The HIV Cycle and the Early Life Cycle: Treatment Effects and Side Effects in Children and Adolescents
Our recent understanding of the effects of highly active antiretroviral therapy (HAART) in children with HIV disease has matured rapidly, and the poster sessions at the 10th Conference on Retroviruses and Opportunistic Infections reflected this growing experience. The number of presentations reporting metabolic complications of antiretrovirals in HIV-infected children was also markedly higher than in past years. This, too, reflected the longer experience with HAART in children as well as greater exposure of children to treatment, albeit in developed countries.
Children and HAART
When antiretroviral use during pregnancy and the peripartum period fails to prevent HIV infection in infants, or in the absence of effective strategies such as vaccines to prevent breast-feeding transmission, the only alternative is to treat the infected infant. The ideal time to initiate treatment in children is currently a subject of debate. Simplified antiretroviral regimens with once-daily dosing undoubtedly facilitate adherence and are attractive alternatives for children, particularly adolescents.
One ongoing phase 1/2 study evaluated emtricitabine (or FTC) in 82 HIV-infected children, age 4 months to 17 years, from Panama, South Africa, Mexico, and the United States. A total of 51 antiretroviral-naive children received a regimen containing 6 mg/kg of FTC combined with stavudine and lopinavir/ritonavir. In addition, 31 children who had received a stable regimen containing 3TC (lamivudine) for ≥ 3 months with an undetectable virus load were also enrolled. Investigators followed the children for 48 weeks. At week 24, 92% of the previously untreated patients and 84% of the drug-experienced children developed or maintained an HIV virus load < 400 copies/mL. The incidence of serious adverse events was 14% in the naive group and 6% in the experienced patients. Serious adverse events included 1 case each of pancreatitis, vomiting, pleural effusion, leukopenia, and anemia. Pharmacokinetic analysis of a subset of 28 patients demonstrated that the 6-mg/kg/day dose produced similar exposure to that of adults receiving 200 mg daily.
Investigators from Pediatric AIDS Clinical Trials Group (PACTG) 1021, an ongoing phase 1/2 trial of emtricitabine, didanosine, and efavirenz in a similar population of children (therapy-naive or minimally treated) ages 3-21 years, also presented preliminary results for week 16 of treatment. Overall efficacy was similar to those in the previously described trial, with a very small number of grade 3/4 laboratory abnormalities, including significant elevations in creatine phosphokinase (CPK) and gamma-glutamyl transferase (GGT) (1 patient each). Based on the preliminary findings of both trials, emtricitabine appears to be an encouraging agent for treating pediatric HIV.
For salvage therapy, PACTG 1013 reported results using indinavir plus 2 doses of ritonavir in antiretroviral-experienced children. Study subjects were 2-18 years of age who had received at least 16 weeks of combination therapy and nonetheless had more than 10,000 copies of HIV RNA/mL of plasma and fewer than 3 primary protease gene mutations.
Patients received either intact or opened indinavir capsules at a dose of 350 mg/m2 twice daily, plus twice-daily ritonavir at 125 mg/m2 or 300 mg/m2 and at least 1 NRTI drug in combination. Pharmacokinetic analyses demonstrated that the ritonavir boost significantly increased indinavir levels, and that the higher-dose ritonavir increased pre-dose indinavir levels 4 times higher than lower-dose ritonavir. Combination therapy was associated with reductions in HIV RNA ≥ 0.75 log10 in 15 of 21 patients by week 16, which was sustained at 50 weeks of therapy. Although the regimens were generally well tolerated, drug discontinuation and virologic failures were essentially attributable to the poor palatability of the regimen and the high pill burden.
Structured Treatment Interruptions in Children
An alternative and unconventional approach to antiretroviral therapy delivery was reported by the PACTG 1015 team. Their ongoing study proposes using progressively increasing structured treatment interruptions for patients with effective viral suppression in order to elicit improved HIV-specific CD4+ and CD8+ cell responses. Candidates for the study are 4-21 years of age and have reached total viral suppression on a HAART regimen that contains at least 1 protease inhibitor (PI), with a CD4% above 20, and without use of any nonnucleoside reverse transcriptase inhibitors (NNRTIs) or abacavir.
The patients receive sequential cycles of structured treatment interruptions, increased in increments of 2 days/cycle, followed by HAART: 3 days of interruption followed by 3 weeks of HAART, with subsequent 5-day interruption followed by 3 weeks of HAART, and so on. After the 5-day interruption, patients receive HAART until virus load levels reach < 50 HIV RNA copies/mL, and then a new cycle of interruption occurs. The protocol allows up to 7 weeks of treatment interruption.
The team presented results for 10 patients. For most of these, plasma viremia rebounded within 5-9 days off therapy. Viral suppression was regained in all of these individuals, with no appearance of new resistance mutations. Changes in CD4+ and CD8+ percentages were not noted; however, transient declines in lymphocytes were seen. Changes in viremia did produce increases in HIV-specific CD4+ and CD8+ cell responses as measured by interferon gamma ELISpot assays.
