Incorporating temporal dynamics of mutations to enhance the prediction capability of antiretroviral therapy's outcome for HIV-1

TL;DR

The H-model demonstrates superior discriminative ability, with a higher ROC-AUC score than the NH-model, and Wilcoxon test results confirm significant improvement of predictive accuracy for treatment outcomes through incorporating historical information.

Abstract

Motivation: In predicting HIV therapy outcomes, a critical clinical question is whether using historical information can enhance predictive capabilities compared with current or latest available data analysis. This study analyses whether historical knowledge, which includes viral mutations detected in all genotypic tests before therapy, their temporal occurrence, and concomitant viral load measurements, can bring improvements. We introduce a method to weigh mutations, considering the previously enumerated factors and the reference mutation-drug Stanford resistance tables. We compare a model encompassing history (H) with one not using it (NH). Results: The H-model demonstrates superior discriminative ability, with a higher ROC-AUC score (76.34%) than the NH-model (74.98%). Significant Wilcoxon test results confirm that incorporating historical information improves consistently predictive accuracy for treatment outcomes. The better performance of the H-model might be attributed to its consideration of latent HIV reservoirs, probably obtained when leveraging historical information. The findings emphasize the importance of temporal dynamics in mutations, offering insights into HIV infection complexities. However, our result also shows that prediction accuracy remains relatively high even when no historical information is available. Supplementary information: Supplementary material is available.

Figures (6)

• Figure 1: Figure (a) represents how different viral loads, measured when a mutation was detected, can impact present drug resistance. Figure (b) represents how different durations of a mutation can impact present drug resistance. Figure (c) represents how different mutation timing can impact present drug resistance.
• Figure 2: Figure (a) slope-intercept pairs clustered for IN-mutations, Figure (b) slope-intercept pairs clustered for PR-mutations and Figure (c) slope-intercept pairs clustered for NNRTI-mutations.
• Figure 3: Therapies and relative patient history considered in the history model and in the no-history model
• Figure 4: Plots of the probability distributions predicted by the models, respectively (a) for successes and (b) for failures. On the x-axis, there is the number of therapies, and on the y-axis, the models' predicted probabilities. The blue and red dotted lines represent the cut-offs for probabilities, respectively, for the H and NH-model. For each model, the portion of the line above its cut-off represents therapy correctly classified by the model, while the one below represents therapy incorrectly classified by the model.
• Figure 5: Graphical representation of how to label an ART therapy as success or failure according to the Standard Datum definition.

Theorems & Definitions (3)

• Definition Patient-treatment episode
• Definition Patient-treatment change episode
• Definition Standard Datum