Abstract 577

Trisomy 8 (+8), the most frequent numerical aberration in AML, occurs in approximately 10% of pts. In one-third of them, +8 is the sole chromosome (chr) anomaly; it is considered to confer an intermediate/adverse prognosis but further studies are required to define the clinical and biologic significance of this cytogenetic abnormality. We assessed the associations of sole +8 with prognostic gene mutations, outcome, and gene and miR expression profiles by comparing a relatively large cohort of sole +8 de novo AML pts (n=80; median, 63 years [y]; range, 18–84 y) with cytogenetically normal (CN) de novo AML pts (n=483; median, 60 y; range, 18–83 y). CN pts constitute the largest cytogenetic group in AML, with an overall intermediate prognosis modified by molecular markers. Markers analyzed in the present sole +8 cohort included mutations in NPM1, FLT3 (FLT3 internal tandem duplication [ITD], FLT3 tyrosine kinase domain), CEBPA, WT1, IDH1/2, N/KRAS and RUNX1. All pts were enrolled on frontline cytarabine/daunorubicin based CALGB protocols. No pt included in outcome analyses received allogeneic stem cell transplant in 1st complete remission (CR). Median follow-up was 7.1 y for pts alive. Compared with CN pts, sole +8 pts had lower platelet counts (P<.001), white blood counts (WBC; P<.001) and % blood blasts (P=.03), and less often NPM1 mutations (25% v 61%; P<.001). Sole +8 pts frequently had IDH1/2 (38%) and RUNX1 (31%) mutations and FLT3 ITD (28%); aside from one pt, RUNX1 and NPM1 mutations were mutually exclusive (P=.007). Compared with CN pts, sole +8 pts had lower CR rates (64% v 76%; P=.05), and shorter disease-free (DFS; P=.002; 3 y rates, 16% v 33%) and overall survival (OS; P=.006; 3 y rates, 24% v 35%). Interestingly, among sole +8 pts, there were no significant differences in CR rates (P=1.0), DFS (P=.31) or OS (P=.19) between pts ≥60 y and pts <60 y. Indeed, in multivariable analyses of sole +8 pts (Table), age added no prognostic information, while lower hemoglobin was associated with higher CR rates, absence of FLT3 ITD and presence of NPM1 mutations with longer DFS, and having an IDH2 mutation and Caucasian race with longer OS. To gain biologic insights, a gene expression signature using Affymetrix U133 plus 2.0 arrays was derived comparing sole +8 and CN pts; 1079 genes were upregulated and 735 downregulated in sole +8 pts. Less than 1% of the downregulated but 258 (24%) of the upregulated genes mapped to chr 8; the latter constitute 62% of 417 genes located on chr 8 studied. MN1 and chr 8 located BAALC were among genes upregulated in sole +8 pts; high expression of these genes associates with poor outcome in CN AML. Using custom made OSUCCC v4.0 arrays we also derived a signature of 23 miRs differentially expressed between sole +8 and CN pts; miR-107 and miR-342, both upregulated during ATRA treatment in t(15;17) AML, and miR-29b, targeting SP1, DNMT3A/B and MCL1, were upregulated in sole +8 pts. Notably, none of the 13 studied miRs located on chr 8 was significantly (P<.005) upregulated in sole +8 pts. In conclusion, compared with CN AML, sole +8 AML is associated with distinctive clinical and molecular characteristics and a poor outcome, indicating that the additional chr 8 confers specific features. The outcome of sole +8 pts is affected by FLT3 ITD and mutations in NPM1 and IDH2 but not by age. The biologic uniqueness of sole +8 AML is supported by the upregulation of many genes located on chr 8. In contrast, miRs located on chr 8 were not significantly upregulated. Functional studies to assess the significance of these findings are underway.

Multivariable Analyses in Sole +8 Pts

CRDFSOS
n=57n=31n=50
POR (95% CI)PHR (95% CI)PHR (95% CI)
Hemoglobin, continuous .05 0.71 (0.61–1.00) – – – – 
FLT3 ITD, present v absent – – .004 4.38 (1.61–11.88) – – 
NPM1, mut v wt – – .05 0.39 (0.16–0.98) – – 
IDH2, mut v wt – – – – .02 0.35 (0.15–0.83) 
Race*, Caucasian v others – – – – .02 0.37 (0.15–0.91) 
Age group, ≥60 y v <60 y .74 – .27 – .14 – 
CRDFSOS
n=57n=31n=50
POR (95% CI)PHR (95% CI)PHR (95% CI)
Hemoglobin, continuous .05 0.71 (0.61–1.00) – – – – 
FLT3 ITD, present v absent – – .004 4.38 (1.61–11.88) – – 
NPM1, mut v wt – – .05 0.39 (0.16–0.98) – – 
IDH2, mut v wt – – – – .02 0.35 (0.15–0.83) 
Race*, Caucasian v others – – – – .02 0.37 (0.15–0.91) 
Age group, ≥60 y v <60 y .74 – .27 – .14 – 

OR, odds ratio; HR, hazard ratio; CI, confidence interval; mut, mutated; wt, wild type.

Variables considered were those significant at α=.20 in univariable models, ie, for CR, hemoglobin, platelet counts (50×109/L increase), +8 metaphases (≥80% v <80%); for DFS, NPM1, FLT3 ITD, RAS (mut v wt), WBC (50×109/L increase), race; for OS, IDH2, RAS, race, age group. Age group was retained regardless of statistical significance.

*

Race did not meet the proportional hazards assumption for OS; HR is at 1 y.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution