 |
|
|
The Office of Health Assessment (OHTA) evaluates the risks, benefits, and clinical effectiveness of new or unestablished medical technologies that are being considered for coverage under Medicare. These assessments are performed at the request of the Health Care Financing Administration (HCFA). They are the basis for recommendations to HCFA regarding coverage policy decisions under Medicare.
Questions about Medicare coverage for certain health care technologies are directed to HCFA by such interested parties as insurers, manufacturers, Medicare contractors, and practitioners. Those questions of a medical, scientific, or technical nature are formally to OHTA for assessment.
OHTA's assessment process includes a comprehensive review of the medical literature and emphasizes broad and open participation from within and outside the Federal Government. A range of expert advice is obtained by widely publicizing the plans for conducting the assessment through publication of an announcement in the Federal Register and solicitation of input from Federl agencies, medical specialty societies, insurers, and manufacturers. The involvement of these experts helps assure inclusion of the experienced and varying viewpoints needed to round out the data derived from individual scientific studies in the medical literature.
After OHTA receives information from experts and the scientific literature, the results are analyzed and synthesized into an assessment report. Each report represents a detailed analysis of the risks, clinical effectiveness, and uses of new or unestablished medical technologies considered for Medicare coverage. These Health Technology Assessment Reports form the basis for the Public Health Service recommendations to HCFA and are disseminated widely. Individual reports are available to the public once HCFA has made a coverage decision regarding the subject technology.
OHTA is one component of the Agency for Health Care Policy and Research (AHCPR), Public Health Service, Department of Health and Human Services.
Thomas V. Holohan, M.D.
Director
Office of Health Technology Assessment
J.Jarrett Clinton, M.D.
Administrator
Questions regarding this assessment should be directed to:
Office of Health Technology Assessment
AHCPR
Executive Office Center
2101 East Jeferson Street, Suite 400
Rockville, MD 20857; (301) 227-8337
Hyperthermia has been used to augment other of forms of cancer treatment for about 100 years. In 1893 W. B. Coley used infections with bacteria and/or injections of bacterial extracts to treat patients with cancer; it is unknown whether the beneficial results he obtained were due to the fever induced or to the patients' production of factors such as tumor necrosis factor (TNF) (1) Increased interest in the used of hyperthermia for the treatment of cancer developed in the 1960s, and various devices to produce either systemic hyperthermia or hyperthermia in selected parts of the body were developed.(2,3) In the past 30 years hyperthermia has been used alone and together with radiotherapy and/or chemotherapy. In addition, local isolated limb perfusion of heated chemotherapeutic agents has also been performed.(4,5) The literature contains several thousand papers on these subjects.
The purpose of the present assessment is to evaluate the clinical effectiveness of hyperthermia alone and hyperthermia combined with chemotherapy. A previous review in 1984 by the Office of Health Technology Assessment of the National Center for Health Services Research (now the Agency for Health Care Policy and Research) dealt primarily with the subject of hyperthermia alone or combined with radiotherapy. That review concluded that hyperthermia combined with radiotherapy may provide useful local palliative effects.(6)
The scientific basis for the use of hyperthermia as an adjunct for the treatment of cancer rests on four main observations. First, there is evidence that some tumor cells may be more sensitive to the effects of increased temperature compared with their normal cellular counterparts. Second, the combination of chemotherapeutic agents and hyperthermia produces additive and synergetic killing effects on tumor cells. These effects appear more dramatic at low pH. Third, hyperthermia produces changes in the blood flow and oxygen levels in tumors that may have beneficial clinical effects, especially in combination with other agents. Fourth, the immune responses of the patient toward his or her tumors may be augmented by hyperthermia. These would include effects on macrophages, natural killer (NK) cells, and T-cell cytotoxic effector functions.
In addition, hyperthermia may have beneficial therapeutic effects on cancer because of actions on systems of which we have no precise knowledge. All these rationales will be discussed in more detail below.
Cancer therapy has been performed since the dawn of recorded history. The earliest treatments involved crude modalities such as burning, caustic agents, and primitive surgery, as well as nonspecific, toxic oral agents such as arsentic. Modern cancer therapy started in the first part of the 20th century with external beam x-ray therapy, radium implantation, and sophisticated surgery. Modern chemotherapy began in the mid-1940s with the use of nitrogen mustard, which was soon followed by the development of many other chemotherapeutic agents. In spite of the many advances made in cancer therapy in the past 100 years, the treatment of cancer still remains unsatisfactory because of the nonspecific and toxic actions of all the standard therapeutic modalities. In an attempt to improve these current standard forms of therapy, various other treatments have also been First, biological agents are used that can act in the patient to stimulate a beneficial immune response directed against the tumor cell.(7) Nonspecific agents such as Calmette-Guegraverin bacillus were used initially. More recently, genetically cloned cytokines and lymphokines, which can be produced in large amounts, have been used. These agents can also combined with infusion of immune-component effector cells to enhance further the immune response directed to killing the tumor cell.(8)
Another form of therapy used in addition to more standard forms of therapy is hyperthermia. Hyperthermia can be administered either systemically or locally, either to superficial (less than 5 cm) or to deeper lesions. Several different types of devices for hyperthermia have now been approved for use by the Center for Devices and Radiological Health, Food and Drug Administration (FDA). The National Cancer Institute (NCI), National Institute of Health (NIH), has provided contractual support for comparative testing of several different types of devices used for hyperthermic therapy. Centers in the United States, Europe, and Japan are currently investigating the clinical usefulness of hyperthermia combined with either radiation, chemotherapy, or both.
