Background: The CD4/CD8 ratio has been associated with the risk of AIDS and non-AIDS events. We describe trends in immunological parameters in people who underwent a switch to monotherapy or dual therapy, compared to a control group remaining on triple antiretroviral therapy (ART). Methods: We included patients in Icona who started a three-drug combination ART regimen from an ART-naïve status and achieved a viral load ≤ 50 copies/mL; they were subsequently switched to another triple or to a mono or double regimen. Standard linear regression at fixed points in time (12-24 months after the switch) and linear mixed model analysis with random intercepts and slopes were used to compare CD4 and CD8 counts and their ratio over time according to regimen types (triple vs. dual and vs. mono). Results: A total of 1241 patients were included; 1073 switched to triple regimens, 104 to dual (72 with 1 nucleoside reverse transcriptase inhibitor (NRTI), 32 NRTI-sparing), and 64 to monotherapy. At 12 months after the switch, for the multivariable linear regression the mean change in the log10CD4/CD8 ratio for patients on dual therapy was -0.03 (95% confidence interval (CI) -0.05, -0.0002), and the mean change in CD8 count was +99 (95% CI +12.1, +186.3), taking those on triple therapy as reference. In contrast, there was no evidence for a difference in CD4 count change. When using all counts, there was evidence for a significant difference in the slope of the ratio and CD8 count between people who were switched to triple (points/year change ratio = +0.056, CD8 = -25.7) and those to dual regimen (ratio = -0.029, CD8 = +110.4). Conclusions: We found an increase in CD8 lymphocytes in people who were switched to dual regimens compared to those who were switched to triple. Patients on monotherapy did not show significant differences. The long-term implications of this difference should be ascertained. © 2018 The Author(s).