Immune status in merkel cell carcinoma: Relationships with clinical factors and independent prognostic value.

e21509 Background: Immunosuppression (IS) is not currently considered in staging for Merkel cell carcinoma (MCC). We performed an analysis of the National Cancer Database (NCDB) to investigate immune status as an independent predictor of overall survival (OS) for patients with MCC and describe the relationship between immune status and other prognostic factors. Methods: The NCDB was queried for patients diagnosed with MCC from 2010 to 2016 with known immune status. Multivariable Cox proportional hazards models were used to define factors associated with OS. Adjuvant radiation and chemotherapy were treated as time-dependent predictors to limit immortal time bias. Secondary models were constructed to assess the association between IS etiology and OS. Multivariable logistic regression models were used to characterize relationships between immune status and other factors. Multiple imputation was used to minimize missing data bias. Results: The overall cohort included 3,882 patients (3,470 patients with known immunocompetence and 412 patients with known immunosuppression). Etiologies for profound IS included chronic lymphocytic leukemia (CLL, n = 118), Other including HIV/AIDS (n = 116), solid-organ transplant (n = 106), and Non-Hodgkin Lymphoma (NHL, n = 72). 2,864 patients (73.8%) underwent surgical nodal examination. The median follow-up time was 33 months (Interquartile Range: 18 to 55 months). The 3-year OS was lower for patients with IS (44.6%, CI 39.8-49.9%) compared to immunocompetent (IC) patients (68.7%, CI 67.1-70.4%, p < .0001). IS was associated with increased adjusted mortality hazard (HR 2.36, 95% CI 2.03-2.75). Etiology of IS was associated with OS ( p = .0015) with lowest 3-year OS (32.7%, CI 24.6%-43.5%) for patients with solid-organ transplantation. IS was associated with increased odds of greater nodal burden (OR 1.70, CI 1.37-2.11) and lymphovascular invasion (OR 1.58, CI 1.23-2.03). Conclusions: Immune status was independently prognostic for OS for patients with localized MCC. Etiology of IS may be associated with differential survival outcomes. Multiple adverse prognostic factors were associated with increased likelihood of IS. Immune status and potentially etiology of IS may be useful prognostic factors to consider for future MCC staging systems.