Paediatric treatment issues at the 8th CROI

Polly Clayden, HIV i-Base

This year’s Retroviruses meeting included an impressive number of sessions and posters focusing on childrens’ care.

Impact of maternal factors on paediatric disease progression

Two presentations at this meeting looked at maternal factors that could determine a child’s rate of disease progression before birth. In the first Dr Ruth Dickover examined the relationship between HIV envelope gene evolution during primary infection and the rate of disease progression in the first year of life in infants infected with HIV [1].

A group of 23 HIV-positive babies were followed from birth in this study. They were tested with DNA heteroduplex mobility assays (HMA) to assess HIV env gene diversity at birth or at the first positive time point, at 4, 8, 12, 26, 39 and 52 weeks of age. Infants with decreased HIV env gene diversity showed high viral loads, rapid decline in CD4 and progressed rapidly to AIDS over the first 12 weeks. In contrast slow progressors showed a very different pattern – either constant or increasing env gene diversity. The investigators observed a correlation between higher CD4 and higher gene diversity, and infants treated with antiretrovirals showed a change in diversity several months after the initiation of therapy. They also noted that the level of HIV env gene diversity can change rapidly in the first 12 weeks. They concluded that ‘The rate of HIV evolution during HIV infection may prove useful in predicting the risk of disease in perinatally infected infants.’

A report from the Perinatal AIDS Collaborative Transmission Study (PACTS), showed an association between high maternal viral load and the rate of disease progression in their HIV-infected infants [2]. This study of 2665 mother/infant pairs, reported that 48% of babies born to mothers with viral loads >100,000 copies/ml had died before 2 years of age. This compared to only 8% born to women with viral load <25,000 copies/ml. Their results showed that HIV-infected children born to women with high viral load are at increased risk for more rapid disease progression. Encouragingly they also reported that treatment with antiretrovirals before 24 weeks of age led to a 67% reduction in relative risk of disease progression. But, as Dr Abrams explained during the presentation, ‘Hopefully women with high viral load will be a thing of the past.’

Early antiretroviral therapy

In the US in particular, early aggressive treatment for perinatally HIV-infected newborns has been the norm. Dr Katherine Luzuriaga presented data supporting this strategy [3], but also reported that infants who are treated early may never develop HIV-specific CD8 responses, unlike adults who are treated in primary infection [4]. She explained the goals of early therapy to be: control of HIV-replication, leading to – long-term survival, absence of clinical symptoms and a preserved immune function. Also that vertically infected infants are a unique population, as it is known exactly when infections occur.

She presented data from two studies – the PACTG 345 [5] and 346, which treated newborns with HAART – AZT/3TC/RTV and AZT/3TC/NVP or AZT/3TC/NVP/ABC or d4T/3TC/NVP/NFV respectively. Findings from both trials showed 60% (22) of infants achieved viral load less than 400copies/ml by 20 weeks. Children treated before 3 months of age were most likely to preserve normal immune response. Similar to adults, lymphoproliferative responses to p24 antigen were not restored although non-HIV responses were. Dr Luzuriaga discussed the possibility of using vaccines to strengthen and broaden HIV-specific responses in combination with early therapy.

During questions after this presentation, the issue of STIs for children with good viral suppression was raised, however Dr Luzuriaga explained that she would not currently be in favour of stopping therapy in this population.

HIV dynamics and latent reservoirs

Although the mechanisms are unclear, adult studies have reported the existence of latently HIV-infected cellular reservoirs. Dr Deborah Persuad’s presentation examined this phenomenon in children [6]. She explained that despite treatment with potent antiretroviral therapy, persistent latent HIV in resting CD4 cells is found at frequencies similar to those in adults. The mechanism for this persistence is now thought to be by cell division, without activation, and therefore beyond the reach of existing drugs. And as with adults, knowledge of the mechanism of viral persistence in HIV positive children is critical to developing more effective therapies in hope that the virus may eventually be eradicated.

Immune reconstitution

Dr William Borkowsky gave an overview of immune reconstitution in children [7]. He explained that due to their greater thymic activity relative to adults, children treated with HAART experience very significant CD4 increases – 320 to 650 cells in the first year of treatment (this compares to 100 to 250 cells in adults). Children also respond with more naive cells than adults, which are often restored, to normal levels. Dr Borkowsky reported that the increase in CD4 shows an inverse correlation with age ie the younger you are the more likely you are to get higher numbers of CD4 cells, which is also thought to reflect thymic size. He reported that children can have a discordant response to therapy and achieve CD4 increases to normal or near normal levels despite incomplete viral suppression. Dr Borkovsky argued strongly for early and aggressive treatment for newborns but for older children and adolescents he felt that the benefits of HAART at a late disease stage might exceed those at an earlier stage.

