Clinical Implications of TP53 Mutations/Allelic State in Patients (Pts) with Myelodysplastic Syndromes/Neoplasms (MDS) Treated with Hypomethylating Agents (HMA)- a Multicenter, Retrospective Analysis from the Validate Database

TK and JPB Co-first; RK, DS, and AMZ are Co-senior authors.

Introduction

MDS with TP53 mutations (TP53^MT) are associated with high-risk disease features and rapid progression to AML. Effective treatment for this aggressive subtype is still not available. As compared to monoallelic TP53^MT, biallelic MT have been associated with adverse outcomes and worse overall survival (OS), but the impact of allelic status on treatment response and OS after HMA therapy are not well established. In this study, we analyzed TP53 allelic state for a large well annotated cohort of pts with MDS who were treated with HMAs.

Methods

The VALIDATE database includes pts with MDS treated with frontline HMA from 14 specialized centers. Pts were considered to have biallelic TP53^MT according to the ICC definition (Arber et al. 2022, Blood) if they meet any of the following: a. two TP53^MT (variant allelic frequency [VAF]>10% for each) b. single TP53^MT with copy neutral loss of heterozygosity (CN-LOH, available for 308 pts) c.TP53^MT with del17p and d. Single TP53^MT with VAF >50%. Time to event analyses were estimated using Kaplan-Meier estimator and groups were compared by the log-rank test from the time of HMA initiation. Only higher risk (HR)-MDS pts (IPSS>1 or IPSS-R>3.5) with BM biopsy within 180 days after HMA initiation were assessed for complete response (CR) and overall response rate (ORR) according to the IWG 2023 criteria (Zeidan et al, Blood 2023) (n=480). OS was calculated from time of HMA therapy initiation. This study was supported by an independent research grant from AbbVie.

Results

A total of 816 pts met eligibility and were included in this analysis. Median age was 68 years (IQR: 61-74) and 64% were men. MDS excess blast (EB) 1/2 was the most common MDS subtype (62%). Overall, 78% of the pts were treated with HMA monotherapy (26% decitabine [DEC] and 74% azacitidine [AZA]) and 22% received HMA as part of combination therapy. In total, 40% of the pts underwent allogenic HSCT. Median HMA treatment time was 5 months (mo), IQR:3-11. Median follow up time was 17 mo (IQR:9-31). TP53^MT were detected in 253 pts (31%) at time of MDS diagnosis, of whom 153 (60%) pts had biallelic TP53^MT and 100 (40%) pts had monoallelic TP53^MT. Missense MTs were identified in 80% of pts with monoallelic TP53^MT. TP53^MT/del17p (33%), TP53^MT/CN-LOH (28%), and multiple TP53^MT (25%) were the most common combinations in pts with biallelic TP53^MT. Pts with TP53^MT had higher rates of MDS-EB (74% vs. 61%, p<0.001). Complex karyotype was observed in 112 (73%) pts with biallelic TP53^MT and 81 (81%) pts with monoallelic TP53^MT (p=0.154). Median number of HMA cycles among pts with TP53^MT was 8 (IQR:5-12) and higher compared to pts without TP53^MT (5 cycles, IQR:3-9), p=0.032 (Panel-A). In a logistic regression model for CR, TP53^MT was not associated with probability of achieving CR (1.4, 0.89-2.29). CR rates for pts with biallelic TP53^MT (18%) and monoallelic TP53^MT (14%) were not significantly different, p=0.37. There were also no significant differences in ORR observed between biallelic TP53^MT (59%) and monoallelic TP53^MT (54%) pts, p=0.39. Among TP53^MT pts treated with HMA monotherapy, pts treated with AZA compared to DEC had similar odds for ORR (OR:1.1, 95%CI: 0.7-1.6)/CR (OR: 0.8, 95%CI: 0.4-1.3) and hazard for OS (HR: 1.1, 0.9-1.4). As expected, the median OS (mo, 95%CI) for pts with TP53^MT (12, 10-13) was significantly lower than pts without TP53^MT (28, 25-31), p<0.001. Importantly, pts with biallelic TP53^MT (11, 9-14) had no significant difference in median OS compared to monoallelic TP53^MT (13, 11-17), p=0.268 (Panel-B). In a multivariable cox proportional hazard regression model for OS (hazards ratio, 95%CI) adjusted for age (1.0, 0.9-1.0), HSCT (0.4, 0.3-0.4), IPSS-M score (1.2, 1.1-1.3), and complex karyotype (1.4, 1.1-1.8), TP53^MT status (1.8, 1.4-2.3) was still associated with OS. However, biallelic TP53^MT had comparable risk to monoallelic TP53^MT (1.2, 0.9-1.7).

Conclusions

To our knowledge, VALIDATE is the largest database of MDS pts with TP53^MT who were treated with HMA. We confirm that TP53^MT have significant negative impact on survival, but not on response to HMA therapy. Importantly, we did not observe OS differences between monoallelic and biallelic TP53^MT. We also show that the type of HMA monotherapy used does not impact outcome of these pts. Overall, pts with TP53^MT have very poor outcomes and are in desperate need of new, effective therapies.