Prevalence and Mutational Make-up of Secondary Hematologic Neoplasms in Multiple Myeloma Patients after CAR-T Cell Therapy

Background:

The approval of chimeric antigen receptor T-cells (CAR-T) represents an important milestone for the treatment of hematologic malignancies, including multiple myeloma (MM). Despite the unprecedented efficacy of CAR-T cell therapy, recent reports have however raised concern about the increased risk of treatment-related myeloid (t-MN, MN-pCT) and lymphoid neoplasms (t-LN) after CAR-T. In this study, we therefore aimed to investigate the prevalence and mutational make-up of post-CART hematologic neoplasms in a multi-center network analysis across Germany.

Methods:

The incidence of t-MN and t-LN was investigated in n=179 CAR-T exposed MM patients treated at three different academic institutions. Molecular characterization of respective specimen was performed by whole-genome and targeted sequencing.

Results:

At a median follow-up of 17.7 months after CAR-T cell treatment we have identified 8/179 (4.5%) cases of clonal hematopoiesis (CH), including 5 patients with secondary AML/ MDS and 3 patients with clonal cytopenia of undetermined significance (CCUS). No cases of treatment-related lymphoid neoplasms (t-LN) have been noted thus far. Onset of t-AML/ MDS after CAR-T was accelerated at a mean latency period of 2.9 months (median 1.3 months, range 0.0-7.8) after CAR-T infusion. This relates to substantially longer time intervals with a median of 56.9 months (range 14.0-239.1) between the diagnosis of MM and secondary t-MN as seen in CAR-T naïve controls (n=26) at our centers. To note, the interval between high-dose melphalan and the onset of t-MN did also not show an instant temporal relationship with a protracted median latency time to t-MN of 53.9 months (range 6.0-148.7) in transplant-exposed, CAR-T naïve controls (n=23). While the pathogenesis of t-MN in MM is well established, and in part driven by prior drug exposure, this observation suggests that CAR-T mediated inflammation within the bone marrow niche may act as an independent stimulus which leads to accelerated outgrowth of t-MN after CAR-T cell therapy. Molecular characterization revealed pathogenic TP53 alterations in 3 patients with MDS/ AML (pt1: TP53^R273H VAF 40%; pt2: TP53^c.102dup VAF 50%; pt3: TP53^p.V272L VAF 45%). The other 2 AML/ MDS cases were characterized by MECOM rearrangement and a pathogenic DNMT3A/ subclonal TET2 alteration.

One CCUS case was likewise driven by a DNMT3A alteration whereas a second case was found to carry a TET2 mutation. Longitudinal studies in a third CCUS case provided more insight into the clonality of a pathogenic PPM1D mutation over time. Interestingly, VAF for PPM1D in this patient flared up to 9% at day 146 after CAR-T and has meanwhile (day 235) decreased to 1% along with stabilizing hematopoiesis suggesting a scenario in which regeneration after CAR-T therapy, with decreasing bone marrow inflammation, removes the competitive advantage of CH and ultimately leads to its decline. At the time of analysis, 3/5 AML/ MDS patients have succumbed to their disease. All 3 CCUS patients remain under active follow-up.

Conclusions:

While the association between t-LN/ t-MN and CAR-T-cell therapy remains elusive to date, our data in MM patients points towards an increased incidence and accelerated onset of t-MN after CAR-T cell exposure. In contrast, absence of t-LN in our analysis seems in line with reports on the low incidence of T-cell leukemia after CAR-T. Enrichment of t-MN may in part be due to the widely established use of alkylating and immunomodulatory agents as well as the increased prevalence of CH in MM patients.

Improved understanding on whether CH confers increased long-term risk for t-MN after CAR-T therapies may prospectively influence our management of patients with MM and other diseases sensitive to CAR-T.