In the past 30 years thousands of patients have been entered into clinical hyperthermia trials using heat alone or in various combinations with radiotherapy and chemotherapy. A whole spectrum of devices, treatment schedules, and drugs in a variety of combinations have been used.
The combined use of hyperthermia and chemotherapy for the treatment of cancer is usually considered to be a "last resort" treatment for patients who have either failed to respond or who are not suitable candidates for conventional therapy.
Several reviews on hyperthermia have appeared in the literature.(2,3,9-12) These reviews discuss both clinical results as well as extensive background material on hyperthermia, effects on cells in vitro, and results of small-animal studies. This assessment presents:
Scientific rationale for the use of hyperthermia
Description of devices used to produce either systemic or local hyperthermia
Reports of toxicity associated with hyperthermia treatment
Review of available information on:
Effects of isolated heated perfusion of melanomas of an extremity
Results of systemic hyperthermia and chemotherapy
Results of local/regional hyperthermia and chemotherapy
Results of a combination of hyperthermia, chemotherapy, and radiotherapy (triple therapy) and the results of hyperthermia alone
Overall assessment of the efficacy of hyperthermia and chemotherapy for the treatment of cancer
If tumor cells were more sensitive to the detrimental effects of heat than normal cells, this would obviously be a rationale for the use of hyperthermia to treat cancer.(13) Indeed, a number of such reports have appeared in the literature. These in vitro studies are rather straightforward because pure populations of either normal or cancer cells can be prepared.(10,14) Doing similar studies in vivo is far more difficult because of the complex admixture of the normal neoplastic cells during heat treatment and the inability to do careful measurement of surviving cell populations. A phenomenon that may have some practical importance is thermotolerance. Thermotolerance is defined as a heat-induced cellular resistance to subsequently elevated temperature.(2) Thus in therapy where the heat treatments are fractionated, careful timing between fractions may be necessary to obtain the desired effects.
The mechanisms of cellular damage after hyperthermia are multiple, complex, and poorly understood. The reasons for the differential sensitivity of tumor cells compared with normal cells are even more obscure. Changes in a number of intracellular biochemical pathways; effects on DNA and RNA, phospholipids, membrane fluidity, and DNA binding proteins; production of heat-shock proteins; and cytoskeletal changes have all been studied, (13). but rigorous studies in this complex area are lacking.
Synergistic
|
Additive
|
Less than additive
|
[a] If heat precedes drug.
The scheduling of hyperthermia and drug administration also is important. In vitro, simultaneous addition of the two modalities appears most effective, whereas in vivo animal experiments show superior results when the drug is given just before hyperthermia. In clinical practice optimal scheduling may be harder to determined and may depend on the method of drug delivery, bolus vs continuous perfusion, and time of peak concentration of the drug in the tissue; again, these parameters may vary from drug to drug.(12,13)
Changes in blood vessels and blood flow have been invoke as yet another explanation for the beneficial effects of hyperthermia on cancer.(15,16) Most of the information in this area is derived from mouse and dog studies. It is not clear whether responses of human blood vessels, either normal or in tumors, are the same.(2) The following observations, made in animals, are purported to explain the beneficial effects of hyperthermia: tumors may have inefficient heat transfer mechanisms because of changes in the microvasculature (therefore, the tumor becomes hotter than the surrounding normal tissues); tumor vasculature may be more easily destroyed by heat, and therefore, the tumor becomes anoxic; vasodilation after hyperthermia id more efficient in normal tissues, and therefore, they remain cooler.
Enhancement of the immune response by hyperthermia is another explanation for the possible therapeutic benefit of this modality. A number of different studies have investigated the role of hyperthermia in the immune response and the results are relevant to the question of how hyperthermia might be beneficial for cancer therapy. Both positive and negative effects of hyperthermia on various aspects of the immune response have been observed.
Some examples of positive results include the following data. Macrophages can exert cytotoxic effects on U937, a lymphoma line. Incubation of the macrophages at 40 degreesC instead of 37 degreesC greatly enchances their cytotoxicity; moreover, lymphokine-stimulated macrophages also demonstrate the same phenomena.(17). When human lymphocytes are cultured at 40 degreesC, their proliferative responses to mitogens are increased. In addition, their ability to kill allogenic cells is also enhanced.(18)
The antigenicity of B 16 melanoma cell membranes is increased by microwave hyperthermia compared with unheated membrances. This may be a useful strategy for preparing melanoma vaccines.(19)
The effect of hyperthermia was investigated in different T-cell subsets. Hyperthermia enhanced the mitogen-induced proliferation of Lyt-1(+) cells but decreased the proliferation of Lyt-23(+) cells. In contrast, B-cell responses to lipopolysaccharide were uniformly inhibited. Thus, the effects of hyperthermia on various subcomponents of the immune response are complex.(20)
The effects of hyperthermia on interleukin-1 (IL-1) production were studied in mice. Mice subjected to hyperthermia showed a transient fall in IL-1 and then a rebound at 16-20 hours. IL-1 levels remained elevated for 5-6 days.(21)
However, other studies have revealed mixed or negative effects of hyperthermia; a few examples are provided below.
Individuals with elevated temperatures were examined for mitogen-induced lymphocyte proliferation, the function of leukocyte-inhibiting factors, and the production of plaque-forming B cells (stimulated by pokeweed mitogen). All these responses were decreased in a dose-response fashion in febrile patients.(22)
Tumor necrosis factor is a potent cytokine capable of killing certain types of tumor cells. When fibrosarcoma cells were heated they became less susceptible to killing by TNF. Also, heat transiently decreased the ability of T cells to secrete TNF, and thus heating of cells can downregulate their cytotoxic functions.(23)
The effects of hyperthermia were investigated in human peripheral blood by heating cells to various temperatures and then testing their function. Natural killer activity and mitogen-induced blastogenesis both were markedly decreased by hyperthermia. This result was in part explained by decreases in NK cell subsets in the heated population of cells. The authors concluded that these facts must be considered when hyperthermia is used for cancer therapy.(24,25)
The effects of hyperthermia on mouse T-lymphocyte cytotoxic function were examined. At temperatures of 43 degreesC for 20 minutes this function was markedly decreased; however, when the cells were later incubated at 37 degreesC the function of these cells recovered.
