Multiple time-point FDG-PET imaging for risk stratification of patients with relapsed/refractory large B-cell lymphoma undergoing CD19-directed CAR-T cell therapy

Abstract

Abstract Introduction CD19-directed CAR-T cell therapy is a curative approach for patients (pts) with relapsed/refractory (r/r) large B-cell lymphoma (LBCL). However, still around 60% of pts relapse post-CAR-T. Thus, the thorough risk assessment of pts in pre- and post-CAR-T setting is of high importance. Here, we investigate the risk stratification of r/r LBCL pts by baseline FDG-PET at CAR-T treatment as well as on day +30 (d30) and day + 100 (d100) post-CAR-T. Material and Methods Standard PET parameters (SUVmax, peak, mean, min), 40% SUVmax threshold-based metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were obtained retrospectively from all PET scans prior and post-CAR-T administered in r/r LBCL pts in two university hospitals. Response in PET measurements of the d100 PET were expressed as percentage relative to the baseline PET. Deauville-Scores (DS) for pts with complete metabolic response (CMR with DS1-3) and partial metabolic response/stable disease (PMR with DS4, and SD included a subset of DS5 pts) were investigated on d30 PET to predict progression-free survival (PFS) and overall survival (OS) using Kaplan-Meier analysis and Cox regression. Patients with progressive disease (PD) (≥30% increase of SUVpeak) were excluded from d30 and d100 PET predictive and prognostic analysis. Baseline and d100 Delta PET parameters were analyzed using the Max.stat statistical test in R to determine the optimal cut-off for low-risk and high-risk groups in predicting PFS and prognosticating OS in Kaplan-Meier analysis and Cox regression. Results 87 pts with r/r LBCL and baseline FDG-PET at CAR-T cell therapy (axicabtagene ciloleucel n=69, 85%; tisagenlecleucel n=11, 13%; lisocabtagen maraleucel n=2, 2%) were included. Post-CAR-T, a subset of 67 pts (77%) and 74 pts (85%) underwent PET on d30 and d100. The median follow-up time post-CAR-T was 14 months (mo). Considering obtained SUVmax values at CAR-T, the cut-off value < 37.5 stratified pts into low- (n=76) and high-risk (n=11) groups for relapse and death post-CAR-T with a hazard ratio (HR) of 0.3 for PFS (P=0.02) and HR of 0.2 for OS (P=0.001) in the low-risk group. Baseline PET MTV was significant for OS prognostication (P<0.001) with a cut-off ≤ 232.5 cm3defining low-risk (n=74) and poor survival for high risk (HR: 3.1, P=0.02; n =13). Patients with PMR or SD on d30 PET (DS4-5) (n=31) exhibited a significantly higher probability of relapse (HR: 3.4, P=0.001) and mortality (HR: 3.1, P=0.02) in subsequent follow up compared with CMR (DS1-3) pts (n = 29). In the DS4-5 group, the median PFS (mPFS) was 3.5 mo (P=0.001) and median OS (mOS) 14 mo (P=0.02). In the DS1-3 (CMR) group the median PFS and OS was not reached at the end of follow-up. In the subgroup who received a d100 PET, 31 (41%) subjects were excluded due to PD for PFS analysis. On d100, those exhibiting a MTV reduction ≥ 40.2% (low-risk, n = 43) demonstrated a significant lower probability of relapse (HR: 0.27, P = 0.002) compared to those without. In detail, pts categorized as high risk (n=13) had a mPFS of 7 mo (P<0.0001), while 89.2% of low-risk pts had no relapse at the end of follow-up. OS analysis was done for the entire subgroup of patients who received a d100 (n = 74) with a cut-off of 14.4% for MTV reduction. Pts showing at least a reduction of 14.4% in MTV (low-risk, n=54) exhibited a significantly reduced likelihood of mortality (HR: 0.26, P=0.006) with 83.4% of them being alive at the end of follow-up (P=0.002) compared with those who had a lower reduction of MTV. Conclusion Higher metabolic activity (SUVmax) before CAR-T cell therapy is associated with worse outcomes post-CAR-T. Moreover, a non-CMR status on d30 post CAR-T was associated with an almost three-fold higher risk for subsequent relapse and mortality. A MTV reduction by 14 % on d100 post-CAR-T is associated with longer survival time. Our results indicate that a sufficient remission prior-CAR-T is a prerequisite for sustained remission and long-term survival post-CAR-T. An early remission status control by FDG-PET on d30 is recommended to identify pts at high risk for treatment failure and helps to guide treatment decisions whereas pts without an evident progression on d100 but insufficient MTV reduction still require a very close monitoring.