Learning from Past Treatments and Their Outcome Improves Prediction of In Vivo Response to Anti-HIV Therapy

Abstract

Infections with the human immunodeficiency virus type 1 (HIV-1) are treated with combinations of drugs. Unfortunately, HIV responds to the treatment by developing resistance mutations. Consequently, the genome of the viral target proteins is sequenced and inspected for resistance mutations as part of routine diagnostic procedures for ensuring an effective treatment. For predicting response to a combination therapy, currently available computer-based methods rely on the genotype of the virus and the composition of the regimen as input. However, no available tool takes full advantage of the knowledge about the order of and the response to previously prescribed regimens. The resulting high-dimensional feature space makes existing methods difficult to apply in a straightforward fashion. The machine learning system proposed in this work, sequence boosting, is tailored to exploiting such high-dimensional information, i.e. the extraction of longitudinal features, by utilizing the recent advancements in data mining and boosting.When applied to predicting the latest treatment outcome for 3,759 treatment-experienced patients from the EuResist integrated database, sequence boosting achieved superior performance compared to SVMs with RBF kernels. Moreover, sequence boosting allows an easy access to the discriminative treatment information.Analysis of feature importance values provided by our model confirmed known facts regarding HIV treatment. For instance, application of potent and recently licensed drugs was beneficial for patients, and, conversely, the patient group that was subject to NRTI mono-therapies in the past had poor treatment perspectives today. Furthermore, our model revealed novel biological insights. More precisely, the combination of previously used drugs with their in vivo response is more in- * Hiroto Saigo and Andre Altmann are joint first authors. Please send correspondence to Hiroto Saigo,

Artificial Intelligence in Pharmacoepidemiology: A Systematic Review. Part 2–Comparison of the Performance of Artificial Intelligence and Traditional Pharmacoepidemiological Techniques

“… Saigo et al (2011) investigated the association between sequences of antiviral pharmacological treatment/virus genotype changes and the occurrence of treatment failure. The data source was the EuResist integrated database which contains the treatment history (e.g., 61,831 different pharmacological treatments) of 18,467 patients with HIV from four different countries (e.g., Germany, Italy, Luxembourg, and Sweden) collected in the period 1987–2007.…”

“…The results suggested that LASSO (mean of the results obtained with the three different datasets—AUC: 0.83; SD: 0.02) and the support vector machine (AUC: 0.79; SD: 0.02) had a higher value of AUC than logistic regression (AUC: 0.77; SD: 0.07). The authors emphasized the LASSO exerted its best performance especially for patients with many treatment changes (≥ 10) ( Saigo et al, 2011 ).…”

Artificial Intelligence in Pharmacoepidemiology: A Systematic Review. Part 1—Overview of Knowledge Discovery Techniques in Artificial Intelligence

“…3) Ravanelli et al used a LASSO regression to assess the predictive value of computed tomography texture analysis on survival in patients with lung adenocarcinoma treated with tyrosine kinase inhibitors (Ravanelli et al, 2018). 4) Saigo et al used a LASSO regression to assess if the history of medical treatments predict anti-HIV therapy response (Saigo et al, 2011).…”

The Individualized Genetic Barrier Predicts Treatment Response in a Large Cohort of HIV-1 Infected Patients

“…In addition to controlled clinical trials, analyzing data from large observational cohort studies is a promising way to identify predictors of treatment outcome, even if the availability of drugs and therapeutic strategies change over time [3]. This approach can be based on modeling the risk of acquiring additional mutations [4], on estimating future drug options [5], on predicting the time to virological failure [6], [7], or on classifying the regimens of treatment change episodes (TCEs) as successful versus failing, depending on the patient's response to therapy.…”