Technology-driven HIV management optimizes treatment and prevention, according to data from the PHOENIX trial. Future studies should focus on long-term AI, genomics, stigma, clade surveillance, comorbidities, implementation, and cost-effectiveness, according to the PHOENIX trial investigators.
HIV transmission in the Middle East and North Africa is rising due to delayed diagnoses, stigma, and limited access to pre-exposure prophylaxis (PrEP) and antiretroviral therapy (ART). Standard strategies are falling behind due to shifting transmission patterns influenced by migration, urbanization, and conflict. Precision-guided, region-specific interventions are urgently needed to reverse this trend and meet global control targets.
The 5-year PHOENIX Trial of 22,000 participants in Egypt, Saudi Arabia, Jordan, Lebanon, Iraq, and Morocco included these cohorts: new HIV (n=6,000), PrEP (n=4,000), ART-naïve (n=5,000), and controls (n=7,000)
The investigators used quarterly reviews and digital tracking with biospecimens (Whole Genome Sequencing (WGS), transcriptomics, and methylation analysis). A nested randomized controlled trial (RCT) with 3,600 participants compared cabotegravir/rilpivirine with tenofovir for suppression and resistance outcomes. The researchers used AI mHealth tools, geo-tagged data, and neural networks to predict outbreaks and analyze suppression, resistance, and PrEP failure. They found that:
• Long-acting cabotegravir/rilpivirine achieved 96.4% virologic suppression in ART-naïve individuals at week 48, surpassing tenofovir (88.1%; P<0.001), with a 58% resistance mutation reduction (P=0.002). This finding supports shifting first-line therapy in eligible populations to long-acting injectable options for better long-term outcomes.
• A 17-gene host signature predicted viral control (area under the curve, 0.93; validated n=620). This finding opens a pathway for precision-guided therapy stratification and personalized treatments.
• 58% reduction in resistance mutations with long-acting regimen, indicating reduced need for costly genotypic testing and complex salvage regimens, improving access to care and cost-effectiveness.
• PrEP adherence improved by 32.5% with digital behavioral tools, validating scalable tech-based adherence supports for real-world application, especially among younger people.
• 46.2% reduction in new seroconversions among digitally supported PrEP users, making a strong case for integrating behavioral health platforms in national PrEP rollouts.
• Neural net outbreak forecasting had 89.7% sensitivity, enabling early deployment of prevention resources in high-risk zones before transmission intensifies.
• Reduced expansion rates in Lebanon and Iraq by 24.1% through predictive response, showing the tangible benefit of precision public health in humanitarian and mobile settings.
• Stigma-associated nonadherence is most marked in males 18–24 (OR 3.7, justifying targeted outreach programs and culturally sensitive messaging for key demographic groups.
• Discovery of three novel recombinant regional clades, justifying targeted outreach programs and culturally sensitive messaging for key demographic segments.
Reference Tanaka H, et al. IAS 2025. Poster exhibition LB29.
