Diagnostic and therapeutic patterns in cutaneous T-cell lymphomas (CTCL): Real-world data from the lymphoma epidemiology outcome-molecular epidemiology resource (LEO-MER) prospective cohort study.

Abstract Introduction: CTCLs are rare, heterogeneous non-Hodgkin lymphomas (NHL) with significant regional variability in treatment patterns. Prospective studies characterizing CTCL management, particularly in racially and ethnically diverse populations, remains limited. The LEO-MER multi-center prospective cohort study (NCT02736357) is designed to address the persistent disparities in outcomes research by enrolling historically underrepresented groups of patients diagnosed with lymphoma across the United States (US). Here, we present a real-world analysis of CTCL patients that reflects US academic practice, evaluating treatment patterns, as well as prognostic factors influencing survival outcomes. Methods: Patients aged ≥18 years with newly diagnosed CTCL were prospectively enrolled in either the University of Iowa/Mayo Clinic MER cohort (2002-2015) or the expanded LEO cohort (2015-2020), which included 8 US academic centers. Standardized protocols were used to abstract clinical, pathologic, treatment, and outcome data from medical records. ANOVA/Kruskal-Wallis and Fisher's/chi-square tests were used to compare patient distribution across demographic and clinical characteristics based on data distribution and sample size. Overall survival (OS) was calculated from diagnosis to death or last follow-up. Survival analyses employed Kaplan-Meier estimation supplemented by Cox proportional hazards modeling. Results: The LEO-MER cohort enrolled 223 CTCL patients (91 MER and 132 LEO). Subtypes included mycosis fungoides (MF, n=155, 69.5%), Sézary syndrome (SS, n=22, 9.9%), primary cutaneous anaplastic large-cell lymphoma (n=17, 7.6%), subcutaneous panniculitis-like T-cell lymphoma (n=11, 4.9%), cutaneous CD4+ small/medium T-cell lymphoproliferative-disorder (n=9, 4%), and other rarer CTCL subtypes (n=9, 4%). Among the MF/SS cohort, median age was 64 years (Range: 18-89 years), 58.2% (n=103) were male and race/ethnicity distribution included White (n=123, 69.5%), Black (n=29; 16.4%) and others/unknown (n=25; 14.1%). Stage distribution included 62.7% with early-stage (ES) MF (IA-IIA, n=111), 37.2% with advanced-stage (AS) MF (IIB-IVB, n=66). To complete diagnostic workup, T-cell receptor (TCR) testing was performed in 52.6% of skin biopsies (n=92) and 62.5% of blood samples (n=110), with clonal detection rates of 75% (69/92) in skin and 51.8% (57/110) in blood. Paired TCR matching between skin and blood samples was performed in 31.6% (n=56) of cases, demonstrating 53.6% (30/56) concordance. Blood flow cytometry data were reported in 80.1% (89/111) and 84.4% (56/66) of ES and AS MF cases, with T-cell population being detected in 7.2% (8/111) and 62.1% (41/66) of ES and AS cases. Nodal biopsies were obtained in 23.4% (n=41) of patients and clonality assessment was reported in 18.6% (33/177) of these cases. The MF/SS cohort's lifetime treatment exposure data (n=160) revealed topical skin-directed therapies alone (n=49, 31.8%) as the most common approach followed by lifetime exposure to topical skin-directed and systemic therapies (n=45, 29.2%) followed by multimodal exposure of topical skin-directed, radiation and systemic treatments (n=30, 19.5%). Only 5.6% of patients were enrolled in a clinical trial. First-line (1L) systemic regimens were predominantly a) immunomodulatory agents (n=47, 29.3%); including oral retinoid (n=20), extracorporeal photopheresis (n=18), interferon (n=9) followed by b) chemotherapy (n=30, 18.7%) and c) targeted therapies (n=17, 10.6%) including Brentuximab Vedotin (n=6), Romidepsin (n=5), Mogamulizumab (n=3), Vorinostat (n=2) and Pralatrexate (n=1). Multivariable analysis predicting 1L systemic treatment initiation demonstrated a significant association with presence of N2/N3 stage (N2/N3 vs N0/N1: Odds Ratio (OR): 5.45) and elevated LDH (OR: 5.30). At median follow-up of 91.1 months (mon), median OS was 122.3 mon (95% confidence interval 67.2-159.2 mon). In subset analyses, Black race (p=0.001) and SS (p=0.0001) were associated with worse OS outcomes. Conclusion: We present initial data from our prospective LEO-MER cohort, a large US-based multicenter consortia. Our findings demonstrate variability in both diagnostic staging and treatment approaches for MF/SS patients. The cohort demonstrated worse outcomes with high-risk disease and Black race/ethnicity. These findings warrant further study on the impact of underlying social determinant factors, given the variability noted in this population.