The structured treatment interruption approach is certainly of scientific relevance; however, the clinical benefit of improved HIV-specific CD4+ and CD8+ cell responses in patients with previous total suppression of HIV RNA levels has yet to be determined. Certainly the theoretical benefits should be weighed against the potential risk of not achieving de novo viral suppression or of seeding cellular reservoirs during the periods of rebound viremia. The practicality of this regimen is also problematic, and it is doubtful that this procedure can be carried outside the realms of a clinical protocol.
Older children and adolescents often practice unstructured treatment interruptions with no evidence of clinical benefit. Furthermore, patients might develop significant adherence issues if placed on progressively interrupted treatment regimens. PACTG 1015 may help elucidate the clinical utility of such a cumbersome and potentially dangerous process.
Less Costly Surrogate Markers for Monitoring HAART
Delivering antiretroviral treatment to children bears a greater expense than the cost of drug. The cost and technology involved in laboratory monitoring of HAART treatment is a partial impediment to antiretroviral dispensation in resource-limited settings. A study by Mofenson and colleagues from the National Institutes of Health in Bethesda, Maryland, evaluated the use of total lymphocyte count and immune complex dissociated (ICD) p24 antigen in lieu of CD4+ cell count and HIV RNA polymerase chain reaction (PCR) in predicting child mortality due to HIV disease. The patient population consisted of antiretroviral-naive HIV-infected children recruited for an intravenous immunoglobulin (IVIG) clinical trial 12-15 years ago. Serial HIV RNA levels, ICD p24 antigen levels, total lymphocyte counts, hematocrit, CD4+ cell counts, and albumin levels were measured in specimens collected every 3 months.
In the current study, investigators assessed the value of simple baseline measures such as ICD p24 antigen, lymphocyte counts, hematocrit, and albumin levels to predict long-term childhood mortality, and compared the predictive value of these measures with that of HIV RNA and CD4+ cell counts. They enrolled 376 children who had a 40% mortality rate during the follow-up period. Using univariate analyses, total lymphocyte counts, ICD p24 antigen, albumin levels, and hematocrit were all associated with mortality risk in children. Using standardized risk ratios, lymphocyte counts proved the best predictor of mortality, and with multivariate proportional hazards model, lymphocyte counts and albumin levels were independently associated with mortality risk. Except for ICD p24 antigen, the sensitivity of all assays in predicting child death was less than 50%, including CD4+ cell count and HIV RNA. The specificity of ICD p24 antigen was poor (32%) compared with all other assays, which were 83% specific or greater in predicting child survival. Given these findings, and the comparability of these assays, “low-tech” approaches could potentially be used in resource-limited settings as prognostic markers or initiation of antiretroviral therapy.
When Therapy Goes Bad: Complications of HAART in Children
A number of studies addressed the metabolic complications of antiretroviral regimens in children. Researchers from Spain determined lactate and alanine plasma levels in 102 HIV-exposed, uninfected infants who received zidovudine following birth at 6 weeks and 3, 6, and 12 months of age. Increased levels of both products were detected in 64% of babies at any given time point, with higher numbers of patients showing abnormalities at 6 weeks of age (60%). The team reported that 3 of the patients experienced mild neurologic symptoms (axial hypotonia), which resolved by 6 months.
The lack of a control group muddies the interpretation of these results. Lactate levels are extremely difficult to interpret and ideally an equal number of non-HIV-exposed infants born to uninfected mothers with the same background risk factors for HIV infection should be screened in the same manner.
The same investigators reported a 22% fat redistribution rate among 80 older HIV-infected children (mean age 9 years) receiving HAART. Dyslipidemia was found in 60% of these patients. Duration of antiretroviral therapy was not statistically different between children with or without fat redistribution. These values are much higher than those previously described in children; however, we now have a population of children with longer exposure to HAART.
Body Composition and Biochemical Changes
Investigators from PACTG 1010 reported results of an interim analysis of 63 subjects in a carefully designed study of body composition and biochemical changes in children starting or switching antiretroviral therapy. Children underwent anthropometric measures, bioelectric impedance analysis, fasting lipid panel, indices of glucose homeostasis, growth factors, and virus load and CD4+ cell subsets at entry and at 4 subsequent study visits between weeks 12 and 48.
Over time, changes occurred in almost every marker, with positive associations with virus load. A modest correlation was found between a greater suppression of HIV RNA and loss of peripheral fat, increased central adiposity, and a greater anabolic response. Increased height-for-age Z-score was associated with a positive change in CD4+ cells rather than virus load, suggesting that such changes are consistent with immunologic host recovery rather than a virologic response. A significant increase in insulin levels and a modest increase in cholesterol levels and IGF-1 were noted with reductions in IGFBP-1.
Anabolic changes appeared to be mediated by reductions in proinflammatory cytokines. Investigators noted these changes at weeks 24 and 36, but no association between these changes and virus load was seen at week 48. Insulin resistance appeared to be a cause rather than a result of fat redistribution. Aside from weight and height, which were reported as Z-scores, age-adjusted comparisons to normal controls have not yet been made, and thus, it is unclear whether many of these changes could be attributable to normal growth.
