Impact of Social Determinants of Health and Geographic Location on Real-World Outcomes in AML Patients Undergoing Low-Intensity Chemotherapy

Introduction: The combination of the BCL-2 inhibitor, venetoclax (Ven), with hypomethylating agents (HMA), such as azacitidine and decitabine, has become the standard of care in patients with acute myeloid leukemia (AML) who are not eligible for intensive chemotherapy due to age or medical comorbidities. This study aims to identify the association between different social determinants of health and geographic data that may contribute to overall survival as well as rates of neutropenic fever, invasive fungal infections (IFIs), and instances and duration of neutropenia which are leading causes of morbidity and mortality in AML.

Methods: This retrospective study identified 80 patients with AML or MDS/AML (per ELN 2022 criteria) treated with at least one cycle of Ven/HMA at the University of Virginia between January 1, 2018, and January 1, 2024. The data collected included baseline characteristics (age, sex, ECOG at diagnosis, bone marrow biopsy results), cumulative doses of Ven/HMA, instances per cycle of neutropenia, average duration of neutropenia per cycle, instances of neutropenic fever, instances of IFIs (proven and probable), and survival data. Additionally, geographic data were collected based on home addresses and data from the Virginia Department of Health. We identified whether the patients lived in a rural or non-rural location, and also grouped them into eight Virginia (VA) demographic regions. The study was censored at the date of last contact and the date of death. We evaluated four different outcomes using a combination of logistic regression, robust linear regression, ANOVA and Kaplan-Meier survival analysis: 1) mortality rate and median overall survival (mOS), 2) incidence of IFIs, 3) average incidence of neutropenic fever per cycle, and 4) average duration and instances of neutropenia per cycle.

Results: The median age at diagnosis was 72.3 years, 36 patients (45%) were female, 44 patients (55%) were male, and median ECOG at diagnosis was 1. Five patients (6.3%) had favorable risk disease, 21 patients (26.2%) had intermediate risk disease, 51 patients (63.8%) had unfavorable risk disease, and 3 patients (3.8%) had an unknown risk category. Of the 77 patients who lived in Virginia, 17 patients (22%) lived in rural areas. Fifteen patients (19.5%) lived in Central VA, 1 patient (1.2%) lived in Eastern VA, 2 patients (2.6%) lived in Hampton Roads VA, 12 patients (15.6%) lived in Northern VA, 5 patients (6.5%) lived in Southside VA, 2 patients (2.6%) lived in Southwest VA, 29 patients (37.7%) lived in Valley VA, and 11 patients (14.3%) lived in West Central VA. Different VA demographic regions were not associated with the death rate, differences in mOS, incidence of IFIs, or incidence of neutropenic fever. However, patients who lived in Southwest VA had a significantly greater average duration of neutropenia per cycle when compared to every other region of the state, except for Eastern VA (all with p <0.001). The designation between rural vs. non-rural location, as well as the patient's sex, were not associated with incidence of IFIs, average incidence of neutropenic fever per cycle or duration of neutropenia per cycle. The mOS for patients in a rural location was 4.9 months compared to 10.2 months in non-rural patients. This association neared significance with p=0.075. An ECOG of ≥2 was significantly associated with mortality rate (OR 5.86, 95% CI 1.71 - 25.9, p = 0.00924). Interestingly, African American patients had significantly fewer neutropenic fever instances per cycle than White patients (difference of -0.54, 95% CI -0.941 - -0.155, p = 0.00371).

Conclusion: We identified a specific demographic region (Southwest VA) that was associated with a greater average duration of neutropenia than almost all other regions in the state. This region also has the lowest median household income in Virginia. Thus, geographic location based on median household income may be an important indicator for screening patients who are at higher risk of developing neutropenia. This may have an impact on laboratory monitoring and decisions to initiate antimicrobial prophylaxis. Additionally, we found that patients who lived in a rural location had a lower mOS than those who lived in a non-rural location; this association neared significance.

El Chaer:DAVA Oncology: Consultancy, Other: Travel Grant; Sobi: Consultancy; FibroGen: Research Funding; Sanofi: Research Funding; Celgene: Research Funding; Bristol Meyers Squibb: Consultancy, Research Funding; Geron: Consultancy; PharmaEssentia: Consultancy, Research Funding; ACCC: Consultancy; Amgen: Consultancy, Research Funding; MorphoSys: Consultancy; CTI Biopharma: Consultancy; AbbVie: Consultancy; Sumitomo Pharma America, Inc.: Consultancy, Research Funding; BioSight: Research Funding; MEI Pharma: Research Funding; Novartis: Research Funding.