Phylogenetic Data Improves HIV Detection

Phylogenetic data collection has been described as one of the most important HIV prevention innovations in recent history. This was the topic of Wednesday’s Plenary Session, “Hugging Phylogenetic Trees: Use of Molecular Analysis for Public Health Intervention,” at the Conference on Retroviruses and Opportunistic Infections 2019.

Presenter Alexandra M. Oster, CAPT, from the US Public Health Service and Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, started the session by giving an overview and background of molecular cluster detection.

“Cluster detection and response can help bring the US closer to the end of the HIV epidemic,” Oster said. “If we can identify networks in which HIV is spreading quickly, we can help people get into care and prevent HIV.”

The networks Oster eluded to include people who are diagnosed with HIV, people who are undiagnosed with HIV, and people who are at risk for HIV. However, the challenges associated with identifying rapid HIV transmission include delayed diagnosis, population mobility, and patterns that can be difficult to detect in high-burden areas.

As a solution to these challenges, Oster said molecular outbreak detection has been a longstanding approach used for detecting foodborne and tuberculosis outbreaks. It is now being used in detecting HIV outbreaks in areas with active transmission.

Next, Oster described the key advances in molecular analysis from 2005 to 2015, outlining the key research findings. In 1997, the United States started using the US National HIV Surveillance System. Data is collected by providers and laboratories during clinical encounters with people with HIV and is reported to state and local health departments. The de-identified data is then sent to CDC for surveillance. State, local, and federal organizations then use this data to prevent infections, improve care, and reduce disparities.

Starting in 2013, this data was used more widely across the country, and providers and laboratories transitioned to electronic reporting, Oster said.

“In 2016, CDC established a routine approach to detecting clusters,” Oster said. “Because of the resources needed to respond to clusters, we focused on recent and rapid transmission. On average in the United States, 4 new HIV infections occur per 100 people living with HIV. We found that clusters identified using a very tight genetic distance threshold had a high rate of transmission.”

She continued to explain the Transmission Network Analysis and prospective cluster detection. In addition, she explained that molecular cluster detection is not the only method available for this work.

“Molecular cluster detection is not the only method, even in areas that do have molecular data. We also monitor for increased diagnoses in a geographic area, which we refer to as time-space analysis. This is particularly useful in areas that have small populations or low HIV burden and for detecting clusters among people who inject drugs,” Oster said.

For more information about this session, click here.

For more CROI 2019 coverage, click here.

—Amanda Balbi



Oster AM. hugging phylogenetic trees: use of molecular analysis for public health intervention. Session presented at: Conference on Retroviruses and Opportunistic Infections 2019; March 4-7, 2019; Seattle, WA. Accessed March 8, 2019.