When treated with antiretrovirals HIV positive adults recover responses to candida but not to tetanus. A study from Dr Weinburg and colleagues, as part of PACTG 366, examined the ability of children to recover immune responses to candida and tetanus [8]. Children with advanced disease, treated with a 4-drug HAART regimen who achieved viral load <400 copies/ml were able to recover responses not only to candida but also to tetanus without need of reimmunisation.

Lipodystrophy, bone mineral loss, avascular necrosis

There were further reports of metabolic changes and lipodystrophy in children [9,10,11,12,13] and also bone mineral changes and avascular necrosis, these topics are covered in the previous issue of HTB [14,15,16,17].

Gender differences

Lower viral loads and higher CD4 levels are reported in adult women compared to men [18] and menstrual cycle studies have suggested that this may be attributed to hormonal differences. Less is known about gender differences in HIV-positive children. Dr Jane Pitt from the Women and Infants Transmission Study (WITS) examined this phenomenon in a large group of 878 (92 infected) female and 892 (94 infected) male children born to HIV positive women [19].

Dr Pitt explained that although previous studies have suggested that there is no difference in surrogate markers in very young children, only small numbers of subjects have previously been studied [20]. She felt findings from this investigation could be particularly useful because: we have an accurate time of infection, hormonal environments are similar (apart from testosterone in the first month of life), social factors are similar, boys and girls ‘lifestyles’ are very similar in the first month of life.

They found that both relative and absolute CD4 counts were significantly higher in females compared to males, both in positive (p<0.05) and negative (p<0.0001) children, relative CD8 counts were significantly lower in uninfected (p<0.0001) and apparently lower in infected (p=0.2) females. Viral load levels were apparently but not significantly lower at all time points to 18 months in HIV-positive girls compared to boys (time averaged mean 0.4 log, this is similar to difference observed in viral load in adults). There were no differences in clinical disease progression.

The investigators concluded that significant and previously unrecognised gender differences in surrogate markers are present in children. Small differences in viral load similar to those seen in adults also appear to be present in HIV-positive children. And ‘Given a more similar hormonal environment in female and male children compared with adults, genetically programmed non-hormonal mechanisms should be considered in seeking explanations for gender differences.’ The clinical implications of these findings remain to be seen.

First paediatric ‘switch’ study

‘Simplification’ strategies are being investigated in adults; this study reported results from the first PI switch study in children [21]. The investigators explained that in general, children have a more difficult time maintaining viral suppression due to many factors including difficulties with adherence and availability of antiretrovirals in paediatric formulations.

15 children were enrolled in this study, all were <400 for 4 months on their PI-containing regimen for 6 months and all were NNRTI na•ve. Their PI was switched to efavirenz and their NRTIs were maintained. All children were heavily pre-treated, 87% had used 3 prior NRTIs and 47% 2 prior PIs. All children remained undetectable at a mean of 32 weeks and their triglyceride and cholesterol levels decreased. The study is ongoing.

Although these data seem encouraging, adult data has provided a clear warning against simplification for anyone with a prior history of treatment with mono or dual nucleosides [22,23]. Although detailed treatment histories were not available for the children it would seem prudent to suggest caution with simplification for any child with a similar prior treatment history of mono or dual therapy or a history of partially suppressive therapy whatever the combination.

T-20 in children

The first paediatric data for the fusion inhibitor T-20 was presented at this meeting. One presentation reported part A of the phase I/II study to evaluate the pharmacokinetics of T-20 following single dosing [24]. Based on preliminary adult PK data doses were chosen that were expected to result in trough plasma concentrations spanning 1000ng/ML. Each dose was administered to the 12 children in this dose escalation study by both subcutaneous (SC) and intravenous (IV) routes. The AUC for the IV route was greater than SC in the two lower dose cohorts (15mg/m2 and 30mg/m2 twice daily). But AUC for the 60mg/m2 twice-daily dose was greater after SC administration. Children receiving the 60mg/m2 SC dose achieved 12 hour T-20 trough levels of >1000ng/mL. This supports the future use of a 60 mg/m2 twice-daily SC dose for children.

