The prognostic proteins identified can serve as the basis of future pathway and network analyses.
Tools for risk prediction are developed and used to identify those who may be at risk for certain diseases and viral infections. Their main function is to facilitate decision making by physicians so that they can choose the most appropriate course of action. However, risk prediction tools in HIV have often been limited by their sole focus on previously known pathways.
Another option may be to analyze unique proteins that can be used to individualize risk assessment and identify new potential therapeutic targets. A recent study applied large-scale proteomics to identify new biomarkers that could inform disease biology as well as develop proteomic risk models predictive of all-cause mortality among HIV positive individuals.
The data was presented during the Conference on Retroviruses and Opportunistic Infections (CROI) 2021virtual sessions.
Investigators behind the study analyzed data from the longitudinal cohort of veterans with and without HIV called the Veterans Aging Cohort Study Biomarkers Cohort (VACS BC). Investigators measured plasma levels of 4,926 proteins from 1,524 HIV positive participants. Using Cox proportional hazards and attendant p-values and q-values, they assessed univariate associations of proteins with mortality (false discovery rates; FDR).
Elastic net was also used to perform multivariate modeling of mortality based on proteins, both with and without adjustment for clinical, demographic and biomarker variables. The participants analysis was divided into two datasets, derivation (80%) and validation (20%).
Findings from the study showed that with an FDR of 5%, 48% of the identified proteins were associated with mortality outcomes. The cross-validation risk model selected 9 prognostic proteins: GNPTG, SVEP1, WFDC2, ADAMTSL1, EGFR, PROC, SET, SPON2 and EFEMP1.
The c-statistic for the 9-protein model was 0.72 in the derivation set and 0.71 in the validation set, results not meaningfully altered by allowing for inclusion of CD4, nadir CD4 count, HIV RNA levels, age, and VACS score.
“Using large-scale proteomics, we identified numerous unique proteins predictive of mortality in HIV,” the authors wrote. “A risk score based on 9 proteins provided moderately good discriminative accuracy, despite heterogeneous causes of death in this population; furthermore, this protein risk score was more predictive than direct measures of HIV infection or age.”