Abstract

Virologic and immunologic responses were examined for 33 human immunodeficiency virus (HIV)–infected children who participated for ⩾96 weeks in a phase 1/2 protocol of 16 weeks of indinavir monotherapy, followed by the addition of zidovudine and lamivudine. At week 96, a median increase of 199 CD4+ T cells/μL and a median decrease of 0.74 log10 HIV RNA copies/mL were observed. The relationship between control of viral replication and CD4+ T cell count was examined. Patients were categorized into 3 response groups on the basis of duration and extent of control of viral replication. Of 21 children with a transient decrease in virus load of ⩾0.7 log10 HIV RNA copies/mL from baseline, 7 experienced sustained increases in CD4+, CD4+CD45RA+, and CD4+CD45RO+ T cell counts. CD4+CD45RA+ (naive) T cells were the major contributor to CD4+ T cell expansion. Continued long-term immunologic benefit may be experienced by a subset of children, despite only transient virologic suppression

The course of human immunodeficiency virus (HIV) infection in children and their response to therapy are often different from those of adults. In general, thymic function wanes with age, and HIV-infected children may have a relatively greater potential for immunologic reconstitution [1]. However, young HIV-infected children often have high HIV loads that cannot be suppressed completely. It therefore is important to evaluate CD4+ T lymphocyte repletion in children who achieve only incomplete or minimal viral suppression while receiving protease inhibitor–containing regimens

Our group has previously reported the initial results of a phase 1/2 study in which HIV-infected children received indinavir monotherapy followed by the addition of zidovudine (ZDV) and lamivudine (3TC) [2]. These patients represent one of the first pediatric cohorts to receive long-term anti-HIV therapy that included a protease inhibitor. Here, we report the long-term results for the 33 patients in this cohort who completed 96 weeks of therapy

Methods

The phase 1/2 trial originally was designed as a dose escalation study of indinavir monotherapy, using a suspension formulation at 3 dose levels (250, 350, and 500 mg/m2 every 8 h) for 16 weeks, followed by the addition of ZDV and 3TC [2]. From July 1995 through August 1996, 54 HIV-infected children, 3–19 years old, were enrolled. Initial results led to a switch to a capsule formulation and then to a switch from the 500-mg dose to a 350-mg dose. Therefore, the children received indinavir at 250, 350, or 500 mg/m2 every 8 h for 16 weeks, followed by the addition of ZDV (dose, 120 mg/m2; maximum, 600 mg/day) every 8 h and 3TC (dose, 4 mg/kg; maximum, 150 mg/day) twice daily. By week 58, all children were receiving 350 mg/m2 of indinavir every 8 h in combination with ZDV and 3TC

Patient evaluationAt entry and at predefined intervals, patients’ medical histories were obtained, and physical examinations and laboratory studies were done [2]. T lymphocyte subsets were determined by flow cytometric analysis, using an Epics XL flow cytometer (Coulter). Total CD4+ (CD4+CD3+), CD4+CD45RA+ (naive), and CD4+CD45RO+ (memory) T cell subsets were determined, by 2-color flow cytometry techniques, for all patients [1]. In some patients, CD4+ naive cells were identified, by 3-color flow cytometry, as CD4+CD45RA+CD45RO or as CD4+CD45RA+CD62L+. Analysis of the results of 2-color and 3-color flow cytometry of paired samples from 17 different patients showed a high degree of correlation between the results (r=.993; P=.0001, by Spearman’s&rank correlation), which suggests that the CD4+CD45RA+ phenotype provided a good approximation for the CD4+ naive T cells in these patients. Plasma HIV RNA quantitation was performed by using the Roche Amplicor polymerase chain reaction assay, with a lower limit of detection of 200 copies/mL

Data analysisAnalysis of the relationships between HIV load and peripheral CD4+ T cell subsets was done by dividing the patients into 3 groups, similar to those described by Deeks et al. [3], on the basis of control of HIV level after the start of triple therapy. The groups were as follows: (1) complete responders (CR), plasma HIV RNA level decreased to and remained ⩽500 copies/mL (4 patients); (2) partial responders (PR), plasma HIV RNA level sustained a decrease of ⩾0.7 log10 copies/mL below baseline but remained >500 copies/mL (8 patients); and (3) transient responders (TR), plasma HIV RNA level transiently decreased to ⩾0.7 log10 copies/mL below baseline (21 patients)

Baseline characteristics for the 33 patients who continued to receive treatment for ⩾96 weeks and for the 21 patients who did not were compared by using the Wilcoxon&rank sum test, the χ2 test, or Mehta’s version of Fisher’s exact test, as appropriate [4]. Paired comparisons between parameters were done by using the Wilcoxon signed&rank test. All P values are 2 sided

Results

Patient characteristicsOf the 54 patients who were enrolled, 33 completed 96 weeks of therapy. Three patients were withdrawn for clinical progression, 2 for immunologic decline, 11 for miscellaneous reasons, and 5 for toxicities that were definitely (2 patients) or possibly (3 patients) related to indinavir. There were no significant differences between the group of 33 patients who completed therapy and the group of 21 patients who withdrew, with regard to age, sex, route of HIV acquisition, race, Centers for Disease Control and Prevention (CDC) clinical category, prior anti-HIV therapy, baseline CD4+ T cell count, or baseline HIV RNA copies/mL. The median baseline age, CD4+ T cell count, and HIV load for the 33 patients (24 boys and 9 girls) who completed therapy were 8.9 years (range, 3–16 years), 267 cells/μL (range, 0–1706 cells/μL), and 4.77 log10 HIV RNA copies/mL (range, 3.46–5.77 log10 copies/mL), respectively. Eighteen patients were classified as CDC clinical category C [5]. Thirty were heavily treated with antiretroviral drugs before entry into the protocol

Immunologic and virologic responsesDespite the initial use of indinavir monotherapy, which is now recognized as suboptimal, and the administration of some doses of indinavir that may have been subtherapeutic, the 33 patients who completed the study had a median decrease in virus load of 0.74 log10 HIV RNA copies/mL from baseline to week 96 (P=.0001, Wilcoxon signed&rank test; figure 1A). The virus loads and CD4 T cell counts of these patients appeared to be similar to those of the entire patient group (N=54; data not shown). All patients experienced an initial decline in virus load after starting indinavir monotherapy [2], and the time to a decrease in virus load of 0.7 log10 HIV RNA copies/mL did not differ among the 3 response groups. Most patients then had partial virologic rebound and a second decrease in virus load after the addition of ZDV and 3TC at week 16. Twenty-eight of the 33 patients had a maximum decline in virus load after the addition of ZDV and 3TC. The median time to the greatest virologic response was 18 weeks (range, 1.9–28 weeks), and the median maximum decrease in virus load was −1.72 log10 HIV RNA copies/mL (range, −0.73 to −3.46 log10 copies/mL)

Figure 1

Median changes from baseline in CD4+ T cells/μL and human immunodeficiency virus (HIV) load (log10 HIV RNA copies/mL) during the 96-week treatment period in HIV-infected children who received indinavir, zidovudine, and lamivudine. A All children who completed 96 weeks of therapy (33 patients); B subset of complete virologic responders (4 patients); C subset of partial virologic responders (8 patients); D subset of transient virologic responders (21 patients). Vertical bars indicate the interquartile range (25th–75th percentile) for each time point

Figure 1

Median changes from baseline in CD4+ T cells/μL and human immunodeficiency virus (HIV) load (log10 HIV RNA copies/mL) during the 96-week treatment period in HIV-infected children who received indinavir, zidovudine, and lamivudine. A All children who completed 96 weeks of therapy (33 patients); B subset of complete virologic responders (4 patients); C subset of partial virologic responders (8 patients); D subset of transient virologic responders (21 patients). Vertical bars indicate the interquartile range (25th–75th percentile) for each time point

The 33 patients had a median increase in CD4+ T cell count of 199 cells/μL (range, −369 to 949 cells/μL; P=.0001, Wilcoxon signed&rank test) from baseline to week 96 (figure 1A). During the first 16 weeks, a rapid median increase in CD4+ T cell count of 141 cells/μL (range, −190 to 1043 cells/μL) was followed by a more gradual increase. The median maximum increase in CD4+ T cell count of 207 cells/μL (range, −89 to 1077 cells/μL) was reached at week 72, with little change after that time. The influence of the initial indinavir dose on changes in CD4+ T cell count and virus load was not statistically significant (data not shown)

