Ultrasensitive Circulating Tumor DNA (ctDNA) Dynamics after Autologous CD30.CAR-T Cell Therapy for Relapsed or Refractory (r/r) Classical Hodgkin Lymphoma (CHARIOT Trial)

Abstract

Introduction Liquid biopsies leveraging circulating tumor DNA (ctDNA) are rapidly emerging as minimally invasive tools to detect, genotype, and monitor diverse tumors. ctDNA dynamics and early responses are well-characterized in non-Hodgkin Lymphomas, especially during frontline immunochemotherapy. However, ctDNA responses and minimal residual disease (MRD) are less well-defined in classical Hodgkin Lymphomas (cHL), especially in the relapsed/refractory setting and after cellular immunotherapy, where significant unmet needs remain. We hypothesized that ultrasensitive ctDNA-based MRD detection by Phased Variant Enrichment & Detection Sequencing (PhasED-Seq; Kurtz et al, 2021 Nat Biotech) can accurately track response and predict progression after investigational therapy of r/r cHL with autologous CD30.CAR-T cells. Method This Phase 2 single arm, multicenter study (NCT04268706) investigates safety and efficacy of CD30.CAR-T cells in cHL patients experiencing progression after at least 3 lines of therapy. In the Pilot part of the study, ctDNA was analyzed as an exploratory biomarker using PhasED-Seq at multiple time-points, including at baseline (pre-treatment), Day 42 post-CD30.CAR-T infusion (D42), and upon progressive disease (PD). Exploratory analyses were performed to assess the predictive role of ctDNA and to compare metabolic tumor response by positron emission tomography (PET) scan with ctDNA-based minimal residual disease (MRD). We profiled 55 serial blood specimens (41 plasma and 14 buffy coat samples) from 14 patients with available samples. Cell-free DNA was profiled by PhasED-Seq in the Foresight Diagnostics CLIA laboratory (Aurora, CO). Baseline plasma and D42 buffy coat specimens were used to genotype each patient's tumor to identify Phased Variants (PVs), and these were used to monitor MRD in subsequent blood specimens. Patients/samples were reported as MRD positive when ctDNA levels exceeded an analytical detection threshold (~1 part per million cfDNA molecules), corresponding to 98% clinical specificity. Result As of 22 July 2022, 17 patients have been screened and 15 patients enrolled in the Pilot segment (median age: 35 years [21-57], 66.7% male). Overall response rate (ORR) after CD30.CAR-T single infusion was 73.3% (n=11), with complete response (CR) rate of 60% (n=9). ORR in 5 patients after the 2nd infusion was 100% (n=5), with CR of 60% (n= 3). ctDNA was successfully genotyped directly from pre-treatment plasma in 12 out of 14 patients where samples were available. Both patients where PV genotyping failed had received bridging cytoreductive therapy before baseline plasma collection. Out of 12 profiled patients, 7 patients achieved CR with median baseline ctDNA concentration of 44.9 hGE/mL. 3 patients had PD with median baseline ctDNA level of 94.0 hGE/mL. One patient with PR and one patient with SD had baseline ctDNA levels of 38.9 hGE/mL and 231.1 hGE/mL, respectively. These results suggest that pre-treatment ctDNA levels could have predictive value on patient response to CAR-T therapy. ctDNA responses largely mirrored radiographic responses. Specifically, among 7 patients achieving radiographic CR, 4 also achieved undetectable ctDNA, and the other 3 had ctDNA responses with low MRD levels (median: 0.2 hGE/mL) at D42. One patient with PR also had low ctDNA concentration at D42 (0.7 hGE/mL). In comparison, 3 patients experiencing PD had relatively higher ctDNA levels (median: 131.4 hGE/mL). Furthermore, ctDNA levels at D42 were significantly correlated with PET Deauville score (p=8.7e-4), and patients with a response had significantly lower ctDNA than those without a response (p=0.004) (Fig B). These results suggest that ctDNA-MRD levels are strongly correlated with metabolic responses determined by PET imaging. Similar results were observed after 2nd infusions of the CAR-T. 3 out of 3 patients achieving CR had low ctDNA concentrations at Day 42 post-2nd infusion (median: 0.2 hGE/mL). Among 2 patients with PR, 1 patient with available sample had 31.9 hGE/mL ctDNA. Conclusion Minimally invasive ctDNA analysis is a viable method to monitor responses, rapidly risk stratify and predict outcomes of patients with r/r cHL treated with CD30.CAR-T therapy. A single infusion of CD30.CAR-T therapy rapidly induces MRD negativity in a significant subset of patients at Day 42. ctDNA as exploratory biomarker will be further evaluated in Pivotal part of the study. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal