Distinct Implications of TP53 Hits for Patients with Treatment-Related MDS and AML

Background:TP53 is one of the most frequently mutated genes across all malignancies. Efforts to describe its role in Myeloid Neoplasms (MN) as well as Clonal Hematopoiesis of Indeterminate Potential (CHIP) have focused primarily on frequency and association with cytogenetic abnormalities and survival. Therapy-related MN (tMN), constituting ~ 10% of MN, show an increased incidence of TP53 mutations (TP53^MT) as well as dismal prognosis. A current hypothesis on the origin of TP53^MT in tMN states that it is present as CHIP prior to initiation of cytotoxic therapy (CT; chemotherapy, radiation) for primary malignancy (PM). However, analysis of differences between pre- and post- therapy TP53^MT is necessary to elucidate mechanisms behind tMN outcomes and potentially offer new therapies.

Methods: Two study groups are used in this analysis: MN patients (pts) seen at CCF and a meta-analysis of CHIP cases from 9 studies with and without PM and prior CT (Tables 1,2). Each group was compared independently as well as to each other. Differences in TP53^MT were assessed including: domains mutated, types of mutations (missense, nonsense, frameshift, splicing), base changes (transitions, transversions). Survival and progression analysis were then examined in our cohort.

We identified 61 MN pts with 79 TP53^MT determined by NGS. Thirty-eight had no PM, termed as having primary MN (pMN); 19 had PM treated with CT (tMN); 4 had PM treated without CT (surgical resection), termed as having spontaneous second MN (sMN). Parallel to this, TP53^MT from the meta-analysis (n=127) were pulled and sorted into the following categories: CHIP no malignancy, having no PM (n=64); CHIP sMN, having PM but not receiving CT prior to sampling (n=18); CHIP tMN, having PM and CT prior to sampling (n=45). We also found published data on 48 tMN pts with 68 TP53^MT, termed tMN external.

Results: In MN and CHIP, ~85% TP53^MT occurred in the DNA binding domain (DBD). The distribution of domains mutated was similar in pMN, sMN, and tMN. Comparison of CHIP TP53^MT to those in MN (irrespective of CT) as well as TP53^MT found in all cases receiving CT to those not receiving treatment (irrespective of occurrence in CHIP or MN), yielded similar results. tMN were 2.5x more likely to have 2 TP53^MTvs. pMN+sMN (p=.19). Transversions were 3x more common in tMN vs. pMN+sMN (p=.07).

Splicing TP53^MT were enriched in tMN (7%) and tMN external (12%) vs. pMN (2%), sMN (0%); enrichment was not as pronounced in CHIP tMN (4%) vs. CHIP no malignancy (2%), CHIP sMN (0%). Overall, treated cases were 6.2x more likely to have a splicing TP53^MT (p=.011). ~7% of tMN TP53^MT and 3% of those in tMN external occurred at residues involved in nuclear trafficking. Interestingly, they were found to be at nearly identical residues in the tetramerization domain: p.K305* in tMN and p.R306fs in tMN external, both in nuclear localization signal bipartite residues; p.F341Y in tMN and p.R342fs in tMN external cohort, which are part of the nuclear export signal.

In our cohort, ~40% of tMN and pMN/sMN had loss of 17p13.1 by karyotyping/SNP analysis. Family history of cancer was present in ~70% in tMN, pMN, sMN. Interestingly, family history of hematologic cancer was only identified in pMN (19%, p=.0365).

The overall survival (OS) of tMN TP53^MT was on average (avg) 8.8 months (m) vs. 19.2m in pMN and 25.2m sMN (p=.018). Avg. OS of nuclear trafficking mutations was 4.7m vs. 17.3m for DBD mutants (p=.0002). Cases in which TP53 was deemed to be an ancestral/founder mutation (by recapitulation of clonal architecture using variant allele frequency and zygosity) had median OS of 8m vs. 17m where it was deemed to be a subclonal/second hit (p=.031).OS for pts with 2 TP53^MT was on avg 8.6m vs. 19m for those with 1 mutation (p=.007). Sex, base change, mutation type, CT modality for PM had no effect on OS. Progression to higher disease state (i.e. MDS to AML) was also assessed. Overall, avg time to progression in tMN TP53^MT was 6.9m vs. 25.8m in pMN/sMN (p=.0058). The avg time to progression in pts with 1 mutation was 25.6m vs. 7.7m in those with 2 mutations (p=.027).

Conclusion:TP53^MT do not all have the same functional consequence. It has been suggested that presence of TP53^MT, type of mutation found, and associated zygosity have novel therapeutic options in addition to current standard of care. Given the differences in pre- and post- therapy cases, stratification beyond simply presence or absence of TP53^MT may be of benefit, particularly in the context of tMN.

View largeDownload slide

View largeDownload slide

Close modal