Another study looked at the effects of hyperthermia on tumor metastases. A rat limb bearing a Yoshida sarcoma was heated to 42 degreesC, and central body temperature was also increased. Under these circumstances there was an increase in metastases and a decrease in lifespan of these animals compared with controls.(26)
In sum, it would appear that hyperthermia can both augment and depress certain aspects of the immune response as well as the biology of the tumor.
Very few studies of these changes have been performed in patients receiving hyperthermia. Therapy using immune modulation together with hyperthermia in patients has not been reported, and no studies were found that described immmunologic studies at the tumor site itself in patients undergoing hyperthermia. Therefore, although enhancement of immune functions has been mentioned often as a rationale for the use of hyperthemia to treat patients with cancer there is little substantial evidence to support this view.
A number of methods are available, each with different advantages and disadvantages, ease of use, and cost. As of August 1990 the FDA's Center For Devices and Radiological Health has given premarket approval for 11 devices for use in hyperthermia treatments of cancer.
The physical parameters involved in the delivery of heat to a particular area of the body are very complex because heat energy diffuses from the location where it is deposited. Moreover, heat continuously modifies in a dynamic and ever-changing manner the diffusion characteristics of the energy in tissues being heated. The definition of a "Thermal Dose" produced by the multitude of devices in use is also a complex problem and no uniformity of agreement exists in this area. The specific absorption rate, which is the absorbed heat per unit mass in tissue, is commonly used to compare the heating ability of different systems.(27,28)
In addition to the devices necessary to produce the heat, other important ancillary pieces of equipment are required. These include computer feedback controls and several typers of probes to measure temperature, pH, and blood flow in the tissue being heated. Coupling devices to transfer the energy to the patient and data storage display systems are also necessary. Therefore, for the proper application of hyperthermic technology, a team consisting of biologists, physicists, engineers, and clinicians is necessary. The goals of the therapy is to produce tumor tissue temperatures between 42.5 degreesC and 43 degreesC for 30-60 minutes per treatment. As many as 10-15 treatments during a course of therapy may be necessary.(16) The hyperthermia treatments have to be done in a coordinated schedule with radiotherapy and/or chemotherapy.
One of the most difficult problem in the delivery of hyperthermia to either superficial or deeper lesions is the lack of uniform temperature distribution within the heated tissues and the difficulties in accurately monitoring the temperatures within large and geometrically complex tumors. Because invasive probes have to be used to measure the temperature in the tumor, only a limited number are feasible. Moreover, the temperatures in a tumor may change within the allotted time of an individual treatment session, which usually lasts 30-60 minutes. Care must also be taken not to overheat normal tissues.
There are several different types of heating systems used. The three physical modalities employed for power deposition in local and regional clinical hyperthermia are ultrasound at frequencies of about 0.3-3 MHz, electromagnetic fields at radiofrequencies of less than 300 MHz, and electromagnetic radiation at microwave frequencies of 300-2,450 MHz. Associated with the use of the above modalities are several physical processes, including radiant, capacitive, inductive, and resistive (ohmic) heating.(16,29)
Mechanical heating with ultrasound. Tissue heating by ultrasound may be carried out with external transducers, appropriately coupled to the surface of the body. Ultrasound is significantly more penetrating in fat than in muscle tissue. For example, at 1 MHz the penetration depths are 4.4 cm in muscle and 31.3 cm in fat. The short wavelength in tissue, less than a few millimeters, makes it possible to focus ultrasound energy in small volumes at large depths. Because of the large impedance mismatch between soft tissue and air and between soft tissue and bone, these interfaces cause almost complete reflection of the ultrasound energy. The anatomic location where ultrasound can be applied are thus limited.(30)
Electrical heating. Capacities radiofrequency techniques: A technique often used in diathermy at the industrial, scientific, and medical (ISM) frequencies of 13.56 and 27.12 MHz is the application of a pair of capacitor plates excited by the radiofrequency generator. The tissue is heated by means of displacement currents produced by the electric field in the tissue between the plates. Theoretically, the technique has a potential for deep heating in a homogeneous medium.(31)
Inductive radiofrequency techniques: When an oscillating magnetic field is applied to a biological medium, eddy currents are generated that produce heating in the tissue. The most conventional type of applicator is the so-called pancake coil often used in diathermy at 13.56 or 27.12 MHz. This applicator is suitable only for heating rather superficial tissue volumes.(31)
Radiative electromagnetic techniques: Techniques based on radiative electromagnetic apertures have been used mostly at the microwave ISM frequencies of 2,450 MHz, 915 MHz, and 434 MHz and at the ISM radiofrequencies of 13 and 27 MHz as well as other radiofrequencies. Radiative apertures working at lower frequencies have, however, also been used for regional deep heating. A technique with a multiple phased applicator in a circular configuration (annular phased array) has also been described.(31)
Because externally applied energy may not be adequate to reach certain anatomical areas, a variety of invasive techniques have also been used to deliver the These include perfusion of the extremities with extracorporeally heated blood (to be discussed in more detail below); the irrigation of the urinary bladder with heated saline; and other intracavitary methods, e.g., the perfusion of heated fluid into the peritoneal cavity. Interstitial techniques are accomplished by placing the heating element directly into the tumor. Thus, a therapeutic temperature can usually be achieved without appreciable heating of normal tissues, regardless of the treatment geometry. Temperature fluctuations caused by nonuniform blood perfusion or other inhomogenecities can be reduced by placing several heating probes in the tumor and adjusting the rate of heat production of each of them to the appropriate level. With this technique one can also generate complex temperature distributions if so desired. In some cases it may be beneficial to heat the center of a lesion, which may be all malignant tissue, to a high temperature in order to deliver a tumor-killing hyperthermia treatment. Several different devices can be used to deliver heat by these invasive interstitial techniques.(32).
