Ragnhild A. Lothe

Cell lines are invaluable biomedical research tools, and recent literature has emphasized the importance of genotype authentication and characterization. In the present study, 24 out of 27 cell line identities were confirmed by short tandem repeat profiling. The molecular phenotypes of the 24 colon cancer cell lines were examined, and microsatellite instability (MSI) and CpG island methylator phenotype (CIMP) were determined, using the Bethesda panel mononucleotide repeat loci and two epimarker panels, respectively. Furthermore, the BRAF, KRAS and PIK3CA oncogenes were analyzed for mutations in known hotspots, while the entire coding sequences of the PTEN and TP53 tumor suppressors were investigated. Nine cell lines showed MSI. Thirteen and nine cell lines were found to be CIMP positive, using the Issa panel and the Weisenberger et al. panel, respectively. The latter was found to be superior for CIMP classification of colon cancer cell lines. Seventeen cell lines harbored disrupting TP53 mutations. Altogether, 20/24 cell lines had the mitogen-activated protein kinase pathway activating mutually exclusive KRAS or BRAF mutations. PIK3CA and PTEN mutations leading to hyperactivation of the phosphoinositide 3-kinase/AKT pathway were observed in 13/24 cell lines. Interestingly, in four cell lines there were no mutations in neither BRAF, KRAS, PIK3CA nor in PTEN. In conclusion, this study presents molecular features of a large number of colon cancer cell lines to aid the selection of suitable in vitro models for descriptive and functional research.

Recent reports have suggested that one or more genes may cause replication errors (RER) during colorectal tumorigenesis. Additional alleles are seen in the tumors when analyzing random microsatellite loci. We have studied seven dinucleotide repeat loci, located on seven different chromosomes, by use of polymerase chain reaction amplification and denaturing polyacrylamide gel electrophoresis. We found that 16.5% (40 of 243) colorectal cancers showed RER at one or several loci (RER+). This includes 31% (4 of 13) among cases with a strong positive family history according to previously published criteria and 17% (35 of 207) among cases with no history of familial cancer. Interestingly, no significant association was found between RER+ tumors and a general familial clustering of cancer. Microsatellite instability was significantly associated with DNA diploid status of the tumor (P < 0.001), with the location of the tumor in the proximal colon (P < 0.001), and with poorly differentiated tumor phenotype (P < 0.001). Patients with RER+ at > or = 2 loci tumors had an increased survival (P = 0.05). We further analyzed 84 breast cancers and 86 male germ cell cancers using the same seven markers. None of the tumors were RER+, indicating that this phenomenon may be specific to certain types of tumors.

Microsatellite instability implying multiple replication errors (RER+ phenotype) characterizes a proportion of colorectal carcinomas, particularly those from patients with the hereditary non-polyposis colorectal carcinoma syndrome. We studied the incidence of microsatellite instability in more than 500 sporadic tumors representing 6 different types of cancer. Apart from colorectal carcinoma [see the paper by Lothe et al. (Cancer Res., 53:5849-5852, 1993)] the RER+ phenotype was found in 18% (6 of 33) of gastric carcinomas and 22% (4 of 18) of endometrial carcinomas. In contrast, no evidence of this abnormality was detected in cancers of the lung (N = 85), breast (N = 84), and testis (N = 86). Importantly, the first three cancers, as opposed to the latter three, are characteristic of the hereditary non-polyposis colorectal carcinoma syndrome. These findings suggest that the cancers belonging to the hereditary non-polyposis colorectal carcinoma tumor spectrum may have essential pathogenetic steps in common, including a tendency to multiple replication errors.

Colorectal cancers (CRCs) can be divided into four gene expression-based biologically distinct consensus molecular subtypes (CMS). This classification provides a potential framework for stratified treatment, but to identify novel CMS-drug associations, translation of the subtypes to pre-clinical models is essential. The currently available classifier is dependent on gene expression signals from the immune and stromal compartments of tumors and fails to identify the poor-prognostic CMS4-mesenchymal group in immortalized cell lines, patient-derived organoids and xenografts. To address this, we present a novel CMS classifier based on a filtered set of cancer cell-intrinsic, subtype-enriched gene expression markers. This new classifier, referred to as CMScaller, recapitulated the subtypes in both in vitro and in vivo models (551 in total). Importantly, by analyzing public drug response data from patient-derived xenografts and cell lines, we show that the subtypes are predictive of response to standard CRC drugs. CMScaller is available as an R package.