Receipt and timing of HIV drug resistance testing in six U.S. jurisdictions

A public health approach to monitoring HIV with resistance to HIV pre-exposure prophylaxis

Background

The risk of HIV pre-exposure prophylaxis (PrEP) failure with sufficient medication adherence is extremely low but has occurred due to transmission of a viral strain with mutations conferring resistance to PrEP components tenofovir (TDF) and emtricitabine (FTC). The extent to which such strains are circulating in the population is unknown.

Methods

We used HIV surveillance data to describe primary and overall TDF/FTC resistance and concurrent viremia among people living with HIV (PLWH). HIV genotypes conducted for clinical purposes are reported as part of HIV surveillance. We examined the prevalence of HIV strains with mutations conferring intermediate to high level resistance to TDF/FTC, defining primary resistance (predominantly K65R and M184I/V mutations) among sequences reported within 3 months of HIV diagnosis and total resistance for sequences reported at any time. We examined trends in primary resistance during 2010–2019 and total resistance among all PLWH in 2019. We also monitored resistance with viremia (≥1,000 copies/mL) at the end of 2019 among PLWH.

Results

Between 2010 and 2019, 2,172 King County residents were diagnosed with HIV; 1,557 (72%) had a genotypic resistance test within three months; three (0.2%) had primary TDF/FTC resistance with both K65R and M184I/V mutations. Adding isolated resistance for each drug resulted in 0.3% with primary TDF resistance and 0.8% with primary FTC resistance. Of 7,056 PLWH in 2019, 4,032 (57%) had genotype results, 241 (6%) had TDF/FTC resistance and 15 (0.4% of those with a genotype result) had viremia and TDF/FTC resistance.

Conclusions

Primary resistance and viremia combined with TDF/FTC resistance are uncommon in King County. Monitoring trends in TDF/FTC resistance coupled with interventions to help ensure PLWH achieve and maintain viral suppression may help ensure that PrEP failure remains rare.

Baseline HIV drug-resistance testing: 12 US jurisdictions, 2014-2019

OBJECTIVE

To understand recent patterns in reported baseline HIV drug-resistance testing over time in the United States.

DESIGN

Data from the National HIV Surveillance System for persons who were aged at least 13 years at the time of HIV diagnosis during 2014-2019 and resided in one of 12 US jurisdictions with high levels of reporting in 2014 and 2015.

METHODS

Among persons included in the analysis, we calculated the total proportion of HIV diagnoses occurring during 2014-2019 with a reported baseline sequence by year of diagnosis and sequence type. A baseline sequence was defined as any protease/ reverse transcriptase (PR/RT) or integrase sequence generated from a specimen collected 90 days or less after diagnosis.

RESULTS

During 2014-2019, reported levels of baseline PR/RT (with or without integrase) testing varied by year from 46.9% to 51.8% without any clear pattern over time. PR/RT with integrase testing increased (8.3-19.4%) and integrase-only testing remained low (1.9-1.3%).

CONCLUSION

While reported levels of baseline PR/RT (with or without integrase) testing have remained sufficiently high for the purposes of molecular cluster detection, higher levels would strengthen jurisdictions' and the Centers for Disease Control and Prevention's ability to monitor trends in HIV drug-resistance and detect and respond to HIV molecular clusters. Efforts to increase levels of reported baseline testing likely need to address both gaps in testing as well as reporting.

Characterization of HIV Risk Behaviors and Clusters Using HIV-Transmission Cluster Engine Among a Cohort of Persons Living with HIV in Washington, DC

Molecular epidemiology (ME) is one tool used to end the HIV epidemic in the United States. We combined clinical and behavioral data with HIV sequence data to identify any overlap in clusters generated from different sequence datasets; to characterize HIV transmission clusters; and to identify correlates of clustering among people living with HIV (PLWH) in Washington, District of Columbia (DC). First, Sanger sequences from DC Cohort participants, a longitudinal HIV study, were combined with next-generation sequences (NGS) from participants in a ME substudy to identify clusters. Next, demographic and self-reported behavioral data from ME substudy participants were used to identify risks of secondary transmission. Finally, we combined NGS from ME substudy participants with Sanger sequences in the DC Molecular HIV Surveillance database to identify clusters. Cluster analyses used HIV-Transmission Cluster Engine to identify linked pairs of sequences (defined as distance ≤1.5%). Twenty-eight clusters of ≥3 sequences (size range: 3-12) representing 108 (3%) participants were identified. None of the five largest clusters (size range: 5-12) included newly diagnosed PLWH. Thirty-four percent of ME substudy participants (n = 213) reported condomless sex during their last sexual encounter and 14% reported a Syphilis diagnosis in the past year. Seven transmission clusters (size range: 2-19) were identified in the final analysis, each containing at least one ME substudy participant. Substudy participants in clusters from the third analysis were present in clusters from the first analysis. Combining HIV sequence, clinical and behavioral data provided insights into HIV transmission that may not be identified using traditional epidemiological methods alone. Specifically, the sexual risk behaviors and STI diagnoses reported in the substudy survey may not have been disclosed during Partner Services activities and the survey data complemented clinical data to fully characterize transmission clusters. These findings can be used to enhance local efforts to interrupt transmission and avert new infections.

