Building Upon the Standard of Care in Adjuvant Therapy of High-Risk Melanoma

Abstract

Melanoma is the fifth most frequent cancer among males and the sixth most frequent cancer in females. It takes a toll in productive life years that is exceeded only by childhood cancers and testicular carcinoma. Melanoma incidence has increased consistently during the past two generations. However, the depth of primary tumor invasion—the cardinal prognostic factor of localized primary melanoma—has decreased. As a consequence, the case fatality rate has decreased from 50% to less than 10% during the past 50 years. Despite these advances, 8,000 deaths in the United States each year are attributable to melanoma. The death rate has increased 157% in the past decade among older men, a subset of the population who present with thicker lesions ( 4.0 mm) that may have evaded detection or may represent a biologically different process. Against this background, it is daunting that no therapy has ever shown a significant prolongation of survival in large, properly conducted, randomized clinical trials for metastatic disease. Only one, interferon alfa-2b (IFN -2b), has shown a significant benefit in terms of overall and relapse-free survival in multicenter randomized United States cooperative group and intergroup trials. These facts suggest that we have much to do in both patient and physician education and in development of better therapies for melanoma. Richtig et al report in this issue of the Journal of Clinical Oncology the closure of the European Cooperative Adjuvant Melanoma Treatment Study Group (ECAMTSG) placebocontrolled trial of isotretinoin and low-dose IFN compared with low-dose IFN in patients with resected stage IIA and IIB melanoma. Unfortunately, like many other recent trials of adjuvant therapy for melanoma, this trial was closed and reported early when an interim analysis demonstrated no significant likelihood of reaching the initial study goals (1.3%). With this negative experience, and a series of other negative trials recently reported in the literature or in abstracts, it is time to ask the question, How may progress be accelerated and the pitfalls of past trials avoided in future melanoma studies? The potential answers to this complex question may be found in our increasing understanding of the requirements of adjuvant clinical trial design and our scientific understanding of tumor progression, along with an analysis of current effective therapies. Only one adjuvant therapy has stood the test of time: the high-dose IFN -2b regimen tested initially in the Eastern Cooperative Oncology Group E1684 trial and reassessed in the United States intergroup E1690 and E1694 trials reported in the Journal of Clinical Oncology in 1996, 2000, and 2001. These trials have taught us many things, including the importance of prognostic factors for melanoma outcome and the necessity of concurrently randomized controlled trials for rigorous assessment of new interventions. They also point to the large time window between new trial initiation and the return of meaningful answers regarding disease-free and overall survival. The evolution of prognostic factors between trial cohorts, even when trial entry criteria have been relatively consistent in the United States cooperative groups, argues for standardization of major prognostic variables as far as is possible to optimize detection of differences related to experimental interventions. Trials conducted in the recent past have taught us that only robust trials that include sufficient numbers of patients to have the power to arrive at unequivocal conclusions are worthy of our patients’ involvement. The design of clinical trials depends on the anticipated rate of relapse and mortality for the population to be treated: the lower the risk of the group selected, the larger the number of patients needed to answer the trial question. Events among patients with resected stage IIIB to IV disease exceed those for stage IIIA and those, in turn, exceed those for stage IIB and IIA. The numbers required for trials designed to detect clinically relevant differences for survival and progression-free interval increase from 600 to 800 for stage IIIB to IV to numbers that may exceed 1,400 for stage IIIA and stage IIA/B disease. The ECAMSTG study JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 34 DECEMBER 1 2005