Pre-CAR-T pulmonary function test and respiratory comorbidities have limited predictive value for survival, toxicities, and respiratory complications

Abstract

Abstract Introduction Pre-hematopoietic cell transplantation pulmonary function tests (PFT) are routinely performed for risk-stratification and post-transplant surveillance. Their abnormalities are associated with increased risk of pulmonary complications, intensive care admission, and mortality. However, their predictive value in chimeric antigen receptor T-cell (CAR-T) therapy remains controversial, with mixed evidence (Sdayoor et al. 2025; Walder et al. 2025). We conducted a retrospective study of PFT, respiratory comorbidities, survival, and 100-day toxicities in B-cell lymphoma. Methods and Results Demographic, disease diagnosis, treatment-related data (line of previous therapy, previous HSCT, lymphodepletion regimen, CAR-T product), respiratory comorbidities (smoking, chronic obstructive pulmonary disease, asthma, interstitial lung disease, previous COVID-19, supplemental oxygen requirement), treatment follow-up (survival, response at 30 days, need for mechanical ventilation and acute hypoxemic respiratory failure [AHRF] within 100 days after CAR-T, immune effector cell-associated neurotoxicity syndrome [ICANS] and cytokine release syndrome [CRS] grades) were collected from 70 patients who received autologous CD-19 CAR-T at 2 NCI-designated cancer centers between year 2020 and 2024. Forced vital capacity % predicted (FVC% pred), forced expiratory volume in one second % predicted (FEV1% pred), and diffusing capacity for carbon monoxide % predicted (DLCO% pred) were collected from patients at Center A (n = 48), where pre-treatment PFT were required, although no treatment was denied based on PFT results. FVC% pred, FEV1% pred were categorized as normal (>80%), moderate (66%-80%), and severe (<66%). DLCO% pred was treated as continuous data. The remaining patients constituted the other cohort. This study was approved by both institutional review boards. Statistical analyses were performed in R. Baseline demographic (sex, age, race, ECOG performance status scale), diagnoses of B-cell lymphoma, respiratory comorbidities, and treatment-related data were comparable between the two cohorts, except that more patients in Center A received CAR-T after first-line of treatment (95.7% vs 39.1%). Axi-cel was used more in Center A (78.7% vs 47.8%), while Liso-cel, Tisa-cel, Brexu-cel were used more in others (14.9% vs 21.7%, 0% vs 13.0%, 2.1% vs 17.4%). Median follow-up was 12.2 and 20.3 months, respectively. Thirty-day survival rates were 100% in both cohorts. Overall survivals were comparable (Log-rank test, p=0.41). Complete remission, partial remission, and progressive disease status at 30 days were comparable among cohorts (65.8%, 21.1%, 13.2% vs 72.7%, 13.6%, 13.6%, p=0.55, chi-squared test). There was no significant association between FVC% pred, FEV1% pred, DLCO% pred and overall survival (Cox proportional hazards model, hazard ratio [HR]: 0.18, 95% confidence interval [CI]: 0.02- 1.92, p=0.16; HR: 4.24, 95% CI: 0.78-23.14, p=0.10; HR: 1.00, 95% CI: 0.96 - 1.04, p=0.87; respectively), or overall response (logistic regression, odds ratio [OR]: 2.15, 95% CI: 0.13-52.21, p=0.99; OR: 3.69, 95% CI: 0.24-55.62, p=0.34; OR: 0.98, 95% CI: 0.92-1.05, p=0.56; respectively). Only 2 patients required mechanical ventilation in 100 days after CAR-T. Univariate logistic regression showed no significant association between PFT measurements and AHRF, CRS or ICANS grades ≥3 in 100 days after CAR-T (for AHRF, FVC% pred [OR 0.46, 95% CI 0.03-7.92, p=0.60], FEV1% pred [OR 2.35, 95% CI 0.24-23.06, p=0.46], DLCO% pred [OR 1.01, 95% CI 0.96-1.07, p=0.70]; for CRS, OR for FVC% pred and FEV1% pred cannot be calculated due to low events, DLCO% pred [OR 1.05, 95% CI 0.95-1.17, p=0.33]; for ICANS, FVC% pred [OR 0.98, 95% CI 0.15 -6.34, p=0.98], FEV1% pred [OR 0.66, 95% CI 0.0.09-4.92, p=0.68], and DLCO% pred [OR 0.99, 95% CI 0.94-1.04, p=0.77]). Using pooled data from both cohorts, univariate logistic regression showed no respiratory comorbidity as a risk factor for AHRF, mechanical ventilation, CRS, or ICANS grades ≥3. Prior documented COVID-19 had higher odds of developing AHRF, although not statistically significant (OR 4.17, 95% CI 0.91-19.10, p=0.07). Conclusion Pre-CAR-T PFTs and respiratory comorbidities did not significantly predict overall survival, CRS/ICANS, AHRF, or mechanical ventilation within 100 days in B-cell lymphoma patients treated with CAR-T. These results suggest against the routine use of PFT for CAR-T workup or eligibility.