Prospective, Randomized, Multicenter, Double-Blind Placebo-Controlled Trial Comparing Adjuvant Interferon Alfa and Isotretinoin With Interfe
http://www.100md.com
《临床肿瘤学》
the Department of Dermatology, Medical University of Graz
Section of Medical Statistics and Department of Dermatology, Medical University of Vienna
Department of Dermatology, Danube Hospital
Department of Dermatology, Wilhelminen Hospital
Roche Austria GmbH, Vienna
Department of Dermatology, Medical University of Innsbruck
Department of Dermatology, Hospital of St P?lten
Department of Dermatology, Hospital of Wels
Department of Dermatology, Hospital of Klagenfurt
Department of Dermatology, Hospital of Salzburg
Department of Dermatology, Hospital of Lainz
Department of Dermatology, Rudolfsstiftung, Wien, Austria
ABSTRACT
PURPOSE: The combination of interferon alfa (IFN) and isotretinoin has shown a direct antiproliferative effect on human melanoma cell lines, but it remained unclear whether this combination is more effective than IFN alone in patients with metastatic melanoma. We evaluated safety and efficacy of IFN and isotretinoin compared with IFN alone as adjuvant treatment in patients with primary malignant melanoma stage IIA and IIB.
PATIENTS AND METHODS: In a prospective, randomized, double-blind, placebo-controlled trial, 407 melanoma patients in stage IIA (301 patients) and IIB (106 patients) were randomly assigned to either IFN and isotretinoin (isotretinoin group; 206 patients) or IFN and placebo (placebo group; 201 patients) after excision of the primary tumor. IFN was administered three times a week at a dose of 3 million units subcutaneously for 24 months. Isotretinoin at a dose of 20 mg for patients 73 kg, 30 mg for patients greater than 73 kg, or placebo daily for 24 months.
RESULTS: A scheduled interim analysis revealed no significant differences in survival rates, with the isotretinoin group and the placebo group showing 5-year disease-free survival rates of 55% (95% CI, 46% to 65%) and 67% (95% CI, 59% to 75%), respectively, and overall 5-year survival rates of 76% (95% CI, 67% to 84%) and 81% (95% CI, 74% to 88%), respectively. The trial was stopped for futility.
CONCLUSION: The addition of isotretinoin to an adjuvant treatment of low-dose IFN in patients with stage IIA and IIB melanoma had no significant effect on disease-free or overall survival and is therefore not recommended.
INTRODUCTION
Interferons are a group of naturally occurring proteins with a large spectrum of biologic activities including antiviral, immunomodulatory, antiproliferative, and differentiation-inducing effects.1-4 Treatment with interferon alfa (IFN) has been shown to prolong disease-free survival in melanoma patients at moderate to high risk of developing metastatic disease after surgery (American Joint Committee on Cancer [AJCC] stage IIA to III)5 and in some studies, to prolong overall survival in this patient population.6-14
As derivates of vitamin A, retinoids also show a wide spectrum of biologic activities, including immune modulation, sebosuppression, and effects on cell proliferation and cell differentiation in both melanoma and nonmelanoma cell lines.15-19 In some clinical trials evaluating isotretinoin and etretinate for treatment of nonmelanoma skin cancers, tumor suppression was achieved,18,20 although isotretinoin was ineffective in preventing basal cell carcinomas in a large study with 981 patients.21 However, in an adjuvant treatment study,22 vitamin A was not significantly better than observation alone in prolonging disease-free survival or overall survival in patients with primary melanomas thicker than 0.75 mm and with clinically negative lymph nodes.
Because the combination of IFN and retinoic acid has shown a direct antiproliferative effect on malignant human cell lines, especially melanoma cell lines,23-26 several phase II clinical trials have been performed to evaluate the combination therapy in patients with nonmelanoma skin cancers and in patients with metastatic melanoma.27-36 In patients with metastatic melanoma disease, the effectiveness of the combination of IFN and retinoic acid was unclear: some studies demonstrated efficacy, with total response from 20% to 30%,27,32,33 whereas other studies reported no or extremely low response to the combination therapy.30,35,36
Motivated by the positive studies, we designed a prospective, randomized, multicenter, double-blind, placebo-controlled trial comparing adjuvant combination therapy of IFN and isotretinoin with adjuvant therapy of IFN alone in patients with primary malignant melanoma stage IIA and IIB AJCC/International Union Against Cancer [UICC] 198837 to investigate whether the addition of isotretinoin to IFN resulted in greater treatment efficacy.
PATIENTS AND METHODS
Study Design
The study was a prospective, 24-month, randomized, multicenter, double-blind, placebo-controlled, parallel-group trial evaluating the efficacy of adjuvant therapy of 3 million units (MU) of IFN given subcutaneously three times a week in combination with daily oral isotretinoin (Roaccutan; Roche, Vienna, Austria; isotretinoin group) compared with IFN alone (placebo group) for treatment of primary melanoma in stage IIA/IIB. The primary end point of the study was disease-free survival, and the secondary end points were overall survival and evaluation of quality of life during treatment. Disease-free survival was chosen as primary end point to get a sufficient number of events for the required power, with a limited sample size and duration of the trial. The study was sponsored by Roche Austria GmbH (Vienna, Austria). Twenty centers in Austria, Hungary, the Netherlands, and Greece participated. Patient recruitment was started on October 16, 1996, and stopped on December 31, 2002.
Four hundred seven melanoma patients who met the inclusion/exclusion criteria were randomly assigned to receive either IFN and isotretinoin (isotretinoin group; 206 patients) or IFN and placebo (placebo group; 201 patients). Treatment was started within 16 weeks after completion of all surgical interventions, including primary and secondary tumor excision.
IFN was administered subcutaneously at a dose of 3 MU 3 times a week for 24 months. Isotretinoin was given daily for 24 months at a dose of 20 mg (two 10-mg capsules) for patients with weight 73 kg and at a dose of 30 mg (three 10-mg capsules) for patients with weight greater than 73 kg. Patients in the placebo group received an equivalent number of matching placebo capsules daily for 24 months.
Radiotherapy or concomitant medication with tetracyclines, other immunotherapy, chemotherapy, or any investigational drug was disallowed during the trial. Patients were permitted to use paracetamol/indomethacine to ameliorate the flu-like symptoms associated with IFN therapy.
The following assessments were made at baseline after randomization, once a month during the first three months, then at 3-month intervals: CBC and blood chemistry (including leukocytes, hemoglobin, WBC differential count, platelets, total bilirubin, liver function enzymes, alkaline phosphatase, creatine phosphokinase, renal function, cholesterol, and triglycerides). Additionally, only at baseline, the following laboratory tests were performed: total protein, blood glucose, lactate dehydrogenase acid, high- and low-density lipoproteins, thyroid-stimulating hormone, serum electrolytes, and urine analysis. Chest x-ray, ultrasonography of the abdomen and the draining lymph nodes, and x-ray of the cervical spine, thoracic spine, lumbar spine, one knee, one ankle, and both forearms were required at baseline and at the end of treatment. Blood pressure and weight were measured at baseline and every 3 months. A visual analog scale was used to assess quality of life every 3 months during treatment and every 6 months after treatment cessation.
The trial protocol and five amendments were reviewed and approved by local ethics committees at the participating sites. The study was overseen by a Data Safety Monitoring Committee consisting of two clinical pharmacologists, a statistician, a dermatologist, and a retired expert on isotretinoin, none of whom was otherwise involved in the study. The Data Safety Monitoring Committee met regularly to review safety data, trial conduct, and protocol violations. The study was conducted in accordance with the Declaration of Helskinki and with Good Clinical Practice guidelines. All patients received a detailed explanation of the potential risks and benefits of the study and provided signed informed consent before participating in the study.
Patients
Initially all patients between 18 and 75 years of age with primary melanoma in stage IIA or IIB as defined by AJCC/UICC 1988 (IIA, localized melanoma with tumor thickness of 1.51 to 4.0 mm or Clark level IV; IIB, localized melanoma > 4.0 mm tumor thickness or Clark level V)37 were eligible for the trial. Because the performance of sentinel node biopsy in routine settings was steadily increasing after the start of the study, the protocol was amended (amendment 2; May 10, 1999) to permit patients to be enrolled after sentinel node biopsy, provided that they were sentinel node–negative or, in case of positivity, that positivity revealed only micrometastasis and that subsequent radical lymphadenectomy showed negative lymph nodes.
The patients had to meet also the following inclusion criteria: baseline laboratory results including leukocyte count (WBC) 3.0 x 109/L, platelets count 100 x 109/L, hemoglobin 10 g/100 mL (1 dL), cholesterol 300 mg/dL, and triglycerides 250 mg/dL. Women of child-bearing potential were informed orally about the teratogenic effect of retinoids and given a patient information sheet. Women of child-bearing potential were required to use an accepted method of birth control for at least 1 month before beginning therapy, during treatment, and for a minimum of 3 months after discontinuation of therapy. Women of child-bearing potential were also required to have a serum pregnancy test performed by a physician within 2 days of treatment initiation and monthly until 3 months after treatment cessation.
Pregnant or lactating women, individuals younger than 18 years, and individuals with psychiatric disorders, depression, or seizure disorders or compromised CNS function were excluded. Persons with history or presence of autoimmune disease, with serious infections within previous 28 days, and with severe systemic disease (such as cardiac disease, severe liver disease, severe renal disease, or myeloid dysfunction) were also excluded. Prior immunotherapy, chemotherapy, or therapy with an investigational drug within 3 months of study initiation was not permitted. Presence of neoplastic disease within the previous 5 years was an exclusion criterion, except for basal cell carcinoma of the skin, actinic keratoses, and carcinoma-in-situ of the skin or cervix, provided that these had been cured by surgery.
