Advanced statistics identification of participant and treatment predictors associated with severe adverse effects induced by fluoropyrimidine-based chemotherapy.

PURPOSE: Adverse effects following fluoropyrimidine-based chemotherapy regimens are common. However, there are no current accepted diagnostic markers for prediction prior to treatment, and the underlying mechanisms remain unclear. This study aimed to determine genetic and non-genetic predictors of adverse effects. METHODS: Genomic DNA was analyzed for 25 single nucleotide polymorphisms (SNPs). Demographics, comorbidities, cancer and fluoropyrimidine-based chemotherapy regimen types, and adverse effect data were obtained from clinical records for 155 Australian White participants. Associations were determined by bivariate analysis, logistic regression modeling and Bayesian network analysis. RESULTS: Twelve different adverse effects were observed in the participants, the most common severe adverse effect was diarrhea (12.9%). Bivariate analysis revealed associations between all adverse effects except neutropenia, between genetic and non-genetic predictors, and between 8 genetic and 12 non-genetic predictors with more than 1 adverse effect. Logistic regression modeling of adverse effects revealed a greater/sole role for six genetic predictors in overall gastrointestinal toxicity, nausea and/or vomiting, constipation, and neutropenia, and for nine non-genetic predictors in diarrhea, mucositis, neuropathy, generalized pain, hand-foot syndrome, skin toxicity, cardiotoxicity and fatigue. The Bayesian network analysis revealed less directly associated predictors (one genetic and six non-genetic) with adverse effects and confirmed associations between six adverse effects, eight genetic predictors and nine non-genetic predictors. CONCLUSION: This study is the first to link both genetic and non-genetic predictors with adverse effects following fluoropyrimidine-based chemotherapy. Collectively, we report a wealth of information that warrants further investigation to elucidate the clinical significance, especially associations with genetic predictors and adverse effects.

Toll-like receptor/interleukin-1 domain innate immune signalling pathway genetic variants are candidate predictors for severe gastrointestinal toxicity risk following 5-fluorouracil-based chemotherapy.

PURPOSE: Severe gastrointestinal (GI) toxicity is a common adverse effect following 5-fluorouracil (5-FU)-based chemotherapy treatment. The presence of severe GI toxicity leads to treatment revisions, sub-optimal therapy outcomes, and decreases to patients' quality of life. There are no adequate predictors for 5-FU-induced severe GI toxicity risk. The Toll-like receptor/interleukin-1 (TIR) domain innate immune signalling pathway is known to be a mediating pathway in the development of GI toxicity. Hence, genetic variability in this signalling pathway may alter the pathophysiology of GI toxicity and, therefore, be predictive of risk. However, little research has investigated the effects of TIR domain innate immune signalling pathway single nucleotide polymorphism (SNPs) on the risk and development of severe GI toxicity. METHODS: This critical review surveyed the literature and reported on the in vitro, ex vivo and in vivo effects, as well as the genetic association, of selected TIR domain innate immune signalling pathway SNPs on disease susceptibility and gene functioning. RESULTS: Of the TIR domain innate immune signalling pathway SNPs reviewed, evidence suggests interleukin-1 beta (IL1B) and tumour necrosis factor alpha (TNF) SNPs have the greatest potential as predictors for severe GI toxicity risk. These results warrant further research into the effect of IL1B and TNF SNPs on the risk and development of severe GI toxicity. CONCLUSIONS: SNPs of the TIR domain innate immune signalling pathway have profound effects on disease susceptibility and gene functioning, making them candidate predictors for severe GI toxicity risk. The identification of a predictor for 5-FU-induced severe GI toxicity will allow the personalization of supportive care measures.