Proteomic patterns associated with response to breast cancer neoadjuvant treatment

Tumor relapse as a consequence of chemotherapy resistance is a major clinical challenge in advanced stage breast tumors. To identify processes associated with poor clinical outcome, we took a mass spectrometry-based proteomic approach and analyzed a breast cancer cohort of 113 formalin-fixed paraffin-embedded samples. Proteomic profiling of matched tumors before and after chemotherapy, and tumor-adjacent normal tissue, all from the same patients, allowed us to define eight patterns of protein level changes, two of which correlate to better chemotherapy response. Supervised analysis identified two proteins of proline biosynthesis pathway, PYCR1 and ALDH18A1, that were significantly associated with resistance to treatment based on pattern dominance. Weighted gene correlation network analysis of post-treatment samples revealed that these proteins are associated with tumor relapse and affect patient survival. Functional analysis showed that knockdown of PYCR1 reduced invasion and migration capabilities of breast cancer cell lines. PYCR1 knockout significantly reduced tumor burden and increased drug sensitivity of orthotopically injected ER-positive tumor in vivo, thus emphasizing the role of PYCR1 in resistance to chemotherapy.