Randomization
Patients were centrally randomized by the Department of Medical Statistics at the University of Vienna. Major inclusion criteria were checked before randomization. A biased coin randomization was applied to balance treatment groups for sex, stage (IIA v IIB), age ( 50 v > 50 years), weight ( 73 v > 73 kg) and study center. The allocation probabilities to the two treatments were calculated on the basis of an imbalance score defined as a weighted sum of the differences of the treatment allocation numbers in the strata defined by the five risk factors. Unconstrained randomization was used in the beginning.38
Blinded medication was prepared by the galenic division of Hoffmann-La Roche, Basel, Switzerland, and stored at Roche (where there was no access to the code). The randomization process generated a patch number of the patient's blinded medication, which was printed and faxed to Roche for distribution to the recruiting center. In this way, all patients and investigators were blinded regarding treatment allocation. Each center received a sealed envelope containing patient number and treatment allocation, which could be opened only in case of a serious adverse event that necessitated disclose of the treatment. A total of 20 envelopes were opened before the end of the study or were missing when the study was terminated.
Compliance
IFN was given monthly under the trade name used in the participating countries and injected by the patients themselves. Study participants were instructed to store IFN in the refrigerator from 2°C to 8°C. Patients were supplied with oral study medication (isotretinoin or placebo) monthly, and instructed to store the medication at room temperature. Any unused capsules were returned at each visit. Compliance was calculated at the end of the trial from pill counts and was defined on the basis of the intake of blinded study medication (isotretinoin/placebo). For each patient, a compliance ratio (actual cumulative medication intake/scheduled cumulative medication intake) was calculated. In the case of missing data, the medication intake was set to zero. Patients with a compliance ratio less than 70% were excluded from the per-protocol population.
Follow-Up Evaluation
After end of treatment patients were evaluated monthly for 3 months and than every 3 months until the third year after study initiation. Afterward, visits took place every 6 months until 5 years after study initiation.
Disease-free status was maintained when there was no hint of disease progression by either clinical assessment or imaging procedures.
Statistics
The main analysis was performed for all randomized patients. Survival and disease-free survival were compared between the treatment groups with the log-rank test (SAS procedure LIFE-TEST; SAS Institute, Cary, NC) at the two-sided level .05. Disease-free survival was defined as the time from start of treatment to the time of progression or death, whichever occurred first.
Patients who were lost to follow-up were censored at the time of the last observation. Kaplan-Meier curves and 5-year survival rates with 95% CIs including CIs for the differences in the 5-year survival rates were calculated. In addition, a multivariate analysis was performed based on the proportional hazards model, with stratification by center and using as covariables melanoma stage, sex, age, and weight. Centers that recruited 20 or fewer patients were combined (summarized) to a group (small centers) per country. Patients from Greece and the Netherlands were combined in a single group. For the subgroup of patients for whom sentinel lymph node biopsy results were available, an additional analysis was performed using a proportional hazards model with the additional factor sentinel lymph-node-positive/negative. Differences between the treatment groups in quality of life during treatment were tested with an independent samples t test, comparing the area under the curve relative to the baseline value of the quality-of-life measurements on a visual analog scale. The rates of patients experiencing adverse events were compared with 2 tests. Missing values were imputed according to the last observation carried forward principle. In the interim analysis a significance level of .01 (two-sided) was applied. Additionally, it was laid down to stop the study for futility if the conditional power (assuming a difference of 10% in the proportion of disease-free patients after 5 years) was less than 20%. The level for the final analysis was set to .05 (two-sided).
Sample Size
A total sample size of 400 patients was planned, which would give a power of approximately 80% for detecting a 10% difference in 5-year disease-free survival rate, assuming a 75% disease-free survival rate in the placebo group, a recruitment period of 4 years, a total study period of 9 years, and exponential disease-free survival.
Collection of Data and Data Management
Regular monitoring of the investigational sites was performed. The data were stored in a relational database (Oracle) located in the Statistical Center at the Department of Medical Statistics at the University of Vienna. The database was closed on February 29, 2004.
RESULTS
Patients
A total of 407 patients were recruited, with 206 patients randomly assigned to the isotretinoin group and 201 patients randomly assigned to the placebo group. Baseline characteristics of the patients are listed in Table 1.
The 407 patients were recruited in 20 centers as follows: 368 patients from 11 centers in Austria, 34 patients from seven centers in Hungary, three patients from one center in Greece, and two patients from one center in the Netherlands. The median number of recruited patients of the six largest centers was 40 patients, and six centers recruited three patients or fewer.
Interim Analysis
A scheduled interim analysis performed 9 months after the end of the recruitment period showed a conditional power of only 1.3%, far below the threshold 20% laid down in the protocol. Consequently, a decision was made to stop the trial immediately for futility, to examine the patients currently enrolled, and to perform a final analysis within the next 3 months.
Final Analysis
Disease-free survival. Disease progression was observed in 67 patients in the isotretinoin group and in 53 patients in the placebo group, and two patients in each group died without progression. There was no significant difference in disease-free survival between the treatment groups (P = .25, log-rank test, two sided). The 5-year disease-free survival rates were 55% (95% CI, 46% to 65%) in the isotretinoin group and 67% (95% CI, 59% to 75%) in the placebo group (Fig 1). The rounded difference in the 5-year disease-free survival rates was –11 percentage points (95% CI, –24% to 1%).
Overall survival. Survival rates did not differ significantly between the treatment groups (P = .8). During the study period, a total of 57 patients (30 in the isotretinoin group and 27 in the placebo group) died, 53 of these patients (28 in the isotretinoin group and 25 in the placebo group) from disseminated melanoma. The 5-year survival rates were 76% in the isotretinoin group (95% CI, 67% to 84%) and 81% in the placebo group (95% CI 74% to 88%). The difference in the 5-year survival rates was –5 percentage points (95% CI, –16% to 6%).
Prognostic factors. In the multivariate analysis, the only significant prognostic factors were stage (P < .0001; hazard ratio for stages IIB/IIA = 2.33; 95% CI, 1.60 to 3.39) and sex (P = .027; hazard ratio for male/female = 1.62; 95% CI, 1.06 to 2.50). The multivariate analysis showed no significant differences between the treatment groups (P = .81; hazard ratio for isotretinoin/placebo = 1.05; 95% CI, 0.72 to 1.52).
Figure 2 shows the Kaplan-Meier estimates by stage (IIA v IIB) and sex. Clearly there was a worse prognosis for stage IIB. Sex seemed to be an independent prognostic factor, with men showing an inferior outcome compared with women.
The analysis for the 158 patients for whom sentinel lymph node biopsy results were available showed no significant differences between treatment groups (P = .39; hazard ratio for isotretinoin/placebo = 1.41; 95% CI, 0.65 to 3.08). Stage was again a significant prognostic factor (P = .02; hazard ratio for stage IIB/IIA = 2.52; 95% CI, 1.16 to 5.48), and significantly worse prognosis was seen for patients with positive sentinel lymph node (P = .02; hazard ratio for sentinel lymph node–positive/negative = 2.31; 95% CI, 1.15 to 4.67).
Per-Protocol Analysis
As specified in the analysis plan, the data were also analyzed in the per protocol population. This analysis confirms the findings from the all randomized patient population. Reasons for exclusion from the per-protocol population were laid down in a blinded review. A total of 173 patients (isotretinoin group, n = 90; placebo group, n = 83) were excluded from the per-protocol population, with no substantial difference in the exclusion pattern under isotretinoin and placebo. A detailed flow chart of the reasons of exclusion from the per-protocol population is shown in Figure 3.
In the per-protocol analysis with the remaining 234 patients, no significant differences were detected in either disease-free survival or overall survival (P = .61 and P = .25, respectively, log-rank test, two sided). The 5-year disease-free survival rates were 65% (95% CI, 54% to 76%) in the isotretinoin group and 71% (95% CI, 62% to 80%) in the placebo group. The difference in the 5-year disease-free survival rates was –6 percentage points (95% CI, –20% to 8%).
The 5-year survival rates were 83% (95% CI, 74% to 92%) in the isotretinoin group and 77% (95% CI, 66% to 88%) in the placebo group. The difference in the 5-year survival rates was 7 percentage points (95% CI, –8% to 21%).
Adverse Events
A higher number of adverse events were reported in the isotretinoin group (889 events) than in the placebo group (740 events). This difference was due to higher incidences in the isotretinoin group versus the placebo group of both cheilitis/xerosis (175 v 44 events, respectively) and hyperlipidemia (56 v 34 events, respectively). 84% percent of the patients in the placebo group and 91% of the patients in the isotretinoin group observed at least one adverse event (P = .03).
Table 2 lists serious adverse events (WHO grades 3 or 4) reported in the study. Twenty-one percent of the patients in both the placebo and the isotretinoin group experienced at least one adverse event of grade 3 or 4 (P close to 1.00). Three life-threatening adverse events (WHO grade 4) were observed during the study. In the placebo group, one patient experienced an ischemic brain stroke after 1 month of treatment, and one patient experienced dyspnea at the start of treatment. In both patients treatment was stopped immediately, and both patients showed complete recovery. In the isotretinoin group, one patient experienced severe depression after 6 months of treatment. Treatment was continued along with antidepressant medication and the patient recovered. Similar numbers of WHO grade 3 serious adverse events were reported in the isotretinoin group (67 events) and in the placebo group (80 events).
Minimal to moderate adverse events (WHO grades 1 to 2), including laboratory disorders, general disorders, skin and tissue disorders, infection, inflammation, gastrointestinal disorders, hormonal disorders, neurologic disorders, cardiac disorders, circulatory disorders, and ophthalmologic disorders were seen in 821 reports in the isotretinoin group and in 658 reports in the placebo group. All disorders were distributed equally within both groups with the exception of cheilitis/xerosis and hyperlipidemia more often seen in the isotretinoin group.
