Proteomic Analysis of Pancreatic Ductal Adenocarcinoma Response to BRCA- Targeted Therapy

Introduction: Pancreatic ductal adenocarcinoma (PDAC) is a highly lethal malignancy with only a few effective treatments. Genomic studies of the cancer genome atlas identified mutations in the tumor suppressors BRCA (1 and 2) in 10-15% of PDAC patients. Based on the known synthetic lethality of BRCA and PARP, BRCA-associated PDAC is proposed to be susceptible to PARP inhibitors (PARPi). However, clinical trials show limited durable anti-tumor responses and accumulating resistance. Previous studies detected PI3K up-regulation in BRCA-associated PDAC samples, suggesting the addition of PI3K inhibitors may enhance sensitivity to PARPi and overcome resistance. Our goal in this study is to identify the differentially expressed proteins upon treatment in order to shed light on the underlying mechanisms of resistance to PARPi therapy.

Materials and methods: we used mass spectrometry-based proteomics to analyze the proteome of 112 patient-derived PDAC xenografts (generated in the Golan lab) from five patients, which were exposed to different treatments of PARPi and/or PI3Ki. We laser capture microdissected the cancer cell regions from FFPE tumor blocks, followed by protein extraction, digestion, chemical labeling and high resolution mass spectrometric analysis. Bioinformatic analysis shows the changes that occur upon treatment and are associated with tumor resistance.

Results and discussion: We examined the changes that occur in each patient upon the various treatments and found the underlying processes involved in drug response. For example, analysis of a BRCA2 tumor, which responded to treatment with PARPi+PI3Ki, showed decrease in various metabolic processes (such as glycine, serine and threonine metabolism) when compared to the control group (untreated). In addition, the combined treatment led to an increase in immune system processes (such as interferon response), as opposed to the mono-treatments. This down-regulation of metabolic processes followed by the combined treatment was not observed in another BRCA2 patient, who developed resistance to PARPi treatment. These results suggest that a certain amino acid metabolic process is involved in the formation of resistance.

Conclusion: Preliminary results of proteomic data analysis identify metabolic processes as a potential focus of resistance mechanism. Further investigation as well as integration with RNAseq and WGS will further expand our understanding of drug resistance.