trials.

In answer to the question "Is there effective adjuvant therapy for patients with rectal cancer?" we conclude the following:

  • No adjuvant therapy is recommended for Stage I patients; in contrast to Stage II colon cancer, we recommend adjuvant treatment of Stage II rectal cancer.
  • Combined postoperative chemotherapy and radiation therapy improves local control and survival in Stage II and III patients and is recommended.
  • At the present time, the most effective combined modality regimen appears to be 5-FU plus methyl-CCNU, and high-dose pelvic irradiation (45 to 55 Gy) but chronic toxicity considerations of methyl-CCNU mitigate against using this regimen outside ongoing clinical trials.
  • Current clinical trials of combined modality therapy are designed to improve the prognosis of Stage II and III patients. Entry of patients into these clinical trials is highly encouraged.

The following directions for future research are suggested:

  • The highest priority for future adjuvant trials in colon cancer should build on the results achieved with 5-FU/levamisole using modulators of 5-FU, modulators of host response, and new regimens of proven efficacy in advanced disease.
  • The highest priority for future adjuvant trials in rectal cancer will be to integrate radiation therapy with newer 5-FU modulated regimens such as 5-FU/levamisole, 5-FU/leucovorin, or other combinations with demonstrated activity in advanced disease.
  • There is a need to identify new determinants of risk to be used to select early stage patients likely to benefit from adjuvant therapy.
  • There is a need to incorporate into intergroup trials the appropriate basic laboratory investigations required to define mechanisms of drug action, especially in trials involving modulators of host immune response.
  • There is a need to address issues of quality of life and the cost benefit of such therapies.
  • There is a need to initiate trials to address questions of differences in disease and outcome observed in various ethnic and socioeconomically disadvantaged groups.

Consensus Development Panel

  • Glenn D. Steele, Jr., M.D., Ph.D.
  • Panel and Conference Chairperson
  • Chairman
  • Department of Surgery
  • New England Deaconess Hospital
  • Boston, Massachusetts
  • Leonard H. Augenlicht, Ph.D.
  • Director
  • Molecular Oncology Program
  • Department of Oncology
  • Albert Einstein Cancer Center
  • Bronx, New York
  • Colin B. Begg, Ph.D.
  • Chairman
  • Department of Epidemiology and Biostatistics
  • Memorial Sloan-Kettering Cancer Center
  • New York, New York
  • Richard D. Gelber, Ph.D.
  • Associate Professor
  • Division of Biostatistics and Epidemiology
  • Dana Farber Cancer Institute
  • Boston, Massachusetts
  • Mark R. Green, M.D.
  • Professor of Medicine
  • Division of Hematology and Oncology
  • University of California at San Diego Medical Center
  • San Diego, California
  • Timothy J. Kinsella, M.D.
  • Professor and Chairman
  • Department of Human Oncology
  • University of Wisconsin Clinical Cancer Center
  • Madison, Wisconsin
  • John E. Niederhuber, M.D.
  • Professor of Surgery and Oncology and Molecular Biology and Genetics
  • The Johns Hopkins University School of Medicine
  • Baltimore, Maryland
  • Theodore L. Phillips, M.D.
  • Professor and Chairman
  • Department of Radiation Oncology
  • University of California, San Francisco
  • San Francisco, California
  • Beth Williams
  • Director of New Initiatives Office
  • Lunar and Planetary Institute
  • Houston, Texas
  • Sidney J. Winawer, M.D.
  • Chief
  • Gastroenterology Service
  • Member and Head
  • Laboratory for Gastrointestinal Cancer Research
  • Memorial Sloan-Kettering Cancer Center
  • New York, New York
  • David P. Winchester, M.D., M.S.
  • Medical Director
  • Cancer Department
  • American College of Surgeons
  • Chicago, Illinois
  • Rodger J. Winn, M.D.
  • Chief, Section of Community Oncology
  • University of Texas M.D. Anderson Cancer Center
  • Houston, Texas
  • Robert C. Young, M.D.
  • President
  • Fox Chase Cancer Center
  • Philadelphia, Pennsylvania