One patient became pregnant during the study. Treatment was immediately stopped, and the treatment assignment was unblinded by the center, revealing that the patient had been in the placebo group. This patient had a spontaneous abortion after 4 months. Another patient, this time in the isotretinoin group, became pregnant 6 months after treatment cessation and delivered a healthy boy.
Four patients (two in each group) died of causes unrelated to melanoma. One patient in the isotretinoin group died of cardiac decompensation 33 months after the end of treatment. A second patient in the isotretinoin group died of myocardial infarction 13 months after start of treatment. In the placebo group, one patient died in a car accident and one patient died of cholangiocellular carcinoma. None of the four deaths were considered drug-related.
Quality of Life
Quality-of-life measurements at baseline, 6 months, and 2 years were available for 68%, 63%, and 31% of the patients with only marginal differences between treatment groups. The area under the baseline values for the visual analog scale to evaluate quality of life showed no significant difference between treatment groups (P = .94, two-sided t test).
Secondary Malignancies
Secondary malignancies reported in the isotretinoin group (five patients) included secondary melanoma (three patients), basal cell carcinoma (one patient), and gastric cancer (one patient). Secondary malignancies reported in the placebo group (10 patients) included basal cell carcinoma (four patients), squamous cell carcinoma (one patient), breast cancer (one patient), cholangiocellular carcinoma (one patient), mixed carcinoma (one patient), multiple myeloma (one patient), and cervical dysplasia (one patient).
DISCUSSION
The optimal care for melanoma patients with moderate and high risk to develop metastasis (stage IIA and IIB AJCC/UICC 1988)37 and for patients with micrometastasis in the sentinel lymph node (stage IIIA AJCC)5 has been a matter of debate.6,10,11,14,39,40 IFN is the drug most frequently prescribed for adjuvant treatment of melanoma, but currently there is no consensus on the optimal regimen. Prolongation of overall survival in patients with moderate and high risk has been demonstrated in studies evaluating high-dose IFN treatment.6,12-14,41 However, although high-dose IFN treatment regimens are widely used in the United States and, according to some, should be the gold standard of adjuvant therapy,4,40 the high cost of high-dose IFN treatment, the toxicity profile (leading in some cases to the death of disease-free patients),42,43 and conflicting efficacy results have created a dilemma for dermatologists and oncologists practicing in Europe. Consequently, low- and intermediate-dose IFN regimens are still under consideration in Europe.14 Some studies evaluating treatment of this patient population with low-dose IFN showed prolongation of disease-free survival,7-9 but others did not.13,14,44 The question "high-dose, low-dose, no dose, which dose?" for the adjuvant treatment39 has not yet been resolved, although low-dose IFN at a dose of 3 MU 3 times a week is registered for adjuvant treatment in the participating countries.
The aim of our study was not to attempt to resolve the issue of optimal IFN dosage, but to look carefully at the question of whether the addition of another active antitumor agent to an adjuvant low-dose IFN regimen widely used in Europe could significantly improve both disease-free survival and overall survival in patients with moderate to high risk of developing metastatic disease. We chose to add isotretinoin, a derivate of retinoic acid, with sebosuppressive, anti-inflammatory, cell differentiation, and immunomodulatory functions, which has been shown to be effective in the treatment of severe acne and other inflammatory diseases.16,18 In both laboratory studies and clinical trials, retinoids have been shown to have antineoplastic activity, inhibiting growth of tumor cells directly or by induction of differentiation, inhibiting cell invasion of melanoma cells, influencing the expression of tumor cell surface markers, and inducing apoptosis in leukemic and solid tumor cells.16-20,45
Our patient population consisted of patients in stage IIA and stage IIB, based on the definition of localized melanoma with tumor thickness from 1.51 to 4.0 mm or Clark level IV (IIA) or localized melanoma with tumor thickness greater than 4.0 mm or Clark level V (IIB) without nodal metastases (AJCC/UICC 1988).37 A total of 301 patients (74%) were in stage IIA and 106 patients (26%) were in stage IIB, a percentage also comparable to both other large studies in those stages of the Austrian and the French groups.7,8 During the course of the study, the definition was modified to include also patients with micrometastasis in the sentinel lymph node and the study protocol was amended to permit the enrollment of those patients. Similar numbers of men and women were enrolled onto the study, with a slight predominance of men (53%), as also seen in other studies.7,8
Patients with stage IIA to IIB disease have an estimated 5-year disease-free survival rate between 45% and 50% without any treatment and between 57% and 66% with low-dose IFN treatment.7,8 Disease-free survival of our study population overall was within the expected range at 61% (95% CI, 55% to 67%), with no significant difference between the isotretinoin and placebo groups (55% and 67%, respectively; P = .25). Similarly, estimated 5-year survival of our study population was within the expected range and was equivalent in the two treatment groups (76%, isotretinoin group; 81%, placebo group; P = .8). The lack of efficacy of isotretinoin in this regimen might reflect the heterogeneous response of melanoma cells to isotretinoin treatment, as shown in variable response rates of melanoma cell lines in laboratory studies24,25 and in conflicting efficacy results seen in clinical trials in patients with metastatic melanoma.27,30,32,33,35,36 Response proportion and survival was also not improved in a study in patients with advanced renal cell carcinoma.46
Sentinel node positivity is found in up to 30% of patients with stage IIA and IIB melanoma.47-49 In our study population, sentinel positivity was detected in 42 (27%) of 158 patients for whom sentinel node biopsy results were available. The multivariate analysis accounting for positive sentinel biopsy confirmed the overall negative study result. Consistent with results reported earlier,50,51 our analysis showed melanoma stage and patient sex to be the only significant prognostic factors. In the subgroup where sentinel node biopsy results were available, sentinel node positivity was an independent negative prognostic factor for disease-free survival.
A limitation of the study is that disease-free survival instead of overall survival has been used as the primary end point. However, when looking at our data, we see that from 120 patients with progression as defined in the study, approximately 44% died in the observation period. Of the 287 patients who had no progression, only four patients died. Hence, our data show a close relationship between progression and survival.
In conclusion, in our study, the addition of isotretinoin to a 2-year low-dose IFN treatment did not have any impact on either disease-free survival or on overall survival in patients with malignant melanoma stage IIA and IIB AJCC/UICC 1988.
Appendix
European Cooperative Adjuvant Melanoma Treatment Study Group. A. Okcu, Department of Dermatology, Medical University of Graz, Graz, Austria; G. Ratzinger, Department of Dermatology, Medical University of Innsbruck, Innsbruck, Austria; A. Seeber, E. Kindermann-Glebowski, Department of Dermatology, Danube Hospital, Vienna, Austria; G. Koglbauer, R. Moshammer, Department of Dermatology, Hospital of St P?lten, St P?lten, Austria; K. Nittmann, Department of Dermatology, Hospital of Wels, Wels, Austria; Ch Kos, Department of Dermatology, Hospital of Klagenfurt, Klagenfurt, Austria; K. Gilde, Department of Dermatology, National Institute of Oncology, Budapest, Hungary; J. Koller, W. Hangler, M. Tritscher, Department of Dermatology, Hospital of Salzburg, Salzburg, Austria; Z. Karolyi, Department of Dermatology, Semmelweis Hospital Miskolc, Miskolc, Hungary; Ch Fellenz, Department of Dermatology, Hospital of Lainz, Vienna, Austria; F. Weihsengruber, S. Orasche, N. Lilgenau, Department of Dermatology, Rudolfsstiftung, Vienna, Austria; J. Olah, Department of Dermatology, Szent-Gy?rgyi Albert Medical School, Szeged, Hungary; Z. Battyani, Department of Dermatology, University of Medicine of Pécs, Pécs, Hungary; D. Tsambaos, Department of Dermatologie, University Hospital of Patras, Patras, Greece; J. Daróczy, Department of Dermatology, Szent István Hospital, Budapest, Hungary; A. Horváth, Department of Dermatology, Semmelweis University of Medicine, Budapest, Hungary; H.P. Sleeboom, Department of Internal Medicine, Leyenburg Hospital, Den Haag, the Netherlands; L. T?r?k, Department of Dermatology, Hollós József Hospital of Bács-Kiskun county, Kecskemét, Hungary.
Principal investigators. H.P. Soyer, Department of Dermatology, Medical University of Graz, Graz, Austria; H. Pehamberger, Department of Dermatology, Medical University of Vienna, Vienna, Austria; P. Bauer, Section of Medical Statistics, Medical University of Vienna, Vienna, Austria.
Safety board. H.G. Eichler, Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria; V. Scheiber, Department of Medical Statistics, Medical University of Vienna, Vienna, Austria; H. Hoenigmann, Department of Dermatology, Medical University of Vienna, Vienna, Austria; W. Bollag, Basel, Switzerland; M. Mueller, Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Acknowledgment
We thank Barbara J. Rutledge, PhD, for editing assistance.
