(PA-047) Melphalan-Induced Enrichment of TP53-Mutant CHIP as a Risk Factor for Subsequent CAR-T Related Myeloid Neoplasms in Multiple Myeloma

CAR-T cell therapy has shown remarkable efficacy in multiple myeloma (MM), but concerns are emerging about an increased risk of secondary myeloid neoplasms (t-MN), potentially linked to the intrinsically high prevalence of clonal hematopoiesis of indeterminate potential (CHIP, ~22%) in MM patients (pts). Most importantly, the interplay between CAR-T induced effects on pre-existing CHIP, including a hyperinflammatory hematopoietic niche, and the mutagenic impact of prior therapy (e.g. melphalan, lenalidomide) remains poorly understood.


Methods:

To better characterize melphalan as a potential driver of CAR-T associated leukemia in MM, we here employed CRISPR-Cas/ sleeping beauty technologies to study melphalan-induced effects on TP53-mutant CHIP clones in THP-1 cell line models. The impact of CAR-T related inflammation was studied by ex vivo pt-derived models using the CoSeedis® platform.


Results:

This study leverages on a prior report from our group which described the accelerated onset of post-CAR-T t-MN in a cohort of n=10 heavily pretreated MM pts (all lenalidomide-exposed, median of 2 (1-4) prior ASCTs). WGS on BMMCs from these pts was positive for the melphalan-associated SBS-MM1 signature, thereby providing clinical evidence that enrichment of CHIP may in part be related to prior ASCT. Consequently, we conducted competition assays on TP53-mutant THP-1 models with and without melphalan using different wildtype (WT):knock-out (KO) ratios. In monoculture, TP53-KO clones showed relative resistance to melphalan compared to WT with an IC50 shift of 1.4-2.1 µM. Interestingly, KO clones had a baseline survival disadvantage in drug-free co-culture, which was reversed upon melphalan exposure, with KO clones rapidly outcompeting WT, and reaching a clonal dominance of >90% after 14 days of treatment; these effects being consistent for different seeding ratios. Upon melphalan withdrawal, recovery of WT cells was observed, while a re-introduction of melphalan again selected for KO clones. This data points towards selection as a key driver for how melphalan contributes to t-MN pathogenesis. To investigate CAR-T as a pathogenic second hit potentially leading to the accelerated expansion of CHIP, we next mimicked CAR-T hyperinflammation using ex vivo models and BMMCs from known CHIP carriers. Preliminary data on these primary samples indicate a ~6 fold increase in clonal expansion as determined by VAF quantification after 14 days for HS-5 co- vs. monocultures. Further investigations on the additional impact of CRS-associated cytokines are ongoing and will be presented at the meeting.


Conclusions:

While the link between t-MN and CAR-T therapy remains unclear, our data suggest melphalan-driven selection of TP53-mutant CHIP as an early leukemogenic event predisposing to subsequent CAR-T-related cytopenia and t-MN. These findings support earlier use of CAR-T to mitigate secondary leukemia risk, particularly in pts with known high-risk TP53CHIP variants.