Plasma proteome molecular sequencing–based prognostic nomogram in metastatic castrate-resistant prostate cancer (mCRPC).

Abstract

e17057 Background: Currently non-specific protein factors (Alkaline Phosphatase-ALK, LDH, albumin) are used for identifying prognostic risk groups in mCRPC. We explored NGS-based proteome sequencing for scalable and high-throughput plasma proteomic profiling for developing an integrated approach of mCRPC tumor-biology associated proteins with existing biomarkers based on patient-centric nomograms that estimate 6-, 12- and 24-months overall survival (OS) probabilities. Methods: Proteomic profiling of plasma for 3,072 proteins using Olink Explore NGS platform was performed in 32 local stage prostate cancer, 123 metastatic hormone-sensitive (mHSPC) and 157 mCRPC state samples (Total N = 312) in a clinically annotated real-world cohort study. Proteins were measured in Normalized Protein eXpression (NPX) units. NPX values of all proteins across all three cancer states were compared to identify Differentially Expressed Proteins (DEPs) exclusively to mCRPC using t-test with multiple corrections. Clinical outcome in the mCRPC was overall survival (OS) and Hazard Ratios (HRs) for mCRPC-DEPs with significant (P<0.05) association with survival at a univariate level were build into an aggregated composite score generated by multiplying each individual DEP’s univariate-level survival HR coefficient with the corresponding NPX value. The “Composite Proteome Prognostic Score (CPPS)” range for the mCRPC cohort (N = 121 pts) was dichotomized above and below the cohort median as “High” and “Low” and evaluated using Cox Proportional Hazard Regression along with current prognostic protein biomarkers (PSA, LDH, Alkaline Phosphatase and Albumin) at the univariate level and included in multi-variable analysis (MVA) for univariate-level significant (P<0.05) variables. Prognostic nomogram integrating MVA significant clinical factors with the CPPS was developed and nomogram performance with and without CPPS for estimating 6-,12- and 24-months survival was determined using Area Under Curve (AUC). Results: Median OS of the mCRPC cohort (N = 157) was 28 months (range: 0.3-45). 866/3072 DEPs were exclusive to mCRPC and 252/866 associated with OS (HR > 1; P < 0.05). 102/252 DEPs were significant after multiple correction (FDR < 0.05). The top 20 DEPs were used to generate CPPS. The CPPS HR value of 3.48 (95% CI: 1.77-6.83) was higher compared to known clinical factors including PSA, LDH , Alkaline Phosphatase, Albumin. AUCs for estimating mCRPC OS probabilities integrating CPPS with clinical factors versus clinical factors alone increased to 0.90 from 0.83 (6-months); 0.86 from 0.72 (12-months) and 0.76 from 0.69 (24-months). Conclusions: A patient-centric nomogram to estimate survival in mCRPC which includes novel protein classifiers can enhance current prognostication models.