NOTES
Supported by Roche Austria GmbH, Vienna, Austria.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Garbe C, Krasagakis K, Zouboulis CC, et al: Antitumor activities of interferon alpha, beta and gamma and their combinations on human melanoma cells in vitro: Changes in proliferation, melanin synthesis and immunophenotype. J Invest Dermatol 95:S231-S237, 1990 (suppl 6)
Parkinson DR, Houghton AN, Hersey P, et al: Biologic therapy for melanoma, in Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma (ed 2). Philadelphia, PA, Lippincott, 1992, pp 522-541
Hauschild A: Adjuvante Therapie des Melanoms, in Garbe C, Dummer R, Kaufmann R, et al (eds): Dermatologische Onkologie. Berlin, Germany, Springer, 1997, pp 352-357
Gray RJ, Pockaj BA, Kirkwood JM: An update on adjuvant interferon for melanoma. Cancer Control 9:16-21, 2002
Balch CM, Buzaid AC, Atkins MB, et al: A new American Joint Committee on Cancer Staging System for Cutaneous Melanoma. Cancer 88:1484-1491, 2000
Hersey P, Morton DL, Eggermont AM: Adjuvant immunotherapy for high-risk primary melanoma and resected metastatic melanoma, in Thomson JF, Morton DL, Kroon B (eds): Textbook of Melanoma. London, United Kingdom, Dunitz, 2004, pp 573-585
Grob JJ, Dreno B, de la Salmonière P, et al: Randomised trial of interferon alfa-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable node metastases. Lancet 351:1905-1910, 1998
Pehamberger H, Soyer HP, Steiner A, et al: Adjuvant interferon alfa-2a treatment in resected primary stage II cutaneous melanoma. J Clin Oncol 16:1425-1429, 1998
Cameron DA, Cornbleet MC, Mackie RM, et al: Adjuvant interferon alpha 2b in high risk melanoma: The Scottish study. Br J Cancer 84:1146-1149, 2001
Hauschild A, Dummer R, Garbe C, et al: Adjuvante Interferon-alfa Therapie beim Melanom. Hautarzt 49:167-169, 1998
Hauschild A: Aktuelle Stellungnahme der Arbeitsgemeinschaft Dermatologische Onkologie zur Hochdosistherapie mit Interferon alfa 2b. Hautarzt 50:395-397, 1999
Kirkwood JM, Strawderman MH, Ernstoff MS, et al: Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 14:7-17, 1996
Kirkwood JM, Ibrahim JG, Sondak VK, et al: High- and low-dose interferon alfa-2b in high-risk melanoma: First analysis of intergroup Trial E1690/S9111/C9190. J Clin Oncol 18:2444-2458, 2000
Eggermont AM: European approach to the treatment of malignant melanoma. Curr Opin Oncol 14:205-211, 2002
Wang Z, Cao Y, D'Urso CM, et al: Differential susceptibility of cultured human melanoma cell lines to enhancement by retinoic acid of intercellular adhesion molecule 1 expression. Cancer Res 52:4766-4772, 1992
Smith MA, Parkinson DR, Cheson BD, et al: Retinoids in cancer therapy. J Clin Oncol 10:839-864, 1992
Helige C, Smolle J, Zellnig G, et al: Inhibition of K1735–M2 melanoma cell invasion in vitro by retinoic acid. Clin Exp Metastasis 11:409-418, 1993
Fink-Puches R, Smolle J, Kerl H: Retinoide in der Chemopr?vention von Haut- und Schleimhauttumoren. Hautarzt 45:671-677, 1994
Garattini E, Gianni M, Terao M: Retinoid related molecules an emerging class of apoptotic agents with promising therapeutic potential in oncology: Pharmacological activity and mechanisms of action. Curr Pharm Des 10:433-448, 2004
Lippman SM, Meyskens FL: Treatment of advanced squamous cell carcinoma of the skin with isotretinoin. Ann Intern Med 107:499-501, 1987
Tangrea JA, Edwards BK, Taylor PR, et al: Long-term therapy with low-dose isotretinoin for prevention of basal cell carcinoma: A multicenter clinical trial. J Natl Cancer Inst 84:328-332, 1992
Meyskens FL, Liu PY, Tuthill RJ, et al: Randomized trial of vitamin A versus observation as adjuvant therapy in high-risk primary malignant melanoma: A Southwest Oncology Group Study. J Clin Oncol 12:2060-2065, 1994
Frey JR, Peck R, Bollag W: Antiproliferative activity of retinoids, interferon alfa and their combination in five human transformed cell lines. Cancer Lett 57:223-227, 1991
Schaber B, Mayer P, Schreiner T, et al: Anti-proliferative activity of natural interferon-alpha, isotretinoin and their combination varies in different human melanoma cell lines. Melanoma Res 4:319-326, 1994
Bearzatto A, Orlandi L, Silvestrini R, et al: Combined effects of interferon-alfa 2a and 13-cis-retinoic acid on human melanoma cell growth and STAT protein expression. Melanoma Res 8:31-38, 1998
Giandomenico V, Vaccari G, Fiorucci G, et al: Apoptosis and growth inhibition of squamous carcinoma cells treated with interferon-alfa, interferon-beta and retinoic acid are associated with induction of the cyclin-dependent kinase inhibitor p21. Eur Cytokine Netw 9:619-631, 1998
Rustin GJ, Dische S, De Garis ST, et al: Treatment of advanced malignant melanoma with interferon alpha and etretinate. Eur J Cancer Clin Oncol 24:783-784, 1988
Lippman SM, Parkinson DR, Itri LM, et al: 13-cis-retinoic acid and interferon alfa-2a: Effective combination therapy for advanced squamous cell carcinoma of the skin. J Natl Cancer Inst 84:235-240, 1992
Lippman SM, Kavanagh JJ, Paredes-Espinoza M, et al: 13-cis-retinoic acid plus interferon alfa-2a: Highly active systemic therapy for squamous cell carcinoma of the cervix. J Natl Cancer Inst 84:241-245, 1992
Dhingra K, Papadopoulos N, Lippman S, et al: Phase II study of alpha-interferon and 13-cis-retinoic acid in metastatic melanoma. Invest New Drugs 11:39-43, 1993
Toma S, Palumbo R, Vincenti M, et al: Efficacy of recombinant alpha-interferon 2a and 13-cis-retinoic acid in the treatment of squamous cell carcinoma. Ann Oncol 5:463-465, 1994
Fierlbeck G, Schreiner T, Rassner G: Combination of highly purified human leukocyte interferon alfa and 13-cis-retinoic acid for the treatment of metastatic melanoma. Cancer Immunol Immunother 40:157-164, 1995
Triozzi PL, Walker MJ, Pellegrini AE, et al: Isotretinoin and recombinant interferon alfa-2a therapy of metastatic malignant melanoma. Cancer Invest 14:293-298, 1996
Lippman SM, Kalvakolanu DV, Lotan R: Retinoids and interferons in non-melanoma skin cancer. J Investig Dermatol Symp Proc 1:219-222, 1996
Rosenthal MA, Oratz R: Phase II clinical trial of recombinant alpha 2b interferon and 13 cis retinoic acid in patients with metastatic melanoma. Am J Clin Oncol 21:352-354, 1998
Sondak VK, Liu PY, Flaherty LE, et al: A phase II evaluation of all-trans-retinoic acid plus interferon alfa-2a in stage IV melanoma: A Southwest Oncology Group Study. Cancer J Sci Am 5:41-47, 1999
Ketcham AS, Moffat FL, Balch CM: Classification and staging, in Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma (ed 2). Philadelphia, PA, Lippincott, 1992, pp 213-220
Pocock SJ, Simon R: Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 31:103-115, 1975
Schuchter LM: Adjuvant interferon therapy for melanoma: High-dose, low-dose, no dose, which dose? J Clin Oncol 22:7-10, 2004
Moschos SJ, Kirkwood JM, Konstantinopoulos PA: Present status and future prospects for adjuvant therapy of melanoma: Time to build upon the foundation of high-dose interferon alfa-2b. J Clin Oncol 22:11-14, 2004
Kirkwood JM, Ibrahim JG, Sosman JA, et al: High-dose interferon alfa-2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIB-III melanoma: Results of Intergroup Trial E1694/S9512/C509801. J Clin Oncol 19:2370-2380, 2001
Reinhold U, Hartl C, Hering R, et al: Fatal rhabdomyolysis and multiple organ failure associated with adjuvant high-dose interferon alfa in malignant melanoma. Lancet 349:540-541, 1997
Carson JJ, Gold LH, Barton AB, et al: Fatality and interferon alfa for malignant melanoma. Lancet 352:1443-1444, 1998
Hancock BW, Wheatley K, Harris S, et al: Adjuvant interferon in high-risk melanoma: The AIM HIGH Study-United Kingdom Coordinating Committee on Cancer Research randomized study of adjuvant low-dose extended-duration interferon alfa-2a in high-risk resected melanoma. J Clin Oncol 22:53-61, 2004
Kraemer KH, DiGiovanna JJ, Moshell AN, et al: Prevention of skin cancer in xeroderma pigmentosum with the use of oral isotretinoin. N Engl J Med 318:1633-1637, 1988
Motzer RJ, Murphy BA, Bacik J, et al: Phase III trial of interferon alfa-2a with or without 13-cis-retinoic acid for patients with advanced renal cell carcinoma. J Clin Oncol 18:2972-2980, 2000
G?hl J, Meyer T, Hohenberger W: Schildw?chter-Lymphknotenbiopsie beim malignen Melanom, in Garbe C, Dummer R, Kaufmann R, Tilgen W (eds): Dermatologische Onkologie. Berlin, Germany, Springer, 1997, pp 344-348
Ulrich J, Otto HJ, Roessner A, et al: Repr?sentative Ergebnisse der Schildw?chter-Lymphknoten-Extirpation (SKLE) beim malignen Melanom, in Plettenberg A, Meigel WN, Moll I (eds): Dermatologie an der Schwelle zum neuen Jahrtausend. Berlin, Germany, Springer, 2000, pp 533-537
Fritsch P, Zelger B, Sepp N: Tumoren der Haut, in Fritsch P (ed): Dermatologie Venerologie. Berlin, Germany, Springer, 2004, pp 585-697
K?lmel KF, Lippold A, Kulle B, et al: Zur Sterblichkeit am kutanen invasiven malignen Melanom. Hautarzt 52:211-218, 2001
Balch CM, Soong S, Thompson JF: The natural history of melanoma and factors predicting outcome, in Thomson JF, Morton DL, Kroon B (eds): Textbook of Melanoma. London, United Kingdom, Dunitz, 2004, pp 181-198(Erika Richtig, H. Peter S)
Section of Medical Statistics and Department of Dermatology, Medical University of Vienna
Department of Dermatology, Danube Hospital
Department of Dermatology, Wilhelminen Hospital
Roche Austria GmbH, Vienna
Department of Dermatology, Medical University of Innsbruck
Department of Dermatology, Hospital of St P?lten
Department of Dermatology, Hospital of Wels
Department of Dermatology, Hospital of Klagenfurt
Department of Dermatology, Hospital of Salzburg
Department of Dermatology, Hospital of Lainz
Department of Dermatology, Rudolfsstiftung, Wien, Austria
ABSTRACT
PURPOSE: The combination of interferon alfa (IFN) and isotretinoin has shown a direct antiproliferative effect on human melanoma cell lines, but it remained unclear whether this combination is more effective than IFN alone in patients with metastatic melanoma. We evaluated safety and efficacy of IFN and isotretinoin compared with IFN alone as adjuvant treatment in patients with primary malignant melanoma stage IIA and IIB.
