Lenalidomide Treatment Is Not Related to AML Progression Risk but Is Associated with a Survival Benefit in RBC Transfusion-Dependent Patients with IPSS Low- or Int-1-Risk MDS with del5q: Results From a Comparative Study

Lenalidomide (LEN) treatment resulted in RBC transfusion independence for ≥ 8 wks in 51–67% of patients (pts) and cytogenetic response in 25–73% of pts with lower-risk MDS and del5q in 2 large multicenter trials (MDS-003 and -004) (List A et al. NEJM 2006;355:1456–65; Fenaux P et al. Blood 2011; doi:10.1182/blood-2011-01-330126). However, these studies were either single-arm or allowed early crossover to LEN, thus data on the influence of LEN on AML progression and overall survival (OS) is lacking. Aims: To assess the risk of AML progression and death in LEN-treated MDS-003 and -004 pts vs untreated MDS pts with del5q from a large multicenter registry, and to determine relevant risk factors. Methods: Data from 459 MDS pts with del5q entered into local or regional MDS registries were retrospectively collected from 9 centers (Europe, USA, Australia) using a uniform minimal data set. Eligible pt controls had IPSS Low-/Int-1-risk MDS and were RBC transfusion-dependent (≥ 1 unit/8 wks), reflecting the relevant inclusion criteria for both trials, and received best supportive care only including ESAs. Incidence of AML progression was assessed using a cumulative incidence estimator in the presence of competing risk (ie, death), and considering left truncation (LT) for the LEN cohort. OS was assessed using a cumulative probability estimator considering LT. Cox proportional hazards (PH) models with LT were used to assess the impact of LEN treatment and baseline factors (ie, age, sex, cytogenetics, bone marrow [BM] blast %, transfusion burden, no. of cytopenias, hemoglobin [Hgb] level [g/dL], and platelet and neutrophil counts) on risk of AML progression and death. LT is a statistical method to correct for different starting points of follow-up (ie, date of first LEN dose in clinical trial pts vs date of diagnosis in registry pts). Results: We analyzed 295 LEN-treated and 125 untreated pts. Baseline characteristics of treated (at first LEN dose) vs untreated (at diagnosis) pts were similar: mean age 65.0 vs 66.2 yrs; female sex 71% vs 68%; IPSS Low-/Int-1-risk 43%/57% vs 43%/57%. Median observation time was 4.3 vs 4.6 yrs. Baseline RBC transfusion burden was higher in the LEN cohort (median [range] units/8 wks: 6 [1–25] vs 2 [1–10]). Two- and 5-yr cumulative AML incidences were 7% and 23% for LEN vs 12% and 20% for the untreated cohort. Two- and 5-yr cumulative OS probabilities were 90% and 54% for LEN vs 74% and 41% for the untreated cohort. Median time to AML progression has not been reached for either cohort. Median OS was 5.2 yrs (95% CI 4.5–5.9) for LEN-treated vs 3.8 yrs (95% CI 2.9–4.8) for untreated pts. In the final Cox PH models, LEN treatment (hazard ratio [HR].939; p =.860) and 1 cytogenetic abnormality (abn) in addition to del5q (HR 1.111; p =.755) did not increase the risk of AML progression. Significant factors associated with an increased risk of AML progression were complex cytogenetics (del5q plus > 1 abn; HR 3.627; p =.002), BM blasts 5–10% (HR 2.215; p =.016), and higher transfusion burden (HR 1.097 [10% increase in risk per unit at baseline]; p =.029); higher Hgb levels were associated with a reduced risk (HR.857; p =.054). Regarding survival, LEN treatment was associated with a reduced risk of death (HR.597; p =.012). Other factors associated with decreased mortality were higher Hgb levels (HR.883; p =.028), higher platelet counts (HR.999; p =.035), and female sex (HR.598; p =.002). Higher transfusion burden (HR 1.056; p =.037) and age (HR 1.049; p <.001) increased the risk of death. In separate Cox PH models considering IPSS risk (Int-1 vs Low) and transfusion dependency for AML progression and OS, as well as age and sex for OS only, IPSS Int-1-risk was associated with an increased risk of AML progression (HR 1.689; p =.041) but not with an increased risk of death (HR 1.056; p =.723). Findings for LEN treatment when considering IPSS risk were similar to the final Cox PH models that considered individual covariates (AML progression: HR.892, p =.741; OS: HR.545; p =.003). Results were similar when Cox PH models were reanalyzed without LT. Conclusions: In this retrospective analysis of RBC transfusion-dependent pts with lower-risk MDS and del5q, LEN treatment was not associated with a higher risk of AML progression but led to a survival benefit vs untreated pts, despite a higher transfusion burden in the LEN cohort. Other significant risk factors for AML progression and death are consistent with previous findings in MDS pts.

Kuendgen:Celgene Corporation: Honoraria. List:Celgene Corporation: Consultancy, Honoraria, Research Funding. Fenaux:Merck: Honoraria; Johnson & Johnson: Honoraria; Amgen: Honoraria, Research Funding; Roche: Honoraria, Research Funding; Celgene Corporation: Honoraria, Research Funding; Cephalon: Honoraria; Novartis: Honoraria. Giagounidis:Celgene Corporation: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees. Brandenburg:Celgene Corporation: Employment, Equity Ownership. Backstrom:Celgene Corporation: Employment, Equity Ownership. Glasmacher:Celgene Corporation: Employment, Equity Ownership. Hasford:Celgene Corporation: Research Funding. Germing:Celgene Corporation: Honoraria, Research Funding.