Introduction Homologous Recombination Deficiency (HRD) score is a measure of genomic complexity caused by disruption/suppression of the homologous recombination repair (HRR) pathway due to specific gene mutations (BRCA1/2, PALB2, RAD51, RAD51C, RAD51D), or epigenetic events. HRD score, which reflects underlying deficiency/mutations in the genes of this pathway, predicts adverse prognosis in ovarian/ breast cancer with favorable responses to PARP inhibitors. In hematological malignancies, although complex karyotype is a known poor prognostic marker, the clinical significance of HRD has not been explored. The goal of this study was to evaluate the clinical utility of HRD score in patients with AML/MDS, specifically those associated with complex karyotype.

Methods We identified consecutive 101 treatment naïve MDS/AML [28 MDS; 73 AML) patients who underwent both clinical grade chromosomal microarray analysis (CMA) testing (Agilent aCGH+SNP 180K) and targeted 81-gene myeloid next-general sequencing mutation analysis (Illumina) in CLIA-certified lab at diagnosis. Following evaluation of CNA(s) and CN-LOH, composite HRD scores [combination of loss-of-heterozygosity (LOH >15Mb but <whole chromosome), telomeric allelic imbalance (TAI) >10Mb with telomere involvement and not crossing centromere, and large-scale state transitions (LST) between CNA(s) and CN-LOH >10Mb, excluding aneuploidy] were generated using algorithms from Nx Clinical v6.2 (BioDiscovery, Bionano Genomics). HRD calculation on the platform was previously validated on 34 hematological malignancies using Illumina SNP 850K. Clinico-pathological information was collected from the medical records. Overall survival was calculated from the time of diagnosis to death or last follow-up.

Results The study cohort included 101 AML/MDS [28 MDS, 73 AML] with a median age of 67 years (range, 1-89). The median bone marrow blast percentage was 38% (0-95). Baseline characteristics are shown. The median HRD score for the entire cohort was 6 (range, 0-39). Of 11 patients with normal karyotype, 8 (73%) patients had HRD score of 0, while rest had a positive score due to CN-LOH. Sixteen had 1-2 abnormalities [median HRD score: 3, range, 0-10].

The remaining 74 patients had complex karyotype (CK, defined as 3 or more abnormalities) which was the focus of the rest of the analysis. The median HRD score was 8 (range, 0-39). Forty-three (58%) CK-AML/MDS cases had TP53 mutations. As expected, CK-AML/MDS cases had a higher median HRD score compared to non-complex cases [median HRD: 3; p=0.005]. TP53mut CK-AML/MDS had a significantly higher number of chromosomal aberrations [16 (1-39)] compared to TP53wt CK-AML/MDS [16 (1-39) vs. 5.5 (2-11); p<0.0001], but 11 (15%) CK-AML/MDS patients had HRD of 0 due to whole chromosome gains/losses (aneuploidy). TP53mut AML/MDS cases had a significantly higher median HRD [15 (0-39)] compared to TP53wt AML/MDS cases [3 (range, 0-11); p=<0.0001].

Over the follow-up time period, the median overall survival (OS) was 7.6 months. A higher HRD score correlated with worse outcome (p=0.001). An HRD score of 12 was the best cut-off based on outcomes (9.4 vs. 2.6 months; p=0.0002; HR 2.46; CI: 1.51-4]. By univariate analysis, age, CK, TP53 mutations and HRD score (both as continuous and categorical variable based on a cut-off of 12) associated with survival, but not gender or diagnosis (MDS vs. AML). By multivariate analysis, only age and HRD score (HR 2.3 95% CI: 1.387-3.666; p=0.001) retained independent prognostic significance.

Conclusions This is the first study to explore a potential role for HRD as a prognostic marker in HM. CK-AML/MDS cases with TP53 mutations have a significantly higher HRD score compared to those with wild-type TP53. HRD score is an independent prognostic biomarker with a potential therapeutic opportunity for treatment using DNA damaging drugs.

Sasaki:Daiichi-Sankyo: Membership on an entity's Board of Directors or advisory committees; Otsuka Pharmaceuticals: Honoraria; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Membership on an entity's Board of Directors or advisory committees. Ravandi:Biomea Fusion, Inc.: Research Funding; Astellas: Consultancy, Honoraria, Research Funding; AstraZeneca: Consultancy; Xencor: Research Funding; Prelude: Research Funding; Abbvie: Consultancy, Honoraria, Research Funding; Novartis: Consultancy; Syos: Consultancy, Honoraria, Research Funding; Amgen: Honoraria, Research Funding; BMS/Celgene: Consultancy, Honoraria, Research Funding; Astex/Taiho: Membership on an entity's Board of Directors or advisory committees, Research Funding; Amgen: Honoraria, Research Funding. Loghavi:QualWorld: Consultancy; PeerView: Honoraria; Astellas: Research Funding; Amgen: Research Funding; Abbvie: Consultancy, Current equity holder in publicly-traded company; GLG: Consultancy. Yilmaz:Daiichi-Sankyo: Research Funding; Pfizer: Research Funding. Garcia-Manero:Astex: Consultancy, Honoraria, Research Funding; Aprea: Honoraria; Novartis: Honoraria, Research Funding; Genentech: Honoraria, Research Funding; Acceleron Pharma: Consultancy; BMS: Consultancy, Honoraria, Research Funding; Gilead Sciences: Research Funding; Curis: Honoraria, Research Funding; AbbVie: Honoraria, Research Funding. Kanagal-Shamanna:Amgen: Consultancy; Novartis: Consultancy; Aptitude Health: Speakers Bureau; Physicians Education Resource: Speakers Bureau.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution