Do mental health disorders modify cytokine release syndrome risk in immune effector cell therapy? Full-cohort analysis with tarlatamab validation.

Abstract

e19002 Background: Cytokine release syndrome (CRS) is a common immune effector cell (IEC) therapy toxicity. Host psychiatric effects on immunotoxicity are underexplored. The stress-immune axis or pre-existing neuro-inflammatory states common in mental health disorders (MHDs) may impact IEC toxicity. We therefore evaluated whether pre-treatment MHDs affect CRS risk, with an independent tarlatamab cohort for validation. Methods: This was a retrospective, single-center study of adults receiving commercial/investigational IECs (CAR-T or bispecific T-cell engagers; N=105). The primary exposure was a documented pre-treatment MHD (anxiety, depression, bipolar, or PTSD). The primary outcome was any-grade CRS per ASTCT criteria. Multivariable logistic regression with stepwise adjustment (therapy alone; therapy+gender+ECOG) was used. Sensitivity analyses included log-transformed baseline CRP adjustment and stabilized inverse probability of treatment weighting (IPTW). Validation was performed in a tarlatamab subgroup (N=33). Secondary outcomes included CRS (grade ≥2), ICANS, and treatment response. Results: Among 105 IEC recipients (MHD+ n=53, MHD− n=52), CRS occurred in 18/53 (34.0%) compared to 34/52 (65.4%; Absolute Risk Reduction [ARR] 31.4%). In the primary adjusted model (n=79), MHD+ remained significantly protective (Odds Ratio [OR] 0.33; 95% CI 0.11–0.98; p=0.046). Sensitivity analyses confirmed this protective association: CRP-adjusted OR 0.31 (95% CI 0.10–0.97); IPTW OR 0.35 (95% CI 0.15–0.79). In the tarlatamab validation cohort, CRS risk was notably lower at 18.2% vs 54.5% (ARR 36.3%; OR 0.18; 95% CI 0.03–0.91). Grade ≥2 CRS trended lower in the MHD+ group, but MHD was not associated with ICANS or peak CRP levels. Importantly, efficacy was preserved (response OR 2.03; p=0.401). Conclusions: Pre-treatment MHD was consistently associated with a lower CRS risk across all statistical models and in the independent tarlatamab validation, without compromising anti-tumor efficacy. Further prospective validation of these novel findings is highly warranted to confirm the clinical implications. Analysis outcomes (MHD+ vs MHD−). Analysis N Rate (+/−) OR (95% CI) P CRS – Unadjusted 105 34.0%/65.4% 0.27 (0.12–0.61) 0.0016 CRS – +Therapy 105 34.0%/65.4% 0.38 (0.16–0.90) 0.0284 CRS – Adjusted 79 30.2%/63.9% 0.33 (0.11–0.98) 0.0463 CRS – CRP adj 81 41.0%/71.4% 0.31 (0.10–0.97) 0.0441 CRS – IPTW 105 34.7%/60.6% 0.35 (0.15–0.79) 0.0119 Tarlatamab – CRS 33 18.2%/54.5% 0.18 (0.03–0.91) 0.0381 CRS (≥Gr2) 79 9.4%/28.8% 0.27 (0.07–1.06) 0.0604 Response 65 74.3%/83.3% 2.03 (0.39–10.62) 0.401 OR<1 indicates lower odds in MHD+ vs MHD− for CRS endpoints; OR>1 indicates higher odds of response in MHD+.