CDC routinely analyzes HIV sequence data to identify priority clusters exhibiting recent and rapid transmission. Prevention efforts for clusters can include improving viral suppression among persons with diagnosed infection and finding undiagnosed infections; the relative importance of these two aims is unknown. We retrospectively identified priority clusters and determined the extent to which future, incident infections in these clusters were associated with presence of cases that, at the time of cluster prioritization, were diagnosed and viremic versus undiagnosed.
Using HIV-1 pol sequences reported to the U.S. National HIV Surveillance System for 6 states with ≥50% sequence data completeness for diagnoses during 2010–2017, we used HIV-TRACE to identify priority clusters among cases diagnosed in 2010–2012 (i.e., pairwise distance ≤0.005 substitutions/site; ≥3 cases diagnosed in 2012). We then identified cases diagnosed through 2017 that were genetically linked to these clusters, representing cluster growth in the 5 years after prioritization. We performed Bayesian molecular clock phylogenetic inference in BEAST on each cluster to estimate the number of 2013–2017 diagnoses that were incident infections (i.e., internal nodes after 2012) and prevalent, undiagnosed infections (i.e., internal nodes in or before 2012). For cases diagnosed in or before 2012, we determined viremia (i.e., viral load ≥200 copies/ml at last lab in or before 2012). These counts were treated as predictors in a cluster-level multivariate logistic regression analysis with incident infection as the outcome.
Of 116 priority clusters (initial size: 3–33 persons), 76 gave rise to ≥1 incident infection after 2012 based on phylogenetic inference. Among priority clusters, both undiagnosed infections and diagnosed, viremic cases were independently and equally associated with incident cluster growth in the following 5 years: odds of cluster growth increased by 57% for each additional viremic person (adjusted odds ratio=1.57, p=0.010) and 51% for each person with undiagnosed infection (adjusted odds ratio=1.51, p=0.019).
These findings suggest that new infections in priority clusters originate equally from diagnosed, viremic cases and undiagnosed infections. These results highlight the importance of promoting retention in care and viral suppression as well as partner notification and other case-finding activities when investigating and intervening on priority molecular transmission clusters.