Abstract
BACKGROUND: Prior studies have demonstrated improved disease-specific survival of mycosis fungoides (MF) patients over the last 50 years. OBJECTIVE: To analyse patterns of survival and incidence from 1973 to 2016 and determine whether apparent improvements in MF-specific survival are due to lead-time bias rather than improvements in treatment. METHODS: We performed an analysis of 10 155 patients diagnosed with MF from 1973 to 2016 in the United States cancer registries of SEER-18. We also performed a literature review of papers including stage data for unselected populations of MF patients prior to 2000. RESULTS: Incidence of MF increased from 3.0 per million person-years in the 1970s to 5.9 in the 2010s. For all cohorts, non-Hodgkin lymphoma (including MF) was the leading cause of death. Survival analysis demonstrated marked improvement in disease-specific and overall survival from the 1970s to 2010s. Based on systematic review of the literature, 32%-73% of patients diagnosed prior to 2000 were diagnosed with early-stage disease, as opposed to 81% of patients in the SEER 2000-2016 cohort (P < 0.035 for all cohorts). CONCLUSIONS: Although there have been improvements in MF-related survival over the last 50 years, these may reflect improvements in our ability to diagnose early-stage disease rather than improved treatment.