Measurable Residual Disease Monitoring By Duplex Sequencing for TP53 in the Post Allogeneic Stem Cell Transplantation Study with Eprenetapopt (APR-246) + Azacitidine Strongly Predicts Outcomes

Introduction

Outcomes are poor for TP53 mutant myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) patients who undergo allogeneic hematopoietic stem cell transplantation (allo-HSCT; ~20% long term survival, 30% 1-year relapse free survival (RFS)). Notably, minimal data exist on the impact of TP53 MRD monitoring post allo-HSCT. We have previously reported our multicenter, open-label phase 2 trial results of the p53 reactivator eprenetapopt (APR-246) + azacitidine maintenance post allo-HSCT with encouraging results (median OS 20.6 months, 62% 1-year RFS; Mishra A et al., JCO 2022). We have now completed duplex TP53 MRD sequencing for all patients on study with at least 2 sequential samples.

Methods

Bone marrow (BM) aspirates were obtained prior to allo-HSCT, prior to the start of investigational therapy (day +30 to day +100) and then every 3 cycles while on combination (maximum 12 cycles including assessment upon completion of maintenance). To assess low allele frequency mutations in TP53, a custom-targeted sequencing panel was used with the TwinStrand Biosciences DuplexSeq technology. Library preparation with the enzymatic fragmentation option was performed starting with 4.5µg of DNA from BM mononuclear cells at all timepoints. Following library quantitation using the Kapa Library Quantification Kit, sequencing was performed in a paired-end 150-base configuration on the Illumina NovaSeq 6000 sequencer to generate 200M read-pairs per sample targeting 30,000-70,000X duplex coverage to detect variants at a frequency as low as .005%. Data analysis including alignment, variant calling, and annotation was performed using the TwinStrand Biosciences DuplexSeq Customer Portal hosted in DNANexus. GraphPad Prism was used for survival analyses and RFS/EFS/OS were calculated from date of allo-HSCT. EFS was defined as either relapse or death. For all analyses, TP53 MRD negativity cutoff was 0.01%.

Results

Fourteen patients had serial samples evaluated by duplex sequencing MRD for TP53 mutations post-allo-HSCT prior and during maintenance therapy. All study patients had significant TP53 positivity immediately prior to allo-HSCT (median variant allele frequency (VAF) 17%, range 5-75%). Pre-transplant VAF (> vs < than median VAF) had no effect on RFS, event free survival (EFS) or overall survival (OS). Of the cohort, 57% of patients (n=8) had positive MRD at the initial assessment post-HSCT prior to the start of maintenance therapy (median VAF 0.14%) although this was not predictive of RFS, EFS or OS. Of the entire cohort (n=14), achievement of TP53 MRD negativity at any timepoint post-HSCT was predictive of RFS (median 30 vs 8.9 months; P=.02) and EFS (median 30.6 vs 8.9 months, P=.005) although was not predictive of OS (33.3 vs 21.3 months; P=.33). Of all time points, MRD evaluation after end of maintenance (12 cycles) was the strongest predictor of outcomes. Specifically, MRD negativity after cycle 12 strongly predicted OS (33.9 vs 20.4 months; P=.005) and EFS (33.9 vs 10.1 months; P=.004) with a trend for RFS (32.6 vs 13.5 months; P=.06). Of the six patients who were MRD negative at the post-maintenance evaluation, 3 of 6 patients relapsed with 2 patients having late relapses (32 and 30 months respectively post-HSCT, one extramedullary only). The only early relapse in a post-maintenance MRD negative patient occurred in a patient with a detectable TP53 clone at .003% (relapsed 6 months after completion of 12 cycles of maintenance therapy). The only patient who did not relapse with +TP53 MRD at the post 12-cycle evaluation was transitioned to continuous maintenance hypomethylating agent therapy off-protocol although died of infection 4 months after completing investigational therapy. Individual clinical TP53 VAF trajectories of all patients sequenced on study will be shown at the meeting (e.g. emergence/regression of very rare TP53 variants).

Conclusions

Many patients with TP53 mutant MDS/AML are MRD positive both before and after allo-HSCT, supporting the need for novel pre- and/or post-HSCT strategies to decrease relapse. Following completion of maintenance therapy with eprenetapopt and azacitidine, TP53 MRD was strongly predictive of outcomes, supporting incorporation of this assay in future novel strategies.