This review assessed the effect of adjuvant treatment with interferon-(IFN)-alpha in patients with high-risk malignant melanoma. While IFN was found to significantly reduce the recurrence of high-risk malignant melanoma and there was some evidence for the effect of IFN dose, the evidence was insufficient to reach definitive conclusions. The evidence presented supports the conclusions.

To assess the effect of adjuvant treatment with interferon-(IFN)-alpha in patients with high-risk malignant melanoma.

The authors report that medical databases (including MEDLINE), oncology journals, proceedings of meetings and databases of protocols were searched. No further details of the search strategy (e.g. the databases or journals searched, the search terms and dates) were given. Studies reported as full publications or abstracts were included.

Study designs of evaluations included in the review

RCTs were eligible for inclusion. One study that was described as randomised was excluded since the methods used for treatment allocation to the treatment groups suggested it was not a RCT. The reviewers made unsuccessful attempts to contact the author of this study for clarification.

Specific interventions included in the review

Studies of adjuvant treatment with IFN-alpha were eligible for inclusion. The included studies used a range of IFN-alpha doses (20 MU/m2; 1, 3, 5 and 10 MU) compared with a no treatment control, except in one study where both the intervention and control groups received a vaccine. The total planned dose ranged from 182 to 3,500 MU, and the duration of treatment ranged from 3 months to 3 years. The review also reports results from one randomised controlled trial (RCT) that included a comparison of IFN alone with vaccine.

Participants included in the review

The inclusion criteria were not explicitly stated in terms of the participants, although it was clear that patients with high-risk malignant melanoma were eligible. The studies included patients with histologically proven AJCC (American Joint Committee on Cancer) stage IIA, IIB or stage III primary or recurrent regional nodal involvement without evidence of systemic metastatic disease, or patients free of disease following complete surgical resection for AJCC stage IIB, III or IV melanoma.

Outcomes assessed in the review

The inclusion criteria were not explicitly stated in terms of the outcomes, although it was clear that the review assessed recurrence-free survival and overall survival. The median duration of follow-up in the included studies ranged from 1.3 to 12.6 years.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

The authors excluded trials reporting non-randomised methods of treatment allocation but, otherwise, did not state that they assessed validity.

Two reviewers independently extracted the data and resolved any disagreements through discussion. The most recent or most complete data set was extracted for studies reported in more than one publication. Data from two unpublished studies were taken from a presentation at the 2001 American Society for Clinical Oncology.

Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated for each study. For studies with treatment arms using different doses of IFN, the HRs were calculated for each dose compared with the control. Data from studies using the same dose of IFN in two treatment arms were combined and compared with the control.

How were the studies combined?

The studies were combined using a meta-analysis. Pooled HRs and 95% CIs were calculated using fixed-effect models for recurrence-free survival and overall survival. Data from the control groups contributed only once to summary estimates. The absolute difference in recurrence-free survival and 95% CI was also calculated. Data from the two studies that directly compared two different doses of IFN were discussed separately.

How were differences between studies investigated?

Statistical heterogeneity was tested using the chi-squared statistic and the results were presented in forest plots. Recurrence-free survival and overall survival were calculated according to the dose of IFN-alpha, based on IFN dose given at each administration, a high dose of 20 MU/m2, an intermediate dose of 5 or 10 MU, a low dose of 3 MU, and a very low dose of 1 MU; and also according to the duration of treatment, i.e. 6 months or less, 12 to 18 months, and 24 months or longer. Tests for trends over dose, treatment duration and total scheduled dose were conducted. Tests for trend were repeated after excluding very low dose studies.

Twelve RCTs (5,206 patients) were included. The authors also report the results of an additional RCT with 880 patients.

Recurrence-free survival (14 comparisons from 12 RCTs).

IFN-alpha significantly reduced the odds of recurrence by 17%. The HR was 0.83 (95% CI: 0.77, 0.90, P=0.000003). The absolute difference in recurrence-free survival was 7% (95% CI: 4, 9). No significant heterogeneity was detected (P=0.4). The test for trend suggested some influence of IFN dose (P=0.02), but after excluding the very low dose studies the trend was no longer significant (P=0.2). Tests suggested an increased effect with increased total IFN dose (P=0.05), but there was no significant trend for longer duration of treatment (P=0.6).

Overall survival (13 comparisons from 11 RCTs).

IFN-alpha improved overall survival but the improvement was not statistically significant. The HR was 0.93 (95% CI: 0.85, 1.02, P=0.1). The absolute difference in survival was 3% (95% CI: 0, 6). No significant heterogeneity was detected (P=0.9). Tests showed no significant trend for increased dose (P=0.8), increased total scheduled dose (P=1.0), or longer duration of treatment (P=0.6).

Studies that directly compared two different doses of IFN (2 RCTs).

One RCT compared a dose of 20 MU decreasing to 10 MU with a dose of 3 MU; the other RCT compared 10 and 5 MU. Neither RCT showed any significant difference between high- and low-dose IFN.

The inclusion of one trial of IFN-alpha versus vaccine with the two trials of high-dose IFN-alpha given for one year showed no clear survival benefit. The HR was 0.85 (95% CI: 0.72, 1.01, P=0.06).

While the studies showed that IFN significantly reduced the recurrence of high-risk malignant melanoma and there was some evidence for the effect of IFN dose, the evidence was insufficient to reach definitive conclusions. The review question was clear in terms of the study design and intervention. It was clear that patients with high-risk melanoma were included and that the outcomes of interest were recurrence-free survival and overall survival. Relevant sources appear to have been searched for published and unpublished data. However, the search strategy was not fully described (i.e. specific databases and journals searched, dates, language restrictions and search terms were not reported). Two reviewers independently extracted the data, which reduces the potential for bias and errors, but the methods used to select the studies were not described; hence, any efforts made to reduce errors and bias cannot be judged. Only RCTs were included, but there was no formal validity assessment or comment on the quality of the included studies.

Some relevant information on the included studies was tabulated. The data were combined in a meta-analysis and statistical heterogeneity was assessed. Subgroup analyses explored the influence of various factors on the results and were defined in advance. The authors discussed some of the limitations of the evidence in the text, including the lack of sufficient long-term data to reach definitive conclusions. The evidence presented supports the conclusions that there was insufficient evidence to reach definitive conclusions.

Practice: The authors did not state any implications for practice.

Research: The authors stated that groups who have conducted RCTs of adjuvant IFN-alpha in patients with high-risk melanoma should collaborate in a meta-analysis of individual patient data. This would provide longer term follow-up results and allow the analysis of different types of patients with different disease categories. They also stated that there is a need for information on the effect of IFN on quality of life and cost-effectiveness.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.