TP53 mutations in the determination of azacitidine response for older patients with AML

Optimal treatment strategies for older patients with acute myeloid leukemia (AML) are elusive and the overall survival (OS) of these patients is poor. The hypomethylating agent (HMA) azacitidine was found to be no better at treating non-proliferative AML compared with intensive chemotherapy in patients older than 65 years in the AZA-AML-001 trial (NCT01074047).¹ Nevertheless, patients with high cytogenetic-risk AML or myelodysplastic syndrome, who are being treated with HMAs, may achieve even better results than those with intermediate cytogenetic risk.¹

It is possible that in patients with adverse cytogenetic risk and a high prevalence of TP53 mutations, the efficacy of HMAs may follow a TP53-independent mechanism of action. TP53 mutations have traditionally been considered as loss-of-function mutations but some recently identified mutants display gain-of-function properties. In solid tumors and diffuse large B-cell lymphoma, classifications of TP53 mutations have been proposed to indicate patient outcome, yet this has not been utilized in patients with AML.¹

Pierre Bories and colleagues recently published their analysis of the impact of TP53 mutations in patients with AML who were treated with azacitidine.¹ Their work, published in PLOS ONE, assessed the utility of TP53 mutation classifications for the identification of specific patient groups who might benefit from azacitidine treatment.

Methods

The study utilized patients from the ONCOMIP registry who had AML and had been treated with azacitidine. TP53 status was determined using bone marrow samples by either next-generation sequencing multiplex polymerase chain reaction or exome sequencing. TP53 mutations were assessed based on three different classifications: (1) disruptive mutations (impact of mutation on protein folding) versus non-disruptive, (2) evolutionary action TP53 score: high or low (a score based on missense mutations that summarizes phenotype to genotype relationship), and (3) relative fitness score (based on the structure and function relationship of mutations with 9,833 DNA-binding domain variants). In cases where patients had more than one mutation, they were classified by the mutation that was predicted to have the highest impact.

Patient characteristics

Patients were characterized according to TP53 status (Table 1), which showed a worse Eastern Cooperative Oncology Group (ECOG) performance status in patients with a TP53 mutation compared to those without (≥ 2 in 42.0% vs 26.4%, respectively; p = 0.037), a higher rate of adverse cytogenetics (96.4% vs 33.1%, respectively; p < 0.001), and also a lower OS (7.9 vs. 12.6 months, respectively; p < 0.001). There were no other significant differences.

Table 1. Significantly different patient characteristics according to TP53 status¹

Patient outcomes

The study found a TP53 mutation prevalence of 24.6% in this cohort, which is higher than previously reported.¹ Following univariate analysis, Bories et al. found several factors associated with poorer OS in patients treated with azacitidine:

• Older age (HR, 1.02; 95% CI, 1.00–1.04; p = 0.040)
• Higher levels of lactate dehydrogenase (HR, 1.08; 95% CI, 1.04–1.11; p < 0.001)
• Adverse karyotype (HR, 1.79; 95% CI, 1.35–2.37; p < 0.001)
• Presence of TP53 mutation (HR, 2.22; 95% CI, 1.60–3.08; p < 0.001)
• TP53 alteration (TP53 mutation and/or -17/17p-; HR,2.53; 95% CI, 1.85–3.45; p < 0.001)

Following multivariate analysis, while age (HR, 1.03; 95% CI, 1.01–1.05; p = 0.001), adverse karyotype (HR, 1.58; 95% CI, 1.15–2.34; p = 0.024), and lactate dehydrogenase levels (HR, 1.07; 95% CI, 1.03–1.11; p < 0.001) remained significant, TP53 mutation was no longer significantly associated with OS (HR, 1.49; 95% CI, 0.95–2.34; p = 0.081).

The cohort of 55 patients with TP53 mutations demonstrated a lower median OS compared to patients with wild-type TP53 (7.9 vs 12.6 months, respectively; HR, 2.22; 95% CI, 1.60–3.08; p < 0.001; Table 2). The 68 patients with TP53 alterations also had a worse median OS (5.4 vs 14.0 months, respectively; HR, 2.53; 95% CI, 1.85–3.45; p < 0.001). Complete response achievement was not impacted by TP53 mutation or alteration status.

TP53 mutation classification status (disruptive, evolutionary action score, relative fitness score) of patients was not associated with azacitidine response or OS.

Table 2. Patient outcomes according to TP53 status¹

Conclusion

The study team noted that their study of 224 patients represents the largest cohort of elderly patients with HMA-treated AML to be assessed for TP53 mutations. Despite finding a high prevalence of TP53 mutation, the study failed to determine which subgroups of patients with TP53 mutations would most benefit from azacitidine treatment, and the authors are uncertain whether this is due to the classification systems used or the biology of the AML subgroups included. TP53 mutation status did not have an association with response, but TP53 mutation presence was significantly associated with worse OS. Bories and colleagues concluded that this study should be repeated for other HMA therapies and for intensive chemotherapy regimens.