UK treatment centres are to join a major international study of HIV treatment in children which will test whether it is better to start treatment with a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor, and whether it is better to switch therapy if viral load rises above 1,000 copies, or wait til it has risen much higher (above 30,000). Dr Gareth Tudor-Williams of St Mary's Hospital, London, spoke to aidsmap about the trial (an edited version of this interview appears in the September edition of AIDS Treatment Update).
AIDS Treatment Update: How did the proposed collaboration between PENTA (Paediatric European Network for Treatment of AIDS) and the US PACTG (Paediatric AIDS Clinical Trial Group) come about?
Gareth Tudor-Williams: Investigators from PENTA/PACTG recognise that there are still many unanswered questions regarding the initial treatment of HIV infected children which justify the development of new treatment protocols. However, both in North America and Europe the number of infected but previously untreated children is fortunately dwindling due in large measure to the success of antenatal screening and prevention of mother/infant transmission. It was recognised that it would not be possible to recruit sufficient numbers of children in a timely fashion in the USA or Europe alone to answer clinically relevant questions.
The PENTA/PACTG have always communicated to ensure that complementary rather than duplicate studies are being undertaken on either side of the Atlantic. Over the past eighteen months a dialogue has developed between several members of both groups and has identified questions of critical importance to the management and treatment of naive children. Potentially available numbers of such children have been assessed. The resulting protocol represents a consensus on the highest priority questions that need answering although inevitably this has involved compromise on both sides and it isn't hard to see other questions that larger numbers could address.
ATU: What are these questions?
GTW: The trial is designed to answer the questions of what treatment to start with and when to switch therapy in children who are previously untreated. It will allow comparison of two different nucleoside reverse transcriptase combinations as initial therapy with and without a protease inhibitor.
The optimal criteria for switching therapy have yet to be defined. There is a logical argument in favour of "tight" control as soon as the virus becomes detectable in the circulation (or if plasma virus load fails to reach undetectable limits by week 24) then a switch should be made to a different and probably more intense regimen.
The problem with this approach given the limitations of drugs available to children is that this policy may rapidly use up all available options for therapy. An alternative approach that may be equally valid over a long period of follow-up would be to try and maximise the benefit of each regimen, and only switch when the viral load is consistently above a much higher threshold. This will preserve options for longer and there is some evidence that virus replicating despite HAART may be less pathogenic in vivo.
Much discussion has taken place around the higher threshold. Below 1,000 copies/ml has generally been associated with slow disease progression in children. An upper threshold needed to be at least 1 log above this but it was felt that 100,000 copies/ml was unacceptably high and a threshold of 30,000 - 50,000 copies/ml is not uncommonly used in current clinical practice.
The long-term nature of this study should clarify whether early switching improves immunological and virological outcome or whether the children simply run out of treatment options more rapidly.
ATU: Can you outline the study design?
GTW: The study is a randomised open-label trial for children between one month and sixteen years of age. They must be naive to antiretroviral therapy except for exposure to drugs used to prevent perinatal transmission (usually AZT for three to six weeks postnatally). There is an initial randomisation to two NRTIs [either d4T/ddI or 3TC/abacavir] plus PI [nelfinavir] versus two NRTIs and an NNRTI [efavirenz]. If a child needs second-line therapy this allows switching to the opposite arm [i.e. switch to the other two NRTIs and swap the PI for the NNRTI or vice versa, with the option of adding a fourth drug at the discretion of the physician]. A second randomisation determines the criteria for such a change in therapy based on viral load. Half will be randomised to ‘tight’ control, i.e. as soon as their viral load rises to above 1,000 copies they would be switched, whilst the other half would stay on their first-line treatment until their viral loads rose above 30,000 copies. The primary outcome measure will be the viral load at four years from randomisation.
This double randomisation results in eight groups, and 256 children will need to be enrolled over a twelve month period (half of whom will be from Europe).
ATU: What are the potential pitfalls in embarking on such a long and large study?
GTW: The trial is designed to run over the next five years and realistically will not start enrolling until the end of this year. The obvious pitfall is that, by the end of the study, new drugs will have become available that will render our initial treatment arms obsolete. Some flexibility will have to be maintained throughout the study particularly for the treatment regimes to which children may be switched although the principal that all three drugs should be changed will be maintained.
Another potential pitfall is that parents and carers of children may be daunted by the prospect of a study that lasts so long. However the reality is that any course of treatment for a child is a long-term proposition whether or not they are enrolled in a clinical trial. It is obviously critical that the trial design incorporates sufficient flexibility to ensure that no child would be prevented from taking advantage of any new developments particularly with regards to second-line therapy during the course of this study.
ATU: What are the key factors which have influenced the development of paediatric care at St Mary's in the recent past?
GTW: With the advent of HAART for children we have witnessed a dramatic decline in hospital admissions and a shift in the focus of the multi-disciplinary team towards finding creative ways of supporting long-term adherence. We are not inclined to under-estimate the difficulties of maintaining children on regular long-term treatment and thus the crucial decision about when to start treatment is highly individual and determined by the multi-disciplinary team. Even when a child presents with symptomatic disease or clear cut evidence of declining CD4 counts or a viral load above 100,000, many other factors may need to be resolved before therapy can be successfully initiated. For many families, the child may be the first individual to present with HIV-related illness and thus the diagnosis has major ramifications for the adults and other siblings. Often the families have very little understanding of HIV and will require patient explanation over more than one visit. Frequently the families are facing severe social disruption and economic and housing problems; if their lives are in turmoil and they don't, for example, have a fridge it is naive to imagine that they can successfully maintain a child on long-term therapy just because we provide the prescriptions!
