RESUMEN
BACKGROUND: Topical imiquimod is sometimes used for lentigo maligna (LM) in situ melanoma instead of surgery, but frequency of cure is uncertain. Pathological complete regression (pCR) is a logical surrogate marker for cure after imiquimod, although residual LM and atypical melanocytic hyperplasia may not be reliably distinguished. A trial comparing imiquimod vs. surgery might be justified by a high imiquimod pCR rate. OBJECTIVES: Primary: to estimate the pCR rate for LM following imiquimod. Secondary: to assess the accuracy of prediction of pCR, using clinical complete regression (cCR) plus negative post-treatment biopsies, tolerability, resource use, patients' preferences and induced melanoma immunity. METHODS: This was a single-arm phase II trial of 60 imiquimod applications over 12 weeks for LM then radical resection. A pCR rate ≥ 25 out of 33 would reliably discriminate between pCR rates < 60% and ≥ 85%. Clinical response was assessed and biopsies taken after imiquimod. Patients recorded adverse events in diaries. Patient preference was measured after surgery using a standard gamble tool. RESULTS: The pCR rate was 10 of 27 (37%, 95% confidence interval 19-58%). The rate of cCR plus negative biopsies was 12 of 28, of whom seven of 11 had pCR on subsequent surgery. The median dose intensity was 86·7%. Of the 16 surveyed patients, eight preferred primary imiquimod over surgery if the cure rate for imiquimod was 80%, and four of 16 if it was ≤ 40%. CONCLUSIONS: The pCR rate was insufficient to justify phase III investigation of imiquimod vs. SURGERY: Clinical complete response and negative targeted biopsies left uncertainty regarding pathological clearance. Some patients would trade less aggressive treatment of LM against efficacy.