A Population Based Study of 408 Cases of Acute Promyelocytic Leukemia Showing Changes in Epidemiology and Prognosis During Two Decades: Impact of Oral Arsenic Trioxide,

Abstract 3597 Background. The incidence of acute promyelocytic leukemia (APL) shows racial variations. Population based epidemiologic studies are feasible because of common diagnostic criteria and treatment. The prognosis of APL has substantially improved since the advent of all trans-retinoic acid (ATRA) and arsenic trioxide (As 2 O 3 ). Material and methods. Data on survival and relapse of consecutive APL patients in Hong Kong from 1991 to 2011 were obtained from the Hong Kong Cancer Registry (with at least 98% reporting and complete follow-up), and verified by hospital records. Data were censored at the end of July 2011. Potential factors impacting on survival including age, platelet count (Plat), white blood cell count (WBC), gender; 5-year cohort and As 2 O 3 maintenance were analyzed by logistic regression. Results. Four hundred and eight cases of APL (198 men, 210 women) at a median age of 41 (3–89) years were registered. There was a rise in cases number with successive 5-year cohorts, but the WHO standardized age incidence rate (WSIR) was unchanged (Table 1). At diagnosis (Dx), the median hemoglobin was 8.4 (2.9–14.9) g/dL, WBC 17.7 (0.3–250) × 10 9 /L (>10 × 10 9 /L in 129 cases) and Plat 35 (3–270) × 10 9 /L ( 9 /L in 282 cases). Early death (within 30 days of Dx) occurred in 88 cases. Complete remission (CR) was achieved in 318 cases. Outcome was unknown in 2 cases. The incidence of early death decreased progressively with each 5-year cohort (p=0.035), but was positively correlated with older age (p 2 O 3 (n=88; since 2001) maintenance. Relapse rates were lower in patients receiving As 2 O 3 than ATRA maintenance (17% versus 38%; p=0.008). Risk factors predicting relapse were older age (p=0.001), no oral-As 2 O 3 maintenance (p=0.003), high WBC (p=0.023) and male gender (trend only, p=0.056). For relapsed patients, 5-year overall survival (OS) from CR1 was improved from 67% with ATRA (1993–1998) to 96% with oral-As 2 O 3 (since 1999–2011) treatment. The causes of deaths after CR1 were APL relapse (n=35), second cancer (n=8), bone marrow transplantation (BMT, n=7), chemotherapy (n=3) and unrelated causes (n=6). Allogeneic BMT was not performed since 1999. For the last 5-year cohort (2006–2011), there were no APL related deaths after CR1. With reduction of induction death, and improvement in treatment and prevention of relapses, the 5-year OS from diagnosis increased from 44% for the first 5-year cohort to 80% for the last 5-year cohort. On multivariate analysis, age (p Conclusions. Population based incidence data showed that APL in Hong Kong was more prevalent than the United States (0.15–0.18/100000/year). Oral-As 2 O 3 has markedly changed the outcome of APL patients. Early death is now the greatest problem curtailing survival of APL patients. Disclosures: No relevant conflicts of interest to declare.