Longitudinal assessment from liquid biopsy of mutations in CD20: A pilot study using a PETE enrichment strategy.

Abstract

7042 Background: In recent years, bispecific T cell Engagers (BsTCE) targeting CD3 and CD20 have emerged as a potent new class of therapeutics for NHL; however, a subset of patients still experience relapsed or refractory disease. Patients undergoing treatment with BsTCEs could benefit from longitudinal screening via liquid biopsy to identify baseline (primary) and treatment-induced (acquired) resistance mutations and tailor treatment accordingly. Here we present the results from a pilot study screening patients treated with Glofitamab (CD20xCD3) using a single gene Primer Extension Target Enrichment (PETE) strategy for MS4A1 (CD20 gene) in liquid biopsies. Methods: Our pilot study was conducted using libraries from the plasma of 134/155 patients with relapsed or refractory large B cell lymphoma (r/r LBCL) who underwent Glofitamab treatment at approved dose in the phase I/II trial NP30179 ( NCT03075696 ) (Dickinson, et al. N Engl J Med 2022), and paired PBMC/PDB available in 91 cases. Primers were designed against the coding regions of the MS4A1 gene. Enrichment was performed using a workflow optimized for the detection of somatic variants in cell-free DNA isolated from plasma. Results were analyzed using a modified AVENIO circulating tumor (ct) DNA (Roche; For Research Use Only) analysis workflow. Variants detected by the analysis pipeline were further filtered based on inclusion criteria: allele fraction > 0.1%, rarity in the cohort, impact on protein function in silico , and if sample was available, absence in germline. Tumor burden as assessed by ctDNA was obtained from retrospective sequencing data. Response to treatment was assessed by PET/CT using the Lugano Criteria. Results: Using inclusion criteria a total of 11/134 (8.2%) patients were identified with a total of 16 unique MS4A1 candidate mutations at baseline or during treatment. The Best Overall Investigator Response (BOR) was progressive disease (PD) for 7 patients and partial metabolic response (PR) for 4 patients. All patients with BOR PR experienced disease progression before treatment completion. Sufficient samples were available to demonstrate expansion of the candidate mutation by the end of treatment (EOT) timepoint for 5 patients. Four of the identified mutations were previously reported in the literature, the remaining mutations were novel, and in vitro characterization demonstrated their functional impact. Conclusions: This work identifies known and novel mutations in CD20 in plasma samples as a potential contributing mechanism to relapsed or refractory cases of NHL. Although the prevalence appears to be low, this is consistent with previous reports and supports investigation of the clinical utility, including utility as a potential predictive biomarker, of sequencing this gene during treatment with CD20xCD3 BsTCE. Future and ongoing work will screen additional cohorts and therapy combinations.