NEW YORK – A group of researchers in Germany have identified a potential mechanism by which multiple myeloma patients treated with CAR T-cell therapy might develop treatment resistance and experience disease relapse.
The finding, described in a paper published Monday in Nature Medicine, suggests the need for immunotherapy approaches to target multiple antigens, so cancer cells cannot evade immune system attacks through the loss of a single antigen, in this case BCMA.
In the paper, researchers, including Leo Rasche of the University Hospital Wurzburg, described an in-depth analysis of one patient who had been treated on the Phase II KarMMa clinical trial of Bristol Myers Squibb and Bluebird Bio's idecabtagene vicleucel (ide-cel). The investigational autologous CAR T-cell therapy is engineered to target the B-cell maturation antigen, BCMA, and the US Food and Drug Administration is currently reviewing a biologics license application for ide-cel as a treatment for patients with multiple myeloma who have received at least three prior therapies.
The researchers' analysis homed in on how this one patient's multiple myeloma cells upon relapse demonstrated loss of expression of TNFRSF17, the gene that encodes for BCMA. The patient initially responded well to ide-cel and had no detectable minimal residual disease three months post-treatment. However, after about five months, the patient relapsed, experienced rapid disease progression, and ultimately, died.
At the time his cancer came back, the patient's multiple myeloma cells expressed barely any soluble BCMA, which is significant, the authors noted in the paper, since antigen levels are usually a clear sign of multiple myeloma tumor load. Although the CAR T cells the patient received were still persistent at relapse and had expanded significantly, there was no BCMA to target, which suggested that the cancer cells evaded ide-cel via loss of BCMA.
To home in on the mechanism of resistance, the researchers compared bone marrow samples collected at baseline and at the time of progression. Using single-cell RNA sequencing, they assessed TNFRSF17 expression, noting that it was expressed strongly at baseline, but then was essentially undetectable at progression. The researchers then used immunohistochemistry to confirm the gene expression results at the protein level. Despite the total loss of BCMA, they noted, most other plasma cell markers that have been identified as potential immunotherapy targets, including CD38 and SLAMF, were still expressed at the time of disease progression. This, they noted, suggested that the CAR T-cell therapy specifically selected for a BCMA-negative clone.
"BCMA is a receptor for APRIL, which is involved in NF kappaB signaling," Rasch said, explaining that this usually results in sustained and improved plasma cell survival, making the BCMA loss variants an unexpected find. "Probably these cells are very rare … but CAR T-cell therapy could select for them," he said.
Following these findings, Rasch and colleagues went on to perform whole-genome sequencing on the patient's multiple myeloma cells at the time of disease progression, with the goal of identifying what was behind the TNFRSF17 expression loss. They identified a number of genomic changes, including a large heterozygous deletion of 21.3 megabases affecting chromosome 16 and homozygous deletion of 91 kilobases, altogether resulting in biallelic loss of TNFRSF17 at 16p13.13.
"These data indicate a genomic mechanism for the observed BCMA-negative progression after CAR T-cell therapy," the authors wrote, going on to explain that, to have a full homozygous deletion of TNFRSF17, "two independent hits" are usually necessary. In other words, the cancer cells potentially needed to have existing heterozygous deletion of TNFSF17 prior to CAR T-cell therapy.
However, whole-genome sequencing and single-cell RNA sequencing both revealed that the patient's cancer cells didn't have any heterozygous TNFSF17 deletions before treatment. That said, they did observe some signs of preexisting genomic instability that pointed to possible branch evolution, and a heterozygous deletion on chromosome 16p could have been present in a subclone at baseline. To further assess the question of heterozygous deletions on chromosome 16p before treatment, the researchers went on to extend their study into 50 other patients with relapsed or refractory multiple myeloma who had not received CAR T-cell therapy or any other anti-BCMA treatment.
Among these patients, they found three who had heterozygous deletions encompassing the TNFRSF17 locus, and identified these patients as being at risk for possible homozygous loss of TNFRSF17 following treatment with anti-BCMA CAR T-cell therapy. This occurrence — that is, multiple myeloma patients with heterozygous TNFRSF17 deletions or monosomy 16 at baseline — was seen in 6.7 percent of 135 patients in a larger dataset. "We found heterozygous BCMA deletions in around seven percent of BCMA naïve patients, which in theory could present a risk factor for BCMA loss after respective immunotherapies," Rasche said, suggesting that it might be possible to identify these patients who are at risk for relapse prior to initiating CAR T-cell therapy. He was careful to note, however, that among these patients, there might also be not-yet-discovered mechanisms leading to relapse or resistance.
As Rasche's group and others pursue further validation of the identified resistance mechanism, the FDA is reviewing the ide-cel BLA, which is based on data from the Phase II KarMMa study. Among the 128 patients who received ide-cel in that study, the overall response rate was 73 percent and the median progression-free survival was 8.6 months at a median follow up of 11.3 months. CAR T cells were detected in 59 percent of patients at six months and in 36 percent of patients at 12 months.
In a previous Phase I study, ide-cel led to an 85 percent response, but the researchers noted that not all responses were durable. As seen with other CAR T-cell therapies, relapses were all too common. Investigations like Rasche's are important for understanding the reason why patients are relapsing and for developing treatment strategies to avoid resistance.
Rasche and his colleagues wrote that should this method of gene expression loss be validated as a reason why patients relapse following CAR T-cell therapy, it might be possible to mitigate the phenomenon by simultaneously targeting other antigens on multiple myeloma cells.
"There is a rich pipeline of multiantigen-targeting approaches including bispecific CAR T products, trispecific antibodies, and combination therapies with monotargeted immunotherapies," they wrote. In directing therapy at several targets at once, clonal selection based on the loss of a single antigen — in this case BCMA — might be avoidable.
Another potential implication of this research, according to Rasche, is that patients who have received BCMA-targeted treatment should probably be tested for BCMA expression before initiating another immunotherapy designed to target the same antigen.
Although Rasche's team identified this potential mechanism of resistance based on in-depth molecular analysis of a single patient's cancer, he said that groups in Boston and Canada have independently described the same BCMA loss phenomenon in other patients. Going forward, he said it will also be important to analyze large immunotherapy clinical trials currently underway in multiple myeloma to discern the frequency of the mechanism.
And because ide-cel is not the only BCMA-targeted therapy for multiple myeloma, Rasche pointed out that it will also be important to see whether the same resistance mechanism is at play when patients relapse after treatment with T-cell-engaging bispecific antibodies or with antibody drug conjugates. One such BCMA-targeted antibody drug conjugate, belantamab mafodotin-blmf (GlaxoSmithKline's Blenrep) garnered accelerated approval from the FDA last summer.