PATIENTS AND METHODS: In a prospective, randomized, double-blind, placebo-controlled trial, 407 melanoma patients in stage IIA (301 patients) and IIB (106 patients) were randomly assigned to either IFN and isotretinoin (isotretinoin group; 206 patients) or IFN and placebo (placebo group; 201 patients) after excision of the primary tumor. IFN was administered three times a week at a dose of 3 million units subcutaneously for 24 months. Isotretinoin at a dose of 20 mg for patients 73 kg, 30 mg for patients greater than 73 kg, or placebo daily for 24 months.
RESULTS: A scheduled interim analysis revealed no significant differences in survival rates, with the isotretinoin group and the placebo group showing 5-year disease-free survival rates of 55% (95% CI, 46% to 65%) and 67% (95% CI, 59% to 75%), respectively, and overall 5-year survival rates of 76% (95% CI, 67% to 84%) and 81% (95% CI, 74% to 88%), respectively. The trial was stopped for futility.
CONCLUSION: The addition of isotretinoin to an adjuvant treatment of low-dose IFN in patients with stage IIA and IIB melanoma had no significant effect on disease-free or overall survival and is therefore not recommended.
INTRODUCTION
Interferons are a group of naturally occurring proteins with a large spectrum of biologic activities including antiviral, immunomodulatory, antiproliferative, and differentiation-inducing effects.1-4 Treatment with interferon alfa (IFN) has been shown to prolong disease-free survival in melanoma patients at moderate to high risk of developing metastatic disease after surgery (American Joint Committee on Cancer [AJCC] stage IIA to III)5 and in some studies, to prolong overall survival in this patient population.6-14
As derivates of vitamin A, retinoids also show a wide spectrum of biologic activities, including immune modulation, sebosuppression, and effects on cell proliferation and cell differentiation in both melanoma and nonmelanoma cell lines.15-19 In some clinical trials evaluating isotretinoin and etretinate for treatment of nonmelanoma skin cancers, tumor suppression was achieved,18,20 although isotretinoin was ineffective in preventing basal cell carcinomas in a large study with 981 patients.21 However, in an adjuvant treatment study,22 vitamin A was not significantly better than observation alone in prolonging disease-free survival or overall survival in patients with primary melanomas thicker than 0.75 mm and with clinically negative lymph nodes.
Because the combination of IFN and retinoic acid has shown a direct antiproliferative effect on malignant human cell lines, especially melanoma cell lines,23-26 several phase II clinical trials have been performed to evaluate the combination therapy in patients with nonmelanoma skin cancers and in patients with metastatic melanoma.27-36 In patients with metastatic melanoma disease, the effectiveness of the combination of IFN and retinoic acid was unclear: some studies demonstrated efficacy, with total response from 20% to 30%,27,32,33 whereas other studies reported no or extremely low response to the combination therapy.30,35,36
Motivated by the positive studies, we designed a prospective, randomized, multicenter, double-blind, placebo-controlled trial comparing adjuvant combination therapy of IFN and isotretinoin with adjuvant therapy of IFN alone in patients with primary malignant melanoma stage IIA and IIB AJCC/International Union Against Cancer [UICC] 198837 to investigate whether the addition of isotretinoin to IFN resulted in greater treatment efficacy.
PATIENTS AND METHODS
Study Design
The study was a prospective, 24-month, randomized, multicenter, double-blind, placebo-controlled, parallel-group trial evaluating the efficacy of adjuvant therapy of 3 million units (MU) of IFN given subcutaneously three times a week in combination with daily oral isotretinoin (Roaccutan; Roche, Vienna, Austria; isotretinoin group) compared with IFN alone (placebo group) for treatment of primary melanoma in stage IIA/IIB. The primary end point of the study was disease-free survival, and the secondary end points were overall survival and evaluation of quality of life during treatment. Disease-free survival was chosen as primary end point to get a sufficient number of events for the required power, with a limited sample size and duration of the trial. The study was sponsored by Roche Austria GmbH (Vienna, Austria). Twenty centers in Austria, Hungary, the Netherlands, and Greece participated. Patient recruitment was started on October 16, 1996, and stopped on December 31, 2002.
Four hundred seven melanoma patients who met the inclusion/exclusion criteria were randomly assigned to receive either IFN and isotretinoin (isotretinoin group; 206 patients) or IFN and placebo (placebo group; 201 patients). Treatment was started within 16 weeks after completion of all surgical interventions, including primary and secondary tumor excision.
IFN was administered subcutaneously at a dose of 3 MU 3 times a week for 24 months. Isotretinoin was given daily for 24 months at a dose of 20 mg (two 10-mg capsules) for patients with weight 73 kg and at a dose of 30 mg (three 10-mg capsules) for patients with weight greater than 73 kg. Patients in the placebo group received an equivalent number of matching placebo capsules daily for 24 months.
Radiotherapy or concomitant medication with tetracyclines, other immunotherapy, chemotherapy, or any investigational drug was disallowed during the trial. Patients were permitted to use paracetamol/indomethacine to ameliorate the flu-like symptoms associated with IFN therapy.
The following assessments were made at baseline after randomization, once a month during the first three months, then at 3-month intervals: CBC and blood chemistry (including leukocytes, hemoglobin, WBC differential count, platelets, total bilirubin, liver function enzymes, alkaline phosphatase, creatine phosphokinase, renal function, cholesterol, and triglycerides). Additionally, only at baseline, the following laboratory tests were performed: total protein, blood glucose, lactate dehydrogenase acid, high- and low-density lipoproteins, thyroid-stimulating hormone, serum electrolytes, and urine analysis. Chest x-ray, ultrasonography of the abdomen and the draining lymph nodes, and x-ray of the cervical spine, thoracic spine, lumbar spine, one knee, one ankle, and both forearms were required at baseline and at the end of treatment. Blood pressure and weight were measured at baseline and every 3 months. A visual analog scale was used to assess quality of life every 3 months during treatment and every 6 months after treatment cessation.
The trial protocol and five amendments were reviewed and approved by local ethics committees at the participating sites. The study was overseen by a Data Safety Monitoring Committee consisting of two clinical pharmacologists, a statistician, a dermatologist, and a retired expert on isotretinoin, none of whom was otherwise involved in the study. The Data Safety Monitoring Committee met regularly to review safety data, trial conduct, and protocol violations. The study was conducted in accordance with the Declaration of Helskinki and with Good Clinical Practice guidelines. All patients received a detailed explanation of the potential risks and benefits of the study and provided signed informed consent before participating in the study.
Patients
Initially all patients between 18 and 75 years of age with primary melanoma in stage IIA or IIB as defined by AJCC/UICC 1988 (IIA, localized melanoma with tumor thickness of 1.51 to 4.0 mm or Clark level IV; IIB, localized melanoma > 4.0 mm tumor thickness or Clark level V)37 were eligible for the trial. Because the performance of sentinel node biopsy in routine settings was steadily increasing after the start of the study, the protocol was amended (amendment 2; May 10, 1999) to permit patients to be enrolled after sentinel node biopsy, provided that they were sentinel node–negative or, in case of positivity, that positivity revealed only micrometastasis and that subsequent radical lymphadenectomy showed negative lymph nodes.
The patients had to meet also the following inclusion criteria: baseline laboratory results including leukocyte count (WBC) 3.0 x 109/L, platelets count 100 x 109/L, hemoglobin 10 g/100 mL (1 dL), cholesterol 300 mg/dL, and triglycerides 250 mg/dL. Women of child-bearing potential were informed orally about the teratogenic effect of retinoids and given a patient information sheet. Women of child-bearing potential were required to use an accepted method of birth control for at least 1 month before beginning therapy, during treatment, and for a minimum of 3 months after discontinuation of therapy. Women of child-bearing potential were also required to have a serum pregnancy test performed by a physician within 2 days of treatment initiation and monthly until 3 months after treatment cessation.
Pregnant or lactating women, individuals younger than 18 years, and individuals with psychiatric disorders, depression, or seizure disorders or compromised CNS function were excluded. Persons with history or presence of autoimmune disease, with serious infections within previous 28 days, and with severe systemic disease (such as cardiac disease, severe liver disease, severe renal disease, or myeloid dysfunction) were also excluded. Prior immunotherapy, chemotherapy, or therapy with an investigational drug within 3 months of study initiation was not permitted. Presence of neoplastic disease within the previous 5 years was an exclusion criterion, except for basal cell carcinoma of the skin, actinic keratoses, and carcinoma-in-situ of the skin or cervix, provided that these had been cured by surgery.