An analysis of 1812 HIV-infected children compared with 187 uninfected subjects, ages 4-21 years, followed in PACTG 219 found a 13% prevalence of hypercholesterolemia, well in excess of the 5% expected. Hypercholesterolemia was associated with virologic suppression, increased in CD4+ cell subsets, good adherence by self-report, younger age, Hispanic ethnicity, and PI use, particularly dual or triple PIs.
Changes in Mitochondrial DNA
One small pilot study measured the mitochondrial DNA (mDNA) content of the placenta and cord blood of 8 women on NNRTIs and compared it with that of 5 uninfected women. The mean mDNA copies per cells from placenta of HIV-infected women was significantly reduced compared with the other group (52 vs 880 copies). Cord blood mDNA levels were also reduced significantly when both groups were compared. At the tissue level, there was evidence of some degree of mitochondrial toxicity, but this did not attain any clinical significance.
Bone Mineral Density
A cross-sectional study from Houston, Texas, examined the effect of HAART on bone mineral density (BMD) in children. Investigators performed lumbar spine and total-body dual-energy x-ray absorptiometry (DEXA) scans in 27 children aged 6-17 years. In all, 66% of children were taking PI-containing HAART. No children were hypocalcemic or vitamin D-deficient. On scanning, however, 52% of the children were osteopenic (Z-score between -1 and -2.5) and 23% had osteoporosis (Z-score under -2.5). There was no difference in mean BMD Z-scores between children taking PIs or not. However, the longer a patient received PIs, the greater the risk for osteopenia. The findings were more prevalent in males. It is extremely important to be aware of the baseline degree of BMD loss among HIV-infected children, since the findings can be attributable to HIV disease itself rather than to treatment. Larger prospective trials are warranted.
Examining Treatment Failure
A detailed study from investigators at the University of California San Diego provided an alternative explanation for treatment failure of pediatric patients receiving nelfinavir. Transport and metabolism of many drugs depend on the multidrug-resistance transporter gene (MDR1), which encodes for P-glycoprotein, and also on genes that encode for the transport of isoenzymes of cytochrome P450 (CYP). Genetic polymorphisms on the MDR1 or CYP genes of patients could be responsible for varying concentrations of antiretroviral drugs in plasma and thus affect virologic and immunologic responses.
In this study, 71 pediatric patients receiving nelfinavir and efavirenz in PACTG 382 underwent DNA genotyping to determine allelic variants. Although no changes specific to efavirenz were noted, children with MDR-3435 C/C genotype experienced slower virologic responses to HAART and demonstrated lower plasma concentrations and higher clearance rates of nelfinavir. P-glycoprotein appears to play an important role in the pharmacokinetics and virologic response to nelfinavir.
- Saez-Llorens X, Violari A, Ndiweni D, et al. Once-daily emtricitabine in HIV-infected pediatric patients with other antiretroviral agents. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 872.
- McKinney R, Rathore M, Jankelovich S, et al. PACTG 1021: an ongoing phase I/II study of once-daily emtricitabine, didanosine, and efavirenz in therapy-naive or minimally treated pediatric patients. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 873.
- Chadwick EG, Rodman JH, Samson P, et al. Antiviral activity, tolerance and pharmacokinetics of indinavir with two doses of ritonavir as salvage therapy in children. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 875.
- Borkowsky W, Yogev R, Muresan P, et al. T-cell and virologic outcomes of a progressively increasing structured treatment interruption study in chronically-infected children and adolescents. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 885.
- McFarland E, Borokowsky W, Muresan P, et al. Increases in HIV-specific CD4+ and CD8+ T-cell mediated responses in children undergoing structured treatment interruption. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 884.
- Mofenson L, Harris DR, Bethel J, et al. Second tier surrogate markers for use in resource-limited settings: association of total lymphocyte count and immune-complex dissociated p24 antigen with mortality in HIV-infected children. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 879.
- Fortuny C, Noguera A, Vilaseca MA, et al. Hyperlactatemia in children exposed to antiretrovirals and its relation with lipodystrophy syndrome in HIV-infected HAART-treated pediatric patients. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 777.
- Chantry C, Cervia J, Hughes M, et al. Body composition and biochemical changes in children starting or switching combination antiretroviral chemotherapy. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 775.
- Farley J, Gona P, Crain M, et al. Prevalence of hypercholesterolemia and associated risk factors among perinatally HIV-infected children (4 -19 years) in PACTG 219C. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 773.
- Shiramizu B, Shikuma K, Kamemoto L, et al. Placenta and cord blood mitochondrial DNA toxicity in HIV-infected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 771.
- Schwarzwald H, Ellis KJ, Evans DL, et al. Effect of HAART on bone density in HIV-infected children. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 778.
- Singh K, Saitoh A, Powell C, et al. Allelic variants of MDR1 alter pharmacokinetics of nelfinavir resulting in higher drug levels and more rapid decline in plasma HIV-1 RNA in children. Program and abstracts of the 10th Conference on Retroviruses and Opportunistic Infections; February 10-14, 2003; Boston, Massachusetts. Abstract 100.
COPIA CON FINES ACADEMICOS