Fried's frailty phenotype predicts survival in pts with relapsed/refractory lymphoma or myeloma undergoing chimeric antigen receptor T-cell therapy – a prospective, observational Study

Abstract

Abstract Chimeric antigen T-cell receptor (CART) therapy is highly effective for pts (pts) with relapsed/refractory (R/R) lymphoma/multiple myeloma (MM). However, due to concerns regarding tolerability, older pts are underrepresented in CART trials and real-world studies indicate that CART is underutilized in older adults. Methods to assess fitness for CART are ECOG, clinician gestalt and age. There is interest in improving risk stratification of older adults using objective measures. Fried's frailty phenotype (FP) uses subjective (exhaustion, reported weight loss, activity level) and objective (gait speed, grip strength) measures to categorize pts into fit, pre-frail, and frail. We have previously shown that FP predicts for overall survival (OS) in older stem cell transplant (SCT) recipients. We hypothesize that FP will be associated with progression-free survival (PFS) and OS in older pts with lymphoma/MM undergoing CART therapy. We prospectively enrolled pts ≥ 60 years planned for CART for R/R lymphoma/MM from May 2019 – 2023 on a clinical trial. We performed FP prior to CART infusion, and at 7 days (d), 14d, 21d, 1 month (mo), 3mo, 6mo and 12mo post-infusion. 36 pts were enrolled with a median age at CART infusion of 69 years (Range 60-81). 53% of pts had MM, of whom 63% had intermediate or high-risk disease by R-ISS. The remainder had lymphoma (diffuse large B-cell or follicular lymphoma) with IPI > 2 at diagnosis in 59%. Idecagtagene vicleucel and tisagenlecleucel were the most frequently administered CART products. Median follow up was 33mo. Median prior lines of therapy (LOT) was 3 (Range 1-7) and 47% had prior auto-SCT. Pre-infusion, majority had low ECOG scores (0-1, 81%), including 71% categorized as frail by FP. At pre-infusion FP, 35% of pts were fit (score 0), 44% were prefrail (score 1-2) and 21% were frail (score 3-5). Frail pts were more likely to be admitted for >7d for their CART infusion (OR 7.0, 95% CI 1.02-47.97, p=0.04). Frailty was not associated with risk of CRS, ICANS or 30-day hospital readmission. 13 pts had died by the time of analysis; all but 2 deaths were related to progressive disease. 2 non-disease related deaths were 1 death from COVID and 1 ICANS-related death from teclistamab after relapse 1 year and 2 years after infusion, respectively. At Day 21 post-infusion, 21% were fit, 57% were prefrail, and 21% were frail. At 1mo post-infusion, 25% were fit, 63% were prefrail, and 13% were frail. Being frail by FP at pre-infusion (p<0.001), Day 21 (p=0.03) or 1 month (p=0.009) post-infusion was associated with inferior OS from that time point. Median PFS in pre-infusion fit, prefrail, and frail pts were 23.4mo (95% CI 17.1-NR), 18.4mo (95% CI 6.8-13.8) and 4.0mo (95% CI 2.5-8.4), respectively. 2-year OS estimates were 100%, 93% and 14%, in fit, prefrail and frail pts respectively. 14 of 36 pts maintained or improved their FP from pre-infusion to 1mo; all but 3 received physical therapy (PT) while in hospital with 5 pts continuing PT outpatient. Notably, pts who maintained or improved their FP from pre-infusion to 1mo post-infusion had significantly better OS (p=0.05) than pts who had declines in their scores. Along with pre-infusion, day 21 and 1mo post-infusion FP scores, LDH (Mean 182 U/L) at the time of CART infusion was significantly associated with OS in the univariate Cox proportional hazards model (HR 5.22, 95% CI 1.43-19.18, p=0.013). Several factors including disease type, number of prior lines of therapy, use of bridging, stage at CART, IPI/RISS at diagnosis, HCT-CI, ECOG, presence of extra-nodal disease, CRS, ICANS, gender, age by decade, and BMI did not correlate with outcome. In pts ≥ 60 with R/R lymphoma/MM undergoing CART, 21% were frail by FP prior to CART. Frailty by FP pre-infusion, day 21 and 1mo post-infusion was associated with inferior OS as opposed to ECOG, HCT-CI, age or several disease related risk factors. FP may improve risk stratification in older adults undergoing CART. Pts with improvement in FP within 1mo post-infusion also had better outcomes. While better disease control could contribute to improved FP scores, most pts received PT to reverse frailty. Our future work aims to implement an exercise regimen to improve outcomes and to determine whether frailty is associated with adverse disease biology. Future work to uncover biologic mechanisms of frailty and adverse disease biology may identify novel targets for intervention to improve outcomes for frail pts.