Randomization
Patients were centrally randomized by the Department of Medical Statistics at the University of Vienna. Major inclusion criteria were checked before randomization. A biased coin randomization was applied to balance treatment groups for sex, stage (IIA v IIB), age ( 50 v > 50 years), weight ( 73 v > 73 kg) and study center. The allocation probabilities to the two treatments were calculated on the basis of an imbalance score defined as a weighted sum of the differences of the treatment allocation numbers in the strata defined by the five risk factors. Unconstrained randomization was used in the beginning.38
Blinded medication was prepared by the galenic division of Hoffmann-La Roche, Basel, Switzerland, and stored at Roche (where there was no access to the code). The randomization process generated a patch number of the patient's blinded medication, which was printed and faxed to Roche for distribution to the recruiting center. In this way, all patients and investigators were blinded regarding treatment allocation. Each center received a sealed envelope containing patient number and treatment allocation, which could be opened only in case of a serious adverse event that necessitated disclose of the treatment. A total of 20 envelopes were opened before the end of the study or were missing when the study was terminated.
Compliance
IFN was given monthly under the trade name used in the participating countries and injected by the patients themselves. Study participants were instructed to store IFN in the refrigerator from 2°C to 8°C. Patients were supplied with oral study medication (isotretinoin or placebo) monthly, and instructed to store the medication at room temperature. Any unused capsules were returned at each visit. Compliance was calculated at the end of the trial from pill counts and was defined on the basis of the intake of blinded study medication (isotretinoin/placebo). For each patient, a compliance ratio (actual cumulative medication intake/scheduled cumulative medication intake) was calculated. In the case of missing data, the medication intake was set to zero. Patients with a compliance ratio less than 70% were excluded from the per-protocol population.
Follow-Up Evaluation
After end of treatment patients were evaluated monthly for 3 months and than every 3 months until the third year after study initiation. Afterward, visits took place every 6 months until 5 years after study initiation.
Disease-free status was maintained when there was no hint of disease progression by either clinical assessment or imaging procedures.
Statistics
The main analysis was performed for all randomized patients. Survival and disease-free survival were compared between the treatment groups with the log-rank test (SAS procedure LIFE-TEST; SAS Institute, Cary, NC) at the two-sided level .05. Disease-free survival was defined as the time from start of treatment to the time of progression or death, whichever occurred first.
Patients who were lost to follow-up were censored at the time of the last observation. Kaplan-Meier curves and 5-year survival rates with 95% CIs including CIs for the differences in the 5-year survival rates were calculated. In addition, a multivariate analysis was performed based on the proportional hazards model, with stratification by center and using as covariables melanoma stage, sex, age, and weight. Centers that recruited 20 or fewer patients were combined (summarized) to a group (small centers) per country. Patients from Greece and the Netherlands were combined in a single group. For the subgroup of patients for whom sentinel lymph node biopsy results were available, an additional analysis was performed using a proportional hazards model with the additional factor sentinel lymph-node-positive/negative. Differences between the treatment groups in quality of life during treatment were tested with an independent samples t test, comparing the area under the curve relative to the baseline value of the quality-of-life measurements on a visual analog scale. The rates of patients experiencing adverse events were compared with 2 tests. Missing values were imputed according to the last observation carried forward principle. In the interim analysis a significance level of .01 (two-sided) was applied. Additionally, it was laid down to stop the study for futility if the conditional power (assuming a difference of 10% in the proportion of disease-free patients after 5 years) was less than 20%. The level for the final analysis was set to .05 (two-sided).
Sample Size
A total sample size of 400 patients was planned, which would give a power of approximately 80% for detecting a 10% difference in 5-year disease-free survival rate, assuming a 75% disease-free survival rate in the placebo group, a recruitment period of 4 years, a total study period of 9 years, and exponential disease-free survival.
Collection of Data and Data Management
Regular monitoring of the investigational sites was performed. The data were stored in a relational database (Oracle) located in the Statistical Center at the Department of Medical Statistics at the University of Vienna. The database was closed on February 29, 2004.
RESULTS
Patients
A total of 407 patients were recruited, with 206 patients randomly assigned to the isotretinoin group and 201 patients randomly assigned to the placebo group. Baseline characteristics of the patients are listed in Table 1.
The 407 patients were recruited in 20 centers as follows: 368 patients from 11 centers in Austria, 34 patients from seven centers in Hungary, three patients from one center in Greece, and two patients from one center in the Netherlands. The median number of recruited patients of the six largest centers was 40 patients, and six centers recruited three patients or fewer.
Interim Analysis
A scheduled interim analysis performed 9 months after the end of the recruitment period showed a conditional power of only 1.3%, far below the threshold 20% laid down in the protocol. Consequently, a decision was made to stop the trial immediately for futility, to examine the patients currently enrolled, and to perform a final analysis within the next 3 months.
Final Analysis
Disease-free survival. Disease progression was observed in 67 patients in the isotretinoin group and in 53 patients in the placebo group, and two patients in each group died without progression. There was no significant difference in disease-free survival between the treatment groups (P = .25, log-rank test, two sided). The 5-year disease-free survival rates were 55% (95% CI, 46% to 65%) in the isotretinoin group and 67% (95% CI, 59% to 75%) in the placebo group (Fig 1). The rounded difference in the 5-year disease-free survival rates was –11 percentage points (95% CI, –24% to 1%).
Overall survival. Survival rates did not differ significantly between the treatment groups (P = .8). During the study period, a total of 57 patients (30 in the isotretinoin group and 27 in the placebo group) died, 53 of these patients (28 in the isotretinoin group and 25 in the placebo group) from disseminated melanoma. The 5-year survival rates were 76% in the isotretinoin group (95% CI, 67% to 84%) and 81% in the placebo group (95% CI 74% to 88%). The difference in the 5-year survival rates was –5 percentage points (95% CI, –16% to 6%).
Prognostic factors. In the multivariate analysis, the only significant prognostic factors were stage (P < .0001; hazard ratio for stages IIB/IIA = 2.33; 95% CI, 1.60 to 3.39) and sex (P = .027; hazard ratio for male/female = 1.62; 95% CI, 1.06 to 2.50). The multivariate analysis showed no significant differences between the treatment groups (P = .81; hazard ratio for isotretinoin/placebo = 1.05; 95% CI, 0.72 to 1.52).
Figure 2 shows the Kaplan-Meier estimates by stage (IIA v IIB) and sex. Clearly there was a worse prognosis for stage IIB. Sex seemed to be an independent prognostic factor, with men showing an inferior outcome compared with women.
The analysis for the 158 patients for whom sentinel lymph node biopsy results were available showed no significant differences between treatment groups (P = .39; hazard ratio for isotretinoin/placebo = 1.41; 95% CI, 0.65 to 3.08). Stage was again a significant prognostic factor (P = .02; hazard ratio for stage IIB/IIA = 2.52; 95% CI, 1.16 to 5.48), and significantly worse prognosis was seen for patients with positive sentinel lymph node (P = .02; hazard ratio for sentinel lymph node–positive/negative = 2.31; 95% CI, 1.15 to 4.67).
Per-Protocol Analysis
As specified in the analysis plan, the data were also analyzed in the per protocol population. This analysis confirms the findings from the all randomized patient population. Reasons for exclusion from the per-protocol population were laid down in a blinded review. A total of 173 patients (isotretinoin group, n = 90; placebo group, n = 83) were excluded from the per-protocol population, with no substantial difference in the exclusion pattern under isotretinoin and placebo. A detailed flow chart of the reasons of exclusion from the per-protocol population is shown in Figure 3.
In the per-protocol analysis with the remaining 234 patients, no significant differences were detected in either disease-free survival or overall survival (P = .61 and P = .25, respectively, log-rank test, two sided). The 5-year disease-free survival rates were 65% (95% CI, 54% to 76%) in the isotretinoin group and 71% (95% CI, 62% to 80%) in the placebo group. The difference in the 5-year disease-free survival rates was –6 percentage points (95% CI, –20% to 8%).
The 5-year survival rates were 83% (95% CI, 74% to 92%) in the isotretinoin group and 77% (95% CI, 66% to 88%) in the placebo group. The difference in the 5-year survival rates was 7 percentage points (95% CI, –8% to 21%).
Adverse Events
A higher number of adverse events were reported in the isotretinoin group (889 events) than in the placebo group (740 events). This difference was due to higher incidences in the isotretinoin group versus the placebo group of both cheilitis/xerosis (175 v 44 events, respectively) and hyperlipidemia (56 v 34 events, respectively). 84% percent of the patients in the placebo group and 91% of the patients in the isotretinoin group observed at least one adverse event (P = .03).
Table 2 lists serious adverse events (WHO grades 3 or 4) reported in the study. Twenty-one percent of the patients in both the placebo and the isotretinoin group experienced at least one adverse event of grade 3 or 4 (P close to 1.00). Three life-threatening adverse events (WHO grade 4) were observed during the study. In the placebo group, one patient experienced an ischemic brain stroke after 1 month of treatment, and one patient experienced dyspnea at the start of treatment. In both patients treatment was stopped immediately, and both patients showed complete recovery. In the isotretinoin group, one patient experienced severe depression after 6 months of treatment. Treatment was continued along with antidepressant medication and the patient recovered. Similar numbers of WHO grade 3 serious adverse events were reported in the isotretinoin group (67 events) and in the placebo group (80 events).
Minimal to moderate adverse events (WHO grades 1 to 2), including laboratory disorders, general disorders, skin and tissue disorders, infection, inflammation, gastrointestinal disorders, hormonal disorders, neurologic disorders, cardiac disorders, circulatory disorders, and ophthalmologic disorders were seen in 821 reports in the isotretinoin group and in 658 reports in the placebo group. All disorders were distributed equally within both groups with the exception of cheilitis/xerosis and hyperlipidemia more often seen in the isotretinoin group.
One patient became pregnant during the study. Treatment was immediately stopped, and the treatment assignment was unblinded by the center, revealing that the patient had been in the placebo group. This patient had a spontaneous abortion after 4 months. Another patient, this time in the isotretinoin group, became pregnant 6 months after treatment cessation and delivered a healthy boy.
