Different degree of immune recovery using antiretroviral regimens with
protease inhibitors or non-nucleosides
The authors findings suggest that for patients with CD4s <300 a PI regimen
may provide better CD4 increases and immune recovery than an NNRTI regimen.
thPablo Barreiroa; Vincent Sorianoa; Esperanza Casasb; Juan Gonzalez-Lahoza
From the a Service of Infectious Diseases, Hospital Carlos III, Instituto de
Salud Carlos III, and the b Hospital Principe de Asturias, Madrid, Spain.
Background: Protease inhibitors (PI) produce significant immune recovery in
most HIV-infected persons, although toxicity and high pill burden often limit
this benefit. Combinations including non-nucleoside reverse transcriptase
inhibitors (NNRTI) result in similar virological success, but data on immune
reconstitution are scarce.
Methods: Baseline plasma viraemia and CD4 cell counts were recorded from 100
patients who began two nucleoside analogue reverse transcriptase inhibitors
plus either one PI or one NNRTI in a case-control study [indinavir (82%) and
nevirapine (80%), respectively, were most frequently prescribed]. Only
patients with baseline CD4 cell counts < 500, good treatment adherence and
plasma HIV RNA < 50 copies/ml sustained for 1 year were recruited.
Results: A rapid CD4 cell gain occurred within 12 weeks on therapy (average
41.8 ? 106 cells/l per month), irrespective of treatment. In contrast, a
trend towards a better CD4 cell gain was noticed between 12 and 48 weeks with
a PI (mean CD4 cell increases per month: 15.2 ? 106 cells/l using PI and 10.4
using NNRTI). During this period, the difference between therapy with a PI
and a NNRTI reached statistical significance for subjects with baseline CD4
counts < 300 , respectively; P < 0.05).
Conclusion: A rapid CD4 cell increase occurred shortly after beginning
antiretroviral therapy using either PI or NNRTI. Late increases in CD4 cell
counts, mostly owing to newly produced cells rather than redistribution, are
more pronounced in therapy using a PI, especially in subjects with lower
initial CD4 cell counts.
A total of 985 HIV-infected patients who began HAART containing either one PI
or one NNRTI during the study period were retrospectively identified. A total
of 245 fulfilled the virological, immunological and treatment adherence
criteria detailed above. Two groups of 50 patients, taking either PI or
NNRTI, were matched by baseline plasma viraemia and CD4 cell counts. Their
respective immunological outcome was examined during the study period.
Demographic, virological and immunological features at baseline were
comparable between the groups. However, the percentage of patients with prior
NRTI experience was significantly higher among subjects in the PI arm than in
those belonging to the NNRTI group (56% and 24%, respectively; P < 0.01).
Indinavir and nevirapine were the drugs most frequently prescribed as PI and
NNRTI in each group (82% and 80%, respectively). The dual NRTI backbone most
frequently prescribed was stavudine plus lamivudine (47%), followed by
stavudine plus didanosine (34%), and zidovudine plus lamivudine (12%). The
use of single agents in each treatment arm was similar, with the exception of
didanosine and lamivudine, which were slightly more frequently prescribed in
subjects taking NNRTI and PI, respectively.
CD4 cells increased significantly in both treatment groups, reaching at 48
weeks mean values of 516 (± 202) and 463 (± 278) in subjects taking a PI and
a NNRTI, respectively. Although those taking a PI experienced a better
immunological outcome during the entire follow-up period, the difference from
the NNRTI group did not reach statistical significance.
As expected, CD4 cell increases were more pronounced during the first 12
weeks of treatment, with a mean increase of 41.8 (± 28.6) ? 106 cells/l per
month irrespectively of the regimen being prescribed. Interestingly, a trend
towards higher increases in CD4 cells between weeks 12 and 48 was noticed
among subjects receiving PI in comparison with those treated with NNRTI. On
average, during this second phase, the CD4 lymphocyte count increased per
month at 15.2 (± 9.8) and 10.4 (± 7.4) ? 106 cells/l in subjects treated with
a PI or a NNRTI, respectively.
In a subgroup of 48 patients who began triple drug therapy when they had a
CD4 cell count of < 300, the mean CD4 lymphocyte values rose from 156 (± 86)
to 433 (± 172) at 48 weeks in the PI group, but just from 145 (± 88) to 300
(± 169) in the NNRTI group (P < 0.05). During the first 12 weeks of
treatment, the mean increase in CD4 cells was 32.3 (± 28.7) and 34.8 (± 23.4)
per month in the PI and NNRTI groups, respectively. Between weeks 12 and 48,
the CD4 lymphocytes rose at a rate of 17.1 (± 14.6) per month in subjects
taking a PI and by 6.4 (± 8.5) ? 106 cells/l per month in those taking a
NNRTI (P <0.05).
The greater immunological benefit provided by PI drugs might be related to
some secondary properties of these compounds. Recent reports have underlined
that PI may inhibit the apoptosis of CD4 T cells [20,24]. It has been
postulated that they might contribute by this mechanism to the CD4 gain
observed in subjects exposed to PI-containing HAART. Furthermore, the
inhibition of apoptotic phenomena driven by PI could be more relevant among
lower CD4 cell counts. Programmed cell death is associated with cell
activation markers [6-8], which are not significantly affected despite
effective virus suppression. It is well known that there is an inverse
relationship between the number of CD4 cells and activation molecules .
Therefore, the direct anti-apoptotic activity of PI drugs might have a higher
impact in subjects with higher immune activation, as generally these patients
have more severe immune depletion.
To our knowledge, this is one of the first studies demonstrating a
therapeutic superiority of PI over NNRTI in the clinical setting. Previous
reports have suggested that more potent virological suppression could be
reached using PI-based regimens, based on data obtained from lymph nodes 
or using ultrasensitive viral load tests  in subjects having undetectable
viraemia in plasma. Our study suggests that the strength of the immunological
response to antiretroviral therapy might be greater using PI- than using
NNRTI-containing regimens, independent of the virological effect; this may be
of particular relevance in subjects with advanced HIV immunosuppression. This
information may help to design the best therapeutic option for individual
patients, although other factors would also need to be considered, such as
the higher pill burden and the greater risk of hyperlipidaemia with the use
of a PI.
In conclusion, although both PI- and NNRTI-containing HAART provide an strong
immunological benefit in patients experiencing complete and sustained
suppression of virus replication, subjects with lower CD4 cell counts might
experience a more pronounced immune recovery when treated with a PI.