When certain of these devices are used, parts of the patient and the operators of the equiptment have to be protected against the excess exposure to the electrical and acoustic energy being produced by these devices. Therefore, proper survey equipment and safety precautions have to be in place to monitor and standardize the many aspects of this type of therapy and to provide quality assurance (QA).(33). In 1983 the NCI established the Hyperthermia Physics Center (HPC) to develop the Hyperthermia Quality Assurance Program. The purpose of this program is to 1)establish a QA protocol, criteria, and guidelines, 2) provide periodic reviews in participating/cooperating facilities, 3) ensure uniformity of data in cooperative clinical trials, and 4) assist individual facilities in implementing their own QA. The HPC is headquartered at Allegheny-Singer Research Institute, Pittsburgh, Pennsylvania.
Some clinicians are also using total body heating combined with chemotherapy for the treatment of cancer. A number of methods have been used to accomplish this. These include placing the patient in a radiant heating chamber, which is heated with copper electrical wires that are wrapped around the metal liner of the chamber.(34) A simpler technique is to place the patient in a large tub of hot water. The patient can also be placed in a special suit containing hot water, or, instead of water, a bath of melted paraffin can be used.(35)
A various kinds of electric blankets have also been used. A more invasive and radial technique to raise whole body temperature is the use of heated blood delivered via a high-flow arteriovenous shunt. A hig temperature can be reached in a few minutes using the device.(36)
Whole body hyperthermia is a complex, labor-intensive technique. The patient may have to be anesthetized and intubated, and careful monitoring is necessary. Thus multiple sessions of hyperthermia may be difficult to accomplish.
In a large phase 1 trial of different types of equipment for the hyperthermia treatment of cancer, treatment-associated toxicity occurred in 56 of all treatment. Forty percent of these toxicities were minor and treatment interruption was not necessary. Types of toxicity noted were pain, erythema, first-and second-degree burns, thrombosis, infection, ulcers, and edema. Superficial burns healed in 2-3 days.(37) Deeper burns healed in several weeks, and long-term toxicity was not noted.() 38) A special case is the hyperthermia treatment of limbs by the use of perfusion of heated blood and chemotherapeutics drugs. Here the toxicity has been moderately severe; edema, phlebitis, nerve damage, emboli, and blistering are often seen. More rarely, arterial obstruction and rupture can take place, and in few cases, amputation of the limb has been necessary.(4,5,39)
Here the toxicities can be significantly greater, especially with temperatures above 42 degreesC. Serious liver damage, brain damage, convulsions, and high-output cardiac failure have been observed in some studies. Other adverse effects were fluid loss, decreased renal blood flow, increases in serum creatinine concentration, neurotoxicity, nausea, vomiting, and diarrhea.(34-36) Patients in more recent trials using systemic hyperthermia had less severe toxicity.
When considered in the context of the toxicities and damages of the other standard forms of cancer therapy (viz, surgery, radiotherapy, and chemotherapy), the toxicities of hyperthermia, especially regional and local hyperthermia, may be considered moderate and are usually reversible in a few days.
The combined use of hyperthermia and chemotherapy will be discussed in four sections: 1) hyperthermia combined with regional perfusion of a limb with a chemotherapeutic agent, 2) systemic hyperthermia combined with chemotherapy, 3) local/regional hyperthermia combined with chemotherapy, and 4) triple therapy, that is, the use of radiotherapy, chemotherapy, and hyperthermia. Finally, the use of hyperthermia alone will also be briefly discussed.
The advantages of regional perfusion are that a high local concentration of the chemotheraputic agent can be achieved in the tumor-bearing limb with minimal systemic toxicity.(40) Hyperthermia was first added to this regimen in 1967.(14)
Briefly, the technique use is as follows: General anesthesia is administered, the major artery and vein supplying the extremity are exposed, the patient is heparinized, and plastic catheters are inserted into the vessels. The catheters are then connected to an extracorporeal perfusion circuit; the chemotherapeutic agent is added to perfusate. A heat exchanger is included in the arterial line to regulate the perfusion temperature. Thermistor probes are placed subcutaneously to monitor the temperature of the extremity, which is maintained at about 41 degreesC. The blood is also oxygenated. An external tourniquet is placed at the roof of the extremity to minimize systemic distribution of chemotherapy. The treatment usually lasts about 60-90 minute.(5) Heated perfusion is usually performed at some time after surgery. The scheduling of surgery and heated perfusion vary with the stage of disease and with the particular clinical situations of each patient.