Part B of the study added T-20 to failing triple combination therapy in 13 children with viral load >10,000 copies/mL [25]. By day 7 the 60mg/m2 group had achieved a mean log drop of 1.03 log, which was maintained in 9 children to 8 weeks. Adverse events were similar to adults ie injection site reactions which occurred in 77% (10/13) of children. No child discontinued due to adverse events but one child discontinued due to an aversion to injections.

Although this would not be an easy option for children (or indeed adults) it will be an invaluable agent for those who are already multi-drug resistant. A small paediatric expanded access programme (currently estimated at 50 patients worldwide) is expected to commence from summer 2001 to run simultaneously to that for adults.


A number of presentations looked at the association between pharmacokinetics and virological response in children. Although use of therapeutic drug monitoring (TDM) is integrated into paediatric care in both France and Holland, in the UK these tests are still under accessed.

Dr Courtney Fletcher presented findings from the PACTG 382 study [26] – an area under the curve controlled phase I/II study in children receiving nelfinavir, efavirenz and at least one NRTI. This study demonstrated significant association between virological effect and efavirenz and nelfinavir concentrations. The AUC of each drug had a significant independent contribution in producing virological response. Dr Fletcher concluded that these findings ‘provide motivation to ensure that paediatric dosing regimens achieve concentrations above threshold values for all children.’

Following the presentation he was asked what proportion of children needed dose adjustment, the reply was a substantial 50% of those on efavirenz and 20% of those on nelfinavir at week 2. Another questioner inquired whether he would advocate using TDM early and adjusting as early as possible. Dr Fletcher replied that ‘We would draw that observation, as starting doses seem to determine response to these two drugs.’

Dr Fletcher’s group also found that trough concentrations of saquinavir were predictive of virological success in a report from PACTG 397 [27]. Dr Luzuriga’s group reported variable concentrations of nelfinavir in children <2 years [28] and lower than predicted concentrations of ritonavir in children 1-24 months of age[29].

These findings continue to strongly substantiate arguments for measuring drug concentrations in paediatric patients of all age groups.

A call for resistance testing

Although there are still persistent question marks concerning the clinical use of both genotype and phenotype resistance testing, for adults these tests are now standard of care in Europe and the US. They are also available for children in some countries. Dr Schmidt’s group from Germany examined the incidence of resistance in their group of children, and took then unusual step of comparing the results to those obtained from an adult cohort [30].

Virological treatment failure appears to be more frequent in children than in adults, and the authors highlighted the following reasons – adherence to therapy seems to be more difficult, different pharmacokinetics may result in subtherapeutic drug levels, viral load is generally higher in children than in adults and the antiviral CTL response is less efficient in the first year of life. Since information on the prevalence of highly resistant viral strains may influence current guidelines for diagnostic and therapeutic management of children, their aim for this study was to evaluate drug resistant profiles in paediatric HIV infection.

Genotype and phenotype tests were performed on 46 samples from 35 children – age 2 months to 16.5 years (median 8.4 years), at the time of their first resistance test. Detailed drug histories were available for all the children. This revealed 15 (42.9%) of samples to be resistant against one group of antiretrovirals, 13 (37.1%) against two and 1 sample (2.9%) against all three classes. 11 follow up samples were obtained from 9 children and in 7 cases resistance had increased. 5 samples including 2 follow ups showed nucleoside multi-drug resistance. Results were then compared adult samples to analyse the frequency of key mutations.

Nucleoside multi-drug resistance was found to be more frequent in children than adults in this study. They believed that this was explained by the past (and sometimes current) common practice of using dual nucleoside therapy for children, which is no longer standard of care for adults.

In addition this was also likely to be provoked by insufficient suppression of viral replication, which appears to be more frequent in children than in adults. The investigators found that NNRTI and PI resistance was lower in children than in adults, although they speculated that, ‘A higher frequency, may only be a matter of time as suggested from the increase in resistance in most of the follow up samples’.

They explained that, although HIV infection may differ considerably between children and adults in its biology as well as treatment, and that this paediatric cohort may not reflect a standard population, however ‘relevant information may be extracted from the comparison of paediatric and adult resistance profiles because reasons for resistance testing were identical for each group’.