Data for the CD4+CD45RA+ and CD4+CD45RO+ T cell subsets were available for all 33 patients for study weeks 0, 16, and 28; for 15 patients for week 96; and for 17 patients for week 108. These latter 2 time points were combined, to enable assessment at approximately the 2-year time point. At entry, the baseline median CD4+CD45RA+ and CD4+CD45RO+ T cell counts were 180 and 106 cells/μL, respectively. At the 2-year time point, CD4+CD45RA T cells (median increase, 72 cells/μL; P<.0001, Wilcoxon signed&amp;rank test), rather than CD4+CD45RO+ T cells (median increase, 27 cells/μL; P=.011), accounted for most of the increase in total CD4+ T cells

Relationship between virologic response and repopulation of CD4+ T cellsChanges in CD4+ T cell count and virus load for the 3 virologic response groups are shown in figure 1B–1D. The median increases in CD4+ T cell count in the CR and PR groups at week 96 were 178 cells/μL (range, −369 to 742 cells/μL; P=.63) and 286 cells/μL (range, 28–949 cells/μL; P=.008), respectively, and were associated with median decreases in virus load at week 96 of −2.22 and −1.52 log10 HIV RNA copies/mL, respectively. The TR group had a median increase in CD4+ T cell count of 157 cells/μL (range, −89 to 1077 cells/μL; P=.0001, Wilcoxon signed&amp;rank test) and 54 cells/μL (range, −66 to 846 cells/μL; P=.0016, Wilcoxon signed&amp;rank test) at weeks 72 and 96, respectively, despite the fact that the median virus load was not significantly different from baseline at time points beyond week 28. CD4+ naive T cells contributed extensively to the increases in total CD4+ T cells in all 3 virologic response groups (figure 2A–2C)

Figure 2

CD4+ T cell subset and virologic data in different response subsets of human immunodeficiency virus (HIV)–infected children who received indinavir, zidovudine, and lamivudine. A–C Median changes in CD4+, CD4+CD45RA+ and CD4+CD45RO+ T cell count from baseline for the patients with complete (A) partial (B) and transient (C) virologic responses. D and E Similar T cell subset data for the patients with transient virologic responses who had no sustained increase in CD4+ T cell count (14 patients; D) and a sustained increase in CD4+ T cell count (7 patients; E). F HIV load data for the same patient subsets depicted in D and E. Vertical bars indicate the interquartile range (25th–75th percentile) for each time point

Figure 2

CD4+ T cell subset and virologic data in different response subsets of human immunodeficiency virus (HIV)–infected children who received indinavir, zidovudine, and lamivudine. A–C Median changes in CD4+, CD4+CD45RA+ and CD4+CD45RO+ T cell count from baseline for the patients with complete (A) partial (B) and transient (C) virologic responses. D and E Similar T cell subset data for the patients with transient virologic responses who had no sustained increase in CD4+ T cell count (14 patients; D) and a sustained increase in CD4+ T cell count (7 patients; E). F HIV load data for the same patient subsets depicted in D and E. Vertical bars indicate the interquartile range (25th–75th percentile) for each time point

Analysis of immunologic responses in patients with transient virologic responseThe 21 patients in the TR group could be subdivided into 2 groups on the basis of their change in CD4+ T cell count from the time that they stopped having a virologic response (i.e., the time after nadir that their plasma HIV RNA level was no longer 0.7 log10 copies/mL below baseline) to week 96 (figure 2D–2F). Fourteen patients had a decrease in CD4+ T cell count during this period, whereas 7 “discordant” patients had a sustained median increase in CD4+ T cell count of 370 cells/μL (range, 80–605 cells/μL). The CD4 T cell count at study entry was somewhat lower in the 7 patients with the sustained increase than in those with an unsustained increase (median, 22 vs. 157 CD4+ T cells/μL), but the difference was not statistically significant (P=.25). In addition, the difference between the 2 groups in the median time until they stopped having a virologic response was not significant (P=.19). During the initial weeks of therapy, the increase in CD4+ T cells in the 7 patients with a sustained response was mostly due to CD4+CD45RO+ T cells. However, after week 28, a relative increase in CD4+CD45RA+ T cells was observed. At weeks 96–108, the 7 patients had an increase above baseline in both CD4+CD45RA+ and CD4+CD45RO+ T cells (363 cells/μL and 109 cells/μL, respectively; figure 2E). There essentially was no correlation between the change in virus load from baseline to week 96 and the change in CD4+CD45RA+ T cell count during the same time period (r=.057; P=.80, by Spearman’s&amp;rank correlation) in the TR group. Thus, these data suggest that a subset of children with transient virologic responses still can experience a long-term robust immunologic response, with CD4+ T cells with a naive phenotype accounting for most of the increase in total CD4+ T cell count

