Abstract 756

Treatment of acute promyelocytic leukemia (APL) has been revolutionized by the introduction of retinoid-based chemotherapy. However, up to 20% of patients fail to achieve or maintain remission. Understanding the mechanisms that underlie treatment failure remains an important goal. We genotyped a common G>A polymorphism at position -1377 (rs2234767) in the core promoter of the CD95 cell death receptor gene in 707 subjects with acute myeloid leukemia (AML) that included 233 patients with APL. We also tested the effect of the promoter variant on transcription factor binding and transcriptional activity. Compared to the GG genotype, carrier status for the A variant at position -1377 was significantly associated with worse prognosis in APL patients, but did not affect outcome in non-APL AML patients. In univariate analysis, APL patients who were GG homozygote were more likely to achieve complete remission (odds ratio (OR) 4.31, 95% confidence interval CI 1.44–12.9, p=0.009, Table) and had a better overall survival at 5 years (78% vs 64%, p=0.04) compared to heterozygotes and AA homozygotes. In multivariate Cox regression analysis that included performance status, WBC count, age, gender and secondary status in the model, carrier status for the CD95 -1377A allele (GA + AA) was a significant predictor of overall survival (OR 2.22 (1.23-4.02) p=0.007, Table) and was the strongest predictor of death within 8 weeks (OR 5.73 (2.26-14.49) p<0.0001, Table). Carriers were also significantly more likely to die from infection compared to GG homozygotes (21% vs 7%, OR 4.18, 95% CI 1.63–10.71, p=0.001). There was some evidence of an interaction between the CD95 variant and WBC count in that penetrance of the rare allele for poor outcome and infection-related death was stronger in patients with a higher WBC count at presentation. The A variant at position -1377 destroys a binding site for the stimulatory protein 1 (SP1) transcriptional regulator and is associated with lower transcriptional activity of CD95, demonstrating functional relevance of this common promoter variant. CD95 is a death receptor that regulates homeostasis in several tissues, including B and T lymphocytes. As such, differences in gene expression mediated by the -1377 variant could affect immune function. In conclusion, the G-1377A substitution represses CD95 expression by disrupting SP1 transcription factor binding and is significantly associated with poor prognosis and particularly early death in APL.

Table 1.

Prognosis of APL cases stratified by CD95 -1377 genotype.

OutcomeCD95-1377 status (Events/Patients (%))Unadjusted OR (95% CI); p valueAdjusted OR (95% CI); p value
GGGA + AA
Complete Remission 175/189 (93) 35/44 (80) 4.31 (1.44–12.9) p=0.009 4.52 (1.48–13.86) p=0.005 
Death within 7 days 5/189 (3) 3/44 (7) 3.46 (0.58–20.5) p=0.17 2.04 (0.35–11.91) p=0.4 
Death within 30 days 11/189 (6) 9/44 (20) 6.16 (1.97–19.3) p=0.002 5.00 (1.92–13.02) p=0.0003 
Death within 8 weeks 11/189 (6) 10/44 (23) 7.45 (2.44–22.7) p=0.0004 5.73 (2.26–14.49) p<0.0001 
Overall survival (deaths) 47/189 (25) 17/44 (39) 2.10 (1.09–4.08) p=0.04 2.22 (1.23–4.02) p=0.007 
Disease–free survival (relapse) 43/175 (25) 13/35 (37) 1.82 (0.88–3.76) p=0.12 1.92 (0.98–3.59) p=0.06 
Death from infection 13/189 (7) 8/44 (18) 4.68 (1.49–14.76) p=0.02 4.18 (1.63–10.71) p=0.001 
OutcomeCD95-1377 status (Events/Patients (%))Unadjusted OR (95% CI); p valueAdjusted OR (95% CI); p value
GGGA + AA
Complete Remission 175/189 (93) 35/44 (80) 4.31 (1.44–12.9) p=0.009 4.52 (1.48–13.86) p=0.005 
Death within 7 days 5/189 (3) 3/44 (7) 3.46 (0.58–20.5) p=0.17 2.04 (0.35–11.91) p=0.4 
Death within 30 days 11/189 (6) 9/44 (20) 6.16 (1.97–19.3) p=0.002 5.00 (1.92–13.02) p=0.0003 
Death within 8 weeks 11/189 (6) 10/44 (23) 7.45 (2.44–22.7) p=0.0004 5.73 (2.26–14.49) p<0.0001 
Overall survival (deaths) 47/189 (25) 17/44 (39) 2.10 (1.09–4.08) p=0.04 2.22 (1.23–4.02) p=0.007 
Disease–free survival (relapse) 43/175 (25) 13/35 (37) 1.82 (0.88–3.76) p=0.12 1.92 (0.98–3.59) p=0.06 
Death from infection 13/189 (7) 8/44 (18) 4.68 (1.49–14.76) p=0.02 4.18 (1.63–10.71) p=0.001 

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution