Abstract

Abstract Introduction: Circulating tumor DNA (ctDNA) can act as a molecular response (MR) biomarker. Non-tumor informed, plasma-only ctDNA assays are an accessible and rapid mechanism to assess ctDNA levels in patients undergoing treatment for lung cancer but may be impacted by clonal hematopoietic (CH) mutations misclassified as tumor derived. Here, we assess the impact of CH on MR measurement in patients receiving durvalumab, with or without tremelimumab, versus chemotherapy in the Phase III MYSTIC trial. Methods: Using a non-tumor informed, mutation based, ctDNA assay that assesses 500-genes (Guardant Health [GH] Omni, subset to the GH360 genomic coordinates [74-genes] for additional analyses) we calculated plasma-only molecular response (MR) based on change in ctDNA levels between baseline and a 4 or 6-week on-treatment timepoint in 713 patients. FoundationOne CDx tissue data subset to GH Omni genomic coordinates enabled “in-silico” tumor-informed MR analysis where only mutations detectable in tumor tissue were followed as response indicators, in 273 of 713 patients. Targeted duplex-sequencing of peripheral blood mononuclear cell (PBMC) DNA was performed to assess the presence of CH mutations misclassified as tumor derived through plasma-only analyses. Results: Plasma-only MR assessment associated with survival outcomes in immunotherapy treatment arms but not the chemotherapy arm; tumor shrinkage on imaging at 6-week radiological evaluation also predicted survival outcomes only in the immunotherapy arms, suggesting that early tumor response is a stronger predictor of outcome for patients treated with immunotherapy. In-silico tumor-informed MR analyses led to an improved association between MR and outcome across all treatment arms and an increased rate in observed ctDNA clearance. Early tumor-informed ctDNA clearance strongly indicated survival benefit in patients receiving durvalumab and tremelimumab. Targeted analyses of mutations categorized as tumor derived in plasma-only analyses, but absent from tumor tissue sequencing analyses in the same patient, revealed that 87/154 (57%) were PBMC-derived highlighting CH as capable of confounding plasma-only analyses. Conclusion: These exploratory analyses from a phase III randomized study support early-ctDNA kinetics as predictive of long-term benefit in metastatic NSCLC patients and support technical optimization of CH-subtraction approaches to maximize utility of ctDNA as a response biomarker in patients with NSCLC. Citation Format: Chris Abbosh, Michael Kuziora, Daniel Stetson, Zhongwu Lai, Yashaswi Shrestha, Paul Labrousse, Xiaojin Shi, Helen Mann, Diana Merino Vega, Matthew D. Hellmann, Solange Peters, J. Carl Barrett, Darren Hodgson. Misclassification of clonal hematopoietic mutations as tumor-derived confound accurate ctDNA-response assessment in patients with metastatic non-small-cell lung cancer: analyses from the phase 3 MYSTIC trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2299.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call