Discussion

The results of this study support previous observations that some HIV-1–infected children and adults have a positive immunologic response despite incomplete HIV suppression or even virologic failure [3, 6,,,10]. The TR group experienced a small median increase in CD4+ T cells at week 96, even though at week 28 and beyond the median virus load in this group was not significantly different from baseline. Also, 7 discordant children experienced sustained increases in CD4+ T cell count with a predominant increase in CD4+CD45RA+ naive T cells through week 96, despite the loss of virologic control

In general, repopulation with CD4+ naive T cells is more limited in adults treated with highly active antiretroviral therapy [11]. We noted here an increase in CD4+ naive T cells after week 16 in all 3 virologic response groups, which possibly reflects the greater thymic function in these younger patients [1]. T cell responsiveness did not increase during the first 28 weeks of this study [12]; however, some patients who were followed until week 120 and beyond had improved proliferative responses to certain recall antigens (C. Chougnet, S. Jankelevich, K. Fowke, D. Liewehr, S. M. Steinberg, B. U. Mueller, P. A. Pizzo, R. Yarchoan, and G. M. Shearer, unpublished data)

What mechanisms are responsible for the sustained increase in CD4+ T lymphocytes in the discordant patients? These patients appear to establish a new equilibrium in which there is an increased number of CD4+ T cells but a similar virus load. One possibility is that the development of resistance to the antiretroviral agents may result in reduced viral replicative fitness or a switch to a less virulent phenotype [6, 13,15]. Another possibility is that protease inhibitors may directly affect T cell apoptosis by inhibiting one of the effector pathways, although a report by Chougnet et al. [12] indicated that this parameter did not change during the beginning of the present study. Finally, even a slight sustained decrease in HIV load may be sufficient to cause immunologic improvement, or patients with immunologic improvement may represent one portion of the normal distribution of CD4+ T cell changes

Our data show that an incomplete or transient virologic response to protease inhibitor–containing therapy may allow certain subgroups of HIV-infected children to experience sustained increases in CD4+ T cell count. However, maintaining patients on a regimen when HIV replication is not suppressed may lead to high-level resistance, and, in general, this approach is not recommended for treatment-naive patients. Additional data will be needed to understand the basis for this discordance and to evaluate whether continuing therapy that does not completely suppress virus replication is warranted for certain patients who have a good immunologic response

Acknowledgments

We thank the patients who volunteered for this study and their parents; Dimiter Dimitrov (National Cancer Institute, Frederick Cancer Research and Development Center, National Institutes of Health [NIH], Frederick, Maryland) for helpful discussions; Paul Jarosinski (NIH Clinical Center Pharmacy Department, Bethesda, Maryland); the nurses, medical staff, social work department, and other affiliated personnel of the HIV and AIDS Malignancy Branch, the Pediatric Oncology Branch, and the NIH Clinical Center (Bethesda, Maryland); Linda A. Hawe (Merck Research Laboratories, Rahway, New Jersey); and Michael Baseler, David J. Waters, and Randy Stevens (Science Applications International Corporation, Frederick, Maryland)