Speakers

  • Robert W. Beart, Jr., M.D.
  • "NCCTG Portal Vein Study"
  • Professor of Surgery
  • Mayo Clinic--Scottsdale
  • Scottsdale, Arizona
  • Martin Brown, Ph.D.
  • "Economic Considerations in Adjuvant Therapy of Colon Cancer"
  • Economist
  • Applied Research Branch
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Marc E. Buyse, M.Sc., M.B.A.
  • "EORTC Studies of Colon Cancer"
  • "EORTC Studies of Rectal Cancer"
  • Assistant Director
  • EORTC
  • Brussels
  • BELGIUM
  • Alfred M. Cohen, M.D.
  • "Adjuvant Therapy for Patients With Colon or Rectum Cancer--Surgical Considerations"
  • Chief
  • Colorectal Service
  • Department of Surgery
  • Memorial Sloan-Kettering Cancer Center
  • New York, New York
  • Harold O. Douglass, Jr., M.D.
  • "Ten-Year Followup of First Generation of Surgical Adjuvant Rectal Cancer Studies of the Gastrointestinal Tumor Study Group"
  • Associate Chief
  • Department of Surgical Oncology
  • Roswell Park Memorial Institute
  • Buffalo, New York
  • Bernard Fisher, M.D.
  • "Postoperative Adjuvant Therapy for Dukes' B and C Adenocarcinoma of Rectum: NSABP Protocols R-01 and R-02"
  • (with Howard Rockette, Ph.D.)
  • Distinguished Service Professor
  • Chairman of National Surgical Adjuvant Breast and
  • Bowel Project
  • University of Pittsburgh School of Medicine
  • Pittsburgh, Pennsylvania
  • Leonard L. Gunderson, M.D.
  • "Rectal Cancer--Overview"
  • Chairman and Professor
  • Department of Radiation Oncology
  • Mayo Clinic
  • Rochester, Minnesota
  • J. Michael Hamilton, M.D.
  • "Future Issues in Adjuvant Therapy of Colorectal Cancer"
  • Senior Investigator
  • Cancer Therapy Evaluation Program
  • Division of Cancer Treatment
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Stanley Hamilton, M.D.
  • "Pathologic Issues"
  • Associate Professor of Pathology and Oncology
  • Department of Pathology
  • Johns Hopkins Hospital
  • Baltimore, Maryland
  • Bernard Levin, M.D.
  • "Review UK Study"
  • "Review MRC Studies"
  • Professor of Medicine
  • Chief
  • Section of Gastrointestinal Oncology and Digestive Diseases
  • University of Texas M.D. Anderson Cancer Center
  • Houston, Texas
  • Robert J. Mayer, M.D.
  • "Adjuvant Colon Cancer Trials of the Gastrointestinal Tumor Study Group"
  • Associate Professor of Medicine
  • Dana Farber Cancer Center
  • Boston, Massachusetts
  • U. Metzger, M.D.
  • "Randomized Multicenter Trial of Adjuvant Intraportal 5-FU + Mitomycin-C Chemotherapy for Colorectal Cancer (SAKK 4-0/81)"
  • Department of Surgery
  • University Hospital--Zurich
  • Zurich
  • SWITZERLAND
  • Charles G. Moertel, M.D.
  • "Studies of Levamisole and 5-Fluorouracil in the Surgical Adjuvant Therapy of Colon Cancer"
  • "An Evaluation of Combined-Modality Radiation and Chemotherapy for Surgical Adjuvant Treatment of Rectal Carcinoma"
  • Chairman of North Central Cancer Treatment Group
  • Professor of Oncology
  • Mayo Clinic
  • Rochester, Minnesota
  • Michael J. O'Connell, M.D.
  • "Historical Overview of Surgical Adjuvant Chemotherapy for Colorectal Cancer"
  • Professor of Oncology
  • Mayo Clinic
  • Rochester, Minnesota
  • Howard Rockette, Ph.D.
  • "Postoperative Adjuvant Therapy for Dukes' B and C Adenocarcinoma of Rectum: NSABP Protocols R-01 and R-02"
  • (with Bernard Fisher, M.D.)
  • Professor of Biostatistics
  • Department of Biostatistics
  • University of Pittsburgh Graduate School of Public
  • Health
  • Pittsburgh, Pennsylvania
  • D.L. Wickerman, M.D.
  • "Postoperative Adjuvant Therapy for Dukes' B and C Adenocarcinoma of Colon: NSABP Protocols C-01, C-03, and C-04"
  • (with Norman Wolmark, M.D.)
  • National Surgical Adjuvant Breast and Bowel Project
  • Associate Research Professor
  • Department of Surgery
  • University of Pittsburgh School of Medicine
  • Pittsburgh, Pennsylvania
  • Norman Wolmark, M.D.
  • "Postoperative Adjuvant Therapy for Dukes' B and C Adenocarcinoma of Colon: NSABP Protocols C-01, C-03, and C-04"
  • (with D.L. Wickerham, M.D.)
  • The Mark M. Ravitch Professor of Surgery
  • Surgeon and Chief
  • Department of Surgery
  • Montefiore Hospital
  • Pittsburgh, Pennsylvania

