Pre-Treatment Pulmonary Assessment Have Limited Utility in B-Cell Lymphoma Patients Treated with Autologous Anti-CD19 CAR T-Cell Therapy

Introduction

Pulmonary function tests (PFT) play a pivotal role in hematopoietic cell transplantation (HCT), with routine evaluation in HCT candidates. However, their predictive value in chimeric antigen receptor T-cell (CAR-T) therapy outcomes remains uncertain. To address this gap, we conducted a retrospective analysis of PFT metrics and pulmonary comorbidity definitions, following the Hematopoietic Cell Transplantation-Comorbidity Index (HCT-CI), in B-cell lymphoma patients treated with autologous CD19-CAR-T cell therapy.

Patients and Methods

In this retrospective analysis, we included adult patients (age ≥ 18 years) with B-cell lymphoma who received treatment at a tertiary medical center with commercially available or point-of-care (POC) anti-CD19 CAR T-cell therapy (Kedmi et al., Transplant Cell Ther. 2022) as their 3rd line of treatment or above, between 2017 and 2022. Patients underwent routine PFT assessment before CAR-T cell therapy, as part of institutional practice, including spirometry, body plethysmography, and single-breath diffusing capacity for carbon monoxide (DLCO). Patients were categorized into three pre-treatment pulmonary comorbidity groups (normal, moderate, severe) based on HCT-CI definitions for forced expiratory volume in 1 second (FEV1) and DLCO expressed as precent of predicted value (>80%, 66-80%, <66%, respectively).

Results

A total of 192 patients (median age, 60 [range 46-69]) were included. Aggressive B-cell lymphoma was the predominant diagnosis (n=166, 86%) followed by indolent B-cell lymphoma (n=18, 9.4%), and Mantle cell lymphoma (n=8, 4.2%). Most patients had a good performance status (KPS ≥ 90, 78%), although with high-risk disease features (pre-lymphodepletion LDH was elevated in 108 [56%]). Seventy-eight (41%) patients received bridging therapy. POC CAR-T were used in 56%, while 44% received commercial CAR-T (26% axi-cel, 18% tisa-cel).

Pre-CAR-T smoking history was documented in 25% of patients. The median pre-treatment measured-to-predicted DLCO and FEV1 were 96% and 92%, respectively. Seventy-four percent of patients had normal DLCO, and 77% had a normal FEV1. The calculated normal, moderate, and severe pulmonary comorbidity levels, per the HCT-CI, were 64%, 23%, and 13%, respectively.

With a median follow-up of 17.6 months (IQR 8.5-29.7), the median overall survival (OS) was not reached, and the median progression-free survival (PFS) was 7.6 months.

There was no correlation between pulmonary comorbidity level and the best overall response rate.

Categorical PFT results (FEV1 and DLCO) and pulmonary comorbidity (per-HCT-CI) were not associated with reduced OS (p=0.3, p=0.4, p=0.6, respectively) or reduced PFS (p=0.058, p>0.9, p=0.2, respectively) in a univariable Cox regression analysis. However, FEV1 as continuous measure and FEV1 categorized into 2 levels (normal vs. moderate-severe) were associated with reduced PFS ( Figure 1). FEV1 (as a continuous metric) association with PFS remained consistent (HR 0.99 [95% CI 0.98-1], p=0.007) in a multivariable Cox regression adjusted for pre CAR T KPS score, elevated LDH level, and co-stimulatory CAR-T domain. Nevertheless, this association is minor, with unclear clinical significance. No association was observed between pulmonary function measures and immune-toxicities, including cytokine release syndrome grade >2 or immune effector cell-associated neurotoxicity grade > 2.

Conclusions

To the best of our knowledge, this is the first study evaluating the relationship between PFTs and the outcomes of CAR T-cell treatment. Our results indicate that among standard measures of pulmonary assessment, only FEV1 was predictive of poor PFS, though with minor association, with unclear clinical significance. We did not observe any association with overall survival. These findings suggest that pre-CAR-T pulmonary assessment may have limited utility in evaluating CAR-T candidates. Nevertheless, further validation in a multicenter setting is required.