Findings highlight ‘critical importance’ of infectious complications after CAR-T
Click Here to Manage Email Alerts
Infections are the primary cause of nonrelapse mortality among individuals who undergo chimeric antigen receptor T-cell therapy, according to study results.
Unique CAR-T-specific side effects — such as cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome (ICANS) — contribute to only a small percentage of nonrelapse deaths, findings showed.
The data highlight the “critical importance” of infectious complications after CAR-T and illustrate the need for comprehensive reporting of nonrelapse mortality, researchers concluded.
Miguel-Angel Perales, MD, chief of the adult bone marrow transplantation service at Memorial Sloan Kettering Cancer Center, and colleagues conducted a systematic review and meta-analysis to analyze nonrelapse mortality among patients with lymphoma or multiple myeloma who underwent CAR-T.
Investigators used MEDLINE, Embase and CINAHL to obtain data from 18 clinical trials and 28 real-world studies that included a combined 7,604 patients.
More than half (50.9%) of the 574 reported nonrelapse deaths were attributable to infections (50.9%). Other leading causes included second malignancies (7.8%) and cardiovascular/respiratory events (7.3%).
CAR-T-specific side effects — including CRS, ICANS and hemophagocytic lymphohistiocytosis — accounted for a combined 11.5% of nonrelapse deaths.
“We tend to think of CRS and ICANS as prototypical CAR-T complications,” Perales told Healio. “However, we are missing the big picture. This is the first time a study has taken a global approach to the deaths related to complications from CAR T cells rather than just focusing on those early CRS and ICANS complications.”
Healio spoke with Perales and study senior author Kai Rejeski, MD, visiting investigator and research fellow with the adult bone marrow transplantation service at Memorial Sloan Kettering, about the findings and how they might inform use of CAR-T.
Healio: Why did you conduct this study?
Rejeski: Nonrelapse mortality hadn’t been comprehensively described in the context of CAR T-cell therapy. We set out to assess the drivers and risk factors that underlie the development of nonrelapse mortality in a large cohort.
Healio: What did you find?
Rejeski: We were able to show that infections were, by far, the main driver of nonrelapse mortality. It is notable that the prototypical side effects of CAR T-cell therapy, like CRS and ICANS, cumulatively only drove about 10% of nonrelapse deaths.
Healio: Did you see any difference between approved CAR T-cell products?
Rejeski: Yes. When we looked at different disease entities, we saw nonrelapse mortality was a bit higher in some — such as mantle cell lymphoma and multiple myeloma — and a bit lower in others, such as large B-cell lymphoma and indolent lymphoma. Within large B-cell lymphoma and multiple myeloma, we had enough studies to take a deeper dive into study-level factors associated with nonrelapse mortality. We found the CAR T-cell product variable was the only independent risk factor for nonrelapse mortality in large B-cell lymphoma and multiple myeloma. In large B-cell lymphoma, patients who received axicabtagene ciloleucel (Yescarta, Kite Pharma/Gilead Sciences), or axi-cel, had increased nonrelapse mortality compared with those who received either tisagenlecleucel (Kymriah, Novartis) — often called tisa-cel — or lisocabtagene maraleucel (Breyanzi, Bristol Myers Squibb), often called liso-cel. In multiple myeloma, ciltacabtagene autoleucel (Carvykti; Janssen, Legend Biotech), or cilta-cel, was associated with a higher nonrelapse mortality rate than idecabtagene vicleucel (Abecma; Bristol Myers Squibb, 2seventy bio), or ide-cel. By performing these disease-specific models where we can account for potential confounders, we’re able to show the CAR-T product used in a given disease probably is driving much of the difference in nonrelapse mortality.
Perales: There is preferential use of different products by treating physicians. The ones that happen to be more toxic are also the ones we think may be more effective. In myeloma, there is a preferential use of cilta-cel over ide-cel because the PFS is much longer. For lymphoma, before the availability of liso-cel, axi-cel was the preferred treatment. This has shifted somewhat. Use of tisagenlecleucel [Kymriah, Novartis] has dropped in the U.S. because of a negative phase 3 trial. So, there are differences, but in this type of analysis you should be able to correct for at least some of these issues.
Healio: Will your findings inform treatment decision-making?
Perales: This will raise awareness about differences between products. It may inform some decisions, but I don’t think it will fundamentally change how we practice. Another consideration is where CAR T cells fit in the general landscape.
It's important to remember that most patients who receive CAR T cells have active disease at the time of treatment. With autologous transplant, most patients either are in complete remission or a very good state of partial remission. We’re not comparing apples to apples in terms of risk profile. Some physicians may look at this and think autologous transplant is safer for their patient, but I think that is the wrong conclusion to draw.
Rejeski: Multiple factors go into choosing one product over another, and nonrelapse mortality can be one of them. Other factors to consider include logistics and cost, the aforementioned differential efficacy between two products, and the patient’s age and functional status. Is this an older, frail patient or someone with a more indolent disease biology? If I have a young patient and we really want to maximize curative potential, we may accept some underlying toxicity. We’re making decisions on a patient-by-patient basis, so I don’t think this analysis fundamentally changes the way we approach selecting one product versus another.
However, this study does raise awareness about the factors that may contribute to nonrelapse mortality and suggests that by changing our management protocols for CRS and ICANS, we’ve made large improvements. Most recent studies report lower incidences and levels of CRS and ICANS, because we have learned to manage patients in a manner that reduces toxicity. Perhaps by considering more structured approaches to managing infections, we can achieve similar improvements in managing infectious complications.
Healio: What are your next steps in research?
Rejeski: We’re looking more closely at second primary malignancies. We were surprised this was the second most common cause of nonrelapse mortality. We’re taking a deeper look into that in a follow-up study.
Reference:
For more information:
Miguel-Angel Perales, MD, can be reached at peralesm@mskcc.org.
Kai Rejeski, MD, can be reached at rejeskik@mskcc.org.
Collapse
Healio Interviews
Click Here to Manage Email Alerts