The investigators concluded that ‘ Since resistance testing recently proved to be beneficial for the management of adult HIV infection, the prognostic relevance of geno and phenotypic resistance testing should be prospectively analysed for optimising therapy in HIV-1 infected children.’ This study was aptly titled ‘A call for resistance testing.’

Dr Susan Eshleman’s group analysed resistance in children experiencing virological failure from the PACTG 377 [31]. In this study experienced children were randomised into 4 treatment arms using different combinations of d4T, 3TC, NVP, NFV or RTV. Children were screened at baseline and at failure.

Amongst their findings they reported that. The selection of NVP mutations was far less common among children receiving 4-drug NVP containing combinations. And that although NFV and RTV resistance was rarely detected at the time of failure, these mutations, as well as additional 3TC and NVP mutations were frequently selected in children who maintained their initial study regimen after virological failure.

Finally to end on an optimistic note Dr Lisa Frenkel and colleagues reported a beneficial effect for children in having the M184V/I at baseline prior to starting salvage therapy in the PACTG 366 trial [32]. Experienced children with and without the M184V/I mutations were evaluated for their response to a new regimen of a combination of 4 drugs. These were dictated by their prior use of PIs and/or NNRTIs and including at least 2 agents that they had not used previously. They investigators found that having the M184V/I mutations was associated with a greater chance of achieving undetectable viral load both at week 12 and week 24. They concluded that ‘The M184V/1 mutations may have a clinically beneficial effect in the suppression of viral replication in children during salvage therapy.’


Research into paediatric HIV has been relatively scant in comparison to that of adults. Although there is no question that children are different in some ways from adults, as many reports presented at this conference reveal – there are many lessons to be learnt from adult studies.

As Lisa Frenkel comments in the forward to the i-Base report “Paediatric HIV providers have the difficult task of evaluating research conducted in adults and deciding what aspects warrant translation, prior to the repetition of similar studies in children. Rapid implementation of certain findings to paediatric care seems warranted when there is a significant potential benefit to children, or when harm might result by ignoring the findings.”

We hope that the trend will continue for research to be conducted in parallel to that of adults and not lag behind sometimes for several years as has often been the case in the past.


  1. Dickover R et al. Early HIVenvGene Evolution in Perinatally Infected Infants Predicts the Rate of Disease Progression. 8th CROI. Feb 2001. Abstract 510
  2. Wiener J et al. Maternal Factors Influencing Pediatric HIV Disease Progression. 8th CROI. Feb 2001. Abstract 512
  3. Luzuriaga K. Early Antiretroviral Therapy and Immunopathogenesis of Vertical HIV-1 Infection. 8th CROI. Feb 2001. Abstract S14.
  4. Rosenberg ES, Walker BD. HIV type 1-specific helper T cells: A critical host defense. AIDS Res Hum Retroviruses 1998; 14 (suppl 2):S143-147.
  5. Scott ZA et al. HIV-1-Specific CD8+T Cells in Vertically Infected Infants: Early Responses and the Effects of Antiretroviral Therapy. 8th CROI. Feb 2001. Abstract 169.
  6. Persaud D. HIV-1 Dynamics and Cellular Reservoirs in Children. 8th CROI. Feb 2001. Abstract S13.
  7. Borkowsky W. Immunologic Reconstitution in Children and Adolescents Treated with HAART. 8th CROI. Feb 2001. Abs. S15.
  8. Weinberg et al. HIV-Infected Children on HAART Reconstitute Tetanus- Specific T Cell Responses without Booster Vaccination. 8th CROI. Feb 2001. Abstract 688.
  9. Amaya RA et al. Antiretroviral-Associated Lipodystrophy Syndrome in HIV- Infected Children. 8th CROI. Feb 2001. Abstract 649.
  10. Meneilly G et al. Metabolic and Body Composition Changes in HIV- Infected Children on Antiretroviral Therapy. 8th CROI. Feb 2001. Abstract 650.
  11. Vigano et al. HAART-Associated Lipodystrophy Is Not Correlated to Mitochondrial Abnormalities in PBLs from HIV-Infected Children. 8th CROI. Feb 2001. Abstract 651.
  12. Vigano et al. HAART-Associated Changes in Body Fat Distribution and Bone Mineral Loss Are Detectable in HIV-Infected Children Even in the Absence of Clinical Evidence of Lipodystrophy. 8th CROI. Feb 2001. Abstract 652.
  13. Danziger-Isakov A et al. Effects of Protease Inhibitors on Insulin, Insulin Resistance, and Lipids in Children and Adolescents. 8th CROI. Feb 2001. Abstract 653
  14. Gaughan DM et al. Avascular Necrosis of the Hip (Leggs-Calve-Perthes Disease [LCPD]) in HIV-Infected Children in Long-Term Follow-Up: PACTG Study 219. 8th CROI. Feb 2001. Abstract 638.
  15. Arpadi S et al. Decreases in Total Body Bone Mineral Content Progress with Age in HIV-infected Children. 8th CROI. Feb 2001. Abstract LB8.
  16. Vigano et al. HAART-Associated Bone Mineral Loss through Increased Rate of Bone Turnover in Vertically HIV-Infected Children.8th CROI. Feb 2001. Abstract LB9.
  17. Collins S. HTB March 2001 Vol. 2 no 2 p.30/31.
  18. Anastos et al. JAIDS May 2000
  19. Pitt J et al. Gender Differences in Immunologic and Virologic Markers in Children Born to Women Infected by Human Immunodeficiency Virus (HIV). 8th CROI. Feb 2001. Abstract 513
    20. Mofenson et al. Effects of Gender on Quantative Serum HIV-1 RNA in HV-Infected Children. 40th ICCAC. Sept 2000. Abstract 569.
    21. McComsey G et al. Is Simplification of HAART Safe in HIV-Infected Children? Frist Pediatric Switch Study.8th CROI. Feb 2001. Abstract679.
    22. Perrin L et al. Prior Treatment with Mono or Dual NRTIs before HAART as Predictor of Virological Failure in Simplified Abacavir-Based Triple NRTI Regimens: Results from the Simplified Maintenance Trial (SMT) and CNA30017 Study Team. 4th international Workshop on HIV Drug Resistance and Treatment Strategies June 2000, Sitges. Abstract 120.
    23. Masquelier et al. Mechanism of Virological Failure after Substitution of a Protease Inhibitor by Nevirapine in Patients with Controlled HIV-1 RNA. 4th international Workshop on HIV Drug Resistance and Treatment Strategies June 2000, Sitges. Abstract 119.
    24. Kosel et al. Pharmacokinetics (PK) of Selected Doses of T-20, a Fusion Inhibitor, in HIV-1- Infected Children. 8th CROI. Feb 2001. Abstract 726.
    25. Church J et al. Safety and Antiviral Activity of Chronic Subcutaneous Administration of T-20 in HIV-1-Infected Children. 8th CROI. Feb 2001. Abstract 681.
    26. Fletcher et al. Pharmacologic Characteristics of Efavirenz (EFV) and Nelfinavir (NFV) Associated with Virologic Response in HIV-Infected Children. 8th CROI. Feb 2001. Abstract 259
    27. Brundage RC et al. Pharmacokinetics of Saquinavir (SQV) with Nelfinavir (NFV) or Ritonavir (RTV) in HIV- Infected Children. 8th CROI. Feb 2001. Abstract 728
    28. Capparelli et al. Pharmacokinetics (PK) of Nelfinavir and Its Metabolite (M8) in HIV-Infected Infants Following BID or TID Administration. 8th CROI. Feb 2001. Abstract 729.
    29. Chadwick G et al. Early Therapy with Ritonavir (RTV), ZDV and 3TC in HIV-1-Infected Children 1-24 Months of Age. 8th CROI. Feb 2001. Abstract 677
    30. Schmidt B et al. Resistance Profiles in HIV-1-Infected Children: A Call for Resistance Testing. 8th CROI. Feb 2001. Abstract 469.
    31. Eshleman SH et al. Analysis of HIV-1 Drug Resistance in a Randomised Controlled Trial of a Combination of Nucleoside Analog Reverse Transcriptase (RT) Inhibitors Plus Nevirapine (NVP), Nelfinavir (NFV), or Ritonavir (RTV) in Stable Antiretroviral Therapy-Experienced HIV-Infected Children. 8th CROI Feb 2001. Abstract 468.
    32. Frenkel LM et al. HIV-1 Reverse Transcriptase (RT) M184V/I Improves the Rate of Suppression of Viral Replication by Salvage Therapy. 8th CROI, Feb 2001. Abstract 463.

Links to other websites are current at date of posting but not maintained.