NEW YORK – Researchers at the University of Miami's Sylvester Comprehensive Cancer Center are launching a research project that will use whole-genome sequencing as well as single-cell transcriptomics to study tumor resistance to cell therapies.
Last month, the team announced a three-year, $1.5 million grant from the US Department of Defense to help study patients with diffuse large B-cell lymphomas (DLBCL) that either did not respond to chimeric antigen receptor (CAR) T-cell therapy or who suffered relapse.
Led by Jonathan Schatz, Francesco Maura, and Juan Alderuccio, all three professors at UMiami, the team will perform whole-genome sequencing on tissue samples from more than 60 patients to search for biomarkers associated with non-response or relapse. The team will also use single-cell transcriptomics to probe the relationship between genotype and phenotype.
"We recently showed that characteristics of tumor genomes detectable by whole-genome sequencing can predict the likelihood of failed CAR-19 responses more strongly than previously reported markers," Schatz said. "This is an aspect of CAR T-cell therapy response which is understudied in our field. Studies have primarily looked at factors related to the CAR T product itself and the tumor microenvironment. These are quite appropriate, but what we haven't looked at is the tumor genome itself — that's where the problem came from, and that dictates and shapes these responses. That's the focus of our project."
The study will be done in parallel to a Genentech-supported clinical trial of a combination of the firm's antibody treatments Lunsumio (mosunetuzumab) and Polivy (polatuzumab), as well as CAR T-cell therapy. That study will provide the patients' samples, as will Michael Green of MD Anderson Cancer Center. Schatz said he also has a pending application for an R01 grant from the National Institutes of Health to support the study. Furthermore, the team plans to create model systems to explore the patient-derived biomarkers in a lab setting.
"The core problem in the field is poor outcomes," Schatz said. "It's still the case that more than half [of patients] will not have a longer term or even any response at all. That's why we focus on resistance."
The project builds on a prior study co-led by Schatz and Maura and Michael Jain, Frederick Locke, and Marco Davila at Moffitt Cancer Center published in Blood in August 2022 that focused on anti-CD-19 CAR T cells. Those cells can treat patients with aggressive B-cell lymphomas; however, they work in less than half of patients. Inflammatory markers and other clinical factors are associated with non-response to treatment, but the study was among the first to look for signs in the genomes of the tumor itself and the first to employ WGS.
The researchers reported whole-genome sequencing results from 49 patients with relapsed/refractory cases, at a median coverage of around 45X, as well as from 50 newly diagnosed DLBCL patients. Their analysis included mutations in known oncogenes, single-nucleotide variants (SNVs), and copy number variants, as well as structural variants and markers of genome complexity.
"We observed that markers of genomic complexity (chromothripsis and APOBEC) and specific genomic alterations (RHOA and RB1 deletion) associate with resistance to CAR-19 immunotherapy for aggressive B-cell lymphomas," the authors wrote. "Fifteen out of 16 patients who relapsed … contained at least one of the described genomic alterations."
"The genomic factors, when you pulled them together, ended up being a stronger predictor of outcome" compared with other clinical features such as tumor volume or patient age, Schatz said. "We think it's very promising."
Doing WGS will provide an opportunity to build on existing molecular tumor classifications, which are based on whole-exome data. "They're good, but not good enough," Schatz said. Two exome-based papers — led by groups based at the National Cancer Institute and Harvard University, respectively — have provided new levels of insight. "But when you think about what you can get out of a whole genome, they leave out a lot of things we've yet to understand. We're building on that, not denigrating it. We think we can learn even more," he said.
Despite an interest in large-scale genomic rearrangements, such as chromothripsis, the study will use Illumina short-read sequencing. "For us, what's more important is the depth of the sequencer," Schatz said. Tumor genomes will be sequenced at coverage between 60X and 80X, and the study will also generate corresponding patient germline genomes.
The researchers will also incorporate single-cell transcriptomics from 10x Genomics, and is considering single-cell ATAC-seq (assay for transposase-accessible chromatin by sequencing) for future work.
The study will perform both WGS and single-cell RNA-seq for about 40 patients. All of these will get the analyses done pre-treatment and 15 — the number expected to relapse, statistically — will have the assays performed a second time when their disease returns.
Starting in July, the DoD grant will power the study for at least three years, but Schatz is hoping to build out a large collaborative research program. "I think that's a model we'd like to further pursue," he said, "where we can really become a center where we focus on the genomics of these therapies."