Development of predictive biomarkers for colorectal cancer molecular subtypes

Strategic priority area:Basic and Clinical Research
Keywords:Subtyping, colorectal cancer, predictive biomarker, patient-derived cells, translational research

Colorectal cancer (CRC) is the second most common cause of cancer death in Europe [1].Although outcomes have improved over the past decades, survival still remains poor, with 5-year survival of 50% across all stages of disease [2].It is becoming increasingly clear that CRC is not just one disease but a heterogeneous group of malignancies that arise within the same organ.Therefore the ability to subtype patient's tumours using a panel of biomarkers in the clinic would be highly favourable.
In 2015 a consortium combined information from 6 studies to produce 4 consensus molecular subtypes [3];however the subtyping focused around genomic testing which is costly to move into a clinical setting. We therefore propose a revision of the subtypes which focuses on IHC-based testing, which can easily be used in a clinical setting.If this method of subtyping proves reliable upon thorough testing, routine subtyping for CRC tumours could revolutionise the treatment of patients, with each subtype receiving a different treatment routine.Therefore subtyping of CRC tumours may be the first step toward precision medicine for these patients and our simplified panel of IHC markers may enable this transition to the clinic.

Aim 1: Validation of Molecular Subtypes (Year 1)
We developed a panel of biomarkers that could easily translate into the clinical setting and provide 4 different CRC phenotypic subtypes (table 1). We have validated the method internally on 1000 patients. In the current study, we intend to externally validate the method on 300 samples from a Norwegian cohort of patients.

In order to validate our panel of clinically translatable markers we require: Klintrup Makinen score (a measure of peritumoural inflammation) and tumour-stroma percentage scored from an H&E slide, and Ki67 proliferation index by IHC. We will also assess these markers in the Norwegian cohort and in a previously described internal cohort of 1000 consecutive patients who have undergone potentially curative resection of CRC [4]. In order to fully validate these markers and subtypes, we will then compare our markers to the original consensus molecular subtypes proposed using a separate external cohort [3]. This section will allow Kathryn to develop translation histopathology and statistical skills, and allow her to work with large clinical datasets.

Aim 2: Delineation of signalling pathways dysregulated in different CRC subtypes (Years 1-2)
In parallel, we will assess the signalling pathways important to each subtype. Firstly, we will use TempO-seq genome sequencing to assess RNA expression for 1000 patients in conjunction with DNA mutational landscaping to assess dysregulated pathways specific to each subtype. We will then choose the two most dysregulated genes/pathways for each subtype & assess the associated protein expression by IHC. We will assess the pathways effects on patient survival & clinicopathological characteristics to establish predictive biomarkers & therapeutic targets specific to each subtype. This section will allow Kathryn to develop experience of basic genomics and NGS.

Aim 3: Utilising patient-derived cell lines to assess therapeutics for each CRC subtype (Years 2-3.5)
Finally we will choose potential therapeutics for each subtype from current treatments, repurposing of available drugs or novel therapeutics & assess their efficacy using patient derived spheroids & orthotopic xenografts. This will be blinded, with the drugs tested on the cells, then FFPE tissue will be subtyped separately, then we will assess if the predicted drug was effective. The aim is to find novel therapeutic regimes for each of the four phenotypic subtypes.This section will allow Kathryn to develop advanced cell culture skills, drug discovery experience and experience using functional assays.