David Schaeffer

Pancreatic ductal adenocarcinoma (PDAC) mortality rates are projected to rise by 2030 due to factors such as delayed diagnosis and resistance to chemotherapy and radiation therapy. A key challenge in treating PDAC is the lack of biomarkers for predicting treatment effectiveness and chemotherapy resistance. Researchers suggest a binary subtype system, basal-like and classical, can predict treatment selection and response, but identifying these subtypes requires costly and time-consuming RNA profiling.Histopathology, which provides an inexpensive and convenient visual readout of disease biology, has been essential in cancer diagnosis and prognosis for over a century. Artificial intelligence (AI) has recently been successfully applied to histopathology data, with AI-based models potentially outperforming traditional pathology assessments. However, an AI expert is needed to utilize and interpret these techniques.This research aimed to: 1) develop an AI-based pipeline to identify and detect histological features for classifying PDAC molecular subtypes, and 2) generalize the pipeline using a “Machine Learning Workflow Engine” and a “Web-based Slide Manager and Annotator” for processing and interpreting histopathology data.The researchers used the developed infrastructures to train and evaluate a deep-learning model for classifying PDAC patients into prognostic subgroups. They used 130 histological slides from the TCGA-PAAD dataset for training and 81 slides from 19 patients from an in-house dataset as the external test dataset. A two-step machine learning model was trained: 1) a classifier distinguishing tumor patches from stroma patches, and 2) a classifier predicting the molecular subtype of a slide based on tumor patches. The tumor/stroma classifier showed excellent performance with an AUC of 96.18% ± 1.84%, while the subtype classifier achieved a balanced accuracy of 96.19% ± 2.45% at the slide level. The model correctly classified 83.03% ± 6.35 of the patients' tumor molecular subtypes in the validation cohort. This classifier is the first to categorize PDAC patients based on biopsy samples.

View record

Pancreatic ductal adenocarcinoma (PDAC) can be stratified into distinct transcriptomebased molecular subtypes, with the ‘basal-like’ (or ‘squamous’) subtype being associatedwith worse prognosis, compared to the ‘classical’ subtype. These subtypesare assigned based on Moffitt genes signature scores where scores above a thresholdvalue are indictive of the basal-like subtype. Furthermore, PDAC tumours have uniquemetabolic transcriptomic profiles based on stratification of glycolytic and cholesterogenicgenes which correlate with basal-like and classical gene expression patterns,respectively. The mitochondrial pyruvate complex (MPC) mediates the transport ofpyruvate into the mitochondria. The mitochondrial pyruvate carrier 1 (MPC1) gene,which encodes one of two subunits of MPC, is deleted in over 60% of metastatic PDACand PDAC glycolytic tumours have lowest levels of MPC1 expression. Using PDACtissue microarray data, our group found that reduced MPC1 protein expression correlateswith reduced survival in patients. Therefore, we hypothesized that targetingMPC1 will alter metabolic reprogramming which may modulate tumour aggressivenessin tumour models. Genomically and clinically annotated patient-derived tumourorganoids (PDOs) were generated from metastatic biopsies from patients enrolled inthe PanGen study. Baseline metabolism and metabolic flux were measured usingSeahorse XFe96 based glycolytic and mito stress tests, these testswere adapted for compatibilitywith PDOs. Baseline glycolysis and oxidative phosphorylation (OXPHOS)rates demonstrated high variability in glycolytic reserves highlighting the extent ofmetabolic reprogramming in PDOs. This variability in glycolytic reserve positivelyassociated with Moffitt gene signature scores where PDOs with larger reserves tendedto have higher Moffitt scores. To alter metabolic activity, eight PDOs were treated for48 hours with UK-5099, a MPC1 inhibitor, or SRT-1720. SRT-1720 is an activator ofsirtuin 1 (SIRT1) which deacetylates peroxisome proliferator-activated receptor gammacoactivator 1-alpha (PGC1?), enhancing its activity. PGC1? has been shown to increasetranscription of MPC1. Treatment with UK-5099 raised glycolysis and glycolyticcapacity in four PDOs tested and reduced maximal respiration rates in seven PDOs.Treatment with SRT-1720 reduced glycolytic capacity in two PDOs but did not alter OXPHOSrates. Taken together, these results elicit the variability in metabolic dependencyin PDOs to meet energy demands and the plasticity of metabolic reprogramming.

View record

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal forms of cancer, with a 5-year survival rate of less than 10%. Current challenges include limited therapeutic options and lack of biomarkers that predict treatment response. Therefore, I sought to determine if a recently rediscovered treatment, pharmacological vitamin C, has clinical utility in PDAC. I determined that PDAC cell lines have differential sensitivity at doses tested. Previous research in colorectal cancer indicated that KRAS mutations infer vitamin C sensitivity, which was a trend in my results. Therefore, I created two isogenic cell line models expressing either KRAS G12D or KRAS G12V. Testing depicted increased sensitivity in one model but none others, suggesting that factors beyond oncogenic KRAS alone may be needed to increase sensitivity to vitamin C treatment. Oncogenic KRAS is known to increase glycolysis through the Warburg effect. Interestingly, pharmacological vitamin C treatment is also hypothesized to affect this pathway. Therefore, I sought to determine the relationship between vitamin C and glycolysis to determine potential markers of vitamin C sensitivity. Testing glycolysis rates demonstrated that vitamin C inhibits glycolysis independent from vitamin C toxicity. Work by Daemen et al. identified that glycolytic inhibitors cause toxicity selective to glycolytic dependant cells, whereas lipogenic cells survive. Furthermore, they characterized our two vitamin C sensitive cell lines as glycolytic. To further understand if glycolytic dependence influences vitamin C sensitivity, I used glucose withdrawal to reduce the cell’s glycolytic dependence. In low glucose conditions, higher doses of vitamin C were needed compared to high glucose conditions, suggesting that glycolytic dependence does influence toxicity to vitamin C. Together, my results suggest that glycolytic dependence may be a good marker for determining vitamin C sensitivity.To test if vitamin C is toxic in KRAS mutated patient-derived models, PDAC-derived organoids were created and treated using vitamin C monotherapy and combination therapy with gemcitabine. Vitamin C showed toxicity as a monotherapy and increased toxicity when combined with gemcitabine. This is the first known use of organoids in testing vitamin C treatment and suggests, congruent with other research, that vitamin C alone and in combination has clinical utility.

View record