Clinicopathologic Features, Prognosis, and Therapeutic Responses in Patients with Granulomatous Mycosis Fungoides: A Single Center Experience From Memorial Sloan Kettering Cancer Center.

Abstract

Abstract 2643 Background:Granulomatous mycosis fungoides (GMF) is a variant of MF/cutaneous T-cell lymphoma characterized by a prominent granulomatous infiltrate in addition to the malignant lymphoid infiltrate. The diagnosis of GMF remains challenging due to variable clinical and histopathologic features. The significance of granuloma formation in CTCL is unclear and whether patients with GMF have a distinct prognosis compared to patients with classic type of MF is controversial. Data are scarce and long term results are not available. Objectives:To evaluate the clinical, histopathologic, prognostic significance and therapeutic responses of patients with GMF and to compare results to age and stage-matched patients with classic type of MF. Methods:A single center retrospective case-case study was performed at Memorial Sloan-Kettering Cancer Center of 430 patients with a diagnosis of MF between January 1981 and April 2012. Available histopathology slides were reviewed for the presence of granulomas or histiocytes comprising ≥ 25% of the atypical lymphoid infiltrate. Each identified case was matched with two classic MF cases via age and TNMB stage. For each case meeting criteria, the medical records, photographs, and histopathology slides were reviewed. Results:Twenty-seven patients with GMF were identified representing 6.3% of all MF patients at our center. Patient demographics were similar between the GMF group and case-control classic MF group, with a male-to-female ratio of 2.1:1 and a median age of 57. Most GMF patients (69%) present at early stages (IA-IIA). Patients with GMF had a longer onset of disease prior to diagnosis (median, 4.5 years vs 3.0 years). Skin manifestations of granulomatous MF showed a great variability, but were not distinguishable from those of classic MF. Histological findings showed an atypical predominantly CD4+ lymphocytic infiltrate with either granuloma formation or histiocytes with giant cells. TCR rearrangement was positive in 70% of cases. Features of GMF were present at initial diagnosis in 17 patients. In 10 patients, granulomatous infiltrates appeared at an average of 5.4 years after MF diagnosis. Secondary malignancies developed in 38% of patients with GMF compared to 23% of patients in the case-control classic MF group. Among the GMF patients, there was a trend towards fewer CRs to any treatment (38% vs 54%; p = 0.21); fewer PRs or CRs with topical therapy (57% vs 83%; p = 0.002) and with UV-light therapy (62% vs 90%; p = 0.008). Significantly more GMF patients received systemic therapy (66.7%) compared to classic MF patients (32.7%; p = 0.006). They also required a longer time to achieve best response (median 35 months vs 9 months, p = 0.002) compared to the control group. Disease progression to higher stage was seen in 46% of cases in the GMF group compared to 30% of cases in the classic MF group (p = 0.23). The 5-year and 10-year progression-free survival rates were significantly lower in the GMF group (59% and 33%, respectively) compared to the control group (84% and 56%, respectively; p = 0.02). However, the 5-year and 10-year overall survival rates were similar between GMF (86% and 72%, respectively) and classic MF patients (85% and 85%; p = 0.54). Conclusion:Our study presents the largest series on GMF in the United States. Most patients with GMF show similar clinical features when compared to patients with conventional MF. The clinical features do not reflect the histologic findings of granulomas. More frequent disease progression and poorer response to skin-directed therapies are seen in GMF patients compared to classic MF; however, overall survival is not significantly different. The majority of GMF patients present with granulomatous features at initial diagnosis and in subsequent biopsies suggesting that this entity has a unique relationship with its microenvironment. Disclosures:No relevant conflicts of interest to declare.