Skip to main content
Premium Trial:

Request an Annual Quote

Whole-Genome Sequencing Elucidates Lymphoma Patients' Response to CD19 CAR T-Cell Therapy


NEW YORK – Researchers demonstrated the ability of whole-genome sequencing to predict large B-cell lymphoma patients' outcomes, including resistance, to CD19-directed CAR T-cell therapies at the American Society of Hematology's annual meeting.

In one study, researchers examined low-pass WGS data from large B-cell lymphoma patients prior to treatment with CD19 CAR T-cell therapy to develop a method for predicting prognosis and risk of treatment failure. In another study, researchers used WGS to identify the genomic drivers of resistance in large B-cell lymphoma patients who had relapsed on Gilead/Kite's CD19 CAR T-cell therapy axicabtagene ciloleucel (Yescarta; axi-cel).

These studies looked at relapsed or refractory large B-cell lymphoma patients who had received CD19 CAR T-cell therapies, specifically axi-cel and tisagenlecleucel (Novartis' Kymriah; tisa-cel). Currently, there are no genomic biomarkers that oncologists can readily use to identify patients who will benefit and those who will experience disease progression on these treatments. Although half of the patients with refractory large B-cell lymphoma relapse within six months of receiving a CD19 CAR T-cell therapy, researchers at the meeting said that the biological causes of this resistance have not been fully explored.

"Additional pretreatment assays are needed to identify high-risk patients before cell infusion to prioritize them for [enrollment in] future studies combining CAR T CD19 with other agents to improve the efficacy of treatment," Hua-Jay Cherng, a hematology and oncology fellow at MD Anderson Cancer Center, said during a presentation at ASH.

Prognostic risk score

Cherng and his colleagues aimed in their study to develop a method for risk stratifying large B-cell lymphoma patients prior to CAR T-cell treatment. The researchers performed low-pass WGS on cell-free DNA from 135 patients before cell infusion; 122 of these patients went on to receive cell therapy treatment. Most patients in this study received axi-cel, and about 10 patients received tisa-cel.

They measured progression-free survival and the rate of durable response at three months or more after treatment. They also analyzed the WGS data using a web-based tool, CNApp, and assigned a value to each altered segment of the genome based on the amplitude of copy number deviation from neutral. The tool generated three scores for each patient — focal, broad, or global — based on their burden of copy number alterations, or CNAs. The researchers then explored each score's association with survival.

The focal CNA score, or FCS, had the strongest association with survival. A score higher than the median score of 52 in the study cohort was associated with inferior progression-free survival and overall survival rates on CAR T-cell therapy. Among patients with an FCS below 52, the median progression-free survival was 11.9 months, compared to 3.9 months for patients with an FCS above 52. A similar trend was seen in overall survival; patients with a low FCS had a median overall survival of 25 months, while high FCS patients had a median overall survival of 10 months.

The researchers then explored other factors associated with inferior survival and found in a multivariate analysis that in addition to an elevated FCS, one extra nodal disease and elevated lactate dehydrogenase in blood were also independent markers of poor outcomes. They integrated these three variables into a risk score, in which patients received one point for each negative prognostic marker.

Cherng reported at the meeting on Saturday that patients who had all three prognostic markers "had very poor outcomes" with one-year progression-free and overall survival rates of 12 percent and 19 percent, respectively. Therefore, using this risk score, Cherng said his group was able to identify a subset of patients with significant clinical unmet need.

Characterizing resistance

In a separate study presented at the meeting, researchers led by Francesco Maura, an assistant professor and co-principal investigator at the Myeloma Genomic Lab at the University of Miami's Sylvester Comprehensive Cancer Center, aimed to identify genomic drivers of resistance in large B-cell lymphoma patients on axi-cel — an important analysis given more than half of the refractory patients receiving CD19 CAR T-cell therapy relapse within the first six months post-treatment.

Previous research has suggested several reasons for the high rate of relapse to CAR T-cell therapy among refractory patients, including tumor metabolic volume, high disease burden, immune exhaustion within the tumor microenvironment, and CD19 loss through acquired mutations, Maura explained.

In the study, Maura's group examined WGS data from 27 diffuse large B-cell lymphoma patients treated with axi-cel, of which nine had complete responses, one patient partially responded, and 16 progressed. Researchers decided to evaluate these patients' samples with WGS instead of more commonly used whole-exome sequencing, hoping to capture more genomic drivers of resistance.

They identified several drivers of resistance to CD19 CAR T-cell therapy in the relapsed or refractory cohort, including markers indicative of genomic complexity like chromothripsis (clustered chromosomal rearrangements) and APOBEC, and RHOA and RB1 deletions.

The researchers also compared sequencing data from their study on relapsed patients against data from 50 newly diagnosed DLBCL patients from the International Cancer Genome Consortium's Pan-Cancer Analysis of Whole Genomes. They found TP53, RHOA, and RB1 gene deletions were significantly enriched in their relapsed or refractory cohort. RHOA and RB1 deletions both independently predicted poor outcomes, with all patients with an RHOA deletion and 75 percent of patient with an RB1 deletion experiencing disease progression after axi-cel. However, TP53 deletions were not associated with progression.

When the four identified resistance drivers — chromothripsis, APOBEC, and RHOA and RB1 deletions — were combined into one signature, that signature identified all patients in the relapsed or refractory cohort who progressed on CAR T-cell therapy within six months post treatment, Maura reported at the meeting. He added that some of these alterations were also associated with immunotherapy resistance in other settings. For instance, the APOBEC signature is associated with immune resistance in lung cancer patients, he said.

"What we should expect for most of the features we see are features that reflect genomic complexity. The more complex your genome, the less efficient the anti-lymphoma [cells] are," Maura said. "A more aggressive tumor is also more immunosuppressive and has more immune exhaustion in the tumor microenvironment, which has been shown in solid cancers but not yet in hematological cancers."

WGS in practice

Both researchers noted that their findings require more study before being implemented in practice, but suggested that more lymphoma patients should undergo WGS.

Maura acknowledged that the sample size in this study could have limited the findings and said he hopes to validate these findings in a large cohort. He also urged colleagues at the meeting to perform WGS on more lymphoma patients to gain a better understanding of their disease, particularly because some of the resistance mechanisms found in his study could have only been identified using WGS.

"In a disease like DLBCL or multiple myeloma, we don't believe there is one single mechanism of resistance, but there are multiple mechanisms that co-occur together," Maura continued. "To really understand CAR T and immunotherapy, we need to combine both the immune system and immune compound and the characterization of the genomics in the cancer cell to really have a complete picture [of resistance]."

Cherng also suggested that low-pass WGS could be "widely implemented" in cancer centers to guide treatment for lymphoma patients. Turnaround time for results using the assay and scoring system in his study was around a week, and Cherng suggested that it could be used while patients' cells are being manufactured for CAR T-cell treatment.

"With a fast-enough turnaround time, this assay could guide which patients should be potentially enrolled in clinical studies with novel agents in combination with CD19 CAR T to improve outcomes before planned cell infusion," he said.