AUTHORS
Michelle Hyunju Lee, MD, Angela Tramontano, MS2, Shanee Yamnitsky, BA3, Evelyn Chen, BS4, Brandon James Aubrey, MBBS PhD FRACP FRCPA5, Philip Amrein, MD6, Andrew M Brunner, MD7, Gabriela Soriano Hobbs, MD8, Hanno Hock, MD, PhD9, Rupa Narayan, MD10, Elyse R Park, PhD11, Areej R El-Jawahri, MD12, Amir T. Fathi, MD13
AFFILIATIONS
1 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
2 Dana Farber Cancer Institute, Boston, Massachusetts, United States
3 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
4 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
5 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
6 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
7 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
8 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
9 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
10 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
11 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
12 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
13 Mass General Brigham Cancer Institute, Massachusetts General Hospital, Boston, Massachusetts, United States
ABSTRACT
Introduction: Health insurance is a critical determinant of cancer disparities, but its influence on acute myeloid leukemia (AML) is unclear. While uninsured status has been associated with worse survival in national claims-based studies (Pulte et al., Leuk Res 2017), whether insurance type affects AML outcomes in the current era following healthcare reform and the emergence of venetoclax (VEN)-based regimens has not been examined. We analyzed sociodemographic factors, clinical characteristics, and AML outcomes by health insurance type.
Methods: This was a retrospective cohort study of adults with AML treated at a Comprehensive Cancer Center (CCC) in Massachusetts between 01/2014-09/2024. Adults 18-64 years at diagnosis were included given their unique exposure to the heterogeneity of pre-Medicare insurance types. Patients were identified through prescription records of cytarabine or venetoclax (VEN)-based induction regimens within the electronic medical records. Patients were excluded if they did not meet diagnostic criteria for AML by blast percentage in the bone marrow (≥20%) or by AML-defining genetic abnormalities per the EuropeanLeukemia Net (ELN) 2022. Baseline demographics, disease characteristics, and prevalence of comorbid mental disorders, substance use disorders (SUDs), and tobacco dependence were compared with descriptive statistics by public (government-funded) versus (vs) private insurance status. Primary outcome was overall survival (OS) via Kaplan-Meier; multivariable Cox regression was used to estimate association of insurance and OS. Secondary outcomes included induction response, rate of stem cell transplantation (SCT), 60-day mortality, and intensive care unit (ICU) utilization.
Results: Of the 203 patients, 139 (68%) were privately insured and 64 (31%) were publicly insured (3 uninsured included in latter given lack of employer-sponsored coverage). Publicly insured patients were more likely to be Hispanic (9% vs 2%) and younger (median 52 vs 55 years), with higher prevalence of mental disorders (43% vs 27%), SUDs (18% vs 4%), and tobacco dependence (59% vs 37%, p<0.05 for all). AML characteristics were comparable; most patients (55%) presented with adverse risk disease and 13% had a preceding myeloid neoplasm. NPM1 (30%), FLT3-ITD (21%), and TP53 (16%) were common. Publicly insured patients often presented with platelets<20,000/µL (27% vs 16%, p=0.05) and serum albumin<3.5g/dL (48% vs 31%, p=0.04). For induction, publicly insured patients were less likely to be treated on a clinical trial (8% vs 21%, p=0.02). Among the 118 patients diagnosed and treated after VEN became commercially available (11/2018), 19 (16%) received VEN (24% of public vs 12% of private, p=0.02); all publicly insured (n=9) had adverse risk AML vs 80% in privately insured (p=0.07).
At a median follow-up of 27 months (interquartile range [IQR]: 10-68.3), the median OS (mOS) was 37 months in publicly insured vs not reached (NR) in privately insured patients (p=0.04). After adjusting for age, sex, race, ELN 2022 risk, SCT, comorbidities, and trial participation, public insurance remained independently associated with higher mortality (hazard ratio=1.73, 95% confidence interval: 1.08-2.75). The groups were similar for rates of complete remission (66% in both), SCT (68% in public vs 72% in private), and 60-day mortality (1.6% in public vs 4.3% in private). ICU utilization (≥1 admission) was higher among publicly insured patients (47% vs 33% p=0.02).
In a subgroup analysis of those treated in the VEN-era (n=118), VEN treated patients had shorter mOS (243 days vs NR, p<0.001) vs those treated with intensive induction (IC), regardless of insurance type. Rates of SCT were 33% in publicly insured+VEN vs 80% in privately insured+IC (p=0.03). Publicly insured+VEN had shorter mOS compared to privately insured+VEN (194 vs 327 days, p<0.001).
Conclusion: Despite comparable disease characteristics and equal access to leukemia care at a CCC, publicly insured AML patients had higher mortality and ICU utilization than privately insured patients. In the VEN era, publicly insured were more often induced with VEN, had lower SCT rates, and worse OS even within this treatment subgroup. These disparities may reflect differences in comorbidity burden and health-related social needs, warranting further study of their influence on treatment selection, trial enrollment, and survivorship care. Post-SCT course and causes of death are under investigation.
DISCLOSURES
Michelle Hyunju Lee, MD: Daiichi Sankyo - Honoraria; Incyte - Honoraria; Novartis - Honoraria; Takeda - Honoraria; Geron - Honoraria; Syndax - Research Funding; PharmaEssentia - Research Funding; PharmaEssentia - Consultancy (Includes expert testimony)