Bone Marrow Fibrosis and Early Hematological Response as Predictors of Poor Outcome in Azacitidine Treated High Risk-Patients With Myelodysplastic Syndromes or Acute Myeloid Leukemia

Azacitidine (AZA) treatment is effective treatment for patients with myeloid disorders, and factors predictive of treatment outcome are under investigation. Little is known about the effect of bone marrow fibrosis on response to AZA therapy. We, retrospectively, evaluated clinical predictors of overall survival (OS) and overall response rate (ORR) for patients treated with AZA in a real-life cohort. We evaluated 94 consecutive patients treated with AZA outside of clinical trials (75 mg/m²/day for 7 days every 28 days; 5 + 2 + 2 schedule), from June 2009 to February 2016. Ninety-three patients were evaluated for response. After a median of 6 cycles, ORR—complete response (CR; including marrow CR) + partial response (PR) + hematological improvement (HI)—was 41.9% (CR = 18.3%; PR = 11.8%; HI = 11.8%). Stable disease was observed in 21.5%, and failure in 36.5%. Pre-AZA bone marrow blast percentage, International Prognostic Scoring System (IPSS) or IPSS-R category, and time from diagnosis to AZA had no effect on response. Median OS from start of therapy was 18.5 months, and was significantly related to higher IPSS category (P =.01), poor cytogenetics according to the IPSS (P =.01), poor and very poor cytogenetics according to the IPSS-R (P =.02), and lower ORR (P =.006). Patients with MF-0 pre-AZA demonstrated significantly higher ORR, (CR + PR + HI) and stable disease, and lower failure rates than those with any grade of fibrosis. Indeed, cases with pre-AZA fibrosis > MF-1 had shorter OS (P =.005). Achievement of HI before 4 cycles of treatment negatively impacted OS (P =.009).