HepatoScore-14: Measures of Biological Heterogeneity Significantly Improve Prediction of Hepatocellular Carcinoma Risk

Abstract

Background and aims: Therapeutic, clinical trial entry and stratification decisions for hepatocellular carcinoma (HCC) are made based on prognostic assessments, using clinical staging systems based on small numbers of empirically selected variables that insufficiently account for differences in biological characteristics of individual patients' disease.

Approach and results: We propose an approach for constructing risk scores from circulating biomarkers that produce a global biological characterization of individual patient's disease. Plasma samples were collected prospectively from 767 patients with HCC and 200 controls, and 317 proteins were quantified in a Clinical Laboratory Improvement Amendments-certified biomarker testing laboratory. We constructed a circulating biomarker aberration score for each patient, a score between 0 and 1 that measures the degree of aberration of his or her biomarker panel relative to normal, which we call HepatoScore. We used log-rank tests to assess its ability to substratify patients within existing staging systems/prognostic factors. To enhance clinical application, we constructed a single-sample score, HepatoScore-14, which requires only a subset of 14 representative proteins encompassing the global biological effects. Patients with HCC were split into three distinct groups (low, medium, and high HepatoScore) with vastly different prognoses (medial overall survival 38.2/18.3/7.1 months; P < 0.0001). Furthermore, HepatoScore accurately substratified patients within levels of existing prognostic factors and staging systems (P < 0.0001 for nearly all), providing substantial and sometimes dramatic refinement of expected patient outcomes with strong therapeutic implications. These results were recapitulated by HepatoScore-14, rigorously validated in repeated training/test splits, concordant across Myriad RBM (Austin, TX) and enzyme-linked immunosorbent assay kits, and established as an independent prognostic factor.

Conclusions: HepatoScore-14 augments existing HCC staging systems, dramatically refining patient prognostic assessments and therapeutic decision making and enrollment in clinical trials. The underlying strategy provides a global biological characterization of disease, and can be applied broadly to other disease settings and biological media.

Figure 1.. HepatoScore Distribution in HCC Patients.

Of note is that the graph shows three distinct patient clusters according to HepatoScore: those with normal-like circulating protein profiles (low HepatoScore, <0.3), those with highly aberrant profiles (high HepatoScore, >0.8), and those with moderately aberrant profiles (moderate HepatoScore, 0.3-0.8).

Figure 2.. Heat Map of Normalized Protein Intensity Measurements in Normal Controls (Left Block) and HCC Patients (Right Block) Ordered by HepatoScore.

Normalization was done by subtracting the mean log₂ intensity across all controls and HCC patients and dividing by the overall standard deviation. The corresponding location parameters of the log₂ normal and aberrant profiles μ^N and μ^T are plotted in the left and right margins, respectively. HCC patients are ordered according to HepatoScore, indicated by the green gradient in the top margin, which was is computed as a score measuring the relative proximity of the protein profile to the normal and aberrant distributions with aggregation across all proteins. The top 14 rows contain the 14 proteins selected for building the HepatoScore-14, the middle 105 rows contain the remaining proteins with no measurements at the detection limits (this subset was used to compute the deconvolution inherent to DeMixT), and the bottom 151 rows contain the remaining panel proteins with some thresholded values that were not used in the deconvolution. The global signal captured by the HepatoScore is apparent in the entire heat map, as HCC patients with a low HepatoScore had more normal-like circulating protein profiles, and those with a higher HepatoScore had increasingly aberrant profiles. This signal is evident in all three groups of proteins.

Figure 3.. OS in HCC Patients According to HepatoScore Within Selected Levels of Prognostic Factors.

Each row includes the number of patients overall as well as in the low (L; <0.3), medium (M; 0.3-0.8), and high (H; >0.8) HepatoScore subgroups; median OS durations and corresponding 95% CIs; P values (log-rank test) for the difference according to low, medium, and high HepatoScore; and Kaplan-Meier plots for a low (green), medium (blue), and high (red) HepatoScore. Each blue box indicates the median OS duration, the size of each box indicates the precision of the median OS estimate, and each arrow indicates that the upper quantile is greater than the maximum value in the scale. The vertical lines indicate the median OS for all patients, and the horizontal lines in the Kaplan-Meier plots indicate median OS survival.