Towards an Understanding of the Growing Incidence of Colorectal Cancer and Appendiceal Neoplasms in Young Adults

Abstract

Colorectal cancer (CRC) incidence and mortality rates are rising in young adults aged <50 years old [referred to as young-onset CRC (YOCRC)] in Australia and many other countries while there has been a steady decline in overall rates of this malignancy in individuals aged ≥50 years old. In addition, the incidence and mortality rates of appendiceal neoplasms (ANs) have also been reported to be on the rise in both age groups (<50 years and ≥50) in the United States, Canada and Netherlands. Currently, the causes of these observations remain largely unknown. Identifying the underlying aetiological factors for YOCRC and ANs is a primary public health priority because addressing these contributing factors is a key prevention strategy. The main aim of this thesis is to explore the observation of the increasing incidence of CRC and ANs in young adults. The aetiology of YOCRC is likely to be heterogeneous, comprising a spectrum of genetic and environmental triggers. To this end, we have investigated the role of type 2 diabetes (T2D) as a marker of increased risk, and explored the exome in YOCRC patients for pathogenic germline variants. Consistent evidence suggests an association between T2D at any age and increased CRC risk. An observational study found that a personal history of T2D was significantly higher in YOCRC patients compared to controls (age- and sex-matched individuals with clear colonoscopies). In addition, analysis of exome sequencing data of YOCRC patients showed that one in six YOCRC patients had clinically actionable germline variants in at least one cancer-predisposing gene, with 35% of these being in genes associated with breast or ovarian cancer. First-degree relatives with CRC were rarely seen in variant carriers and three patients with variants in polyposis associated genes (MUTYH (bi-allelic), RNF43 and BMPR1A) showed no polyposis. In addition, two individuals with CRC were identified from a single-family carrying a likely-pathogenic germline variant in RNF43:c.375+1G>A. Tumours from both carriers were BRAFV600E-mutated and mismatch repair-proficient indicating that the CRCs arose in sessile serrated lesions. However, the proband did not meet the clinical criteria for serrated polyposis. Both studies taken together suggest that phenotype was a poor predictor of genotype. Trends in incidence and mortality rates of ANs in Australia were explored by performing a retrospective analysis on national data obtained from the Australian Institute of Health and Welfare from 1982 to 2013. Similar to the observed trend in other countries, this work has demonstrated that the incidence and mortality rates of ANs are alarmingly on the rise in Australia in both age groups (<50 years and ≥50), both genders, and within diverse histological subtypes. In conclusion, findings from this work suggest that there is an enrichment for personal history of T2D in patients with YOCRC, and that carriers of variants in breast/ovarian cancer-related genes might need to receive surveillance tests for CRC earlier than the general population, and importantly, that multigene panel testing is warranted for all YOCRC patients regardless of family history or phenotype. The findings also lend weight to further consideration for a hereditary role for RNF43 as a tumour suppressor gene in colorectal tumorigenesis outside the setting of individuals meeting the clinical criteria for serrated polyposis. In addition, an apparent rise in the incidence and mortality rates of ANs in Australia was demonstrated, the causes of which remain unclear. Further research exploring the risk factors for YOCRC and ANs is warranted, to stem the rising trend of both these malignancies.