Clonal Evolution of TP53 Configurations with Treatment Predict Prognosis in Myeloid Neoplasms

Background:

TP53 is mutated in ~5-10% of de novo MDS and AML and in >30% of therapy-related MDS and secondary AML (sAML). Novel therapies have not improved outcomes yet. TP53 alterations are heterogeneous, and they affect the outcomes based on recent reports from larger registry studies. While these TP53 alterations are prognostically impactful, it is still not well studied in a longitudinally followed-up cohort with serial samples to evaluate the influence of existing standards of care treatment on the different TP53 mutational configurations. We took advantage of analyzing serially collected and sequenced TP53^MT AML/MDS/MPN samples at our institution to characterize the clonal trajectory of TP53^MT across various time points in their clinical course and the impact of treatment on it to predict prognosis.

Methods:

Targeted next-generation sequencing was performed (Illumina’s MiSeqDx) using a 54-gene Illumina TruSight Myeloid Panel. Clinical configurations of TP53 were classified based on our prior publication by Bahaj et al. (Journal of Hematology & Oncology 16.1 (2023): 91). Patients with single-hit TP53^MT and variant allelic frequency (VAF) >50% were classified as obligatory biallelic, VAF between 23-50% as probable biallelic and <23% as probable monoallelic. Combined VAF was taken for double-hit TP53^MT. TP53^MT was considered cleared when it was undetectable or had a VAF of < 3% or persistent if VAF ≥3%. The trend in TP53^MT configuration over time/treatment was studied based on this schema. Baseline clinical and molecular characteristics were noted. Chi-square and Fisher tests were used to study parameters described, and Kaplan-Meier curves were used to estimate overall survival (OS).

Results:

From 851 sequenced patients between 2016-24 with myeloid neoplasms at Karmanos Cancer Institute, 448 had serial samples. Of them, 49 had TP53^MT at diagnosis. 31 patients had sAML (63%), 10 pAML (20%), 7 MDS (14%) and 1 had MPN (2%). The median age at diagnosis was 66 (61-71) ys and 27 (55%) patients were male. 9 patients (18%) had del17p. 33 patients underwent HSCT (73%) [matched unrelated donor was common (24/33,72%)]. TP53 was cleared in 15 patients who underwent transplant and 5 with chemotherapy alone (58% v 33%; p=0.32). Overall, TP53^MT persisted in 24 (53%), while cleared in 21 (47%) patients. Based on the classification schema as described, TP53^MT myeloid neoplasm diagnostic sampling had 20 with obligatory biallelic (41%), 16 with probable biallelic (33 %), and 13 patients with probable monoallelic (26%) configurations. Clonal tracking with serial samples revealed 8 patients (18%) uptrending while 28 (62%) downtrending to a lower configuration. 9 (20%) remained in the same pattern despite treatment. 21 out of 28 downtrended patients cleared TP53^MT, with transplant being numerically but not statistically associated with clearance compared to chemotherapy alone (79 % vs 33 %; p=0.16). Patients with the same TP53 configuration on the serial samples received chemotherapy only (5/9), and 4/9 underwent transplant. TP53 was the only mutation in 26 (58%) and dominant in 33 (73%), co-dominant in 6, and a secondary event in 6 patients (13% each) in the first sample. 22 patients of 33 with the dominant clone at diagnosis remained dominant at relapse despite associated mutations.

DNMT3A, TET2, SF3B1, JAK2 and ASXL1 were the most common co-mutations observed. DNMT3A was frequently lost with TP53 (4/6 pts), while ASXL1 persisted (2/2) with TP53 over time. Patients with dominant or co-dominant TP53^MThad numerically worse OS than those who had as a secondary event (13 vs 66 mos; p=0.19). Patients with TP53^MTpersistence had significantly worse OS than those who cleared (12 vs 52 mos; p=0.007). Patients with an uptrend or the same configurational class had significantly lower OS than those who down-trended a class (mOS 11 vs. 22 mos; p=0.03).

Conclusion:

Serial NGS-based studies in patients with TP53^MT inform important clues to prognosis. While being a dominant or persistent TP53^MT clone either at the same configuration or up-trending clonal burden post treatment equates to poor prognosis, it is crucial to study further those who display better clinical outcomes with standard-of-care treatment. Such information will be invaluable for making treatment decisions in this grave prognostic disease otherwise. Ongoing work will dissect the exact treatment modalities and their influence on the TP53^MT configurational switch.