The use of circulating tumor DNA (ctDNA) in diagnosing and monitoring treatment responses in hematologic malignancies has multiple advantages: (1) ctDNA is noninvasive, allowing for frequent, serial, real-time tracking of multiple tumor clones; (2) ctDNA is highly sensitive for clone-specific genomic aberrations, and may rival traditional minimal residual disease (MRD) methodologies in detecting low-volume disease; and (3) ctDNA “integrates” the tumor genome across multiple body compartments, allowing the detection of emerging resistant clones that would otherwise be missed by single-site biopsies. Since 2014, our group has evaluated the use of ctDNA in patients with hematologic cancers initially as an investigational tool, and more recently as an accredited (ISO15189) hybridization-based NGS panel that can perform sequence variant detection, genome-wide copy number calling, and structural variant detection from ctDNA. In this talk, we will discuss different clinical scenarios and the evolution of ctDNA at our center. Our initial proof-of-concept study was conducted in patients receiving BTK inhibitor (BTKi) and/or BCL2 inhibitor (BCL2i) therapy for chronic lymphocytic leukemia (CLL) or mantle cell lymphoma (MCL). Clinical challenges in this population include response assessment being confounded by BTKi redistribution lymphocytosis, differential compartmental responses to different classes of novel agents, and emergence of resistant clones in disparate disease sites. In our study of CLL, serial samples from 32 patients treated with either ibrutinib (BTKi, n=22) or venetoclax (BCL2i, n=10) were analyzed, including 3 patients who developed RS. Using a bespoke TS panel of 7 commonly mutated genes in CLL (SF3B1, NOTCH1, ATM, TP53, MYD88, KRAS, and BIRC3), somatic mutations comprising 0.1% to 90% of total ctDNA were detected in 25 of 32 (78%) patients. Serial monitoring of MAF was concordant with disease response as determined by CT scans and, importantly, ctDNA was able to “look beyond” the artificial rise in circulating tumor cells caused by BTKi, showing an early and consistent fall in MAF. ctDNA was sensitive to the detection of disease clones in occult body compartments—of 88 serial timepoints assessed by both ctDNA and MNL DNA after the initiation of novel therapy, mutations were found to be detectable only in ctDNA in 18 (20%) timepoints. In 30 timepoints where MRD testing was performed by both ctDNA and high-resolution flow cytometry, there was 100% concordance in results. In a second study, serial ctDNA analyses were performed in a cohort of 24 patients with MCL receiving combination ibrutinib and venetoclax therapy. Using TS and a panel of 42 genes known to be recurrently mutated in MCL, trackable ctDNA mutations were detected in 71% of patients, with baseline MAF ranging from 0.4-60%. Similar to our experience with CLL, dynamic changes in ctDNA closely mirrored tumor responses as determined using traditional staging methods of CT, PET, endoscopy, and bone marrow examination. In one patient, ctDNA started rising several weeks before relapse was evident by traditional MRD methods. A further patient had progression in an occult site (pleura), with acquisition of new CNAs on chromosome 6, including 6p loss. In this patient, these new CNAs were readily identified by LC-WGS of plasma. Finally, in a patient with primary refractory disease, an acquired loss of chromosome 9p was identified in plasma 8 weeks before disease progression was evident on imaging. We have since optimized and validated ctDNA testing for routine use into our diagnostic laboratory NGS workflow and we will discuss recent cases of BTK and BCL2 mutations being readily detectable in patients in early stages of novel agent resistance, facilitating early transition to next-line therapy. In aggressive lymphomas, we will demonstrate the value of ctDNA in detecting CD274(PD-L1)/PDCD1LG2 (PD-L2) copy number abnormalities and structural variants in patients with lymphomas in inaccessible tissue sites, guiding the rational use of checkpoint inhibitors. Other important structural variants that are able to be detected in ctDNA through this methodology are IGH-MYC and IGH-BCL2 translocations, thus allowing rapid categorization of patients with diffuse large B-cell lymphoma as “double-hit” lymphoma. Taken together, these studies underscore the value of ctDNA as a noninvasive, highly sensitive tool for assessment of genomic landscape and tracking clonal dynamics across multiple body compartments in patients with hematologic malignancies.

Citation Format: Constantine S. Tam, Piers Blombery. ctDNA in indolent lymphoid cancers treated with targeted therapies [abstract]. In: Proceedings of the AACR Virtual Meeting: Advances in Malignant Lymphoma; 2020 Aug 17-19. Philadelphia (PA): AACR; Blood Cancer Discov 2020;1(3_Suppl):Abstract nr IA44.