ABSTRACT Background Panitumumab shows limited clinical benefit in colorectal cancer (CRC), and reliable predictive biomarkers to guide patient selection remain lacking. To address this gap, we investigated molecular determinants of therapeutic response using tumor samples from patients with primary and metastatic CRC. By integrating PIMS‐based metastatic classification, NPOT interaction profiling and quantitative proteomics, this study aimed to identify response‐associated pathways and potential prognostic biomarkers that could support improved stratification for panitumumab therapy. Method Twenty‐one tumor resection samples from twenty CRC patients from primary site ( n = 12) and from intrahepatic metastasis ( n = 9), female ( n = 6) and male ( n = 14) were analyzed. Clinical metadata of donors was associated with molecular properties of each sample. Patients’ cryostored tumor material was blinded before subjecting to PIMS analysis. PIMS analysis was at first performed to separate metastatic and non‐metastatic patients. After uncovering the blind, tumors were all challenged by 1 μg of panitumumab in PIMS to identify responder and non‐responder with or without metastasis. Tumors from metastatic ( n = 3) and non‐metastatic ( n = 2) patients were thereafter analyzed by NPOT to identify EGFR‐related signaling pathway. All group tumors were analyzed using label‐free quantitative proteomics. Results PIMS identified with 82% accuracy metastatic ( n = 9) from non‐metastatic ( n = 7) tumor. The metastatic tumor had higher resonance volumes (2948–5094) compared to non‐metastatic (1076–2759) tumor. NPOT identified EGFR only in metastatic tumor. The metastatic interactome was composed of 34 proteins (EGFR, PTPN1, CTNNB1, CTNND1, YWHAZ, CD44, FN1, ITGB1, GADPH, ENO1, HSPA4, HSPA8, HSP90AA1, HSP90AB1, ANXA2, A1BG, DNTM1, TSR1, RPS27, PPM1G, SMC2, LIG1, NCAPD2, POLD1, PRKDC, YBX1, ANK1, FTL, NCL, ITGB2, SERPINA7, HP, and A2M). The first 10 proteins (underlined) were shared also with non‐metastatic tumors. Label‐free quantitative proteomics identified 145 differentiated protein, 15 of which were enriched and 130 impoverished specifically in metastatic tumors. Evidence suggests that HSPA4, HSP90AB1, DNTM1, RPS27, FTL, NCL, A2M are implicated in the pathogenesis and progression of colorectal cancer, positioning them as potential prognostic biomarkers for the onset of metastasis. Conclusion Combination of PIMS and NPOT coupled to label‐free quantitative proteomics point towards the distinct panitumumab mode of action in CRC patients and highlights specific proteins as prognostic biomarkers which need further validation in a bigger cohort and multicentric investigation, ideally involving patient registry follow up data. Novelty and Impact This study presents an integrative molecular profiling strategy that combines PIMS, NPOT, and proteomics to uncover mechanistically relevant biomarkers of therapeutic response in colorectal cancer. By identifying an EGFR‐centered interactome and responder‐specific protein signatures, the research offers a novel approach to stratify panitumumab response and supports advancement of precision oncology in clinical settings.
Nagelschmitz et al. (Sun,) studied this question.