The great majority of the tumors treated have been melanomas. However, in a few cases sarcomas and other less common tumors have also been treated. The chemotherapeutic agent most commonly used is L-phenylalanine mustard (L-PAM), also called mephalan. It should be noted that the parenteral formulation of this drug is not approved by the FDA. However, other agents such as dacarbazine, cisplatin, actinomycin, and interferon alfa have also been used. In some cases these agents have been combined.
| Stage | Description |
| IA | Intact primary melanoma |
| IB | Primary melanoma excised |
| IC | Multiple primary melanoma |
| II | Local recurrence within 3 cm of primary site |
| IIIA | Satellite or intransit metastases excluding regional nodes |
| IIIB | Positive regional nodes |
| IIIAB | Satellite or intransit metastases with positive regional nodes |
| IV | Distant metastases |
The role of perfusion, normothermic and hyperthermic, in the curative treatment of melanoma remains controversial. Survival appears to be somewhat improved over that of surgery alone in all stages, especially with hyperthermic perfusion, but all comparisons have been retrospective and uncontrolled. In addition to the usual problems with historical controls, melanoma presents its own special problems because of its unpredictable natural history in any given individual and the multiplicity of factors known prognostic factors. It is unfortunate that hundreds of patients have been treated in uncontrolled studies. Randomized trials continue to be necessary to define this role and even then careful attention will have to paid to the distribution of known prognostic factors in each group to insure a comparable cohort of patients.
The response of melanoma to perfusion is clearly significant, however, and the response rate seems to be improved with hyperthermic perfusion. Hyperthermic perfusion appears to be a useful palliative treatment for locally advanced melanoma of the extremity, especially for which the alternative surgical therapy would be amputation.
| Year | Reference | No. of patients | Type of study | Criteria for evaluation | Results |
| 1986 | Krementz(42). | 897 [b] | Retrospective; no control group | 10-y survival | Survival at stage I 79, stage II 60, stage III 42, stage IV 8 |
| 1986 | Shiu(43). | 42 [ c] | Only patients with intransit metastases were studied; no control group | Early response; survival at 60 mo | Complete response 8, partial response 6, minor response 5, no response 10; survival at 60 mo: 10 of group 1, 62 of group 2 |
| 1986 | Fletcher(44). | 23 | Prospective nonrandomized in acral lentiginous melanoma | 5-and 10-y survival | Survival at 5-y 75.4, 10-y 58.7 |
| 1987 | Hartley(45). | 65 | Nonrandomized | Recurrence of disease; 5-and 10-y survival | Disease recurrence 61; 5-y survival 58; 10-y survival 40 |
| 1987 | Klein(46). | 15 [d] | Retrospective; no control group | Survival at 36 mo | 3 patients died; 12 patients are alive and free of disease |
| 1987 | Vaglini(47). | 24 [e] | Consecutive study; no control group | Response of tumor at 30 d | Complete response 3, partial response 7; less than a 50 response 10, no response 4 |
| 1988 | Krige(48). | 93 | Prospective nonrandomized | 5-y survival | 83 survived |
| 1988 | Ghussen(39). | 107 | Prospective and randomized [f] | Recurrences at 3.5 y | 26/54 recurrences in control group; 6/53 in perfusion group (P=.001) |
| 1988 | Stehlin(5). | 162 | Retrospective; no control group | Survival time of patient | Data shown as several curves according to stage; effective for stages II, IIIA; and IIIAB |
| 1989 | Santinami(49). | 140 [b ] | Retrospective; no control group | 6-y survival time of patient | For stage IIIA 51 vs 16 in the control group; for stage IIIAB 34 vs 16 in the historical control group |
| 1989 | Bass(50). | 14 | Retrospective; no control group | 10-y survival (range 5-14 y) | 81 survived 10 y |
a Unless otherwise noted melphalan was the chemotherapeutic drug used.
b Several different types of drug were used.
c All patients were treated with nitrogen mustard. Only 29 patients had measurable disease. Twelve of 42 patients were treated at lower temperatures and lower doses of nitrogen mustard (group 1). The other 30 patients had more optimal treatment (group 2).
d All patients were treated with cisplatin.
e All patients were treated with imidazole carboxamide (also called DTIC or dacarbazine).
f Control group had surgery of local lesions and lymph node dissection; perfusion group had same surgery followed by heated perfusion with melphalan.
As can be seen from this table, since the 1985 review(41). a number of other papers have been published, but only the one prospective controlled trial by Ghussen et al, (39). appeared in the literature. However, this single prospective controlled trial clearly demonstrated a significant advantage (increase in disease-free survival and overall survival) of surgery followed by heated perfusion with melphalan for the treatment of melanoma of the limb compared with surgery alone.
With the exception of Ghussen et al, (39). the investigators cited in Table 3(5,42-50). claimed, on the basis of uncontrolled clinical studies, that their results with this technique were superior to results --both their own and those described in the literature --with conventional treatment(surgery and/or chemotherapy). However, these claims are anecdotal and retrospective impressions and not based on rigorous prospective randomized trials.(42) In some of these trials, the earlier patients were treated with perfusion at normothermic temperatures and the later patients in the same trials were treated with hyperthermic perfusion. In several such clinical studies, these later patients had a superior result.
Hyperthermic perfusion of the limb is an expensive and complicated therapy requiring a team approach, as several hours of operating time by a vascular surgeon and a perfusionist are necessary. In addition, equipment to oxygenate and heat the blood is required. Moreover, this type of therapy appears to be most effectively performed in institutions with considerable experience in this field. As noted above, a moderate degree of toxicity is associated with this form of therapy.
The use of heated intra-arterial perfusion with melphalan (FDA-unapproved formulation) as the chemotherapeutic drug for the treatment of melanoma of the limb appears to be a useful, albeit complex and expensive, therapy. It should be performed only under an investigational new drug (IND) protocol in institutions with considerable experience in this modality of therapy.