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

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.

Identifying Clusters of Recent and Rapid HIV Transmission Through Analysis of Molecular Surveillance Data

BACKGROUND

Detecting recent and rapid spread of HIV can help prioritize prevention and early treatment for those at highest risk of transmission. HIV genetic sequence data can identify transmission clusters, but previous approaches have not distinguished clusters of recent, rapid transmission. We assessed an analytic approach to identify such clusters in the United States.

METHODS

We analyzed 156,553 partial HIV-1 polymerase sequences reported to the National HIV Surveillance System and inferred transmission clusters using 2 genetic distance thresholds (0.5% and 1.5%) and 2 periods for diagnoses (all years and 2013-2015, ie, recent diagnoses). For rapidly growing clusters (with ≥5 diagnoses during 2015), molecular clock phylogenetic analysis estimated the time to most recent common ancestor for all divergence events within the cluster. Cluster transmission rates were estimated using these phylogenies.

RESULTS

A distance threshold of 1.5% identified 103 rapidly growing clusters using all diagnoses and 73 using recent diagnoses; at 0.5%, 15 clusters were identified using all diagnoses and 13 using recent diagnoses. Molecular clock analysis estimated that the 13 clusters identified at 0.5% using recent diagnoses had been diversifying for a median of 4.7 years, compared with 6.5-13.2 years using other approaches. The 13 clusters at 0.5% had a transmission rate of 33/100 person-years, compared with previous national estimates of 4/100 person-years.

CONCLUSIONS

Our approach identified clusters with transmission rates 8 times those of previous national estimates. This method can identify groups involved in rapid transmission and help programs effectively direct and prioritize limited public health resources.

Trends in HIV-1 Drug Resistance Mutations from a U.S. Reference Laboratory from 2006 to 2017

Trends in resistance to antiretroviral drugs for HIV-1 may inform clinical support and drug development. We evaluated drug resistance mutation (DRM) trends for nucleoside reverse transcriptase inhibitor (NRTI), non-nucleoside reverse transcriptase inhibitor (NNRTI), protease inhibitor (PI), and integrase strand transfer inhibitor (INSTI) in a large U.S. reference laboratory database. DRMs with a Stanford HIV Drug Resistance Database mutation score ≥10 from deidentified subtype B NRTI/NNRTI/PI specimens (2006-2017; >10,000/year) and INSTI specimens (2010-2017; >1,000/year) were evaluated. Sequences with NRTI, NNRTI, or PI single- or multiclass DRMs declined from 48.9% to 39.3%. High-level dual- and triple-class resistance declined from 43.3% (2006) to 17.1% (2017), while sequences with only single-class DRMs increased from 40.0% to 52.9%. The prevalence of DRMs associated with earlier treatment regimens declined, while prevalence of some DRMs associated with newer regimens increased. M184V/I decreased from 48.3% to 29.4%. K103N/S/T declined from 42.5% in 2012 to 36.4% in 2017. Rilpivirine and etravirine DRMs E138A/Q/R and E138K increased from 4.9% and 0.4% to 9.7% and 1.7%, respectively. Sequences with ≥1 darunavir DRM declined from 18.1% to 4.8% by 2017. INSTI DRM Q148H/K/R declined from 39.3% (2010) to 13.8% (2017). Prevalence of elvitegravir-associated DRMs T66A/I/K, E92Q, S147G, and the dolutegravir-associated DRM R263K increased. For a subset of patients with serial testing, 50% (2,646/5,290) of those who initially had no reportable DRM subsequently developed ≥1 DRM for NRTI/NNRTI/PI and 49.7% (159/320) for INSTI. These trends may inform the need for baseline genotypic resistance testing. The detection of treatment-emergent DRMs in serially tested patients confirms the value of genotypic testing following virologic failure.

