Treatment Related Mortality of AML Pediatric Patients in a Single Center in Saudi Arabia

Abstract 4291

Despite minimal changes in therapeutic approach, the outcome of acute myeloid leukemia (AML) in pediatric patients has improved significantly in the past two decades. Supportive care measures may have contributed to this success by reducing treatment related mortality (TRM) and thereby improving the overall survival (OS) of patients. Yet their impact on outcome remained unknown and masked under protocol effect. To assess the impact of supportive care measures on outcome, we undertook a retrospective review of all pediatric patients diagnosed with AML between 1986 and 2011and treated in our institution, the Princess Norah Oncology Center, King Abdulaziz Medical City, Jeddah. A total of 87 patients were reviewed. Of these, two patients whose parents refused treatment and one lost to follow-up were excluded. A total of 84 patients were qualified for the study. These patients were treated with two different protocols based on treatment eras. Patients diagnosed between 1986 and 1995 (era 1) were treated following AML-BFM-78 protocol while patients diagnosed between 1996 and 2011 (era 2) were treated following the MRC AML10. The cumulative TRM incidence was 76% in era1 compared to 11.5% in era 2 (P = 0.0001). This resulted in an improved 5-year OS from 10.5% in era 1 to 56% in era 2 (P = 0.007). The protocols used in both eras were different and may have improved OS. Significant difference in TRM however, suggests that other factors contributed to the improved OS. To gain further insight of the contributing factors, patients who received only MRC AML10 protocol (in era2) were partitioned into two sub-eras based on supportive care measures introduced sequentially in our institution as follows: 1996 to 2002 (era 2a) and 2003 to 2011 (era 2b).The cumulative TRM incidence was 48.6% in era 2a and 4.7% in era 2b (P = 0.001). This also resulted in an improved OS from 33.3% in era 2a to 56.2% in era 2b despite using the same protocol (P = 0.007). Our findings highlight the importance of supportive care as a significant factor in outcome of children. Comparing protocols per se masks the importance of supportive care measures in impacting outcome. We suggest devising a standardized scoring system to evaluate center-specific supportive care measures to quantify the impact of supportive care on TRM and survival outcomes while simultaneously allowing us to distinguish the effect of supportive care from that of protocol and other factors such as ethnicity.