A personalised approach to detect and treat endometrial cancer recurrence through mutation tracking in circulating tumour DNA and bioinformatic analysis

Despite the overall good survival of patients with endometrial cancer (EC), those with recurrent or metastatic disease have significantly poorer prognosis. There is currently a lack of prognostic and diagnostic biomarkers for EC, and particularly in the case of longitudinal monitoring of patients with EC. Circulating tumour DNA (ctDNA) has shown promise in several other cancers enabling longitudinal monitoring and accurate determination of tumour recurrence, earlier than imaging. This thesis aimed to identify prognostic biomarkers which can be utilised in earlier detection of EC recurrence, and which both inform and guide therapeutic intervention.

A systemic review was conducted to identify blood-based biomarkers for identification and longitudinal monitoring of recurrent EC with risk of bias assessed using QUADAS-2 criteria. To identify potential RNA markers of relapse, differential gene expression (DE) analysis, gene ontology (GO), and survival analysis was conducted on The Cancer Genome Atlas (TCGA) data for EC (TCGA-UCEC). Next generation sequencing using an off-the-shelf Oncomine and personalised ctDNA panels was performed on 13 patients’ tumours, with subsequent analysis in plasma.

The systematic review revealed poor specificity in identifying EC recurrence for biomarkers such as CA125, despite being widely used in clinical scenarios. Bioinformatic analysis revealed several prognostic candidate markers including CEACAM1, CTLA4, TNFRSF14 and VTCN1, which may allow earlier identification of patients with poorer prognosis. Longitudinal monitoring by liquid biopsy demonstrated that ctDNA can be detected using both off-the-shelf kits such as the Oncomine Pan Cancer Cell-Free assay and using a targeted approach with WES followed by a custom tNGS panel with ctDNA was detected in 81.25% (13/16) of patients.

In conclusion this study has shown that more sensitive circulating biomarkers are required for earlier detection of recurrence, with ctDNA showing potential as an ideal candidate.