Four patients (two in each group) died of causes unrelated to melanoma. One patient in the isotretinoin group died of cardiac decompensation 33 months after the end of treatment. A second patient in the isotretinoin group died of myocardial infarction 13 months after start of treatment. In the placebo group, one patient died in a car accident and one patient died of cholangiocellular carcinoma. None of the four deaths were considered drug-related.
Quality of Life
Quality-of-life measurements at baseline, 6 months, and 2 years were available for 68%, 63%, and 31% of the patients with only marginal differences between treatment groups. The area under the baseline values for the visual analog scale to evaluate quality of life showed no significant difference between treatment groups (P = .94, two-sided t test).
Secondary Malignancies
Secondary malignancies reported in the isotretinoin group (five patients) included secondary melanoma (three patients), basal cell carcinoma (one patient), and gastric cancer (one patient). Secondary malignancies reported in the placebo group (10 patients) included basal cell carcinoma (four patients), squamous cell carcinoma (one patient), breast cancer (one patient), cholangiocellular carcinoma (one patient), mixed carcinoma (one patient), multiple myeloma (one patient), and cervical dysplasia (one patient).
DISCUSSION
The optimal care for melanoma patients with moderate and high risk to develop metastasis (stage IIA and IIB AJCC/UICC 1988)37 and for patients with micrometastasis in the sentinel lymph node (stage IIIA AJCC)5 has been a matter of debate.6,10,11,14,39,40 IFN is the drug most frequently prescribed for adjuvant treatment of melanoma, but currently there is no consensus on the optimal regimen. Prolongation of overall survival in patients with moderate and high risk has been demonstrated in studies evaluating high-dose IFN treatment.6,12-14,41 However, although high-dose IFN treatment regimens are widely used in the United States and, according to some, should be the gold standard of adjuvant therapy,4,40 the high cost of high-dose IFN treatment, the toxicity profile (leading in some cases to the death of disease-free patients),42,43 and conflicting efficacy results have created a dilemma for dermatologists and oncologists practicing in Europe. Consequently, low- and intermediate-dose IFN regimens are still under consideration in Europe.14 Some studies evaluating treatment of this patient population with low-dose IFN showed prolongation of disease-free survival,7-9 but others did not.13,14,44 The question "high-dose, low-dose, no dose, which dose?" for the adjuvant treatment39 has not yet been resolved, although low-dose IFN at a dose of 3 MU 3 times a week is registered for adjuvant treatment in the participating countries.
The aim of our study was not to attempt to resolve the issue of optimal IFN dosage, but to look carefully at the question of whether the addition of another active antitumor agent to an adjuvant low-dose IFN regimen widely used in Europe could significantly improve both disease-free survival and overall survival in patients with moderate to high risk of developing metastatic disease. We chose to add isotretinoin, a derivate of retinoic acid, with sebosuppressive, anti-inflammatory, cell differentiation, and immunomodulatory functions, which has been shown to be effective in the treatment of severe acne and other inflammatory diseases.16,18 In both laboratory studies and clinical trials, retinoids have been shown to have antineoplastic activity, inhibiting growth of tumor cells directly or by induction of differentiation, inhibiting cell invasion of melanoma cells, influencing the expression of tumor cell surface markers, and inducing apoptosis in leukemic and solid tumor cells.16-20,45
Our patient population consisted of patients in stage IIA and stage IIB, based on the definition of localized melanoma with tumor thickness from 1.51 to 4.0 mm or Clark level IV (IIA) or localized melanoma with tumor thickness greater than 4.0 mm or Clark level V (IIB) without nodal metastases (AJCC/UICC 1988).37 A total of 301 patients (74%) were in stage IIA and 106 patients (26%) were in stage IIB, a percentage also comparable to both other large studies in those stages of the Austrian and the French groups.7,8 During the course of the study, the definition was modified to include also patients with micrometastasis in the sentinel lymph node and the study protocol was amended to permit the enrollment of those patients. Similar numbers of men and women were enrolled onto the study, with a slight predominance of men (53%), as also seen in other studies.7,8
Patients with stage IIA to IIB disease have an estimated 5-year disease-free survival rate between 45% and 50% without any treatment and between 57% and 66% with low-dose IFN treatment.7,8 Disease-free survival of our study population overall was within the expected range at 61% (95% CI, 55% to 67%), with no significant difference between the isotretinoin and placebo groups (55% and 67%, respectively; P = .25). Similarly, estimated 5-year survival of our study population was within the expected range and was equivalent in the two treatment groups (76%, isotretinoin group; 81%, placebo group; P = .8). The lack of efficacy of isotretinoin in this regimen might reflect the heterogeneous response of melanoma cells to isotretinoin treatment, as shown in variable response rates of melanoma cell lines in laboratory studies24,25 and in conflicting efficacy results seen in clinical trials in patients with metastatic melanoma.27,30,32,33,35,36 Response proportion and survival was also not improved in a study in patients with advanced renal cell carcinoma.46
Sentinel node positivity is found in up to 30% of patients with stage IIA and IIB melanoma.47-49 In our study population, sentinel positivity was detected in 42 (27%) of 158 patients for whom sentinel node biopsy results were available. The multivariate analysis accounting for positive sentinel biopsy confirmed the overall negative study result. Consistent with results reported earlier,50,51 our analysis showed melanoma stage and patient sex to be the only significant prognostic factors. In the subgroup where sentinel node biopsy results were available, sentinel node positivity was an independent negative prognostic factor for disease-free survival.
A limitation of the study is that disease-free survival instead of overall survival has been used as the primary end point. However, when looking at our data, we see that from 120 patients with progression as defined in the study, approximately 44% died in the observation period. Of the 287 patients who had no progression, only four patients died. Hence, our data show a close relationship between progression and survival.
In conclusion, in our study, the addition of isotretinoin to a 2-year low-dose IFN treatment did not have any impact on either disease-free survival or on overall survival in patients with malignant melanoma stage IIA and IIB AJCC/UICC 1988.
Appendix
European Cooperative Adjuvant Melanoma Treatment Study Group. A. Okcu, Department of Dermatology, Medical University of Graz, Graz, Austria; G. Ratzinger, Department of Dermatology, Medical University of Innsbruck, Innsbruck, Austria; A. Seeber, E. Kindermann-Glebowski, Department of Dermatology, Danube Hospital, Vienna, Austria; G. Koglbauer, R. Moshammer, Department of Dermatology, Hospital of St P?lten, St P?lten, Austria; K. Nittmann, Department of Dermatology, Hospital of Wels, Wels, Austria; Ch Kos, Department of Dermatology, Hospital of Klagenfurt, Klagenfurt, Austria; K. Gilde, Department of Dermatology, National Institute of Oncology, Budapest, Hungary; J. Koller, W. Hangler, M. Tritscher, Department of Dermatology, Hospital of Salzburg, Salzburg, Austria; Z. Karolyi, Department of Dermatology, Semmelweis Hospital Miskolc, Miskolc, Hungary; Ch Fellenz, Department of Dermatology, Hospital of Lainz, Vienna, Austria; F. Weihsengruber, S. Orasche, N. Lilgenau, Department of Dermatology, Rudolfsstiftung, Vienna, Austria; J. Olah, Department of Dermatology, Szent-Gy?rgyi Albert Medical School, Szeged, Hungary; Z. Battyani, Department of Dermatology, University of Medicine of Pécs, Pécs, Hungary; D. Tsambaos, Department of Dermatologie, University Hospital of Patras, Patras, Greece; J. Daróczy, Department of Dermatology, Szent István Hospital, Budapest, Hungary; A. Horváth, Department of Dermatology, Semmelweis University of Medicine, Budapest, Hungary; H.P. Sleeboom, Department of Internal Medicine, Leyenburg Hospital, Den Haag, the Netherlands; L. T?r?k, Department of Dermatology, Hollós József Hospital of Bács-Kiskun county, Kecskemét, Hungary.
Principal investigators. H.P. Soyer, Department of Dermatology, Medical University of Graz, Graz, Austria; H. Pehamberger, Department of Dermatology, Medical University of Vienna, Vienna, Austria; P. Bauer, Section of Medical Statistics, Medical University of Vienna, Vienna, Austria.
Safety board. H.G. Eichler, Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria; V. Scheiber, Department of Medical Statistics, Medical University of Vienna, Vienna, Austria; H. Hoenigmann, Department of Dermatology, Medical University of Vienna, Vienna, Austria; W. Bollag, Basel, Switzerland; M. Mueller, Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria.
Authors' Disclosures of Potential Conflicts of Interest
Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Acknowledgment
We thank Barbara J. Rutledge, PhD, for editing assistance.
