Simulation of Full HIV Cluster Networks in a Nationally Representative Model Indicates Intervention Opportunities

Area of Science:

• Epidemiology
• Virology
• Public Health

Background:

• Molecular sequence data analysis identifies rapid HIV transmission clusters in the US.
• Understanding the full scope of these networks is crucial for effective intervention strategies.
• Undiagnosed infections represent a significant challenge in observing complete HIV transmission networks.

Purpose of the Study:

• To model and assess the full extent, composition, and transmission rates of HIV molecular cluster networks.
• To differentiate between ongoing and newly emerging priority clusters of rapid HIV transmission.
• To quantify the proportion of undiagnosed infections within these priority clusters.

Main Methods:

• Utilized a stochastic dynamic network simulation model for HIV sexual transmission in the US (2015-2017).
• Defined clusters based on a 0.5% genetic distance threshold.
• Identified ongoing priority clusters (≥3 diagnoses/year for multiple years) and new priority clusters (≥3 diagnoses/year in 2017).

Main Results:

• Full HIV clusters were 3-9 times larger than detected clusters.
• Median detected cluster sizes were 4 (new) and 11 (ongoing), with full median sizes of 14 and 94, respectively.
• A median of 36.3% of infections in new priority clusters were undiagnosed, with transmission rates over 4 times the overall simulation rate.

Conclusions:

• Priority clusters represent networks characterized by rapid HIV transmission.
• The significantly larger size of full clusters, high rates of undiagnosed infections, and elevated transmission rates present critical public health intervention opportunities.
• These findings underscore the importance of comprehensive surveillance and targeted interventions to control HIV spread.