Predicting Response to Selinexor-Based Therapy in Relapsed/Refractory Multiple Myeloma: A Multicenter Prospective Study

Background: Relapsed/refractory multiple myeloma (RRMM) remains a major clinical challenge due to treatment resistance and disease heterogeneity. Selinexor-based regimens have demonstrated promising efficacy; however, predictive biomarkers for treatment response are still lacking. This study aims to evaluate the predictive value of differential multigene expression in RRMM patients treated with selinexor-based therapy. Methods: This is a multicenter, prospective clinical study enrolling 127 patients with RRMM. Eligible patients are aged ≥18 years, have a confirmed diagnosis of multiple myeloma based on International Myeloma Working Group (IMWG) criteria, and have received 2-4 prior lines of therapy. Key exclusion criteria include unresolved toxicities from prior treatments, severe comorbidities, prior bone marrow transplantation within 6 months, hypersensitivity to study drugs, HIV infection, pregnancy or lactation, and participation in other clinical trials within 30 days. All patients receive a selinexor-based regimen consisting of selinexor 60 mg orally once weekly (Day 1 of each week) in combination with standard agents. Each treatment cycle lasts 28 days. Treatment response is assessed every two cycles using bone marrow examination, M-protein quantification, and imaging studies. After treatment completion, patients are followed every 3 months for up to 2 years. Endpoints: The primary endpoint is overall response rate (ORR) after two treatment cycles. Secondary endpoints include progression-free survival (PFS), overall survival (OS), and safety. Exploratory endpoints focus on the association between the expression levels of candidate genes (hnRNPU, IRF3, ALB2RP, ZBTB17, ATRX, ABCC4, ASB8, and E2F1) and ORR following two cycles of selinexor-based therapy. Statistical Analysis: Statistical analyses are performed using Excel 2019 and SPSS 26.0. Continuous variables are tested for normality using the Shapiro-Wilk test and Q-Q plots. Normally distributed data