Several reviews of this subject have been published in the past few years.(2,3,9-12) The problems in interpretation and evaluation of the literature on this topic are the same as that mentioned above; that is, there is great heterogeneity of the patients and their types of cancer, variations in techniques and drugs used, and lack of standard criteria for assessment of response. In addition, most patients treated with systemic hyperthermia and chemotherapy previously had had multiple forms of therapy, and their performance status was usually quite low.
| Year | Reference | No. of patients | Type of study | Criteria for evaluation | Results |
| 1979 | Larkin(51). | 77 [a] | Not a randomized or controlled trial | Responses were classified as objective, subjective | Objective response 43 subjective response 15, no response 42 |
| 1979 | Bull(52). | 14 | Phase 1 trial | Not provided | No tumor responses |
| 1983 | Parks(36). | 102 [b ] | Not a randomized or controlled trial | Survival and "objective evidence of regression" | 71 patients survived for 2 mo; 43 patients had regression (11 of these were complete; in 18 of patients these regressions lasted more than 6 mo) |
| 1985 | Koga(53). | 17 [c ] | Not a randomized or controlled study | Regression of tumor | Partial response 23, no change 69, not evaluable 8 |
| 1987 | Maeta(54 [d]). | 168 | Not a randomized or controlled study | Regression of tumor for at least 4 wk | In 36 patients results could not be evaluated; 24 patients died of complications of treatment; complete response 1.5, partial response 22, minor response 5. 3, no change 43; progressive disease 26.5 |
| 1988 | Robins(55). | 24 [e] | Not a randomized or controlled trial | Not stated in detail | Of 24 patients, 9 had some type of response; those ranged from partial responses, and less than partial responses to disease stabilization |
| 1989 Robins(56). | 17 [f ] | Not a randomized or controlled trial | Not stated in detail | Of the 2 partial responses, one lasted 7 mo, the other lasted 5 mo | |
| 1989 | Willnow(57). | 17 [g] | Not a randomized or controlled study | Regression of tumor | Complete response 1, partial response 5, incomplete partial response 3, no response 3, not evaluable 5 |
| 1990 | Robins(58). | 8 [h] | Not randomized | Responses of tumor | Complete responses 3, partial response 4, less than partial 1 |
a Patients also were treated with either radiation or a wide variety of chemotherapeutic agents or hormones.
b Criteria for regression or responses were not provided.
c All patients had far advanced gastrointestinal cancer; they also received 5-fluorouracil, mitomycin-C, or cyclophosphamide.
d All patients were treated with concomitant chemotherapy of several different types.
e All patients were treated with Lonidamine.
f All patients were treated with interferon alfa.
g All patients were children; they were made hyperglycemic and were treated with several different types of chemotherapeutic agents.
h All patients had B-cell neoplasms and all received total body radiation in addition to hyperthermia.
In summary, systemic hyperthermia together with chemotherapy has been used over the past 15 years to treat several hundred patients. These patients were almost always in advanced stages of cancer, and no randomized controlled studies have been performed. The technology is cumbersome and difficult to use and has a significant degree of toxicity. At the present time, this form of therapy must be considered to be of indeterminate effectiveness. Devices so employed are considered to be investigational by the FDA and are subject to the Investigational Device Exemption (IDE) regulations.
Compared with the types of trials discussed above, the use of local/regional hyperthermia combined with chemotherapy presents the evaluator with an even more complex and heterogeneous group of factors to be considered.
| Year | Reference | No. of Patients | Type of Study | Device used to produce hyperthermia | Quantification of Thermal Dose | Criteria for Evaluation | Results |
| 1979 | Arcangeli(59). | 15 [a ] | "Random controlled study" | "High-frequency wave apparatus | Probe measured temperature of lesion; 42 -43 degreesC was reached | Response of individual nodes in the patients [b] | After 4 mo 0/11 nodes treated with ADM or BLM alone showed a response; 4/11 nodes treated with either ADM or BLM and hyperthermia showed a response |
| 1982 | Storm(60). | 10 [c] | Not a randomized controlled trial; results were compared with historical controls | Magnetrode, magnetic loop induction device | Indirectly estimated [d] | Regression of tumors duration of response | Completed response 10. partial response 20, no change 50, progressive disease 20, duration of response ranged from 3-14 mo, median survival time 8 mo (this was not different than that of patients treated with DT |
| 1984 | Moffat(61). | 14 [e] | Not a randomized or controlled trial | 11 ms-10 model, 13.56 MHz | Temperature measured in tumor | Regression of tumor | Completed response 21.4, partial regression 7.1, stable disease 0, progression, 71. 4 |
| 1984 | Kohno(62). | 65 [f] | Prospective randomized study | Microwave 2.45 GHz | "Local tissue temperature raised to about 43 degreesC" | Regression of tumor [g] | In patients with primary cancer: chemotherapy group 40, combination therapy 81.3, in patients with recurrent cancer: chemotheraphy group 14.3, combination therapy 42.9, patient prognosis was however poor in both group, most of the patients were dead after 2 y |
| 1985 | Storm(63). | 405 [h] | Not a randomized or controlled trial [i] | Magnetrode 13.56 MHz | Highest temperature sustained during therapy | Regression of tumor, change in performance status | Complete response 3.75, partial response 12.5, minimal response 6.5, stabilization 39.5 progression 37.5; mean duration of pain and activity improvement 2 mo and 3 mo |
| 1987 | Cruciani(64). | 13 [j ] | Not a randomized or controlled trial | 434MHz with a waveguide applicator | 43 degreesC for 30 minutes at a depth of 3 cm | Regression of tumor, duration of disease response | Complete response 8, partial response 46, stabilization of disease response 3 mo |
