ABSTRACT
Background: Dysregulation of immunohematologic function (IHF) promotes cardiovascular disease and impairs protective responses to cancer and infection. A pragmatic method to identify those as risk due to IHF could improve the precision of preventive interventions and provide insight into the heterogeneity of immunologic capacity. We developed and validated a method to distill complete blood cell count data into distinct IHF profiles of prognostic relevance. Methods: We adapted latent profile analysis methods to simultaneously identify distinct groups of patients with respect to 10 immunohematologic indicators and regress time to all-cause mortality on this latent IHF profile. The model was developed using data from 30274 National Health and Nutrition Examination Survey participants and externally validated in 49851 outpatients in the Veterans Heath Administration (VHA) system and 44142 SARS-CoV-2 positive VHA patients.Findings: Ten distinct IHF profiles were identified. Profile 1, with relative mild pan-leukopenia in absence of red cell abnormalities, was associated with the best long term survival in each setting. Profiles 8-10, featuring anemia/anisocytosis especially in the setting of lymphopenia (Profiles 9-10) were associated with adjusted hazard ratio (HR) estimates of 1.76-2.62 for mortality across the three cohorts, compared to Profile 1. Profiles 6-7, featuring relative neutrophilia, were less common but also independently associated with mortality risk, especially after COVID-19 infection (Profile 7 HR [95% CI]: 2.51 [1.63 – 3.86]). The magnitude of adjusted risk conveyed by IHF profiles was greater than individual clinical risk factors (i.e., smoking, diabetes) or prevalent co-morbidities.Interpretation: Distinct immunohematologic endotypes can be identified during routine blood panels which project to mortality risk on par with a decade of life, additive to demographic and clinical factors. Applications that consider immunohematologic dysfunction may improve prevention of common fatal diseases, including COVID-19.Funding Information:This study was funded in part by The National Institute on Aging (R01AG055480; Dalton and Perzynski), the National Cancer Institute (U01CA260513; Zidar and Chan), and the United States Veteran Administration (COVID19-8900-05; Zidar). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Department of Veteran Affairs or the National Institutes of Health.Declaration of Interests: No conflict of interest exists between any of the authors and the contents of this paper.Ethics Approval Statement: The study was approved by the Institutional Review Board of the Louis Stokes Cleveland VAMC.