Exome sequencing in diagnostic evaluation of colorectal cancer predisposition in young patients

Abstract

Objective. Early-onset colorectal cancer (CRC), defined here as age of onset less than 40 years, develops frequently in genetically predisposed individuals. Next-generation sequencing is an increasingly available option in the diagnostic workup of suspected hereditary susceptibility, but little is known about the practical feasibility and additional diagnostic yield of the technology in this patient group. Materials and methods. We analyzed 38 young CRC patients derived from a set of 1514 CRC cases. All 38 tumors had been tested in our laboratory for microsatellite instability (MSI), and Sanger sequencing had been used to screen for MLH1 and MSH2 mutations in MSI cases. Also, gastrointestinal polyposis had been diagnosed clinically and molecularly. Family histories were acquired from national registries. If inherited syndromes had not been diagnosed in routine diagnostic efforts (n = 23), normal tissue DNA was analyzed for mutations in a comprehensive set of high-penetrance genes (MLH1, MSH2, MSH6, PMS2, APC, MUTYH, SMAD4, BMPR1A, LKB1/STK11, and PTEN) by exome sequencing. Results. CRC predisposition syndromes were confirmed in 42% (16/38) of early-onset CRC patients. Hereditary nonpolyposis colorectal cancer was diagnosed in 12 (32%) patients, familial adenomatous polyposis in three (7.9%), and juvenile polyposis in one (2.6%) patient. Exome sequencing revealed one additional MLH1 mutation. Over half of the patients had advanced cancers (Dukes C or D, 61%, 23/38). The majority of nonsyndromic patients had unaffected first-degree relatives and microsatellite-stable tumors. Conclusions. Microsatellite instability positivity or gastrointestinal polyposis characterized all patients with unambiguous highly penetrant germline mutations. In our series, exome sequencing produced little added value in diagnosing the underlying predisposition conditions.

Key Words::

• age of onset
• colorectal neoplasms
• exome
• genetic predisposition to disease
• mutation

Acknowledgements

We are grateful to Sini Nieminen, Sirpa Soisalo, Inga-Lill Svedberg, Mairi Kuris, Sini Karjalainen, Kirsi Pylvänäinen, Tuula Lehtinen, and Iina Vuoristo for technical assistance. We acknowledge professor Aarno Palotie (FIMM and Wellcome Trust Sanger Institute, UK) and docent Maija Wessman (FIMM and Folhälsan Research Center, Helsinki) for providing Finnish, population-specific exome sequences for filtering.

Declaration of interest: The authors declare no competing interests. This work was supported by grants from the Academy of Finland (Finnish Center of Excellence Program 2012–2017), the Finnish Cancer Society, the Sigrid Juselius Foundation and SYSCOL (EU FP7 Collaborative Project).