| 1988 | Koga(65). | 47 [k] | Prospective randomized | Mitomycin-C infused into peritoneal cavity | Temperature measurement of inflow and outflow tubes | Survival at 30 mo | Mitomycin-C group had a survival of 83; control group had a survival of 67.3 (these numbers are not significantly different) |
| 1989 | Pilepich(66). | 12 [l] | Clinical pilot study | Microwave 915 MHz or ultra sound 1 MHz | Probe measured temperature in tumor; aim was to reach a temperature of 43 degreesC [m] | Tumor response | Complete response 33; partial response 33; no response for remainder; most patients were dead within 4 mo |
| 1989 | Petrovich(67). | 14 [n] | Not a randomized or controlled trial | BSD100 annular phased array 55 MHz | Thermal dose was defined as the number of minutes at 42 degreesC [o] | Regression of tumor measured at 1 mo | Complete response 7, nominal response 7, progressive disease 36, not evaluable 50 |
| 1989 | Green(68). | 9 [p] | Not a randomized or controlled trial | External microwave device 915 MHz or annular phased array 70 MHz | Temperature measured in tumors was averaged | Regresional of tumor | Partial response 44, minor response 11, no response 34, progressive disease 11 |
| 1989 | Petrovich(69). | 42 [q] | Not a randomized or controlled trial | BSD annular phase array | Thermal dose was defined as the number of minutes at 42.5 degreesC | Regression of tumor | Complete 2, partial 10, nominal 17, none 0, progression 26, not evaluable 45 |
| 1990 | Nagata(70). | 67 [r] | Not a randomized or controlled trial | Eight MHz RF capacitive heating equipment | Average maximum temperature was measured with a probe [s | Regression of tumor | Complete response 17, partial response 22, nonresponse 61; progressive disease 10 |
a These patients had a total of 29 neck node metastases. Patients with tumor with a doubling time of less than 30 days were treated with adriamycin (ADM), those with tumore with doubling time of more than 30 days were treated with bleomycin (BLM); 8 patients were in first group, and 7 patients were in second group.
b Eight nodes treated with ADM nd hyperthermia, 8 nodes with ADM alone, 7 nodes with BLM and hyperthermia, 6 nodes with BLM alone.
c All patients had metastatic melanoma to liver and all received DTIC in the hepatic artery.
d Probes could be placed in only 3 patients.
e All patients had head and neck tumors. Methorexate and other types of chemotherapy were used.
f Twenty-seven patient with gynecologic malignancies were treated with chemotherapy alone, 38 were treated with combination of chemotherapy (BLM and mitomycin) and hyperthermia.
g Criteria for regression were not stated in detail.
h A total of 1,170 patients with a wide variety of tumors were treated; of these 405 received hyperthermia nd chemotherapy
i Was a 5-year multi-institutional national cooperative trial.
j Patients also received cisplatin, BLM, and methotrexate.
k All patients had gastric cancer removed surgically; they all had macroscopic serosal invasion. Thirty-two patients received heated perfusion with mitomycin-C. Twenty-eight patients did not receive mitomycin-C.
l Patients also received BLM.
m Not stated whether this aim was reached.
n All patients had metastatic disease to liver. Patients also given 5-fluorouracil or cisplatin.
o Optimal temperature was not reached in most patients.
p Patients with variety of tumors treated with
q Part of a multi-institutional trial, 1980-1986; total number of patients treated was 353, of which 42 received hyperthermia and chemotherapy; drug most frequently used has a combination of 5-fluorouracil and cisplatin; patients had many different types of malignancies.
r Tumors consisted of 41 heptocellular carcinomas, 6 cholangiocarcinomas, and 20 metastatic tumors. All patients were treated with some form of chemotherapy. Some patients also were treated with arterial embolization with starch microspheres. Other treated with doxorubicin or OK -432, an immunomodulator.
s In many cases thermometry of liver was not satisfactory.
Most of these trials were not randomized controlled trials and are therefore difficult to interpret. However, there were three trials that could be considered as prospective controlled trials. The first involved only 15 patients with head and neck tumors; 7 of these patients received chemotherapy and hyperthermia.(59). In these patients only 11 lymph nodes were evaluated for regression after therapy for a period of 4 months.
In another study, Kohno et al(62). evaluated 65 patients with gynecologic malignancies. Patients receiving chemotherapy and hyperthermia had a higher frequency of regressing lesions than patients receiving chemotherapy alone. However, the duration of responses in both groups of patients was short; most patients in both groups were dead within 2 years.
In the third controlled study of patients with surgical removal of gastric cancer, (65). one group of postoperative patients also received heated perfusion with mitomycin into the peritoneal cavity. The other group of patients had surgery only. At 30 months there was no difference in survival between the two groups.
In summary, most of the trials did not allow firm conclusions regarding effectiveness because they were not randomized controlled trials. The few controlled trials cited showed marginally beneficial responses of short duration.
The clinical effectiveness of local/regional hyperthermia combined with chemotherapy should be considered as unestablished at the present time.
Triple therapy employs a combination of radiation, chemotherapy, and hyperthermia. To assess the usefulness of triple therapy, controlled prospective studies would require studies in large groups of patients because, to get interpretable results, groups of comparable patients would have to be treated separately with either one modality, two modalities, or all three modalities of therapy. Since this has not been done yet, one cannot make as informed judgment about the efficacy of triple therapy. Only small uncontrolled trials using therapy have been published.(37,69,71) Triple therapy has not been demonstrated to be clinically effective.
