TP53 Mutation Status Defines a Distinct Clinicopathological Entity of Therapy-Related Myeloid Neoplasm, Characterized By Genomic Instability and Extremely Poor Outcome

BACKGROUND: The prognostic and therapeutic impact of TP53-mutated (TP53^mut) myeloid neoplasm (MN) is an area of intense interest, with the 5^th edition of the World Health Organization, the International Consensus Classification, and the European LeukemiaNet guidelines, all recommending major changes MN classification based on TP53^mut. However, studies driving these changes predominantly included de novo MDS (Bernard et al. Nat Med 2020), AML and high risk MDS treated with intensive chemotherapy (Grob et al. Blood 2022), or MN with complex karyotype (CK, Weinberg et al. Blood Adv 2022). Therefore, the significance of these findings as they relate to t-MN is not known.

METHODS: We analyzed therapy-related myeloid neoplasm (t-MN) (n=377) patients evaluated at Mayo Clinic, Rochester and South Australian MDS Registry with clinical, integrated genomic profiling, and long-term follow-up to characterize the genomic profile and outcome of TP53^mut t-MN.

RESULTS: Our cohort of 377 t-MN included t-MDS (n=245, 65%) and t-AML (n=132, 35%), with a median age at t-MN diagnosis being 65.7 years. 184 TP53^mut at variant allele frequency (VAF) ≥2% were detected in 131 (34.74%) patients. Median overall survival (OS) of TP53^mut VAF <10% (n=14) was similar to wild-type TP53 (TP53^wt), hence further analysis was restricted to TP53^mut with VAF >10% (n=111). TP53^mut patients without VAF (n=6) were not included for further analysis.

Genomic instability was highly evident in TP53^mut t-MN compared to TP53^wt t-MN. CK, monosomal karyotype (MK), chromosome 5q deletion, and marker chromosomes were enriched in TP53^mut t-MN. In contrast, driver oncogenic gene mutations such as IDH2, SRSF2, RAS, and TET2 were less frequent in TP53^mut. TP53^mut also dictated the clinical presentation, longer latency (interval between primary cancer and t-MN diagnosis), and inferior OS compared to TP53^wt (8.2 vs. 19.4 months; P <0.001). On multivariate Cox regression analyses, TP53^mut was associated with inferior OS (HR 1.43 95% CI 1.01-2; P =0.03) independent of CK and bone marrow blast %.

A significantly higher proportion of CK patients had TP53^mut compared to non-CK patients (74.4% vs. 5.4%, P <0.001). Even within the CK cohort, TP53^mut was associated with structural genomic instability, with a higher proportion of TP53^mut harbouring MK (P =0.013), marker chromosome (P =0.033), deletion 5q (P <0.001), abnormal chromosome 17 (P =0.001), and other structural chromosomal aberrancies compared to TP53^wt-CK. In contrast to the predominance of structural chromosomal aberrancies in the TP53^mut-CK cohort, mutations driving MN were sparse in TP53^mut-CK compared to TP53^wt-CK. Finally, TP53^mut-CK was associated with significantly poor survival compared to the TP53^wt-CK (8.2 vs. 17 months; P <0.001).

Though bone marrow (BM) % predicted poor outcome in the whole t-MN cohort, it failed to predict outcome in TP53^mut t-MN patients. We compared the clinical and molecular profiling of TP53^mut t-MDS (<5%, 5-9%, 10-19% blasts) and t-AML (≥20% blasts) at t-MN diagnosis. Genomic instability, TP53^mut allelic status, number of TP53^mut, and the clone size were similar across all four cohorts. Remarkably, the survival of TP53^mut patients were equally dismal irrespective of blast percentage (P=0.2, Figure 1A). Together, these findings suggest that TP53^mut t-MN represent a homogenous group irrespective of BM blast percentage.

Integrating the mutation, conventional cytogenetic, copy number variation, FISH, and SNP array analysis, TP53^mut were classified as single- or multi-hit (Grob et al. Blood 2022). In total, 31 (25.2%) of the 123 patients with TP53^mut (n=111) and/or loss of TP53 (n=12) locus were considered single-hit, and 92 (74.8%) were considered as multi-hit. In contrast to prior reports, we did not observe enrichment of genomic instability, or differences in degree of cytopenia, bone marrow blast % and t-MN phenotype in multi- vs single-hit TP53^mut patients. Importantly, no OS differences were observed when TP53^mut were stratified by type of TP53^mut (truncation vs. missense), site of missense mutation (hot- vs. non-hot spot) and single- vs multi-hit (Figure 1B).

CONCLUSION: The classification of t-MN based on TP53^mut status is clinically and biological relevant, irrespective of blast % or single- or multiple-hit status. TP53^mut t-MN is a molecularly defined unique group enriched for genetic instability and is associated with extremely poor outcomes.