Azacitidine In Compassionate Use: Response to Therapy, Survival, and Prognostic Factors In 200 Patients Diagnosed with MDS or AML

Survival analysis need of large period of time for getting results. Survival subrogates variables provide with earlier data for clinical decisions. The current treatment paradigm is nowadays how the quality of the different treatment responses impact in patient's survival with the new treatment options. Azacitidine (AZA) is a hypomethilating agent which was available for clinical trials or compassionate use in Spain before receiving it marketing authorization in May 2009. We present the final analysis from those patients diagnosed with mielodysplastic syndromes (MDS) or acute myeloid leukemia (AML) selected from a longitudinal, multicenter Spanish patient registry.

This analysis retrospectively gathers clinical data about the treatment, disease progression and survival of patients with MDS or AML who had received AZA 75mg/m2 in compassionate use conditions, with a dosage regimen documented. Three different dosage regimens at the beginning of each 28-day cycle were used; group A: days 1–5 (M-F)/group B: days 1–5, 8–9 (M-F, M-Tu)/group C: days 1–7 (M-Su). Survival analysis was stratified by patients' basal conditions, dosage schedule and best response after 4^th and 6^th cycles.

Data were collected from 200 patients with MDS or AML according to the WHO diagnostic criteria. Basal data, effectiveness results and data from haematological tolerance are mainly summarized in table 1. Median survival time was 706 days (95% CI 588–1093) in those patients who achieved a response versus 225 days (95% CI 156–319) in those who did not. Survival analysis showed differences based on the best response achieved (p<0.0001), the response achieved after the 4^th cycle (p<0.0001) and the response achieved after the 6^th cycle (p<0.0001). Survival Hazard Ratios based on the best response, the responses after 4^th and 6^th cycles are summarized in table 2. Figures 1 and 2 shows survival curves based on best response after 4 and 6 cycles of therapy.

These in routine clinical practice data confirms those survival data previously published. Although results of AZA-001 show CR is sufficient but not necessary to prolong OS; our data shows an increase in the risk of not being alive when decreasing the quality of the best response achieved by patients during the treatment. To achieve the best response after the 4^th cycle could be an AZA treatment goal.

Percentages (Chi-Square Test, Exact's Fisher Test, LR Test according to application criteria met), Median +/− IQR (Kruskal-Wallis Test) or Median +/− Std (T-Test) when applicable.

Figure 1.

View largeDownload slide

Best Response after 4 cycles (R4) and Survival

Figure 1.

View largeDownload slide

Best Response after 4 cycles (R4) and Survival

Close modal

Figure 2.

View largeDownload slide

Best Response after 6 cycles (R6) and Survival

Figure 2.

View largeDownload slide

Best Response after 6 cycles (R6) and Survival

Close modal

García:Celgene: Research Funding. de Miguel LLorente:Celgene: Speakers Bureau. Bargay:Celgene: Research Funding. Ramos:Celgene: Speakers Bureau. Sanz:Celgene: Speakers Bureau. Sánchez:Celgene: Speakers Bureau.