Challenges in Identifying Cancer Associated Changes in the Blood Microbiome

Abstract

In this study, we employed 16S rRNA gene sequencing to explore whether individuals with a prior cancer diagnosis exhibit alterations in their blood microbiome relative to cancer-free controls. We successfully sequenced 707 blood samples from Atlantic Canadian participants and characterized the healthy blood microbiome, which was dominated by Proteobacteria and Actinobacteriota at the phylum level and Cutibacterium and Staphylococcus at the genus level. There was a wide range of feature counts and alpha diversity metrics, but stratifying samples by age, sex, geographic location, or smoking status revealed no discernible patterns or significant associations.. Cancer samples mirrored the taxonomic profile of controls, showing only minor shifts in relative abundance and no significant differences in alpha or beta diversity across subtypes. Contamination analyses produced conflicting results for Cutibacterium-associated Amplicon Sequence Variants (ASVs), whereas batch effects were confidently ruled out. Integration with paired saliva-blood data from Nearing et al. (2023) identified shared ASVs, but their distribution did not exceed random expectation. Although bacterial DNA was detectable in samples, these profiles did not recapitulate patterns seen in microbiomes from other well-established body sites like the gut and exhibited high variability. Given the possibility of residual contamination, our results remain inconclusive regarding a healthy blood microbiome or cancer-associated compositional shifts, and reinforce the importance of incorporating stringent controls in future studies.