Association of elevated red cell distribution width and overall survival in multiple myeloma.

Abstract

8534 Background: The most widely used multiple myeloma (MM) staging system, the Revised International Staging System (R-ISS), is based on lactate dehydrogenase, albumin, cytogenetics, and beta-2 microglobulin (B2M). B2M has limited clinical utility, is often only obtained to compute R-ISS, and is often a send-out test. Prior studies have shown that elevated red cell distribution width (RDW), a common test that is part of the routine complete blood count, is associated with reduced overall survival (OS) in MM. We hypothesized that a MM staging system could be constructed replacing B2M with RDW. Methods: Patients treated at Vanderbilt University Medical Center from 2000 to 2018 with cancer registry-confirmed MM diagnoses were included. OS was computed by registry-curated death and diagnosis dates; living patients were censored at date last known alive or of last follow up. A Cox proportional hazards model determining the independent effects of R-ISS, age, and RDW on OS was built. An alternate staging system (RDW-SS) was developed, replacing the B2M cutoffs in R-ISS with RDW (stage 1: RDW < 14.0% stage 2: RDW 14.0-15.5%; stage 3: RDW > 15.5%). Cox models comparing RDW-SS and R-ISS after adjustment for age and category of induction therapy (proteasome inhibitor [PI] and immunomodulatory drug [IMiD]), PI only, IMiD only, other) were built. Results: In 604 MM patients with available data, RDW was independently associated with OS after adjustment (HR = 1.069 per 1% RDW increase, p < 0.001). RDW-SS stages were associated with reduced OS (RDW-SS 2 HR 1.52, 95% CI 1.09 – 2.11, p = 0.01; RDW-SS 3 HR 2.13, 95% CI 1.37 – 3.28, p < 0.001) after adjustment; R-ISS stages were not (Table). Conclusions: RDW at MM diagnosis was independently associated with reduced OS, after adjustment for the clinical factors of age and treatment exposure; notably, R-ISS was not. The RDW-SS is simpler to obtain than R-ISS and may have improved prognostic value pending independent validation. If confirmed, mechanistic study of the etiology of this relationship is warranted. [Table: see text]