Grant: Master's Award: Canada Graduate Scholarships

Lung neuroendocrine neoplasms (NENs) are an increasingly common group of cancers that are poorly understood and difficult to diagnose. Most lung NENs are highly aggressive tumors, making early diagnosis critical to improving patient outcomes. My proposed research will lay groundwork for an inexpensive and minimally invasive real-time PCR blood test that could reduce time-to-diagnosis for lung NENs and enable diagnosis for those with limited access to care. microRNAs (miRNAs, miRs) are small RNA molecules that regulate most physiological processes and have different patterns of expression in disease states such as cancer. They also make great blood-based biomarkers due to their abundance and stability in plasma. Our group demonstrated that miR-375 expression is elevated in lung NEN patient tissue samples compared to controls. I hypothesize that plasma miR-375 can be used to accurately diagnose lung NENs. To test this, I will assess if plasma miR-375 is elevated in lung NEN patients compared to controls and evaluate its diagnostic utility. Using machine learning and statistics, I will analyze miR-375 expression data from plasma samples of patients with and without lung NENs. This will allow me to determine a plasma miR-375 expression cutoff value that corresponds to a NEN or non-NEN diagnosis. However, to optimize the diagnostic utility of miR-375, it may be necesary to create a classification model using multiple discriminant miRNAs. To test this, I will use machine learning to uncover the group of miRNAs, including miR-375, that are best at discriminating lung NEN plasma samples from controls. I will then construct a classification model that scores the likelihood of a plasma sample belonging to a lung NEN patient based on its expression of these discriminant miRNAs.