At the opposite end of the spectrum there are some families who are extremely well-informed and even though the child may be asymptomatic with relatively low viral loads and preserved immune function we would consider starting treatment with a view to maintaining long-term complete viral suppression which offers the child the best hope for long-term preservation of the immune system and probably for survival.
As for the choice of drugs we almost invariably start with three different agents and only in exceptional circumstances have we recommended quadruple therapy. To fit better with family lifestyles treatment regimens are all constructed on a twice daily basis except for the newborn infants for whom zidovudine is recommended in the first few weeks of life, four times a day, in order to help prevent mother/infant transmission of HIV.
As a class, the protease inhibitors are generally quite insoluble and difficult to manufacture in a palatable and stable solution. This is a real problem for the younger children who cannot swallow pills or capsules. We are therefore tending to adopt PI-sparing first line treatment regimens by either using two NRTI's plus an NNRTI (usually nevirapine), or more recently we have had good experiences using a triple NRTI combination of zidovudine, 3TC and abacavir. We have to recognise the low but significant risk of abacavir hypersensitivity in view of the serious consequences of failing to recognise such an event, and rechallenging the child, which involves careful parent education.
Either of these options still leaves the way open for switching to two new nucleoside analogues and adding a protease inhibitor as second-line therapy. Third-line and subsequent treatment switches are obviously very limited for children and sometimes involve radical solutions such as the insertion of gastrostomy tubes which may be needed anyway for nutrition but which also provide a route for giving highly unpalatable solutions such as ritonavir. It is in recognition of these longer-term treatment problems that so much of the effort of the multi-disciplinary team is directed towards helping families give first and second-line treatments on an absolutely regular basis so that the children can hopefully gain as long-term benefit as possible from these regimens.
ATU: The UK is relatively fortunate to have only a small number of HIV-infected children. How does this influence the care they receive?
GTW: It is true that the number of HIV-infected children in the UK is small in relation to the global epidemic. This has appropriately led to just a few centres developing comprehensive services for children and their families. These centres however now look after sizeable numbers of children. For example, here at St Mary's we are involved in the care of over 100 infected children which is as large a cohort as many US centres. By actively participating in the design and execution of international trials, and by attending clinical meetings, we have been able to keep abreast of developments and improvements in the management of children and adults with HIV infection world-wide. The available treatment for children in the UK is therefore comparable to that available in the USA or any other industrialised country. In many respects the fact that we are not overwhelmed with numbers has enabled us to provide more individualised care and achieve better long-term follow-up rates than in the larger US centres.
In keeping with many sub-specialities within paediatrics such as cardiology and oncology, we have had to develop a pattern of shared care with paediatric colleagues at district hospitals who look after small numbers of HIV infected children. This typically involves a joint clinic in which a Consultant from our centre will see children jointly with a Consultant at the local hospital every one to three months. The children may be reviewed by the multi-disciplinary team at the centre once a year but interim care and drug supply can be maintained locally. At St Mary's we currently have such an arrangement with the North Middlesex Hospital, Northwick Park, Royal Free, Lewisham, Ealing and the Chelsea & Westminster Hospitals. We provide care for children from a very much wider area outside London and typically these children will travel down with their families once every three months to be seen at St Mary's. Such need for travel is obviously a disadvantage but it is obviously impractical to develop multi-disciplinary teams to provide care for only one or two children.
ATU: Are there any other trials open to children in the UK at present? Can you say anything about PENTA 5's progress?
GTW: PENTA 7, which is a trial for HIV-infected infants is imminently about to open to enrolment but there are no other trials currently open for children in the UK. PENTA 5 is fully recruited and the last children enrolled have now reached week 24. The trial is therefore scheduled to finish at the end of September with a view to completing data analysis by the end of the year, with a hope of presenting this as a late breaker at the Retroviruses meeting in February 2000.
In addition there are plans for a trial for treatment of experienced children and in particular focusing on the question of the role of genotypic and phenotypic resistance assays in determining second and third-line treatment.
ATU: What’s the most important thing that AIDS researchers have learnt about HIV paediatric management in recent years?
GTW: From the point of view of HIV in children it is the advances in prevention of mother/infant transmission that have had the most dramatic impact in recent years. As the September issue of AIDS Treatment Update highlights we are now in a position to reduce mother/infant transmission from approximately 30% to 1% or less in developed countries.
As far as the management of infected children go, I think recent recognition that children metabolise these drugs differently to adults, and we therefore need to pay more attention to therapeutic drug monitoring and improved individual and population pharmacokinetics, will be an important advance.
ATU: What is the most important thing that AIDS researchers have failed to learn?
GTW: Over and over again in the clinic we face the reality that however potent the inhibitors of HIV may be in vitro they will not work in children unless they can be formulated in a way that is tolerable for very long-term use. Thus, AIDS researchers, which in its broadest sense I take to include pharmaceutical industry, need to continue the efforts at finding ways to formulate drugs, particularly the protease inhibitors, for use in younger children. It would be a tragic mistake to consider this a diminishing market simply in view of current trends in developed countries. The reality is that well over 1,600 children are infected each day world-wide and finding effective and affordable strategies to help them remains a very serious challenge.