HIV, HCV epidemics among PWID driven by large social networks
Click Here to Manage Email Alerts
“Explosive” HIV and hepatitis C epidemics among people who inject drugs in New Delhi are primarily driven by exposure to large, complicated networks of viremic individuals, according to findings from CROI.
Researchers found that HIV incidence was associated with the number of viremic contacts in study recruits’ egocentric — or immediate — injection network, as well as their sociometric network — the people their network injected with.
They said the findings emphasize the need to achieve far-reaching viral suppression in order to stop transmission of the virus.
“We have had ongoing cluster randomized trials among people who inject drugs (PWID) in multiple cities in India, including Delhi, since 2012. As part of this trial, we measured cross-sectional HIV incidence using data from large serial population-based [and] cross-sectional surveys conducted in 2013 and 2016-2017,” Sunil S. Solomon, MBBS, PhD, MPH, associate professor of medicine and epidemiology at Johns Hopkins University School of Medicine, told Healio. “These surveys demonstrated dramatic increases in HIV incidence among PWID in multiple settings and concomitant changes in drug use patterns (eg, heroin vs. pharmaceutical opioids). Delhi was one of these cities and, therefore, we decided to explore in-depth factors associated with HIV and HCV transmission among PWID in New Delhi.”
Solomon added that there has been also been a great deal of discussion about network-based treatment and prevention and the role of networks in HIV and HCV transmission, but most of these data were from high-income settings.
Solomon and colleagues enrolled a cohort of 2,512 PWID in New Delhi between 2017 and 2019 through a referral chain approach in which participants were asked to name and recruit people they recently injected with. The researchers used biometrics to find duplicates and cross-network linkages.
Surveys and blood draws were conducted semiannually. The blood was tested for HIV and HCV antibodies, as well as HIV RNA and HCV RNA. A Poisson regression model was then used to detect predictors of incident HIV.
According to Solomon, the results showed that there was an “alarmingly high incidence” of HIV and HCV in this population, with baseline HIV and HCV prevalences of 37% and 65%, respectively, in the cohort of 2,512 PWID. The study demonstrated 159 HIV seroconversions in 712 person-years of follow-up, for an incidence rate of 22.3 per 100 person-years. The primary HCV incidence was 25.3 per 100 person-years.
“When we think about a person’s risk, we usually think about their individual risk factors and sometimes about their immediate connections because of the shared risk of the people they are directly injecting with,” Solomon said. “In this study, we looked at the individual factors, their immediate network, their larger sociometric network and the space they were injecting in.”
According to Solomon, this approach demonstrated that although individual level factors, including frequency of injection and history of sharing, were predictive of incidence, the network factors — in particular, the number of people with detectable virus in the their immediate (egocentric) injection network and the network distance from someone who had detectable virus in their larger sociometric network — correlated with HIV incidence.
The strongest predictor, he added, was the space where people injected. The study showed that injecting at a hotspot correlated with a 3.1 times higher risk of seroconverting, even after considering individual and network risk factors.
“HIV incidence in this cohort is an order of magnitude higher than what we are seeing in populations across the globe. We frequently talk about ending the HIV epidemic or achieving epidemic control. This will not be possible without reaching populations such as PWID in this setting and other [low- and middle-income country] settings that currently account for some of the fastest growing HIV epidemics globally,” Solomon concluded. “Our findings indicate that programs will need to take into account the larger sociometric network, as well as space and time, to truly impact transmission.” – by Caitlyn Stulpin
Reference:
Clipman SJ, et al. Abstract 147. Presented at: Conference on Retroviruses and Opportunistic Infections; March 8-11, 2020; Boston.
Disclosures: Solomon reports receiving consulting fees from Gilead Sciences and pending research grants for his institution from Abbott Diagnostics and Gilead Sciences.