NOTES
Supported by Roche Austria GmbH, Vienna, Austria.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Garbe C, Krasagakis K, Zouboulis CC, et al: Antitumor activities of interferon alpha, beta and gamma and their combinations on human melanoma cells in vitro: Changes in proliferation, melanin synthesis and immunophenotype. J Invest Dermatol 95:S231-S237, 1990 (suppl 6)
Parkinson DR, Houghton AN, Hersey P, et al: Biologic therapy for melanoma, in Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma (ed 2). Philadelphia, PA, Lippincott, 1992, pp 522-541
Hauschild A: Adjuvante Therapie des Melanoms, in Garbe C, Dummer R, Kaufmann R, et al (eds): Dermatologische Onkologie. Berlin, Germany, Springer, 1997, pp 352-357
Gray RJ, Pockaj BA, Kirkwood JM: An update on adjuvant interferon for melanoma. Cancer Control 9:16-21, 2002
Balch CM, Buzaid AC, Atkins MB, et al: A new American Joint Committee on Cancer Staging System for Cutaneous Melanoma. Cancer 88:1484-1491, 2000
Hersey P, Morton DL, Eggermont AM: Adjuvant immunotherapy for high-risk primary melanoma and resected metastatic melanoma, in Thomson JF, Morton DL, Kroon B (eds): Textbook of Melanoma. London, United Kingdom, Dunitz, 2004, pp 573-585
Grob JJ, Dreno B, de la Salmonière P, et al: Randomised trial of interferon alfa-2a as adjuvant therapy in resected primary melanoma thicker than 1.5 mm without clinically detectable node metastases. Lancet 351:1905-1910, 1998
Pehamberger H, Soyer HP, Steiner A, et al: Adjuvant interferon alfa-2a treatment in resected primary stage II cutaneous melanoma. J Clin Oncol 16:1425-1429, 1998
Cameron DA, Cornbleet MC, Mackie RM, et al: Adjuvant interferon alpha 2b in high risk melanoma: The Scottish study. Br J Cancer 84:1146-1149, 2001
Hauschild A, Dummer R, Garbe C, et al: Adjuvante Interferon-alfa Therapie beim Melanom. Hautarzt 49:167-169, 1998
Hauschild A: Aktuelle Stellungnahme der Arbeitsgemeinschaft Dermatologische Onkologie zur Hochdosistherapie mit Interferon alfa 2b. Hautarzt 50:395-397, 1999
Kirkwood JM, Strawderman MH, Ernstoff MS, et al: Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 14:7-17, 1996
Kirkwood JM, Ibrahim JG, Sondak VK, et al: High- and low-dose interferon alfa-2b in high-risk melanoma: First analysis of intergroup Trial E1690/S9111/C9190. J Clin Oncol 18:2444-2458, 2000
Eggermont AM: European approach to the treatment of malignant melanoma. Curr Opin Oncol 14:205-211, 2002
Wang Z, Cao Y, D'Urso CM, et al: Differential susceptibility of cultured human melanoma cell lines to enhancement by retinoic acid of intercellular adhesion molecule 1 expression. Cancer Res 52:4766-4772, 1992
Smith MA, Parkinson DR, Cheson BD, et al: Retinoids in cancer therapy. J Clin Oncol 10:839-864, 1992
Helige C, Smolle J, Zellnig G, et al: Inhibition of K1735–M2 melanoma cell invasion in vitro by retinoic acid. Clin Exp Metastasis 11:409-418, 1993
Fink-Puches R, Smolle J, Kerl H: Retinoide in der Chemopr?vention von Haut- und Schleimhauttumoren. Hautarzt 45:671-677, 1994
Garattini E, Gianni M, Terao M: Retinoid related molecules an emerging class of apoptotic agents with promising therapeutic potential in oncology: Pharmacological activity and mechanisms of action. Curr Pharm Des 10:433-448, 2004
Lippman SM, Meyskens FL: Treatment of advanced squamous cell carcinoma of the skin with isotretinoin. Ann Intern Med 107:499-501, 1987
Tangrea JA, Edwards BK, Taylor PR, et al: Long-term therapy with low-dose isotretinoin for prevention of basal cell carcinoma: A multicenter clinical trial. J Natl Cancer Inst 84:328-332, 1992
Meyskens FL, Liu PY, Tuthill RJ, et al: Randomized trial of vitamin A versus observation as adjuvant therapy in high-risk primary malignant melanoma: A Southwest Oncology Group Study. J Clin Oncol 12:2060-2065, 1994
Frey JR, Peck R, Bollag W: Antiproliferative activity of retinoids, interferon alfa and their combination in five human transformed cell lines. Cancer Lett 57:223-227, 1991
Schaber B, Mayer P, Schreiner T, et al: Anti-proliferative activity of natural interferon-alpha, isotretinoin and their combination varies in different human melanoma cell lines. Melanoma Res 4:319-326, 1994
Bearzatto A, Orlandi L, Silvestrini R, et al: Combined effects of interferon-alfa 2a and 13-cis-retinoic acid on human melanoma cell growth and STAT protein expression. Melanoma Res 8:31-38, 1998
Giandomenico V, Vaccari G, Fiorucci G, et al: Apoptosis and growth inhibition of squamous carcinoma cells treated with interferon-alfa, interferon-beta and retinoic acid are associated with induction of the cyclin-dependent kinase inhibitor p21. Eur Cytokine Netw 9:619-631, 1998
Rustin GJ, Dische S, De Garis ST, et al: Treatment of advanced malignant melanoma with interferon alpha and etretinate. Eur J Cancer Clin Oncol 24:783-784, 1988
Lippman SM, Parkinson DR, Itri LM, et al: 13-cis-retinoic acid and interferon alfa-2a: Effective combination therapy for advanced squamous cell carcinoma of the skin. J Natl Cancer Inst 84:235-240, 1992
Lippman SM, Kavanagh JJ, Paredes-Espinoza M, et al: 13-cis-retinoic acid plus interferon alfa-2a: Highly active systemic therapy for squamous cell carcinoma of the cervix. J Natl Cancer Inst 84:241-245, 1992
Dhingra K, Papadopoulos N, Lippman S, et al: Phase II study of alpha-interferon and 13-cis-retinoic acid in metastatic melanoma. Invest New Drugs 11:39-43, 1993
Toma S, Palumbo R, Vincenti M, et al: Efficacy of recombinant alpha-interferon 2a and 13-cis-retinoic acid in the treatment of squamous cell carcinoma. Ann Oncol 5:463-465, 1994
Fierlbeck G, Schreiner T, Rassner G: Combination of highly purified human leukocyte interferon alfa and 13-cis-retinoic acid for the treatment of metastatic melanoma. Cancer Immunol Immunother 40:157-164, 1995
Triozzi PL, Walker MJ, Pellegrini AE, et al: Isotretinoin and recombinant interferon alfa-2a therapy of metastatic malignant melanoma. Cancer Invest 14:293-298, 1996
Lippman SM, Kalvakolanu DV, Lotan R: Retinoids and interferons in non-melanoma skin cancer. J Investig Dermatol Symp Proc 1:219-222, 1996
Rosenthal MA, Oratz R: Phase II clinical trial of recombinant alpha 2b interferon and 13 cis retinoic acid in patients with metastatic melanoma. Am J Clin Oncol 21:352-354, 1998
Sondak VK, Liu PY, Flaherty LE, et al: A phase II evaluation of all-trans-retinoic acid plus interferon alfa-2a in stage IV melanoma: A Southwest Oncology Group Study. Cancer J Sci Am 5:41-47, 1999
Ketcham AS, Moffat FL, Balch CM: Classification and staging, in Balch CM, Houghton AN, Milton GW, et al (eds): Cutaneous Melanoma (ed 2). Philadelphia, PA, Lippincott, 1992, pp 213-220
Pocock SJ, Simon R: Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics 31:103-115, 1975
Schuchter LM: Adjuvant interferon therapy for melanoma: High-dose, low-dose, no dose, which dose? J Clin Oncol 22:7-10, 2004
Moschos SJ, Kirkwood JM, Konstantinopoulos PA: Present status and future prospects for adjuvant therapy of melanoma: Time to build upon the foundation of high-dose interferon alfa-2b. J Clin Oncol 22:11-14, 2004
Kirkwood JM, Ibrahim JG, Sosman JA, et al: High-dose interferon alfa-2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIB-III melanoma: Results of Intergroup Trial E1694/S9512/C509801. J Clin Oncol 19:2370-2380, 2001
Reinhold U, Hartl C, Hering R, et al: Fatal rhabdomyolysis and multiple organ failure associated with adjuvant high-dose interferon alfa in malignant melanoma. Lancet 349:540-541, 1997
Carson JJ, Gold LH, Barton AB, et al: Fatality and interferon alfa for malignant melanoma. Lancet 352:1443-1444, 1998
Hancock BW, Wheatley K, Harris S, et al: Adjuvant interferon in high-risk melanoma: The AIM HIGH Study-United Kingdom Coordinating Committee on Cancer Research randomized study of adjuvant low-dose extended-duration interferon alfa-2a in high-risk resected melanoma. J Clin Oncol 22:53-61, 2004
Kraemer KH, DiGiovanna JJ, Moshell AN, et al: Prevention of skin cancer in xeroderma pigmentosum with the use of oral isotretinoin. N Engl J Med 318:1633-1637, 1988
Motzer RJ, Murphy BA, Bacik J, et al: Phase III trial of interferon alfa-2a with or without 13-cis-retinoic acid for patients with advanced renal cell carcinoma. J Clin Oncol 18:2972-2980, 2000
G?hl J, Meyer T, Hohenberger W: Schildw?chter-Lymphknotenbiopsie beim malignen Melanom, in Garbe C, Dummer R, Kaufmann R, Tilgen W (eds): Dermatologische Onkologie. Berlin, Germany, Springer, 1997, pp 344-348
Ulrich J, Otto HJ, Roessner A, et al: Repr?sentative Ergebnisse der Schildw?chter-Lymphknoten-Extirpation (SKLE) beim malignen Melanom, in Plettenberg A, Meigel WN, Moll I (eds): Dermatologie an der Schwelle zum neuen Jahrtausend. Berlin, Germany, Springer, 2000, pp 533-537
Fritsch P, Zelger B, Sepp N: Tumoren der Haut, in Fritsch P (ed): Dermatologie Venerologie. Berlin, Germany, Springer, 2004, pp 585-697
K?lmel KF, Lippold A, Kulle B, et al: Zur Sterblichkeit am kutanen invasiven malignen Melanom. Hautarzt 52:211-218, 2001
Balch CM, Soong S, Thompson JF: The natural history of melanoma and factors predicting outcome, in Thomson JF, Morton DL, Kroon B (eds): Textbook of Melanoma. London, United Kingdom, Dunitz, 2004, pp 181-198(Erika Richtig, H. Peter S)