SAN DIEGO – A method for single-cell whole-genome sequencing of circulating tumor cells (CTCs), in the future, could help doctors pick the right treatment for relapsed and refractory multiple myeloma patients following BCMA- or GPRC5D-targeted immunotherapies.
At the American Society of Hematology's annual meeting on Sunday, Rosa Toenges, a research fellow at Dana-Farber Cancer Institute, presented results from a study, in which researchers conducted whole-genome sequencing of CTCs collected from multiple myeloma patients before and during treatment with a BCMA- or GPRC5D-targeted therapy. Her team developed a novel method that involves enriching CTCs from peripheral blood samples, extracting and amplifying DNA, and sequencing that DNA using Illumina's NovaSeq X instrument with 10B flow cells. The researchers' goal was to develop a method that could inform clinical treatment decisions without having to perform repeat bone marrow biopsies.
In the relapsed and refractory multiple myeloma setting, the aim is to provide treatment that offers patients symptom relief, prevents new organ damage, and achieves remission again. With dozens of available treatments to choose from, including proteasome inhibitors, immunomodulatory agents, monoclonal antibodies, bispecific antibodies, and CAR T-cell therapies, doctors pick treatments for refractory patients based on factors such as their disease stage, treatment history, and drug tolerability. Biomarkers that elucidate a resistance mechanism or indicate the likelihood of response to therapy could help doctors narrow down the options for heavily pretreated patients.
Toenges and her colleagues hypothesized that the use of whole-genome sequencing in this setting could inform multiple myeloma patients' next choice of treatment after they relapse on BCMA- or GPRC5D-directed therapy.
In her presentation, Toenges said mutations in genes encoding the target cell-surface proteins BCMA and GPRC5D have been shown to drive selection and expansion of relapsed tumor cells. "Characterizing the mutational status of these targets could play an important role not only in potentially guiding therapy decisions in the future but also in enhancing our understanding of the biology," she said. "However, [assessing] mutational status is not currently a routine practice."
One reason for that, according to Toenges, is that genomic profiling in multiple myeloma patients is routinely conducted by fluorescence in situ hybridization (FISH), not next-generation sequencing, and a second bone marrow sample is rarely collected from patients with relapsed or refractory disease. However, CTCs offer a potential alternative to invasive and painful repeat bone marrow biopsies. "Our group, along with others, has shown that the vast majority of [RRMM] patients, even those at a low tumor burden, have CTCs detectable from just a single tube of blood," Toenges said.
In a previous study, she and her colleagues demonstrated that standard whole-genome sequencing not only has the potential to replace FISH but also gives a complete mutational profile when performed on CTCs, compared to bone marrow. This led them to wonder whether sequencing CTCs could provide a comprehensive mutational profile of multiple myeloma and whether there was a feasible method to do so.
In the first part of the new study, the investigators sequenced the whole genomes of CTCs from 16 patients who had received a median six lines of therapy before receiving a BCMA- or GPRC5D-targeted therapy. Eight of those patients had received a BCMA-directed CAR T-cell therapy. Five had received a BCMA-targeted T-cell engager, and another three had received a GPRC5D-targeted T-cell engager.
The researchers sequenced a median 50 cells per 10 mL sample of blood collected from each patient. In 13 out of the 16 patients, WGS identified point mutations in known multiple myeloma driver genes that researchers expected to find. None of the patients, not even those who had received prior anti-BCMA therapy, had a mutation in TNFRSF17, the gene that encodes BCMA. One patient who was refractory to an anti-GPRC5D therapy had a mutation in a single copy of the GPRC5D gene prior to treatment. A majority of the patients had a complete response or a stringent complete response, which is deeper than a complete response.
Toenges said that findings from this part of the study suggest that profiling mutations before treatment using whole-genome sequencing of CTCs is feasible and can detect alterations in target genes.
The researchers then studied whether resistant clones were present prior to relapse on BCMA- and GPRC5D-targeted therapies in three patients, who eventually relapsed on those therapies. In one patient that initially had stable disease on an anti-BCMA T-cell engager, they found a clone with a TP53 nonsense mutation that was present before relapse. That patient's cancer cell fraction also increased from 16 percent before therapy to 45 percent just prior to progression, despite an overall 65 percent decrease in CTCs.
A second patient on an anti-BCMA CAR T-cell therapy at first had a stringent complete response before relapsing. Toenges and her team retrieved and sequenced seven individual CTCs, and of those, five came from the tumor clone, but none had a mutation in the BCMA-encoding TNFRSF17 gene. The third patient was treated with an anti-BCMA T-cell engager and also did not have any mutations in TNFRSF17, suggesting a different mechanism of treatment escape leading to disease progression.
"Clonal selection correlates with mutations in known drivers other than BCMA and GPRC5D, which are rare in this patient population," Toenges concluded. "Sequencing tumor cells during a complete response revealed that late, persistent cancer cells may drive relapse."
Toenges concluded that the novel method she and her colleagues developed to profile relapsed and refractory multiple myeloma using fewer than 50 cells from peripheral blood was feasible and that her team identified driver mutations at baseline and during therapy that could affect treatment in the future. "Clonal selection correlates with mutations in known drivers other than BCMA and GPRC5D, which are rare in this patient population," Toenges said, adding, "sequencing tumor cells during a complete response revealed that late persistent cancer cells may drive relapse."
Toenges and her team believe that various driver mutations identified at baseline and during therapy could impact treatment decisions in the future, and that genomic characterization of tumor cells at the single-cell level via MRD can be used to predict response to immunotherapy, monitor for emergence of high-risk subclones, and inform treatment decisions for patients with disease relapse without the need for a bone marrow biopsy.