Pb1825: treatment patterns and survival outcomes of acute myeloid leukemia patients managed in the community setting in croatia

Background: In Croatia, patients with acute myeloid leukemia (AML) who are considered eligible for allogeneic bone marrow transplantation are usually treated at academic centers with high-intensity chemotherapies; however, no data exist regarding clinical outcomes of patients who remain treated in the community setting. Aims: To analyze treatment patterns and survival outcomes of AML patients treated in the community setting in Croatia. Methods: This was a multicenter study conducted at three community hospitals in Croatia. Clinical and laboratory data regarding AML patients who were considered unfit for allotransplant (due to advanced age and/or comorbidities) was retrospectively collected through the medical chart review. Best suportive care (BSC) included transfusion support and/or cytoreduction with hydroxycarbamide/thioguanine. Lower-intensity chemotherapies included low-dose cytarabine (LDAC) and azacytidine (AZA) with or without venetoclax (VEN), whereas higher-intensity chemotherapies included the co-administration of cytarabine and anthracyclines. Antimicrobial prophylaxis considered the administration of fluoroquinolones/co-trimoxazole. Treatment response was defined according to 2017 European Leukemia Net (ELN) guidelines. Overall survival (OS) was measured as the time from diagnosis until death or the last follow-up. Results: A total of 117 AML patients were included; median age was 76 years (56-91), 47% were females, and the median Charlson Comorbidity Index (CCI) was 5 (2-12). Seventy-two patients (61.5%) had de-novo AML, 41 (35%) secondary AML and four (3.4%) therapy-related AML. Fifty (42.7%), 43 (36.8%), and 24 (20.5%) of patients received BSC, lower-intensity (AZA=31, AZA/VEN=4, LDAC=8) and higher-intensity chemotherapies (3 + 7=12, 1 + 5=7, 2 + 5=2, 3 + 5=1, 1 + 4=1) as front-line treatments, respectively. Ninty-one patients (77.8%) prophylactically received antibiotics. Seventy-one patients (60.6%) had data available for response assessment. There was a similar proportion of CR (29.4% vs 27.8%) but more PR in the lower-intensity group when compared to a higher-intensity group (41.2% vs. 22.22%). The median OS for the entire cohort was 3 months (95% CI 2-5) and early death rate (<1 month after the diagnosis) was substantial (40.2%). Patients receiving lower-intensity treatments had longer OS (median 12 months) when compared to BSC (median OS 1 month) and higher-intensity treatments (median OS 6 months, p<0.001), as shown in the Figure 1A. Similarly, patients achieving CR (median OS 13 months) or PR (median OS 12 months) had better outcomes when compared to refractory patients (median OS 2 months, p<0.001), as depicted in Figure 1B. Other variables univariately associated with an inferior OS were: age ≥75 years (p<0.001), CCI>5 (p=0.004), no antimicrobial prophylaxis (p<0.001), total leukocyte count >3x10e9 (p=0.003), hemoglobin ≤85g/L (p=0.003) and platelets ≤66x10e9 (p=0.049). In the sex- and AML-type-adjusted Cox regression model, BSC/high-intensity treatments (p<0.001), no antimicrobial prophylaxis (p=0.006) and leukocyte count>3x10e9 (p=0.041) were independent predictors of an inferior OS. Image:Summary/Conclusion: Our real-life data shows that younger age, lower comorbidity burden, hypoproliferative disease, azacytidine/LDAC, treatment response and antimicrobial prophylaxis are associated with an improved OS of AML patients treated in the community setting in Croatia. Nevertheless, high early mortality and poor AML-related survival highlight the urgent need for clinical trials focused at this patient population.