Sequence-Based Characterization of Intratumoral Bacteria-A Guide to Best Practice

Abstract

Tumors are hospitable environments to bacteria and several recent studies on cancer patient samples have introduced the concept of an endogenous tumor microbiome. For a variety of reasons, this putative tumor microbiome is particularly challenging to investigate, and a failure to account for the various potential pitfalls will result in erroneous results and claims. Before this potentially extremely medically-significant habitat can be accurately characterized, a clear understanding of all potential confounding factors is required, and a best-practice approach should be developed and adopted. This review summarizes all of the potential issues confounding accurate bacterial DNA sequence analysis of the putative tumor microbiome, and offers solutions based on related research with the hope of assisting in the progression of research in this field.

Keywords: FFPE; cancer; low biomass; microbiome; sequence analysis.

Figure 1

The scale of the problem. Low biomass environments are considerably more susceptible to biological signal alteration arising from contaminant DNA than high biomass samples, along with the increased likelihood of PCR bias.

Figure 2

Tumors are uniquely hospitable environments for bacteria. (i) Leaky vasculature allows circulating bacteria to embed in tumor tissue; (ii) Tumors are immune privileged regions; (iii) Solid tumors possess low oxygen regions suitable for the proliferation of facultative and anaerobic bacteria; (iv) High-turnover regions of tumors can be nutrient rich, promoting bacterial growth.

Figure 3

Workflow of biological considerations prior to bioinformatic sequence analysis.

Figure 4

Overview of suggested sample preparation with appropriate control for contamination and bias.

Figure 5

Suggested bioinformatics workflow for bacterial sequence analysis from tumor tissue.