Estimating Effects of HIV Sequencing Data Completeness on Transmission Network Patterns and Detection of Growing HIV Transmission Clusters

HIV nucleotide sequence data can identify clusters of persons with genetically similar strains suggesting transmission. We simulated the effect of lowered data completeness, defined by the percent of persons with diagnosed HIV with a reported sequence, on transmission patterns and detection of growing HIV transmission clusters. We analyzed HIV surveillance data for persons with HIV diagnosed during 2008-2014 who resided in Michigan or Washington. We calculated genetic distances, constructed the inferred transmission network for each jurisdiction, and compared transmission network characteristics and detection of growing transmission clusters in the full dataset with artificially reduced datasets. Simulating lower levels of completeness resulted in decreased percentages of persons linked to a cluster from high completeness (full dataset) to low completeness (5%) (Michigan: 54%-18%; Washington, 46%-16%). Patterns of transmission between certain populations remained robust as data completeness level was reduced. As data completeness was artificially decreased, sensitivity of cluster detection substantially diminished in both states. In Michigan, sensitivity decreased from 100% with the full dataset, to 62% at 50% completeness and 21% at 25% completeness. In Washington, sensitivity decreased from 100% with the full dataset, to 71% at 50% completeness and 29% at 25% completeness. Lower sequence data completeness limits the ability to detect clusters that may benefit from investigation; however, inferences can be made about transmission patterns even with low data completeness, given sufficient numbers. Data completeness should be prioritized, as lack of or delays in detection of transmission clusters could result in additional infections.

Transmission Patterns in a Low HIV-Morbidity State - Wisconsin, 2014-2017

Public health interviews (i.e., partner services), during which persons with diagnosed human immunodeficiency virus (HIV) infection name their sexual or needle-sharing partners (named partners), are used to identify HIV transmission networks to guide and prioritize HIV prevention activities. HIV sequence data, generated from provider-ordered drug resistance testing, can be used to understand characteristics of molecular clusters, a group of sequences for which each sequence is highly similar (linked) to all other sequences, and assess whether named partners are plausible HIV transmission partners. Although molecular data in higher HIV-morbidity states have been analyzed (1-3), few analyses exist for lower morbidity states (4), such as Wisconsin, which reported 4.6 HIV diagnoses per 100,000 persons aged ≥13 years in 2016 (5). The Wisconsin Division of Public Health (DPH) analyzed HIV sequence data generated from provider-ordered drug resistance testing and collected through routine HIV surveillance to identify molecular clusters and describe demographic and transmission risk characteristics among pairs of persons whose sequences were highly genetically similar (i.e., molecular linkages). In addition, overlap between partner linkages identified during public health interviews and molecular linkages was assessed. Overall, characteristics of molecular clusters in Wisconsin mirrored those from states with more HIV diagnoses, particularly in that most molecular linkages were observed among persons of the same race (78.2% of non-Hispanic blacks [blacks] linked to other blacks), the same transmission risk (90.2% of men who have sex with men [MSM] linked to other MSM), and the same age group (59.2% of persons aged 20-29 years linked to other persons aged 20-29 years). Among named partner linkages identified during interviews in which both persons also had a reported sequence, overlap of named partner and molecular linkages was moderate: 33.8% of named partners were plausible transmission partners according to available molecular data. Analysis of HIV sequence data is a useful tool for characterizing transmission patterns not immediately apparent using traditional public health interview data, even in a state with lower HIV morbidity. Prevention recommendations generated from national data (e.g., targeting preexposure prophylaxis for HIV-negative persons at high risk and implementing measures to maintain viral suppression among persons with HIV infection) also are relevant in a lower HIV-morbidity state.

Completeness of HIV nucleotide sequence ascertainment and its potential impact on understanding HIV transmission - Maryland, 2011-2013

HIV nucleotide sequences generated through routine drug resistance testing (DRT) and reported to Maryland's Molecular HIV Surveillance system are most effective for elucidating transmission patterns and identifying outbreaks if DRT is ordered promptly and sequences are reported completely. Among reported cases of HIV infection newly diagnosed during 2011-2013 in Maryland residents aged ≥13 years, we assessed sequence ascertainment completeness. To better understand which populations were most likely to have a sequence, we examined associations between sequence ascertainment and clinical and demographic characteristics. During 2011-2013, 4423 new HIV infection diagnoses were reported; sequences were ascertained for 1282 (29.0%). Among 3267 cases with complete data, odds for having a sequence ascertained were highest for cases in persons living inside Maryland's Central Region with initial CD4 counts ≤500 cells/mm³ (adjusted odds ratio [aOR] 2.4, 95% confidence interval [CI] 1.9-3.1). Sequence ascertainment did not vary significantly by patient age, sex, race/ethnicity or HIV transmission category. Educational interventions, policy changes and improved processes to increase timely DRT and subsequent sequence reporting with a focus on testing at entry to care, particularly for those with higher CD4 counts and those living outside the Central Region, might improve ascertainment completeness.