Impact of a Validated Composite Comorbidity Score on Outcomes in Patients Treated with CAR T-Cell Therapy for Diffuse Large B Cell Lymphoma (DLBCL): A Multicenter Real-World Evidence (RWE) Study
washington; swedish; swedish cancer
Introduction: Chimeric Antigen Receptor T-cell therapy (CART) has dramatically improved outcomes for patients (pts) with relapsed/refractory (r/r) DLBCL, but the majority of pts still have poor outcomes due to progressive disease and toxicities. Tools quantifying frailty and comorbidities have not been verified in large patient cohorts. The Cumulative Illness Rating Scale (CIRS) is a comprehensive tool that has been found to predict outcomes in various B cell malignancies. We used a machine learning algorithm to rank the prognostic impact of specific comorbidities measured by CIRS on progression-free survival (PFS) in DLBCL pts undergoing leukapheresis for CART in a multicenter learning cohort (LC; Shouse et al, ASH 2021). In this study, we establish that these comorbidities also predict overall survival (OS) and verify their prognostic significance in a separate validation cohort (VC).
Methods: We conducted a retrospective RWE analysis of pts with r/r DLBCL who underwent leukapheresis for CART at 10 academic centers. CIRS was assessed at the time of T-cell collection and calculated per Salvi (2008). High comorbidity burden was defined using a published cutoff of CIRS score ≥7.
PFS and OS were measured from T-cell collection. Random survival forest (RSF) modeling of PFS and OS was repeatedly applied to random subsets of the LC to determine the most important CIRS categories and comorbidity levels in the presence of known prognostic factors (IPI at diagnosis, concurrent indolent lymphoma, age, ECOG performance status, number of prior therapies, prior transplant, number of medications, cell of origin subtype, complex karyotype, MYC rearrangement by FISH, and MYC+BCL2/BCL6 rearrangement). Cox proportional hazards models were fit to quantify the association between survival and significant pt features. Associations between comorbidities and CART adverse events were evaluated with Fisher's exact test.
Results: We analyzed data from 577 pts. The median CIRS score was 7 (range, 0-25) with 54% (n=309) having CIRS ≥7. The median PFS was 11 months (95% CI: 8 - 15) and OS 30 months (95% CI: 23 - NA), with a median follow-up of 20 months. Although CIRS ≥7 was significantly associated with inferior PFS (HR=1.26) and OS (HR=1.35) in univariable analysis, it did not remain significant in multivariable models.
According to RSF variable importance and node splits, key CIRS categories which adversely impacted OS were respiratory, upper GI, renal, and hepatic. 9% of pts had a score of ≥3 in at least one of the above comorbidities, which we designated Severe4. When accounting for other significant variables, Severe4 was independently associated with inferior PFS (HR=2.45, p<0.001) and OS (HR=2.30, p<0.001) in the LC.
This finding was recapitulated in a single-center VC of 218 patients. In this cohort, median follow-up was 35 months, median age 61 years, median 3 prior therapies, ECOG 0-1 in 74%. As in the LC, axicabtagene ciloleucel was the predominant CART (93%). A CIRS comorbidity score of ≥3 in any of the Severe4 categories was found in 16% of pts. Severe4 remained independently associated with inferior PFS (HR=1.84, p=0.003) and OS (HR=1.82, p=0.007) in the VC (Figure).
Pts with Severe4 also had a higher rate of grade ≥3 CRS (16% vs 6%; p=0.013), while development of grade ≥3 ICANS was associated with CIRS ≥7 (26% vs 12%; p<0.001).
Conclusions: In this large RWE study, we demonstrate that the presence of comorbidity within the Severe4 composite index had prognostic significance for OS in CART recipients for r/r DLBCL. Importantly, Severe4 was validated in a separate cohort. Severe4 is predictive of severe CRS and CIRS ≥7 is predictive of severe ICANS. The underlying mechanism leading to this observation is an area of active investigation. Given these results, CIRS evaluation and Severe4 should be calculated prior to CART in DLBCL to identify those at highest risk for morbidity and mortality and to counsel pts about their expected risk. These results may be product specific given the enrichment of axi-cel patients and may need additional verification with other products.
Shouse, Geoffrey; Bailey, Neil; Patel, Krish; and See all authors in comments, "Impact of a Validated Composite Comorbidity Score on Outcomes in Patients Treated with CAR T-Cell Therapy for Diffuse Large B Cell Lymphoma (DLBCL): A Multicenter Real-World Evidence (RWE) Study" (2022). Articles, Abstracts, and Reports. 7442.