TP53 mutation variant allele frequency of ≥10% is associated with poor prognosis in therapy-related myeloid neoplasms

Abstract

Revised diagnostic criteria for myeloid neoplasms (MN) issued by the International Consensus Classification (ICC) and the World Health Organization (WHO) recommended major change pertaining to TP53-mutated (TP53^mut) MN. However, these assertions have not been specifically examined in therapy-related myeloid neoplasm (t-MN), a subset enriched with TP53^mut. We analyzed 488 t-MN patients for TP53^mut. At least one TP53^mut with variant allele frequency (VAF) ≥ 2% with or without loss of TP53 locus was noted in 182 (37.3%) patients and 88.2% of TP53^mut t-MN had a VAF ≥10%. TP53^mut t-MN with VAF ≥ 10% had a distinct clinical and biological profile compared to both TP53^mut VAF < 10% and wild-type TP53 (TP53^wt) cases. Notably, TP53^mut VAF ≥ 10% had a significantly shorter survival compared to TP53^wt (8.3 vs. 21.6 months; P < 0.001), while the survival of TP53^mut VAF < 10% was comparable to TP53^wt. Within TP53^mut VAF ≥ 10% cohort, the inferior outcomes persisted irrespective of the single- or multi-hit status, co-mutation pattern, or treatments received. Finally, survival of TP53^mut patients was poor across all the blast categories and MDS patients with >10% blasts had inferior survival compared to <5%. In summary, TP53^mut VAF ≥10% signified a clinically and molecularly homogenous cohort regardless of the allelic status.

Fig. 1. Landscape of TP53 mutation (TP53^mut) in therapy-related myeloid neoplasms (t-MN).

A Oncoplot showing cytogenetic and mutational landscape of TP53^mut and wild-type TP53 (TP53^wt) t-MN. Patient-related factors (t-MN phenotype, abnormal karyotype, primary disease, treatment status, and gender) are shown in the upper panel, and the distribution of somatic gene mutations (including TP53 mutation status) in the lower panel. Each column represents an individual patient, and the presence of the aberration or mutation is colored as indicated above; B Consort diagram of the mutant TP53 status of 488 t-MN patients analyzed by integrated analysis employing next gene sequencing (NGS), conventional cytogenetics, FISH, SNP-array and CNA analysis. *Of the 48 patients with single TP53^mut VAF 10–50% LOH information was available in 33 patients. Importantly, 80% of the remaining TP53^mut patients (n = 15) without LOH information has complex karyotype and are considered equivalent to multi-hit by ICC; C Distribution of TP53^mut along the gene. Mutations from single-hit patients are shown at the bottom and those from multi-hit patients are shown at the top. Missense mutations are shown as blue circles, truncated mutations corresponding to nonsense mutations as orange circles, frameshift deletions or insertions as red circles, and splice site variants are shown as purple circles. Other types of mutations are shown as green circles. Functional protein domains are indicated in yellow (transactivation motif), brown (transactivation domain 2), blue (DNA binding domain), and green (tetramerization motif); D Summary of TP53^mut separated by mutation type and frequency of the mutations.

Fig. 2. TP53^mut drive genomic instability and was associated with poor overall survival in therapy-related myeloid neoplasms (t-MN).

A Distribution of cases according to TP53^mut VAF; B Volcano plot comparing cytogenetic aberration and somatic mutations in TP53^mut and TP53^wt t-MN. Chromosomal aberrancies highly prevalent in TP53^mut (red) and somatic mutations enriched in TP53^wt cohort (green). Genomic changes that are not differentially expressed between the two groups are shown in gray color; C Frequency of cytogenetic aberrations or driver oncogenic gene mutations in TP53^wt and TP53^mut t-MN; D Number of co-mutations in TP53^wt and TP53^mut t-MN; E Overall survival (OS) of TP53^mut with VAF ≥10% or loss of TP53 locus was significantly poor compared to wild-type TP53 (TP53^wt) and TP53^mut with VAF < 10% t-MN; F Multivariate Cox-regression analysis of factors predicting overall survival in t-MN; G OS of TP53^mut t-MN according to VAF cut-offs; H Frequency of loss of heterozygosity (LOH) and copy neutral LOH (cnLOH) according to number of TP53^mut; I Density estimation of VAF of single-hit and multi-hit TP53^mut; J OS is equally poor in single- and multi-hit in t-MN.

Fig. 3. TP53^mut was enriched in complex karyotype (CK) and is associated with significantly poor outcome of CK t-MN.

A Frequency of TP53^mut according to number of structural cytogenetic abnormalities; B High frequency of TP53^mut in typical-CK compared to atypical-CK t-MN; C TP53^mut cases were enriched for CK; D In the absence of CK, TP53^mut were prevalent in cases with 5q loss compared to 7q loss (+, present; -; absent; +/-, present or absent). Loss of 5q was defined as monosomy 5 or del 5q, while loss of 7q was defined as monosomy 7 or del 7q; E CK is associated with significantly poor survival in t-MN; F TP53^mut status further stratify CK with very poor outcome.

Fig. 4. Interaction between TP53^mut and bone marrow (BM) blast percentage.

A Distribution of t-AML and t-MDS according to TP53^mut status; B TP53^mut frequency in t-MDS was similar across blast categories at diagnosis; C Chromosomal aberrancies and TP53^mut allelic status across BM blast %; D Multivariate Cox-regression analysis showing BM blast >10%, chromosomal 18/19 abnormalities and disease modifying therapies were independent predictors of TP53^mut OS; E In TP53^mut t-MN, BM blast 10–19% and ≥20% were associated with poor OS compared to BM blast <5%; (F) OS of TP53^mut t-MN according to ICC blast categories (0–9% vs. 10–19% vs. ≥20%).