Epidemiological Characteristics and the Development of Prognostic Nomograms of Patients With HIV-Associated Cutaneous T-Cell Lymphoma

Abstract

BackgroundThe incidence of human immunodeficiency virus (HIV) associated cutaneous T-Cell lymphoma (HIV-associated CTCL) is very low, and there is a lack of relevant epidemiological and clinical prognostic studies. Therefore, we aimed to study the epidemiological characteristics of HIV-associated CTCL and to construct and validate a nomogram predicting patient survival.MethodsDemographic, clinical characteristics, and incidence data from the Surveillance, Epidemiology and End Results (SEER) database were screened for patients with HIV-associated CTCL. Independent prognostic factors in patients with HIV-associated CTCL were analyzed to establish nomograms of overall survival (OS) and disease-specific survival (DSS) rates of patients. The performance of the prediction model was validated by the consistency index (C-index), the area under the receiver operating characteristic curve (AUC), and calibration plots.ResultsA total of 883 eligible patients were screened for inclusion in this study and randomized to the training cohort (70%, n = 619) and the validation cohort (30%, n = 264). The age-adjusted average incidence rate per 100,000 persons per year for HIV-associated CTCL was 0.071 for the period 2004-2017, with an increasing incidence rate. The median age of the included patients was 59 years, of which male Caucasian held a majority. 99.5% of the patients had a tumor tissue subtype of mycosis fungoides, while the other tumor subtypes were sézary syndrome. The median OS for patients with HIV-associated CTCL was 162 months, and the OS rates at 1, 3, 5, and 10 years were 0.964, 0.904, 0.835, and 0.766, respectively. Univariate and multivariate COX regression analysis were performed, and prognostic indicators such as “Age”, “Radiation”, “Chemotherapy”, “Summary stage”, “Sequence number” were ultimately incorporated and used to establish nomograms of OS and DSS rates at 1, 3, 5 and 10 years for the training cohort. The C-index, AUC, and calibration plot confirmed that our prediction model had good accuracy.ConclusionWhile HIV-associated CTCL is very rare, its incidence has been on the rise in the last decade or so. We described the epidemiological characteristics and prognostic factors in patients with HIV-associated CTCL.