Organoid icon
5-Fluorouracil treatment induces characteristic T>G mutations in human cancer

5-Fluorouracil (5-FU) is a chemotherapeutic drug commonly used for the treatment of solid tumours including colorectal and breast cancers. Since its primary synthesis in 1957 the anticancer agent became routine practice in cancer treatment both on its own and in combination with other compounds with response rates ranging between 10 and 50%, respectively. It is proposed that 5-FU, which is structurally similar to thymidine and uracil nucleotides, interferes with nucleotide synthesis though formation of 5-FdUMP that binds covalently to thymidylate synthase (TYMS) inhibiting its essential action in DNA synthesis, but also though direct incorporation of 5-FU into DNA. As a result, the compound impairs genome replication, with negative consequences for rapidly dividing cells such as cancer cells. Considering the above properties, it is conceivable that 5-FU has mutagenic potential, both on surviving tumour and healthy cells, however mutational consequences of 5-FU treatments are poorly understood.

In this report authors assessed the mutational consequences of 5-FU by exposing cultured organoids derived from healthy human small intestinal stem cells to several rounds of drug treatment, followed by single cells genome-wide analysis to look for any mutational patterns that may have developed during experimental time course.

In total 1324 highly confident induced single base substitutions (SBSs) were identified in the autosomal genome that were accumulated during 5-FU treatment. When these SBSs were analysed in trinucleotide context, authors observed a distinct mutation profile for 5-FU exposed organoids when compared to the untreated SI organoids controls. The most striking differences were specific T>G mutations in CTT trinucleotide context (converting CTT to CGT) and to a lesser extent T>C in CTT and T>G in GTT contexts (converting CTT to CCT and GTT to GGT respectively).

The observed “5-FU mutational signature” was also detected in vivo in human cancer samples. Authors explored a whole genome sequencing (WGS) database (Hartwig Medical Foundation) containing sequencing data from metastatic colorectal and breast cancer patients treated with 5-FU at any time prior to biopsy. They found similar mutation signature that they observed in in vitro organoids, with the predominance of T>G substitution in CTT context in the cancer biopsies. This indicated that 5-FU has the same mutagenic effect in vivo as in vitro.

Authors were also able to identify the same “5-FU mutational signature” in other types of solid cancers including pancreas, biliary tract and head and neck cancers confirming that the 5-FU mutational process is tissue independent and widespread.

When “5-FU mutational signature” was compared to known COSMIC cancer signatures, high molecular similarity was found with COSMIC signature 17, which is predominantly found in treatment-naïve oesophagus and gastric cancers. In agreement with COSMIC signature 17, detail analysis showed wider seven-base mutation context for C [T>G] T mutation in 5-FU pretreated colon and breast cancer patients which was predominated by A/T bases at the -4, -3 and -2 positions from the mutated base position.

Interestingly, authors observed an extensive variation in the number of 5-FU mutations per 5-TU treated patients, ranging from 0 to 15,000 mutations in both colon and breast cancer patients. This phenomenon may be explained by variation in pharmacodynamic between patients, differences in the dosing and duration of the 5-FU treatment schedules, as well as other molecular and environmental characteristics of tumours. Indeed, analysis of TP53 mutated cancers revealed that they accumulated more 5-FU mutations then TP53 wild-type cancers, both in colon and breast. However, no significant oncogenic driver genes were identified as the genes that are particularly susceptible to 5-FU exposure.

Modelling of accumulation of 5-FU induced cancer driver mutations estimated that about 300 oncogenic mutations were introduced in vivo in 108 colon stem cells per 5-FU treatments, which is 50-fold higher then under normal conditions as a result of mutational processes associated with ageing. This has a clear implication in clinical decisions whether the 5-FU treatment is advisable for a relatively young cancer patients, as therapy introduces increased risk to develop a secondary malignancy due to accumulation of cancer driver mutations in healthy cells.

In summary, through utilising a human SI organoid in vitro model treated with 5-FU, authors revealed a highly specific mutational pattern that was dominated by T>G substitution in a CTT nucleotide sequence context in treated cultures. The mutational signature was also identified in vivo in biopsies from colorectal and breast cancer patients who received 5-FU treatment. The signature showed strong resemblances to COSMIC signature 17. Although no specific effects on known oncogene genes has been observed, it has been proposed that 5-FU contributes to the mutational landscape of human cancer cells through accumulation of cancer driver mutations over time. Mutational consequences of treatment with 5-FU therefore should be taken in account when a cancer treatment plan is proposed for an individual patient.

Reference

Christensen et al., 5-Fluorouracil treatment induces characteristic T>G mutations in human cancer. Nature Communications 10:4571 (2019)

https://doi.org/10.1038/s41467-019-12594-8

Want to ask a question about

Organoids?

Organoid icon
Epistem's Organoids and Biomarker Platform

Epistem offers validated small intestinal in vitro organoid models as a robust screening tool to aid the selection of candidate treatments with minimum GI toxicity and maximum efficacy preventing or treating GI mucositis. We have extensive experience with using 5-FU chemotherapeutic drug in our organoid models.

Organoid cultures, derived from a variety of species, display cellular architecture that closely resembles that observed in vivo. Cultures mimic the stem cell niche allowing cell proliferation and differentiation to occur. The established organoid protocols allow MOA determination and provide in vitro to in vivo biological and PD linkages through the following readouts:

  • Viability assays (visual or with MTS)
  • Histology/IHC or RNAscope to assess the effect on specific protein expression
  • FACS analysis to identify specific cell populations effected
  • ELISA to identify and quantify cytokines of interest within organoid's spend media
  • Gene expression (NGS, Microarray or qRT-PCR) to evaluate the effect of treatment on expression of signalling pathways genes

About Epistem

Epistem's contract research service is committed to providing reliable, innovative and transferable pre-clinical models and services to support decision making throughout the drug discovery and development pipeline.

Tel: +44 (0)161 850 7600  Email: info@epistem.co.uk