Key Risk Factors Identified
Key factors found after multivariate adjustment to be independently predictive of early relapse or progression included extramedullary disease or plasma cell leukemia, being refractory to lenalidomide, having high-risk cytogenetics, and having increased age- and sex-adjusted ferritin at the time of lymphodepletion.
With each of the risk factors valued at 1 point, the MyCARe model ranked scores of 0-1 points as low-risk, 2-3 as intermediate risk, and a score above 4 was considered high-risk.
Based on the model, the risk of early relapse within 5 months among those scored as low risk was 7%, for intermediate risk, 27% (hazard ratio [HR], 3.27 vs low-risk; P < .001), and for high risk, 53% (HR, 7.89 vs low-risk; P < .001), with outcomes overall comparable between the two geographic groups. Importantly, the model maintained utility for patients who did and did not receive salvage therapies; however, “more studies are needed to identify the optimal post–CAR-T approach,” the authors write.
Dr. Gagelmann added that older age was significantly associated with improved progression-free survival in the US cohort, with a 12-month progression-free survival of 27% among patients under 65 versus 43% for those over 65 (P = .03). However, age was not found to be associated with similar outcomes in the European cohort.
The authors note that the MyCARe model outperformed the CAR-HEMATOTOX and more recent disease-specific R2-ISS risk-stratification tools regarding prediction of relapse/progression and progression-free survival.
However, with CAR-HEMATOTOX developed to predict side effects and non-relapse mortality, “our results demonstrate that both scores independently predict different outcomes after anti–BCMA CAR-T in RRMM,” the authors report. Therefore, “they can be used complimentarily to predict complications (CAR-HEMATOTOX) and relapse/progression-free survival (MyCARe model).”
Importantly, the authors add that the tool may help in patient selection for earlier treatment.
“As ide-cel and cilta-cel have shown astonishing efficacy for earlier treatment lines, our model might also be validated for such patients,” the authors note in the study. They conclude that the study provides “the first Euro-American cartography of the efficacy and safety profile of current CAR-T, showing comparable results.”
“We also built the MyCARe model, which can predict early relapse, response, and survival and may facilitate patient selection in this very challenging setting,” the authors report.
Hope for Interventions Based on Patients’ Risk
Commenting on the study, Rahul Banerjee, MD, an assistant professor with the Division of Hematology and Oncology, University of Washington, Seattle, underscored that “we need more cross-border research like this in the myeloma field.”
“Clinically, my hope that this will help us tailor post–CAR-T interventions according to each patient’s risk profile,” he said.
Risk factors such as the presence of extramedullary disease, plasma cell leukemia, or high-risk cytogenetics are expected; however, Dr. Banerjee said the inclusion of increased ferritin before CAR-T was “an interesting new risk factor that we’ve also heard about from our colleagues in the lymphoma space.”
Ferritin perturbations can indicate many things, but high ferritin can be a sign of elevated inflammation at baseline,” he explained. “These patients may have a hyperinflammatory phenotype of their myeloma which can predispose T-cells to exhaustion,” Dr. Banerjee said.
“Exhausted T-cells at collection mean exhausted CAR T-cells at infusion, and so the negative prognostic significance of elevated ferritin — which we don’t always check before CAR-T — makes sense.”
While the authors suggest a potential benefit of the MyCAR3 model in identifying patients who could benefit from other novel therapies at relapse, Dr. Banerjee suggests another possibility. “I’d take this a step further and suggest future studies of this MyCARe model to identify patients who might benefit from post–CAR-T maintenance,” he said.
“The ‘one-and-done’ nature of CAR-T in terms of not requiring further myeloma therapy after infusion is a powerful benefit for patients, but there are some patients who may benefit from low-dose pomalidomide or iberdomide/mezigdomide maintenance to help keep the myeloma at bay and to promote T-cell fitness,” Dr. Banerjee explained. “This risk model may identify patients to prioritize for such types of clinical trials in the future.”
Caveats include that factors beyond the baseline features (used for the risk model) can further influence outcomes,” Dr. Banerjee noted.
“Risk stratification is inherently a dynamic process over time,” he said, questioning, for instance, “what about patients who achieve measurable residual disease negativity [MRD] at day +28 after CAR-T cell? Does the achievement of MRD negativity ‘erase’ a high-risk MyCARe score? We’ll need future studies to tell.”
An overriding take-home message for clinicians should be to simply refer eligible patients to a CAR-T capable center as soon as possible for evaluation.
“In the lymphoma world, they have a nice adage for this: ‘If they recur, you should refer,’ ” he said. “I’d suggest the same here. By no means will we move to CAR-T therapy for every patient at first relapse. However, based on their MyCARe score and other risk factors, there may be patients we prioritize for CAR-T first versus CAR-T with maintenance versus clinical trials.”
Dr. Gagelmann reported relationships with BMS, Pfizer, Stemline, MorphoSys, and Kite. Dr. Banerjee disclosed ties with BMS, Caribou Biosciences, Genentech, Janssen, Karyopharm, Pfizer, Sanofi, SparkCures, Novartis, and Pack Health.