Characteristics, genomic landscape, and prognostic impact of trisomy 8 on contemporary outcomes in patients with newly diagnosed Acute Myeloid Leukemia (AML)

Abstract

Abstract Background: Trisomy 8 (+8) is a recurrent cytogenetic (CG) abnormality in acute myeloid leukemia (AML), occurring either as a sole chromosomal change or associated with other CG abnormalities. It is considered intermediate risk by the European LeukemiaNet (ELN) classification, but its prognostic significance in the context of coexisting CG and molecular abnormalities or with the use of venetoclax (VEN) has not yet been established. Methods: We retrospectively analyzed patients (pts) with newly diagnosed AML at 2 comprehensive cancer centers with available karyotype and molecular data at diagnosis. Pts with APL and CBF AML, or those with adverse KT (by ELN 2022) were excluded. +8 was required to have been present in ≥ 2 metaphases. Pts with +8 and hyperdiploid karyotype, but without structural abnormalities were included. Results: From 2013- 2025, 239 pts with AML and +8 were identified with a median age of 67 years (range, 21-89). Compared to pts without +8 (n=1285), those with +8 were more likely to be categorized as secondary AML (41% vs. 33%, p=0.02);other baseline characteristics did not differ significantly between the groups. 161 pts (67%) had +8 as the sole chromosomal aberration. The most common co-occurring mutations were ASXL1 (31%), SRSF2 (28%), TET2 (27%), and RUNX1 (25%). After applying the Benjamini-Hochberg method for multiple testing correction, +8 was significantly associated with mutations in ASXL1 (OR: 1.9, FDR=0.003) and KRAS (OR: 2.1, FDR=0.04), and negatively associated with mutations in NPM1 (OR: 0.4, FDR<0.01) and FLT3-ITD (OR: 0.4, FDR=0.002). To assess the prognostic impact of +8 in AML, pts were stratified by treatment intensity. Among the entire cohort, 915 pts (60%) received low-intensity therapy (LIT), including 493 (32%) in combination with VEN, while 562 (37%) received intensive chemotherapy (IC), of whom 166 (11%) received IC + VEN. Among pts treated with any LIT, pts with +8 (n=117) had inferior median overall survival (OS) compared to those without +8 (n= 798) (11.0 vs. 14.4 months, p=0.04). Using the ELN 2024 risk classification for the subgroup of pts who received LIT + VEN, pts with +8 within the favorable-risk group had a median OS of 9.5 months compared to 32.2 months in those without +8 (p=0.001). When pts with NPM1-mutated AML were excluded, the median OS was 11.2 months in those with +8 vs. 24.6 months for those without +8 (p=0.01). In the intermediate-risk group, median OS was 7.5 and 17.0 months for pts with and without +8, respectively (p=0.01). Given the association of +8 with both ASXL1 and KRAS mutations, we evaluated the impact of its co-occurrence with each on survival. In pts with ASXL1 mutations, OS was similar regardless of +8 status (median OS 7.1 vs. 11.8 months for those with [n=44], or without +8 [n=180], respectively; p=0.75). In contrast, pts with co-occurring KRAS mutations and +8 (n = 12) had significantly inferior OS compared to those with KRAS mutations without +8 (n = 55) (median OS 3.5 vs. 8.0 months, p = 0.01). Among pts treated with IC, +8 karyotype (n=103) was associated with significantly worse survival (median OS 18.6 vs. 58.1 months, p=0.01). When excluding pts with NPM1 mutated AML, however, there was no significant difference between the groups. Median OS was 18.4 with +8 vs. 31.7 months without, p=0.13. Trisomy 8 was not associated with worse outcome in pts with ASXL1 (n=89) or KRAS (n=36) mutations treated with IC. Median OS for ASXL1-mutated AML was 9.1 months with +8 vs. 13.8 months without +8 (p=0.3). Similarly, for KRAS mutations, median OS was 10.6 vs. 14.9 months with and without +8, respectively (p=0.92) when treated with IC. We then examined the impact of dual-nucleoside analogue therapy in IC treated pts with +8, and found trend toward better OS between for pts who received cladribine-based therapy vs those who did not, although this did not meet statistical significance (median OS: 52.6 vs. 17.5 months, p=0.6). Conclusion: Trisomy 8 in non-adverse risk AML conferred an adverse prognosis among pts treated with LIT + Ven, most prominently observed among pts with ELN 2024 intermediate risk. ASXL1 and KRAS mutations are overrepresented in AML with +8 and may have an impact on outcomes. Pts with concurrent +8 and KRAS mutations treated with LIT had a particularly poor outcome, highlighting the need to understand the underlying biology and to seek new therapeutic strategies.