Since 2012 treatment guidelines in the Netherlands recommend starting combination antiretroviral treatment (cART) immediately after HIV diagnosis, irrespective of CD4 cell counts. At the same time, cART effectiveness and tolerability improved considerably. We investigated to what extent the observed decrease in HIV diagnoses among men who have sex with men (MSM), from 766 in 2011 to 558 in 2015, could have been the result of earlier diagnosis and treatment.
We used an existing mathematical model describing HIV transmission, disease progression, and the effects of cART to study changes in the HIV epidemic among MSM. Treatment-related parameters were derived from data in the national ATHENA cohort, including time to viral suppression and rates of viral rebound. Viral suppression was defined as HIV RNA <1000 copies/ml; below this threshold HIV transmission is unlikely. We estimated changes over calendar time in duration from HIV infection to diagnosis, per-capita transmission rate (a proxy for risk behaviour), and annual number of newly acquired infections needed to explain annual data on HIV and AIDS diagnoses up to 2015. In a hypothetical scenario, we assumed treatment guidelines did not change in 2012.
Median time to viral suppression in MSM diagnosed with asymptomatic HIV and CD4 counts ≥200 cells/mm³ decreased from 0.97 (interquartile range, 0.29-3.30) years in 2000 to 0.76 (0.24-1.99) years in 2011 and 0.16 (0.10-0.25) years in 2015, and was shorter in MSM with CD4 <200 cells/mm³, symptomatic HIV, or AIDS. During the same time, the proportion with viral rebound decreased from 55% to 11%. Our model estimated that HIV-positive individuals were diagnosed earlier: average time from infection to diagnosis decreased from 2.86 (95% confidence interval [CI], 2.72-3.02) years in 2000-2003 to 2.07 (1.93-2.19) years in 2012-2015, while the transmission rate increased by 20% (9-36). Annual number of newly acquired HIV infections decreased from a peak of 710 (95% CI, 650-780) in 2007 to 590 (540-650) in 2011 and 350 (320-400) in 2015. In 2012-2015, there were 1864 (1722-2074) estimated new HIV infections, which is 832 less than in a counterfactual scenario with no earlier treatment after 2011.
Our model suggests that immediate treatment, in combination with earlier diagnosis and less viral rebound, contributed to a substantial decrease in the annual number of new HIV infections in MSM in the Netherlands despite an increase in risk behaviour.