|
Protease inhibitors and NNRTIs have a comparable effect on the CD4 cell change after switching to tenofovir-based regimens
|
|
|
CORRESPONDENCE
AIDS: Volume 19(15) 14 October 2005 p 1722-1723
van Leth, Franka,b; Prins, Jan Ma; Lange, Joep MAa,b; Geerlings, Suzanne Ea
aDepartment of Internal Medicine, Academic Medical Center, University of Amsterdam, the Netherlands
bInternational Antiviral Therapy Evaluation Center, Amsterdam, the Netherlands.
Barrios et al. [1] described a CD4 cell decline in patients who change their antiretroviral regimen to one containing tenofovir and didanosine, while having an adequate viral suppression. In their analyses, the authors exclude patients who, besides these two drugs, used a protease inhibitor (PI). This was because of the reported benefit that PI-containing regimens might have on CD4 cell gains compared with regimens containing a non-nucleoside reverse transcriptase inhibitor (NNRTI) [2,3].
To assess whether such a selection of patients is warranted, we examined whether the use of PI or NNRTI influences the immunological efficacy of regimens containing tenofovir with or without didanosine. We assessed the CD4 cell changes up to one year after the start of a tenofovir-containing regimen in patients who had an adequate viral suppression of at least 6 months. For this, we used a generalized linear model that incorporates repeated measurements. All analyses were adjusted for differences in the number of CD4 cells at the start of the tenofovir-containing regimen. Only CD4 cell estimates were included while the patient still used the tenofovir-containing regimen and had an HIV-1-RNA concentration below 50 copies/ml.
In total, 190 patients in our hospital who had a plasma viral load of less than 50 copies/ml for at least 6 months changed to a regimen including tenofovir. Of these, 34 patients were not included in the analyses because the new regimen had less than three drugs, they used a PI and an NNRTI concurrently, or they did not have a CD4 cell measurement within one year of follow-up after the start of the new regimen. Of the remaining 156 patients, 136 used tenofovir without didanosine (TDF/ddI-), of which 29 were with a PI and 107 were with an NNRTI, and 20 patients used tenofovir with didanosine (TDF/ddI+), of which seven were with a PI and 13 were with an NNRTI.
The decrease in CD4 cells in the TDF/ddI+ group as reported by Barrios et al. [1] was confirmed in the present analyses. These patients had a decrease of 71 cells/mm3 after one year of follow-up while having a plasma viral load of less than 50 copies/ml, compared with an increase of 61 cells/mm3 for patients using TDF/ddI-. The difference of 132 cells/mm3 was statistically significant (P = 0.026).
In a multivariable model we included, apart from the baseline CD4 cell value and a variable for the type of tenofovir regimen (ddI+ versus ddI-), a variable denoting PI or NNRTI use. In this model, the difference in CD4 cell increase comparing patients using a PI with those using an NNRTI was 13 cells/mm3 in favour of the PI group (P = 0.558). This small difference did not modify the estimates of the mean increase in CD4 cells in the TDF/ddI+ and the TDF/ddI- groups. The difference between these two groups after adjusting for the baseline level and PI use (130 cells/mm3) remained statistically significant (P = 0.028).
Looking at the absolute number of CD4 cells over time after the start of the tenofovir-containing regimen, there was a decline in the TDF/ddI+ group in the second half of the follow-up period. This is in accordance with the findings of Barrios et al. [1]. The number of CD4 cells remained stable in the TDF/ddI- group (Table 1).
![tableAbso-1.gif](../images/092905/tableAbso-1.gif)
The present analyses were based on a small number of patients, especially in the TDF/ddI+ group. However, the P value for the effect of PI use on CD4 cell change in the multivariable analysis was far greater than the conventional 0.05 for statistical significance. Furthermore, the estimates for the CD4 cell changes in the TDF/ddI+ and TDF/ddI- groups are robust after adjusting for the use of PI or NNRTI. This indicates that the use of PI or NNRTI does not have a clear impact on the immunological efficacy of a tenofovir-containing regimen, but that the concurrent use of didanosine does.
The results cannot be influenced by potential differences in antiretroviral efficacy between the TDF/ddI+ and the TDF/ddI- group, because only CD4 cell data were analysed while the patient had an adequate viral suppression. Therefore, one has to assume that an unfavourable immunological effect can be anticipated with a regimen containing both tenofovir and didanosine, even with a plasma viral load below 50 copies/ml. Such an unfavourable effect cannot be overcome by adding a PI to the regimen.
It is important that these findings are confirmed by larger studies, in order to guide physicians in the use of optimal simplification regimens.
|
|
|
|
|
|
|