by In an effort to address the debilitating side effects associated with combination antiretroviral therapy (cART) in individuals with HIV, a team of researchers at Johns Hopkins University has devised a novel method to optimize HIV treatments. Led by Yanxun Xu, an associate professor in the Department of Applied Mathematics and Statistics, and postdoctoral fellow Wei Jin, the team’s approach aims to strike a balance between viral suppression and minimizing side effects, ultimately enhancing the quality of life for patients undergoing treatment.
By personalizing cART regimens based on individual patient characteristics, the researchers anticipate improved health outcomes, leading to an overall enhancement in well-being and quality of life. For instance, individuals suffering from depression could potentially see a substantial 22% enhancement in their depression scores by following the medication recommendations generated by the team’s model, surpassing the benefits of their original treatment plans.
Navigating the vast array of potential drug combinations posed a significant challenge for Xu’s team. To address this, they introduced a two-phase strategy that leverages patient data to tailor treatment plans. The first phase involved utilizing a Bayesian statistical method known as a multivariate Gaussian process (MGP) to monitor changes in patients’ health trajectories. Subsequently, this information was integrated into an offline reinforcement learning framework to determine the most effective sequence of treatments (cART regimens) based on the evolving health conditions of the patients.
The team put their method to the test using the Women’s Interagency HIV Study (WIHS) database, focusing on a group of 29 HIV-positive individuals experiencing severe depression. Implementing their approach resulted in a significant 22% improvement in depression scores for the participants, with 14 individuals reporting no signs of depression in the subsequent two years.
Wei Jin emphasized the broader implications of the study, highlighting the challenges posed by the limited number of HIV care providers, especially in a country like the US, where approximately one million individuals are living with HIV and almost 40,000 new cases are diagnosed each year. The scarcity of updated knowledge on HIV prevention, screening, and diagnosis among primary care clinicians, who handle a significant portion of HIV care, further underscores the need for innovative solutions.
The proposed method not only serves as a valuable tool to assist physicians in treatment decision-making but also has the potential to optimize therapy management for better patient outcomes. The researchers are now focused on developing a user-friendly online software that will recommend optimal cART regimens to healthcare providers using this approach.
Xu expressed optimism about the impact of this software on transforming the clinical management of HIV. In contrast to traditional treatment guidelines that primarily focus on viral suppression, their approach takes into account the potential comorbidities stemming from both HIV infection and its medications, with the aim of reducing additional health burdens for individuals living with HIV.
*Note:
1. Source: Coherent Market Insights, Public sources, Desk research
2. We have leveraged AI tools to mine information and compile it
