Abstract 4956

Background:

About 50% of pts with MDS ultimately progress to AML with no clear pattern in underlying parameters leading to progression. This analysis sought to identify predictive factors in pts with MDS that are associated with progression to AML after treatment with decitabine.

Methods:

In a combined analysis of MDS pts treated with either a 3- or 5-day dosing schedule of decitabine, pts were stratified on the basis of those who progressed to AML (P) versus those who did not (NP). CR rates by IWG2006 criteria for each of the baseline factors were compared by using the chi-square test. Factors that were significant in the univariate analysis (p<0.05) were included in a multivariate analysis. Logistic regression analysis was conducted for progression to AML. The following variables were identified as predictors of progression: study effect; age; secondary or de novo MDS; prior MDS treatment, including growth factor use only and prior chemotherapy; del 5Q or 7; time to diagnosis; baseline ANC (< 500, 500 – 1000, or >1000 U/L); baseline HGB, plt and BM blast count ≤20; FAB classification; ECOG status; and IPSS (low, int-1, int-2). A Cox proportional-hazards analysis was conducted for survival.

Results:

A combined analysis of prognostic factors for MDS was completed in 163 pts with a complete response to decitabine; 37 (22.6%) had progressed to AML and 126 (77.3%) had not. Baseline characteristics in pts that progressed had significantly less time since diagnosis (<3mos), more RAEB, RAEB-t and IPSS classification of Int-2 and high-risk, lower baseline HGB levels and fewer prior treatments. From the multivariate analysis, the following factors were selected as predictors of progression: No history of prior treatment, time from diagnosis of MDS <3 months, hemoglobin levels under 10 g/dL, and a 3-day schedule of decitabine (p=0.005, 95% CI 1.4, 6.8). Survival estimates were determined for Del 5q or 7, baseline HGB <10, plt <50 and the 3-day dosing schedule approached but did not achieve a significant effect (p=0.08).

Conclusions:

MDS pts with a deep anemia (HGB <10), thrombocytopenia (plts <50), and a cytogenetic risk profile that includes a del5 and/or 7 anomaly are at a statistically higher risk of transformation into AML and should be considered for additional treatment options.

Multivariate Analysis and Odds Ratio Estimates
ParameterEstimateErrorChi-quarePr > ChiSqPoint Estimate95% Wald Confidence Limits
Intercept −1.1553 0.2214 27.2218 <.0001    
Prior Therapy 0.1062 0.2113 0.2525 0.6153 1.237 0.540 2.831 
Diagnosis <3 mos 0.2985 0.2179 1.8754 0.1709 1.817 0.773 4.269 
Baseline HGB <10 0.00847 0.2178 0.0015 0.9690 1.017 0.433 2.388 
3-day Dosing ( 0007) 0.5672 0.2019 7.8915 0.0050 3.110 1.409 6.862 
Multivariate Analysis and Odds Ratio Estimates
ParameterEstimateErrorChi-quarePr > ChiSqPoint Estimate95% Wald Confidence Limits
Intercept −1.1553 0.2214 27.2218 <.0001    
Prior Therapy 0.1062 0.2113 0.2525 0.6153 1.237 0.540 2.831 
Diagnosis <3 mos 0.2985 0.2179 1.8754 0.1709 1.817 0.773 4.269 
Baseline HGB <10 0.00847 0.2178 0.0015 0.9690 1.017 0.433 2.388 
3-day Dosing ( 0007) 0.5672 0.2019 7.8915 0.0050 3.110 1.409 6.862 
Survival estimates by Cox model
Standard EstimateqErrorChi-SquarePr > ChiSqHazard Ratio
Del 5Q or 7 −0.84016 0.25203 11.1131 0.0009 0.432   
IPSS: INTERMEDIATE 1 −0.19448 0.31913 0.3714 0.5423 0.823   
IPSS: INTERMEDIATE 2 0.03408 0.29180 0.0136 0.9070 1.035   
Baseline HGB <10 0.57219 0.24546 5.4340 0.0197 1.772   
Baseline PLT <50 0.50562 0.21548 5.5060 0.0190 1.658   
Prior Therapy −0.28537 0.21227 1.8073 0.1788 0.752   
3-day Dosing ( 0007) 0.36508 0.21525 2.8766 0.0899 1.441   
Survival estimates by Cox model
Standard EstimateqErrorChi-SquarePr > ChiSqHazard Ratio
Del 5Q or 7 −0.84016 0.25203 11.1131 0.0009 0.432   
IPSS: INTERMEDIATE 1 −0.19448 0.31913 0.3714 0.5423 0.823   
IPSS: INTERMEDIATE 2 0.03408 0.29180 0.0136 0.9070 1.035   
Baseline HGB <10 0.57219 0.24546 5.4340 0.0197 1.772   
Baseline PLT <50 0.50562 0.21548 5.5060 0.0190 1.658   
Prior Therapy −0.28537 0.21227 1.8073 0.1788 0.752   
3-day Dosing ( 0007) 0.36508 0.21525 2.8766 0.0899 1.441   
Disclosures:

Borthakur: Eisai Inc.: Research Funding. Kantarjian: Novartis: Consultancy, Research Funding; Pfizer: Research Funding; Bristol Myers Squibb: Research Funding. Jabbour: Eisai Inc.: Editorial and statistical support, Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution