Interventions for mycosis fungoides: critical commentary on a Cochrane Systematic Review

Background: The authors performed a systematic review of randomized controlled trials (RCTs) on interventions for any stage of typical mycosis fungoides (MF). They searched electronic databases including the Cochrane Central Register of Controlled Trials, Medline, Embase, and the Latin American and Caribbean Health Science Information database, and included reports from conference proceedings and unpublished data without language restrictions. The authors also searched trial registries affiliated with the U.S.A., Australia, the World Health Organization and the European Organisation of Research and Treatment of Cancer for studies on 'mycosis fungoides' or 'cutaneous T-cell lymphoma'. These searches were supplemented by correspondence with the groups or individuals who conducted the RCTs.

Methods: The authors included RCTs with participants who were 18 years of age or older, that had staging information, and in which > 90% of patients had biopsy-proven typical CD4+ MF. Data on treatment and outcome of participants, including information on stage of MF, therapy, quality of life, remission or improvement, duration of remission, survival, adverse effects and toxicity were obtained from included studies. Primary outcomes were adverse effects and quality of life. Secondary outcomes were clearance of at least 90% of surface area involvement, improvement of at least 50% of surface area involvement, survival rate, relapse rate and disease-free interval. The authors also recorded potentially significant participant-related prognostic factors, such as age and sex, and tumour-related prognostic factors, such as histological subtype and systemic involvement.

Findings: From 407 unique references, 14 RCTs were included with a total of 675 patients. These trials included skin-directed therapies [topical peldesine, topical imiquimod, topical hypericin, intralesional interferon (IFN)-α, psoralen ultraviolet A (PUVA) therapy, electron-beam therapy (EBT) and local radiation], systemic therapies [extracorporeal photopheresis (ECP), denileukin diftitox, bexarotene] and combination therapies (injected transfer factor with concomitant topical nitrogen mustard use). Only one meta-analysis of two studies comparing PUVA with IFN-α vs. PUVA alone could be performed, and no significant differences between the two therapies were found. Two studies on intralesional IFN-α vs. placebo were included in the review and provided opposing results, but were not examined by meta-analysis due to differences in their study design. The remainder of the Cochrane analysis reviewed outcomes of individual RCTs. There were statistically significant differences in improvement or clearance for five therapeutic regimens. One trial of topical hypericin vs. placebo found a relative benefit of hypericin, risk ratio (RR) for improvement 7·00, 95% confidence interval (CI) 1·01-48·54, P ≤ 0·028. A trial comparing ECP with PUVA demonstrated significantly better improvement in the PUVA group (RR 0·07, 95% CI 0·00-1·00, P ≤ 0·002). An RCT examining 'conservative', stepwise escalation from topical nitrogen mustard to 'combination therapy' with EBT and cyclophosphamide, doxorubicin, etoposide and vincristine chemotherapy found that combination therapy was superior in clearance (RR 2·18, 95% CI 1·10-4·33, P ≤ 0·03) and improvement (RR 1·40, 95% CI 1·12-1·74, P ≤ 0·003). However, there were no statistically significant differences in survival rates at a median follow-up of 75 months. A comparison of subcutaneously injected IFN-α and acitretin vs. subcutaneously injected IFN-α and PUVA found increased clearance with IFN-α and PUVA (RR 0·54, 95% CI 0·35-0·84, P ≤ 0·005). There were also significant reductions in grade III, severe adverse events on the World Health Organization scale; events requiring discontinuation; and neurological disorders in the IFN-α plus PUVA group. Finally, a trial comparing active vs. inactivated transfer factor found significant differences between the groups, favouring inactivated transfer factor (Fisher's exact test, P ≤ 0·03, RR 0·09, 95% CI 0-0·61). The original study authors speculated that their results reflected a better initial prognosis for the group receiving inactivated transfer factor. None of the interventions assessed showed significant long-term benefit. Despite significantly superior clearance rates in four trials, participants in those studies had high relapse rates.

Interpretation: This review of RCTs for MF interventions led to more questions than answers due to a dearth of adequately powered RCTs. Only one meta-analysis could be performed. The remaining review was based on single trials, many of which assessed infrequently used treatments or regimens and are not reflective of current clinical practices. Only two of the 14 RCTs assessed patient health-related quality-of-life outcomes.