Comparing Real-World Versus Clinical Trial Outcomes for Acute Promyelocytic Leukemia

A recent study published in Leukemia & Lymphoma examined one-month mortality and overall survival (OS) data of patients diagnosed with acute promyelocytic leukemia (APL) between 2004 and 2015 who were treated in real-world practices across the U.S. The study, led by hematology/oncology fellow Prajwal Dhakal, MD, of the University of Nebraska Medical Center, analyzed data from the National Cancer Database on 7,190 adult patients with APL. The investigators sought to determine whether real-world outcomes mirrored those observed in clinical trials.

Since 2009, there has been a marked improvement in OS for patients with APL, a rare yet aggressive form of acute myeloid leukemia. Thanks to the availability of all-trans retinoic acid and arsenic trioxide, APL has become a highly curable illness, Dr. Dhakal and coauthors noted. However, key challenges remain in the care of patients with APL, including those related to age, type of insurance used to obtain treatment, and the setting in which care is received.

The main variables of interest in this analysis included age, sex, race, education, annual household income, insurance type, Charlson Comorbidity Index (CCI) score, treatment facility type, distance traveled to treatment facility, year of diagnosis, and receipt of therapy or antileukemic agents.

The median age of the overall population was 50 years (range = 0-98 years). Approximately 5% of patients were under the age of 18, and 82% of the overall cohort was white. More than half of the population (53%) had private insurance, and most patients (92%) had a CCI score of 0 or 1.

Approximately 53% of patients in the database study were treated at an academic center. Of the 97% of patients who received treatment, 64% were treated with multi-agent therapy and 29% received single-agent therapy.

Median follow-up duration for survivors was 5.3 years (range = 3 days to 13.9 years). Overall, the one-month mortality rate was 12%, and the investigators noted no significant improvements in this rate in recent years. Across all age groups, the one-month mortality rate was 6% for those ages 18 or younger, 6% for patients ages 19 to 40, 10% for ages 41 to 60, and 21% for those over age 60.

A multivariable analysis found three variables associated with lower one-month mortality: lower CCI score (p<0.0001), private insurance (p<0.0001), and treatment at a comprehensive community cancer center versus an academic center (p=0.01). The authors also observed a trend toward worse one-month mortality in patients older than age 60 compared with patients age 18 or younger (p=0.05).

In the entire cohort, the one-year OS rate was 81%, the three-year OS rate was 75%, and the five-year OS rate was 71%. For patients age 18 and younger and ages 19 to 40, median OS was not yet reached. The median OS was 13.6 years for those between ages 41 and 60, and 4.5 years for those older than age 60 (p<0.001). According to multivariable analysis, factors associated with significantly improved OS included age 40 and younger (p<0.0001), female sex (p=0.0001), and private insurance (p<0.0001).

Factors linked with worse OS included receipt of single-agent therapy versus multi-agent therapy (hazard ratio [HR] = 1.27; p<0.0001) and no therapy (HR=3.49; p<0.0001).

The analysis of the data by year found that patients diagnosed with APL after 2009 had better survival outcomes than those diagnosed before or during 2009 (HR=0.85; p=0.001).

“Although outcomes have continued to improve over time, there are challenges associated with translating clinical trial findings to real-world practice,” the authors concluded. “Our findings highlight the need for continued research and studies aimed at improving outcomes of APL in the real world, particularly among older adults.”

A limitation of this retrospective study was its reliance on data from the National Cancer Database, which the investigators noted did not offer details on low- versus high-risk APL or information on patients’ performance statuses. In addition, the study focused its analysis on practices in the U.S., generating results that may not be fully applicable to practices in other countries with different health care systems.

The authors report no relevant conflicts of interest.


Dhakal P, Lyden E, Rajasurya V, et al. Early mortality and overall survival in acute promyelocytic leukemia: Do real-world data match results of the clinical trials? [published online ahead of print, 2021 Mar 12]. Leuk Lymphoma. doi:10.1080/10428194.2021.1894651.