Comparative accuracy of alternative early measures of residual disease after car19 therapy of relapsed/refractory lbcl

Introduction: Over half of relapsed/refractory large B-cell lymphoma (rrLBCL) patients receiving anti-CD19 chimeric antigen receptor (CAR19) T-cells experience subsequent relapse. Improved methods are therefore needed to predict outcomes, and to accurately measure minimal residual disease (MRD). While PET/CT metabolic response ∼4 weeks after CAR infusion is established for such early assessments, several liquid biopsy MRD methods (including IgHTS and CAPP-Seq) have been reported to noninvasively measure responses (Frank et al. JCO, 2021; Sworder et al. ICML-16). Phased variant enrichment and detection sequencing (PhasED-seq) allows further improved sensitivity to detect circulating tumor-derived DNA (ctDNA) by monitoring of phased-variants (PVs), comprising multiple independent somatic mutations in individual cell-free DNA fragments. Here, we compared ∼week-4 residual disease evaluations by PET/CT, CAPP-Seq, and PhasED-Seq in patients receiving standard of care axicabtagene ciloleucel (axi-cel) CAR19 therapy (Sworder et al., Cancer Cell, 2023). Methods: Tumor or baseline plasma and matched PBMC samples were used to genotype SNVs and PVs, with subsequent evaluation of MRD status in on-treatment samples by CAPP-Seq (SNVs) or PhasED-Seq (PVs). We empirically determined an analytical sensitivity of at least 1:1e5 to result in optimal performance for PhasED-Seq for MRD detection after CAR19. Patients were considered evaluable for such analyses by having week +4 and adequate baseline samples available allowing 1:1e5 analytical sensitivity. PET/CT at week +4 was assessed by Deauville 5-point score per Lugano criteria. Results: A total of 29 patients were evaluable for all 3 evaluations at week +4 (PET/CT, CAPP-Seq, PhasED-Seq). Patients with detectable MRD by PhasED-Seq at this landmark had inferior event-free survival (EFS) [log-rank p = 0.00018, Cox HR = 7.88 (95% CI: 1.9–33.3); Figure 1A]. MRD positivity by CAPP-Seq [log-rank p = 0.0028, Cox HR = 3.02 (95% CI: 1.4–6.6), Figure 1B] and failure to achieve a metabolic complete response (Deauville 1–3) by PET/CT [log-rank p = 0.015, Cox HR=2.78 (95% CI: 1.1–6.6); Figure 1C] were also associated with inferior EFS, however, PhasED-Seq outperformed both of these methods in predicting clinical outcomes. When considering patients with discordant ctDNA detection between CAPP-Seq and PhasED-Seq at week +4, PhasED-Seq detected ctDNA in an additional 3 patients that ultimately relapsed (Figure 1D). Among patients that failed to achieve a complete metabolic response at week +4, PhasED-Seq demonstrated 100% specificity in identifying which of these patients would not experience disease progression (Figure 1E). Keywords: cellular therapies, diagnostic and prognostic biomarkers, minimal residual disease Conflicts of interests pertinent to the abstract B. J. Sworder Consultant or advisory role: Foresight Diagnostics S. K. Alig Honoraria: Takeda M. Shahrokh Esfahani Consultant or advisory role: Foresight Diagnostics J. H. Baird Research funding: Kite Pharma Y. Natkunam Consultant or advisory role: Leica Biosystems, Roche Research funding: Kite Pharma R. G. Majzner Consultant or advisory role: Lyell Immunopharma, NKarta, Arovella Pharmaceuticals, Innervate Radiopharmaceuticals, GammaDelta Therapeutics, Aptorum Group, Zai Labs, ImmunAI, Gadeta, FATE Therapeutics (DSMB), Waypoint Bio Stock ownership: Link Cell Therapies, CARGO Therapeutics C. L. Mackall Consultant or advisory role: Lyell, CARGO, Link, Apricity, Nektar, Immatics, Mammoth, Ensoma. Stock ownership: Lyell Immunopharma, CARGO Therapeutics, Link Cell Therapies D. B. Miklos Consultant or advisory role: Pharmacyclics, Kite Pharma, Adaptive Biotechnologies, Novartis, BMS, Janssen Pharmaceuticals, Roche, Genentech, Precision Bioscience, Allogene, Miltenyi Biotec, Fate Therapeutics, 2Seventy, Adicet M. Diehn Consultant or advisory role: AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, Genentech, Gritstone Oncology, Illumina, Novartis, Roche Stock ownership: CiberMed, Foresight Diagnostics Research funding: AstraZeneca, Genentech, Varian Medical Systems, Illumina M. J. Frank Consultant or advisory role: Adaptive Biotechnologies Stock ownership: Allogene, Roche/Genentech Research funding: Adaptive Biotechnologies, Kite/Gilead D. M. Kurtz Consultant or advisory role: Roche, Adaptive Biotechnologies, Genentech Stock ownership: Foresight diagnostics A. A. Alizadeh Employment or leadership position: Foresight Diagnostics Consultant or advisory role: Celgene, Chugai, Genentech, Gilead, Janssen, Pharmacyclics, Roche Stock ownership: CiberMed Inc., Foresight Diagnostics, FortySeven Inc., CARGO Therapeutics. Research funding: Celgene, Bristol Myers Squibb, Pfizer