Skip to Main Content
An official website of the United States government
Scheduled Maintenance: CDAS will be undergoing maintenance on Saturday, June 27th from 7:30 AM to 9:30 AM U.S. Eastern Standard Time. During this time you may experience intermittent downtime.

ABO Blood Type and Secretor Status as Determinants of Oral Microbiome Composition and Cancer Risk: Evidence from the PLCO Cohort

Principal Investigator

Name
Emily Vogtmann

Degrees
Ph.D., M.P.H.

Institution
National Cancer Institute

Position Title
Earl Stadtman Investigator

Email
emily.vogtmann@nih.gov

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCO-2058

Initial CDAS Request Approval
Jun 18, 2026

Title
ABO Blood Type and Secretor Status as Determinants of Oral Microbiome Composition and Cancer Risk: Evidence from the PLCO Cohort

Summary
We propose to use the PLCO data to evaluate whether ABO blood type and FUT2-determined secretor status are associated with (1) oral microbiome composition and (2) subsequent cancer risk. ABO antigens and secretor status shape mucosal glycosylation, which influences microbial colonization, and both have been independently linked to cancer susceptibility. By integrating genetically predicted blood type and secretor status, oral microbiome, and cancer risk data in a large prospective cohort, we will have the novel opportunity to better understand these interrelationships.

Study Design: This is a study nested within the PLCO cohort, using existing genome-wide genotype data, oral microbiome data, questionnaire data, and prospectively ascertained cancer outcomes. ABO blood type and FUT2-related secretor status will be derived from genotype data using established tag SNPs (e.g., FUT2 rs601338 for secretor status; ABO-defining variants for blood group). The blood type-microbiome analysis is cross-sectional while the blood type-cancer analysis is prospective, leveraging incident cancer ascertainment during follow-up.

Analysis Plan: Associations of ABO blood type and secretor status with overall oral microbiome composition will be tested using MiRKAT (Microbiome Regression-based Kernel Association Test), with kernels derived from multiple beta-diversity distances (e.g., Bray–Curtis and weighted/unweighted UniFrac), adjusting for relevant covariates. Alpha diversity will be assessed with linear regression, and differential abundance of individual taxa and pathways with ANCOM-BC, applying false discovery rate correction.
We will estimate associations of ABO blood type and secretor status with incident cancer overall using Cox proportional hazards models. If sample size permits, we will examine site-specific cancers of interest (e.g., colorectal, pancreatic, gastric). The proportional hazards assumption will be assessed and addressed as needed.
Covariate adjustment. Models will adjust for age, sex, race and ethnicity, education, marital status, smoking, alcohol use, body mass index, diet, and other relevant PLCO covariates.

Sensitivity analyses: We will assess robustness to alternative SNP definitions, additional covariate sets, exclusion of early follow-up cancers (to address reverse causation), and complete-case versus imputed analyses.

Sample Size and Limitations: Analytic power will depend on the number of participants with both genotype and microbiome data and on case counts for site-specific outcomes; we will report achievable precision and interpret underpowered analyses cautiously.

Aims

Aim 1: Determine the association between ABO blood type and FUT2-determined secretor status with oral microbiome composition. Using PLCO participants with both genome-wide genotype and oral microbiome data, we will derive ABO blood type and secretor status from established tag SNPs (e.g., FUT2 rs601338 for secretor status; ABO-defining variants for blood group) and test their associations with overall microbiome community composition, alpha diversity, and differential abundance of individual taxa, using compositionally appropriate methods (e.g., ANCOM-BC) with false discovery rate correction. All models will adjust for relevant covariates and genetic principal components.

Aim 2: Evaluate the association of the ABO blood type and secretor status with incident cancer risk. We will estimate associations of ABO blood type and secretor status with incident cancer overall and, where sample size permits, with selected site-specific cancers of interest (e.g., colorectal, pancreatic, gastric).

Collaborators

Emily Vogtmann National Cancer Institute
Giovanny Herrera National Cancer Institute
Jianxin Shi National Cancer Institute