Pioneering studies have demonstrated that the gut microbiomes of mice, rats, and worms shape the efficacy and toxicity of cancer therapy, but the mechanisms responsible and their clinical relevance remains poorly understood. This is particularly true for fluoropyrimidines, including 5-fluorouracil (5-FU), the backbone of colorectal cancer (CRC) therapy for nearly 40 years. The increasing use of oral fluoropyrimidines has motivated efforts to better understand how host and/or microbial enzymes contribute to their first-pass metabolism prior to reaching general circulation. Hepatic dihydropyrimidine dehydrogenase (DPYD) is the rate- limiting enzyme in 5-FU clearance; however, whereas ~35% of patients will experience severe 5-FU toxicities, only ~5% of cases are explained by deficiency in host-encoded DPYD. Remarkably, gut bacteria also encode a homolog of DYPD that is capable of inactivating 5-FU. Here, we propose complementary studies in humans and gnotobiotic animals aimed at testing the hypothesis that the abundance and expression of gut bacterial DPYD will predict oral fluoropyrimidine response in CRC patients. There is a strong premise for this hypothesis based on our Preliminary Results and the published literature, but definitive studies in humans and rodent models are lacking.
In Aim 1, we will identify gut microbial biomarkers of cancer drug outcomes. Stool samples are being prospectively collected from CRC patients prior to initiation of oral fluoropyrimidine therapy and at designated time-points during therapy as part of our existing clinical study protocol. We will assess the following endpoints: microbial gene abundance (metagenomics) and gene expression (metatranscriptomics). Together with extensive clinical metadata, we will identify microbial predictors of clinical outcomes, including response, resistance, and toxicity.
In Aim 2, we will establish causal links between specific gut bacterial genes and drug levels. Genetic deletion and complementation of bacterial DPYD will validate its proposed role in drug inactivation during in vitro growth and following the colonization of germ-free mice and oral administration of fluoropyrimidines. This work directly addresses Provocative Question 10 (RFA-CA-17-018), ?How do microbiota affect the response to cancer therapies?? and will lay the foundation for therapeutic manipulation of patients? microbiota to optimize drug response and minimize toxicity. Although we are currently focused on the direct inactivation of 5-FU by gut bacterial enzymes, our approach will also enable unbiased associations that could reveal unexpected links between the gut microbial metabolites and anti-cancer drug response.
Here, we propose complementary studies in humans and animal models aimed at answering the provocative question, ?how do microbiota affect the response to cancer therapies??. While our focus is on the direct inactivation of 5-FU by gut bacterial enzymes in patients with colorectal cancer, our unbiased approach could also reveal unexpected links between the gut microbial metabolites and anti-cancer drug response. If successful, this research would lay the foundation for therapeutic manipulation of patients? microbiota to optimize drug response and minimize toxicity.