Premium
A comparative epidemiological analysis of epithelioid, spindle cell, and mixed epithelioid‐spindle cell tumor subtypes of uveal melanoma, 1973–2016
Author(s) -
Tadrosse Abanoob,
Mikhael Sandra,
Tadrosse Marina,
Mikhael Mina,
Eloy Jean A.
Publication year - 2021
Publication title -
acta ophthalmologica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.534
H-Index - 87
eISSN - 1755-3768
pISSN - 1755-375X
DOI - 10.1111/j.1755-3768.20200203
Subject(s) - incidence (geometry) , melanoma , medicine , epidemiology , epithelioid cell , hazard ratio , surveillance, epidemiology, and end results , proportional hazards model , pathology , gastroenterology , cancer registry , immunohistochemistry , cancer research , confidence interval , physics , optics
Purpose To compare the epidemiology of epithelioid (ED), spindle cell (SC), and mixed epithelioid‐spindle cell (MES) tumor subtypes of uveal melanoma (UM). Methods Data were extracted from the Surveillance, Epidemiology, and End Results U.S. cancer database from 1975 to 2016. Incidence (IR) was estimated in number of cases/million/year. Incidence (IR) trends across the study's timeframe were assessed by measuring the annual percent change (APC). Disease‐specific survival (DSS) was calculated using the Kaplan‐Meier method. Cox regression analyses were conducted to evaluate hazard ratios (HR). Results Two thousand nine hundred thirty‐eight cases were identified: MES‐UM (38%), SC‐UM (49%), ED‐UM (13%). Diagnostic age in years was significantly different (MES‐UM, 63; SC‐UM, 58; ED‐UM, 63; p < 0.01). Incidence (IR) also differed: MES‐UM, 0.8; SC‐UM, 0.7; ED‐UM, 0.2 (p < 0.01). Incidence (IR) declined over the study's time period with APC values of −4.5 (p < 0.01; MES‐UM), −2.8 (p < 0.01; SC‐UM), and −2.7 (p = 0.04; ED‐UM). Five‐year DSS for patients diagnosed during 1995–2000 were 74.6% (MES‐UM), 86.4% (SC‐UM), and 56.2% (ED‐UM). In comparison, Five‐year DSS during 2006–2011 was similar for SC‐UM (91.8%; p = 0.07) and ED‐UM (58.4%; p = 0.35), but significantly declined for MES‐UM (57.6%; p < 0.01). MES‐UM patients aged 30‐50 years old exhibited favorable prognosis compared to older or younger ones (HR: 0.53; p = 0.023). Asian and black races portended poor prognoses for ED‐UM (HR: 2.7; p = 0.04) and SC‐UM (HR: 2.9; p = 0.03), respectively. Advanced stage at diagnosis was not a survival predictor in MES‐UM patients, but it predicted poor prognosis in ED‐UM (HR: 4.8; p < 0.01) and SC‐UM (HR: 5.7; p < 0.001). Conclusions Diagnostic age is youngest for SC‐UM. Incidence (IR) for all subtypes has declined. Disease‐specific survival (DSS) was stable over the study’s timeframe for SC‐UM and ED‐UM but substantially worsened for MES‐UM. Age impacts MES‐UM prognosis. Asian and black races are negative prognostic factors for ED‐UM and SC‐UM, respectively. Disease stage is a survival predictor only for ED‐UM and SC‐UM.