Characterization of Molecular Cluster Detection and Evaluation of Cluster Investigation Criteria Using Machine Learning Methods and Statewide Surveillance Data in Washington State

Abstract

Molecular cluster detection can be used to interrupt HIV transmission but is dependent on identifying clusters where transmission is likely. We characterized molecular cluster detection in Washington State, evaluated the current cluster investigation criteria, and developed a criterion using machine learning. The population living with HIV (PLWH) in Washington State, those with an analyzable genotype sequences, and those in clusters were described across demographic characteristics from 2015 to2018. The relationship between 3- and 12-month cluster growth and demographic, clinical, and temporal predictors were described, and a random forest model was fit using data from 2016 to 2017. The ability of this model to identify clusters with future transmission was compared to Centers for Disease Control and Prevention (CDC) and the Washington state criteria in 2018. The population with a genotype was similar to all PLWH, but people in a cluster were disproportionately white, male, and men who have sex with men. The clusters selected for investigation by the random forest model grew on average 2.3 cases (95% CI 1.1–1.4) in 3 months, which was not significantly larger than the CDC criteria (2.0 cases, 95% CI 0.5–3.4). Disparities in the cases analyzed suggest that molecular cluster detection may not benefit all populations. Jurisdictions should use auxiliary data sources for prediction or continue using established investigation criteria.

Increasing Capacity to Detect Clusters of Rapid HIV Transmission in Varied Populations—United States

“…First, our priority cluster definition for this analysis used a threshold of five diagnoses in the preceding 12 months. However, clusters with as few as three diagnoses in the preceding 12 months have equally high growth and transmission rates [ 3 , 16 ]. The CDC currently applies the lower threshold to areas with lower numbers of HIV diagnoses, and some health departments in higher burden areas that have capacity to respond earlier also use lower thresholds.…”

The Role of Phylogenetics in Unravelling Patterns of HIV Transmission towards Epidemic Control: The Quebec Experience (2002–2020)

“…Our findings show “short-lived” small cluster lineages epidemiologically resemble solitary transmissions while being distinct from “super-spreader” larger cluster variants. Stable large cluster lineages have been recently shown to fuel the Rhode Island, North Carolina, and New York statewide epidemics and may contribute to more transmissible drug-resistant epidemics [ 16 , 27 , 33 , 41 ].…”

“…This shorter genetic distance is estimated to evolve over approximately 2–3 years [ 44 ]. Our findings found that this lower cut-off may not always assist in forecasting cluster growth [ 26 , 27 , 33 ]. Analyses at genetic distances of 1.5% and 2.5% thresholds assisted in distinguishing patterns of spread of large cluster networks implicated in the Quebec epidemic.…”

“…Clustering patterns were assessed at genetic distance thresholds of 1.5% and 2.5%, using an ambiguity of 0.15 to resolve ambiguous nucleotides [ 19 , 20 , 25 , 26 ]. The genetic distance threshold of 1.5% is commonly used in research studies [ 25 , 27 ]. The longer 2.5% genetic distance threshold was included to describe the growth trajectories of several long-lived large cluster variants [ 25 , 26 ].…”

MicrobeTrace: Retooling Molecular Epidemiology for Rapid Public Health Response

“…A more stringent TN93 threshold of 0.5% is often used to identify the most recent and rapid clusters of HIV-1 transmission ( Fig 2B). Threshold determinations are often informed by cluster size and growth rate criteria (Erly, et al, 2020;France and Oster, 2020;Oster, et al, 2018). MicrobeTrace offers the ability to filter by genetic distance and cluster size thresholds in the same 'Global Settings' menu.…”