Planning Committee

  • J. Michael Hamilton, M.D.
  • Planning Committee Chairperson
  • Senior Investigator
  • Cancer Therapy Evaluation Program
  • Division of Cancer Treatment
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Alfred M. Cohen, M.D.
  • Chief
  • Colorectal Service
  • Department of Surgery
  • Memorial Sloan-Kettering Cancer Center
  • New York, New York
  • Jerry M. Elliott
  • Program Analyst
  • Office of Medical Applications of Research
  • National Institutes of Health
  • Bethesda, Maryland
  • John H. Ferguson, M.D.
  • Director
  • Office of Medical Applications of Research
  • National Institutes of Health
  • Bethesda, Maryland
  • Michael Friedman, M.D.
  • Associate Director
  • Cancer Therapy Evaluation Program
  • Division of Cancer Treatment
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Leonard L. Gunderson, M.D.
  • Chairman and Professor
  • Department of Radiation Oncology
  • Mayo Clinic and Mayo Medical School
  • Rochester, Minnesota
  • William H. Hall
  • Director of Communications
  • Office of Medical Applications of Research
  • National Institutes of Health
  • Bethesda, Maryland
  • Kathleen Isner
  • Conference Coordinator
  • Prospect Associates
  • Rockville, Maryland
  • Barney Lepovetsky, Ph.D.
  • Office of Technology Development
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Bernard Levin, M.D.
  • Professor of Medicine
  • Chief
  • Section of Gastrointestinal Oncology and Digestive Diseases
  • University of Texas M.D. Anderson Cancer Center
  • Houston, Texas
  • Michael E. Newman
  • Science Writer
  • Report Section
  • Office of Cancer Communications
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Lawrence Rubinstein, Ph.D.
  • Biometric Research Branch
  • Cancer Therapy Evaluation Program
  • Division of Cancer Treatment
  • National Cancer Institute
  • National Institutes of Health
  • Bethesda, Maryland
  • Glenn D. Steele, Jr., M.D., Ph.D.
  • Panel and Conference Chairperson
  • Chairman
  • Department of Surgery
  • New England Deaconess Hospital
  • Boston, Massachusetts

Conference Sponsors

  • National Cancer Institute
  • Samuel Broder, M.D.
  • Director
  • Office of Medical Applications of Research
  • John H. Ferguson, M.D.
  • Director

Supplemental Information for NIH Consensus Statement on Adjuvant Therapy for Patients With Colorectal Cancer

Since the NIH Consensus Statement on Adjuvant Therapy for Patients With Colorectal Cancer was issued, additional information has become available that supplements the original statement.

For colon cancer, barring other mitigating factors, patients who have resection for cure of stage III should receive adjuvant chemotherapy. This therapy may consist of 5FU + levamisole as given in the intergroup study (as stated in the consensus statement) or 5FU + leucovorin given for 6 cycles. Levamisole added to 5FU +leucovorin does not appear to add to efficacy.

For stage II and III rectal cancer, standard adjuvant therapy remains radiation + either continuous infusion or bolus 5FU regimens. Ongoing studies are evaluating which may be the optimal regimens.

References:

  • O'Connell MJ, Laurie JA, Kahn M et al. Prospectively randomized trial of postoperative adjuvant chemotherapy in patients with high risk colon cancer. J. Clin Oncol 1998:16:295-300.
  • O'Connell MJ Mailliard JA, Kahn MJ et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J. Clin Oncol 1997:15:246-250.
  • Haller DG, Catalano PJ, Macdonald JS et al. Fluorouracil (FU), leucovorin, and levamisole (LEV) adjuvant therapy for colon cancer: four years results of INT-0089. Proc Amer Soc Clin Oncol 1997:16:A940.

Tables

Table 1. Tumor/Node/Metastasis (TNM) Staging

Table 1. Tumor/Node/Metastasis (TNM) Staging.

Table

Table 1. Tumor/Node/Metastasis (TNM) Staging.

Table 2. TNM Staging

Table 2. TNM Staging.

Table

Table 2. TNM Staging.

This statement was originally published as: Adjuvant Therapy for Patients with Colon and Rectum Cancer. NIH Consens Statement 1990 Apr 16-18;8(4):1-25.

For making bibliographic reference to the statement in the electronic form displayed here, it is recommended that the following format be used: Adjuvant Therapy for Patients with Colon and Rectum Cancer. NIH Consens Statement Online 1990 Apr 16-18 [cited year month day];8(4):1-25. [PubMed: 2077398]

NIH Consensus Statements are prepared by a nonadvocate, non-Federal panel of experts, based on (1) presentations by investigators working in areas relevant to the consensus questions during a 2-day public session; (2) questions and statements from conference attendees during open discussion periods that are part of the public session; and (3) closed deliberations by the panel during the remainder of the second day and morning of the third. This statement is an independent report of the consensus panel and is not a policy statement of the NIH or the Federal Government.

PubReader format: click here to try

Views

  • PubReader
  • Print View
  • Cite this Page

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...