|
Abstract INTRODUCTION:
Breast cancer is the leading cause of new cancer cases and second leading contributor to cancer-related deaths among Canadian women. In 2025, ~32,000 Canadian women were diagnosed with breast cancer, and over 5,400 women died due to breast cancer related complications. Of those cases, between 12-17% of all breast cancer cases diagnosed were classified as triple negative breast cancer (TNBC). TNBC is the most aggressive subtype of breast cancer (BC) with over a third of patients developing distant metastases throughout the course of their disease. Further it is associated with a poor prognosis, high risk of recurrence, and limited treatment options. A fruitful study area of research involves defining metabolomic signals of TNBC tumors to shed light on novel prognostic/predictive biomarkers to overcome these challenges.
To assess the TNBC metabolome, we use desorption electrospray ionization-mass spectrometry imaging (DESI), a rapid, non-destructive assay that allows for the detection of small molecules and lipid metabolites based on mass to charge ratios (m/z), as well as colocalization of metabolite expression to pathologically annotated tumor and non-tumor regions. Interestingly, we showed peroxisome proliferator-activated receptor γ (PPARγ), a transcription factor critical to normal sugar and lipid metabolism, suppresses BC progression when expressed and activated. The role of PPARγ in TNBC is unknown. Further studies of TNBC for metabolomic differences, including those dependent on protective PPARγ signaling, that distinguish aggressive from non- aggressive tumors has the potential to unveil important prognostic information and lay a foundation for the development of more targeted treatments for patients with TNBC.
OBJECTIVES:
Given the ability of DESI to detect small molecule and lipid metabolites and their abundance, and colocalize metabolite location to pathologically assessed samples, the aim is to elucidate key metabolic signaling pathways in TNBC. Ultimately, the goal is to elucidate potential therapeutic targets or prognostic markers for TNBC patients. With PPARγ being a known regulator of sugar and lipid metabolism and its implication in the role of suppressing breast tumour progression, further exploring the role PPARγ plays in metabolically dysregulated TNBC tumours has the potential to unveil important prognostic information and set the foundation for the development of more targeted treatments for these patients.
METHODS:
For these studies, I used human TNBC derived MDA-MB-231 metastatic cells that normally express very low endogenous PPARγ levels, and stably transduced them to express inducible wildtype PPARγ (231PPARγWT) in the presence of doxycycline (DOX). Next, mammary xenograft tumors (n=10 mice/treatment group) were generated by orthotopically injecting 8-12 week old female immunocompromised Rag2-/-IL2Rgc-/- knockout mice randomly assigned to one of the following treatment groups: Group 1-Mice were injected with noninduced MDA-MB-231 cells (231P) with mice maintained on normal chow diet for study duration (6 weeks); Group 2- Mice were injected with 231P cells and maintained on a PPAR activating drug Rosiglitazone (ROSI, 4mg/kg BW/day p.o.)-supplemented normal chow diet (231P+ROSI); Group 3-Mice were injected with 231PPARγWT cells and maintained on DOX diet alone; and Group 4-Mice were injected 231PPARγWT cells and maintained on a ROSI-supplemented DOX diet (231PPARγWT+ROSI).
Primary tumors collected at necropsy from were frozen in optimal cutting temperature compound, cut into 10μm sections, and metabolically profiled using DESI. The top 2000 most intense peaks with m/z ratio between the range of 50 to 1200 were recorded in effect with a lock mass of 554.2615 m/z and a tolerance of +/-0.1. All samples were run under the same optimized conditions. The same DESI slide was H&E stained, classified, and annotated for tumour, non-tumour, and necrotic zones by our collaborating pathologists blinded to treatment groups.
RESULTS:
Luminescence from in vivo imaging plotted to represent tumor growth over time (6 weeks) as a function of average area by average intensity, revealed that control mice exhibited more drastic and exponential growth compared to treatment mice. Further, measurement of tumor volumes and fluorescent imaging at endpoint showed that the expression and activation of PPARγ significantly decreased primary mammary tumor volumes and lung metastases compared to the control, respectively.
Model training using PCA/LDA demonstrates successful classification of ROIs based on tissue type (tumor, non-tumor, and necrotic) and treatment group. Of the 15,243 spectra, 15,062 were correctly classified and 182 were incorrectly classified giving a balanced accuracy of 97.56%. Therefore, the model was able to predict three pathological zones within four treatment groups with a balanced accuracy of 97.56%. Initial analysis examined the ions significantly elevated in the group 1 compared to group 4 to assess the effects of PPARγ expression and activation on metabolite abundance compared to controls. Among the top 25 differential ions were long/very long-chain fatty acids putatively identified as arachidic acid, euraic acid, gondoic acid and its C13 isomer, and cerotic acid. This could reflect a more aggressive phenotype in the control TNBC tumors. Arachidic, euraic, and cerotic acid are all fatty acids that were found to be elevated in group 1 tumor regions compared to group 4. Interestingly, these three fatty acids are all intermediates of saturated and mono-unsaturated fatty acid elongation and synthesis, a phenomenon that has been shown to contribute to cancer cell proliferation.
CONCLUSION:
Understanding the role of PPARγ in TNBC metabolomic dysregulation has the potential to unveil important prognostic information and provide the basis for developing more effective treatment for these patients.
|
= Discovery stage. (55.92%, 2026)
= Translation stage. (23.22%, 2026)
= Clinically available. (20.85%, 2026)