References

1
Mackall
CL
Fleisher
TA
Brown
MR
, et al.  . 
Age, thymopoiesis, and CD4+ T-lymphocyte regeneration after intensive chemotherapy
N Engl J Med
 , 
1995
, vol. 
332
 (pg. 
143
-
9
)
2
Mueller
BU
Sleasman
J
Nelson
RP
Jr
, et al.  . 
A phase I/II study of the protease inhibitor indinavir in children with HIV infection
Pediatrics
 , 
1998
, vol. 
102
 (pg. 
101
-
9
)
3
Deeks
SG
Barbour
JD
Martin
JN
Swanson
MS
Grant
RM
Sustained CD4+ T cell response after virologic failure of protease inhibitor–based regimens in patients with human immunodeficiency virus infection
J Infect Dis
 , 
2000
, vol. 
181
 (pg. 
946
-
53
)
4
Mehta
CR
Patel
NR
A network algorithm for performing Fisher’s exact test in r × c contingency tables
J Am Stat Assoc
 , 
1983
, vol. 
78
 (pg. 
427
-
34
)
5
Centers for Disease Control and Prevention
1994 Revised classification system for human immunodeficiency virus infection in children less than 13 years of age
MMWR Morb Mortal Wkly Rep
 , 
1994
, vol. 
43
 (pg. 
1
-
10
)
6
Kaufmann
D
Pantaleo
G
Sudre
P
Telenti
A
CD4-cell count in HIV-1–infected individuals remaining viraemic with highly active antiretroviral therapy (HAART). Swiss HIV Cohort Study
Lancet
 , 
1998
, vol. 
351
 (pg. 
723
-
4
)
7
Piketty
C
Castiel
P
Belec
L
, et al.  . 
Discrepant responses to triple combination antiretroviral therapy in advanced HIV disease
AIDS
 , 
1998
, vol. 
12
 (pg. 
745
-
50
)
8
Casado
JL
Perez-Elias
MJ
Antela
A
, et al.  . 
Predictors of long-term response to protease inhibitor therapy in a cohort of HIV-infected patients
AIDS
 , 
1998
, vol. 
12
 (pg. 
F131
-
5
)
9
Mezzaroma
I
Carlesimo
M
Pinter
E
, et al.  . 
Clinical and immunologic response without decrease in virus load in patients with AIDS after 24 months of highly active antiretroviral therapy
Clin Infect Dis
 , 
1999
, vol. 
29
 (pg. 
1423
-
30
)
10
Thuret
I
Michel
G
Chambost
H
, et al.  . 
Combination antiretroviral therapy including ritonavir in children infected with human immunodeficiency
AIDS
 , 
1999
, vol. 
13
 (pg. 
81
-
7
)
11
Connors
M
Kovacs
JA
Krevat
S
, et al.  . 
HIV infection induces changes in CD4+ T-cell phenotype and depletions within the CD4+ T-cell repertoire that are not immediately restored by antiviral or immune-based therapies
Nat Med
 , 
1997
, vol. 
3
 (pg. 
533
-
40
)
12
Chougnet
C
Fowke
KR
Mueller
BU
, et al.  . 
Protease inhibitor and triple-drug therapy: cellular immune parameters are not restored in pediatric AIDS patients after 6 months of treatment
AIDS
 , 
1998
, vol. 
12
 (pg. 
2397
-
406
)
13
Mammano
F
Petit
C
Clavel
F
Resistance-associated loss of viral fitness in human immunodeficiency virus type 1: phenotypic analysis of protease and gag coevolution in protease inhibitor–treated patients
J Virol
 , 
1998
, vol. 
72
 (pg. 
7632
-
7
)
14
Yuste
E
Sanchez-Palomino
S
Casado
C
Domingo
E
Lopez-Galindez
C
Drastic fitness loss in human immunodeficiency virus type 1 upon serial bottleneck events
J Virol
 , 
1999
, vol. 
73
 (pg. 
2745
-
51
)
15
Stoddart
C
Mammano
F
Moreno
M
, et al.  . 
Lack of fitness of protease inhibitor–resistant HIV-1 in vivo [abstract 4]
Program and abstracts of the 6th Conference on Retroviruses and Opportunistic Infections
 , 
1999
Alexandria, VA
Foundation for Retrovirology and Human Health
pg. 
67
 
Presented in part: 6th Conference on Retroviruses and Opportunistic Infections, Chicago, January 1999 (abstract 438)
The protocol for this study was approved by the institutional review board of the National Cancer Institute. Informed consent was obtained from all study participants, in accordance with the human experimentation guidelines of the US Department of Health and Human Services
Financial support: National Cancer Institute intramural support and Clinical Trials Agreement between Merck Research Laboratories and the National Cancer Institute
B.-Y.N. is an employee of Merck Research Laboratories, the manufacturer of indinavir, and L.S. owns shares of Merck stock. Otherwise, no author has a commercial or other association that may pose a conflict of interest
Present affiliations: Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland (S.J.); Children’s Hospital, Boston, Massachusetts (B.U.M and P.A.P.); Indiana University School of Medicine, Indianapolis (R.P.N.)