A large multicenter trial has shown that the use of heat alone as a form of therapy for advanced cancer in even less effective than hyperthermia combined with either chemotherapy or radiotherapy. The author claimed, however, that hyperthermia alone can produce some degree of stabilization of disease but was less effective than when combined with other modalities of treatment.(63)
There have been a number of recent reports, but no prospective controlled trials, describing hyperthermia alone in the treatment of prostatic cancer, (72). tumors of the liver, (73). brain tumors, (74). ocular tumors, (75). and head and neck tumors.(76) The benefits of such treatment in general were minimal and of short (63,72-76). Thus limited evidence exists for the effectiveness of hyperthermia alone in the treatment of cancer, and its proper role in the spectrum of therapeutic interventions is uncertain.
As part of the assessment process, other agencies of the Public Health Service were consulted. The following is a response from the NIH:
The field of clinical hyperthermia is plagued by technical problems. Many companies and universities have invented or modified their own devices. There are many different devices available, but they have an inconsistent capacity to uniformly heat tissues to the desired temperature, except at superficial sites.
Hyperthermia when used as a localized therapy (aimed at only one region of the body at a time) can control only regional disease. However, systemic hyperthermia and chemotherapy are being investigated. Quality of life of local control end-points have usually been assessed in hyperthermia clinical trials. Survival, while being the most crucial endpoint, cannot usually be satisfactorily evaluated.
Hyperthermia by itself can result in antitumor responses in patients with breast cancer, melanoma, and other cancers metastasizing to the skin and lymph nodes. Heat can control superficial cancers.
Combinations of hyperthermia with chemotherapy or irradiation have a solid scientific rationale. However, combination studies that include other active agents or modalities are difficult to interpret. It is not possible to identify the therapeutic contribution of hyperthermia alone. In addition, the clinical trials that have been performed to date have been flawed because of relatively small sample size or heterogeneous patient populations (with respect to disease, site, stage, prior therapy, etc). Therefore, it has been difficult, if not impossible, to compare one series with another.
For nearly all diseases, there have been no randomized studies of sufficient size to isolate a hyperthermia question (relative to other reasonable treatment options) or to demonstrate a substantial additional benefit for a combination including heat.
Specifically, with respect to limb perfusion in patients with melanoma, the use of heated blood and L-PAM chemotherapy results in a large number of objective responses that seem clinically meaningful. It is not known how best to integrate this therapy into a strategy of care for the melanoma patient. Data from Germany suggest that one role would be as an adjuvant treatment in the setting of high recurrence risk for melanoma.
The following are responses from the FDA:
Center for Devices and Radiological Health:
There are no hyperthermia devices approved by FDA for combined use with chemotherapy. There are, however, investigational studies in progress being conducted at several institutions under the Investigational Device Exemption (IDE) regulations. These clinical trials have been approved to determine the safety and effectiveness of the combination treatment with hyperthermia and a specific chemotherapeutic agent.
Center for Drug Evaluation and Research:
Three IND protocols for chemotherapy and hyperthermia were submitted. One was closed and submitted as an IDE. Studies under the other IND protocols were never initiated.
It should be noted that there is no formulation for melphalan for parenteral administration (as for melanoma of the limb) currently approved by the FDA. The only approved product is an oral dosage form. However, the NCI is currently conducting trials with this product under an IND protocol.
In response to the Federal Register notice of this assessment and the solicitation of information and opinion from individuals and groups having experience with hyperthermia for the treatment of cancer, the Office of Health Technology Assessment has received a number of letters and comments concerning this technology.
Dr. John S. Stehlin (written communication, 1990, Houston, Texas) indicated that he has had 30 years of experience in treating local/regional recurrence of melanoma of the limb using heated perfusion. He is convinced that this is an effective form of therapy and that it works by causing an "internal vaccination" leading to a favorable immune response on the part of the patient against his or her own tumor.
The American College of Radiology (ACR) stated the following:
The efficacy of hyperthemia when used in conjunction with radiation therapy has been proven over time and through clinical investigation. For this reason, the ACR supports this usage and the reimbursement for such a treatment mortality. Regarding your current assessment, however, we feel that hyperthermia used alone or in conjunction with chemotherapy has not yet been proven to be effective. We believe that more clinical efficacy studies would be necessary to validate that approach before the use of such treatment techniques could be considered efficacious, and therefore, should not be considered reimbursable.
Despite an enormous effort and many studies by basic scientists, clinicians, and physicists devoted to the study of hyperthermia alone or combined with chemotherapy for cancer therapy, the usefulness of this modality of therapy remains unclear. The reasons for this are several: the complexity and heterogeneity of the patients being studied, the heterogeneity of the devices used to produce hyperthemia, the lack of standardized methods to measure a Thermal Dose, and the almost complete lack of well-done, large, rigorous, prospective, randomized, controlled trials. The problem is as follows: The reason for the lack of such large prospective randomized trials is that they are extremely expensive and complicated and the proponents of hyperthermia have not yet produced sufficient preliminary evidence in its favor to convince peer review bodies to support such trials. In the absence of definitive trials, substantial evidence is lacking that hyperthemia alone or combined with chemotherapy has an established role in the spectum of available therapies for malignant disease. The only exception is the use of hyperthermic regional limb perfusion with melphalan to treat melanomas; however, it should be noted that although melphalan has been the most widely used agent for this purpose, the intravenous formulation of this agent has not been approved by the FDA.
What is the outlook fo future controlled trials in this area? The Cancer Therapy Evaluation Program of the Division of Cancer Treatment, NCI, funds cooperative trial groups in a consortium of universities in the United States. One such group that is involved in hyperthermia research is the Radiation Therapy Oncology Group. This group is not now funding any trials of chemotherapy and hyperthermia but they stated that they may do so in the next few years. They do not plan to fund any trials of hyperthermia alone (personal communication, B.N. Emani, M.D., August 15, 1990, Washington University School of Medicine).
