Effect on CD4 T-cell count of replacing protease inhibitors in patients with successful HIV suppression

Protease inhibitor (PI) therapy may be associated with adverse events. Switching the PI component to a non-PI may reduce some side-effects, but the immune effects are unclear. In a meta-analysis using strict inclusion criteria, a randomized switch from PI was associated with impaired CD4 T-cell gains compared with maintaining PI (PI arm +32 cells, effect size 0.130,P= 0.05). Therefore, PI-based therapy results in superior CD4 T-cell gains compared with switching to a non-PI regimen. Combination antiretroviral therapies promote immune recovery and viral suppression in HIV-infected patients [1]. Unfortunately, the side-effects of protease inhibitors (PI) often prompt the replacement of the PI with other drug classes. Studies have now demonstrated that switching the PI component to a non-nucleoside reverse transcriptase inhibitor (NNRTI) or to abacavir maintains viral suppression [2–7]; however, the immunological effects of such switches have not been fully evaluated. An emerging body of literature suggests that PI therapy may have superior immunological outcomes compared with therapies that do not include a PI [8]. To evaluate whether such a difference occurs, we performed a meta-analysis of prospective studies that replaced the PI component of virologically suppressive therapy with either an NNRTI or abacavir and analysed the CD4 T-cell outcome of such a switch. A MEDLINE search was conducted from 1996 to 2002 using the MeSH terms: HIV-1, anti-HIV agents, HIV protease inhibitors, reverse transcriptase inhibitors, nevirapine, switch.mp and replace.mp. The inclusion criteria were: (i) randomized, controlled clinical trials; (ii) HIV-1-infected adults treated with two nucleoside reverse transcriptase inhibitors and one PI with suppressed viral replication for greater than 6 months before the switch; (iii) random assignment to either the continuation of PI therapy, or switch to an NNRTI or abacavir while maintaining the two nucleoside reverse transcriptase inhibitors; and (iv) CD4 cell counts recorded at baseline (i.e. before the switch) and at the conclusion of the study. We evaluated the mean increase in the CD4 lymphocyte count using the standard deviation (SD) associated measure of variation. In the cases in which only the medians of the CD4 cell increases were reported, these were used as estimates for the mean increases. Estimates of SD in other cases were imputed based on studies in which the SD or standard error (SE) estimates were available. The effect size, which reflects the standardized mean difference between the two groups, was calculated as follows: Effect size together with an estimate of SD was used to calculate the t-statistic using a two-sided P-value and a 95% confidence interval. Thirty-eight potential studies were identified, of which seven met all the inclusion criteria. Of the seven eligible studies, five studies, which included data on 757 patients (80.5% of the total patients) demonstrated a superiority of PI continuation on CD4 cell counts (Fig. 1). In contrast, two studies with 183 patients (19.4% of the total patients) demonstrated superior CD4 T-cell gains by switching to a NNRTI (nevirapine or efavirenz) as opposed to continuing PI therapy. We questioned whether differences in CD4 T-cell outcomes were governed by different lengths of follow-up, but linear regression analysis of CD4 T-cell change versus the length of follow-up did not reveal co-association (R2 = 0.0051, P = NS).Fig. 1.: CD4 T-cell outcomes in patients randomly assigned to continue protease inhibitor-based therapy (solid bars) or to switch to non-protease inhibitor-based therapy (open bars), stratified by study. ABC, Abacavir; EFV, efavirenz; NVP, nevirapine. ▒ Randomly assigned to continue therapy; □ switch to non-protease inhibitor-based therapy.We therefore performed a meta-analysis of PI switch using all seven randomized studies. The calculation of weighted average effect size was performed for all seven trials. The mean gain of CD4 T cells in the patients who were maintained on PI was 66 ± 11 (mean ± SE) cells/ml, whereas the increase was 34 ± 7 cells/ml in the patients switched to NNRTI or abacavir. Pooled results from all seven trials revealed that switching from a PI-based therapy to a non-PI-based therapy (involving either NNRTI or abacavir) in patients who have achieved successful viral suppression on multi-drug therapy does result in diminished CD4 cell increases. The effect size (0.130) was statistically significant [P = 0.05, 95% confidence interval (CI) −0.002 to 0.261] with a mean difference of 32 cells in the PI maintenance arm. The success of the PI class of antiretroviral agents has transformed HIV infection into a chronic but manageable disease for many patients [1]. Unfortunately, despite their success, the utility of PI-based regimens is often limited by a high daily pill burden, dietary restrictions and treatment-associated adverse events, often prompting a switch to more tolerable agents [9]. However, the immunological effect of such changes is unknown. Our data demonstrate that discontinuing a PI in favor of another drug is associated with impaired CD4 T-cell recovery. The comparative gain of 66 ± 11 CD4 T cells/ml in the PI-treated patients compared with 34 ± 7 cells/ml in the patients switched to alternative therapy must be viewed within the appropriate context. First, this difference was apparent after as little as 4 months of follow-up. It is important that studies with longer follow-ups be reported in order to determine whether such differences will continue to expand, or stabilize. Second, it is unclear whether these changes vary inversely with baseline CD4 T-cell counts. Although others have suggested that this effect may be greatest in patients with lower baseline CD4 T-cell counts [9], our study did not permit such subgroup analyses. Third, even small increases in the CD4 T-cell number enhance clinical outcomes [10], even when immunological improvement is not accompanied by virological response [10]. Our results, combined with th in-vitro data and non-randomized clinical studies, support the suggestion of a superior effect of PI on the CD4 T-cell number, and indicate the need for further research to define how this information may be used to the benefit of patients infected with HIV. Although this difference may not be of clinical significance for patients with high baseline CD4 cell counts, it is possible that patients with severe immune depletion would experience clinical benefit from the additional CD4 T-cell increase associated with PI. Prospective analyses are required to assess the clinical benefit of PI versus non-PI-based therapy.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.