BACKGROUND: We aimed to compare the non-AIDS event (nADE) risk between normal progressors using antiretroviral therapy (NP-ART) and people with human immunodeficiency virus (HIV, PWH) who naturally control HIV infection (HIV controllers), as well as the risk of nADE following ART in HIV controllers. METHODS: The primary end point was the composite of cardiovascular disease, non-AIDS malignancy, or all-cause mortality, whichever came first. The role of ART in HIV controllers was assessed as a time-varying covariate. RESULTS: We included 1007 ART-naive HIV controllers (60 of them were elite controllers), 1510 early-ART (<
6 months after negative HIV test), and 15437 NP-ART (reference group), contributing 3813, 11 060, and 160 050 years of follow-up, respectively. HIV controllers had lower risk of the primary end point (hazard ratio [HR], 0.55
95% confidence interval [CI]: .38-.81
P = .0023), all-cause mortality (adjusted HR [aHR], 0.45
95% CI: .25-.79
P = .0054), and cardiovascular disease (aHR, 0.47
95% CI: .22-.99
P = .046), but not non-AIDS malignancy (aHR, 0.74
95% CI: .41-1.35
P = .33), compared with NP-ART. Among HIV controllers, each log10 lower baseline viral load further decreased the risk of a nADE (aHR, 0.54
95% CI: .29-.99
P = .045). ART in HIV controllers did not reduce the risk of any nADE (aHR, 1.22
95% CI: .66-2.29
P = .53). CONCLUSIONS: HIV controllers had a lower n ADE risk than NP-ART, especially in those with low plasma viral loads. ART did not alter the nADE risk in HIV controllers. Our findings help clinicians to decide on prescribing ART in HIV controllers.