Changing Antiretroviral Therapy

When to Change Antiretroviral Therapy
The following three reasons warrant a change in antiretroviral therapy: a)failure of the current regimen with evidence of disease progression based on virologic, immunologic, or clinical parameters; b) toxicity or intolerance to the current regimen; and c) new data demonstrating that a drug or regimen is superior to the current regimen (Table 9). When therapy must be changed because of treatment failure or suboptimal response to treatment, clinicians should work with families to assess the possible contribution of adherence problems to the failure of the current regimen. Issues regarding adherence should be addressed to increase the likelihood of a successful outcome when initiating a new therapy. These issues are best addressed before therapy is instituted and need to be reinforced during therapy.

Intensive family education, training in the administration of prescribed medications, and discussion of the importance of adherence to the drug regimen should be completed before initiation of new treatment. In addition, frequent patient visits and intensive follow-up during the initial months after a new antiretroviral regimen is started are needed to support and educate the family and to monitor adherence, tolerance, and virologic response to the new regimen.

Virologic Considerations for Changing Therapy
Information is limited regarding HIV RNA response to antiretroviral therapy in infants and young children. However, the general virologic principles underlying the use of antiretroviral therapy are similar for all HIV-infected persons. Because HIV RNA monitoring is critical for the management of infected children, Working Group members used the available data and clinical experience when definitive data were not available to make the following recommendations. These recommendations may require modification as new information becomes available.

Ideally, antiretroviral therapy should maximally suppress viral replication to below levels capable of being detected with HIV RNA assays -- which may not always be achievable in HIV-infected children. Perinatally infected children generally have high HIV RNA levels, and clinical benefit may be observed with decrements in HIV RNA levels that do not result in undetectable levels. However, failure to maximally suppress replication may be associated with increased probability of viral mutations and the emergence of drug resistance. Consideration of the implications of changing regimens and the choice of new drugs should include an acknowledgment of the potential for limiting the patient's future options for potent therapy.

Consensus recommendations have been developed using plasma HIV RNA measurements to guide changes in antiretroviral therapy for HIV-infected adults (5). The recommendations for adults state that health-care providers should consider changing therapy if a) HIV RNA levels drop less than threefold (0.5 log₁₀) after 4 weeks of therapy and less than 10-fold (1.0 log₁₀) after 8 weeks of therapy or b) HIV RNA has not decreased to undetectable levels after 4-6 months of therapy. Because HIV RNA levels in infants who are perinatally infected are high compared with levels observed when therapy is initiated in most infected adults, the initial virologic response of infected infants and young children to initiation of antiretroviral therapy may take longer than observed in adults (i.e., 8-12 weeks). In addition, suppression of HIV RNA to undetectable levels may be achieved less often than has been reported for infected adults despite potent combination therapy with two NRTIs and a protease inhibitor. Therefore, virologic indications for changing therapy in infected children differ slightly from those recommended for infected adults. Adult guidelines should be followed for infected adolescents.

Virologic response should be initially assessed 4 weeks after therapy is initiated. However, the time required to achieve maximal virologic response to therapy may vary depending on the specific baseline HIV RNA value at the time of starting therapy. If baseline HIV RNA levels are high (i.e., >1,000,000 copies/mL), virologic response may not be observed until 8-12 weeks after initiating antiretroviral therapy. However, if baseline HIV RNA levels are more similar to those observed in untreated infected adults (i.e., <100,000 copies/mL), initial response should be observed within 4 weeks following initiation of therapy. After a maximal virologic response is achieved, HIV RNA levels should be measured at least every 3 months to monitor continued response to therapy. At least two measurements (taken 1 week apart) should be performed before considering a change in therapy. The following situations may indicate a need for change in therapy in infected children:

• Less than a minimally acceptable virologic response after 8-12 weeks of therapy. For children receiving antiretroviral therapy with two NRTIs and a protease inhibitor, such a response is defined as a <10-fold (1.0 log₁₀) decrease from baseline HIV RNA levels. For children who are receiving less potent antiretroviral therapy (i.e., dual NRTI combinations), an insufficient response is defined as a less than fivefold (0.7 log₁₀) decrease in HIV RNA levels from baseline.
• HIV RNA not suppressed to undetectable levels after 4-6 months of antiretroviral therapy. However, although suppression of HIV RNA to undetectable levels and maintenance for prolonged periods is desirable, few data among children indicate that such suppression is always achievable. In addition, the number of alternative therapeutic regimens for children is limited. The initial HIV RNA level of the child at the start of therapy and the level achieved with therapy should be considered when contemplating potential drug changes. For example, an immediate change in therapy may not be warranted if there is a sustained 1.5 to 2.0 log₁₀ fall in HIV RNA copy number, even if RNA remains detectable at low levels.
• Repeated detection of HIV RNA in children who initially had had undetectable levels in response to antiretroviral therapy. The presence of repeatedly detectable RNA suggests the development of resistance or problems with adherence or drug bioavailability. More frequent evaluation of HIV RNA levels should be considered if the HIV RNA increase is limited (e.g., if using an HIV RNA assay with a lower limit of detection of 1,000 copies/mL, there is a <0.7 log₁₀ increase from undetectable to approximately 5,000 copies/mL in an infant aged <2 years). If adherence to therapy has been inconsistent, renewed efforts to educate the caregivers and patient and closer follow-up from members of a multidisciplinary care team may improve adherence.
• A reproducible increase in HIV RNA copy number among children who have had a substantial HIV RNA response but still have low levels of detectable HIV RNA. Such an increase would warrant a change in therapy if, after initiation of the therapeutic regimen, a greater than threefold (>0.5 log₁₀) increase in copy number is observed in children aged >2 years. Because of the greater biologic variability in RNA in young children, a change in therapy is warranted when a greater than fivefold (>0.7 log₁₀) increase is observed for children aged <2 years.

Immunologic Considerations for Changing Therapy
CD4⁺ T-lymphocyte count and percentage are independent predictors of disease progression and mortality in HIV-infected children (35,36). The association of HIV RNA and CD4⁺ percentage with long-term mortality risk in HIV-infected children has been evaluated; for each absolute decline of five percentiles in CD4⁺ percentage at baseline or during follow-up, the mortality risk ratio increased by 1.3 (95% CI=1.2-1.5), independent of the child's HIV RNA level (35). For children with CD4⁺ percentages of <15% (i.e., those in immune category 3), prognosis also was correlated with the degree of depression of CD4⁺ percentage (i.e., life expectancy was less for children with CD4⁺ percentages of <5% compared with children with CD4⁺ percentages of 10%-14%) (Table 3).

Before considering changing antiretroviral therapy because of a decline in CD4⁺ lymphocyte values, a minimum of one repeated measurement of CD4⁺ values should be obtained at least 1 week after the initial test. The following are immunologic indications that may warrant a change in antiretroviral therapy for HIV-infected children:

• Change in immune classification (Table 1). However, minimal changes in CD4⁺ percentile that may result in a change in immune category (e.g., from 26% to 24% or from 16% to 14%) may not be as concerning as a rapid substantial change in CD4⁺ percentile within the same immune category (e.g., a decrease from 35% to 25%).
• For children with CD4⁺ percentages of <15% (i.e., those in immune category 3), a persistent decline of five percentiles or more in CD4⁺ cell percentage (e.g., from 15% to 10% or from 10% to 5%).
• A rapid and substantial decrease in absolute CD4⁺ T-lymphocyte count (e.g., a >30% decline in <6 months).

Clinical Considerations for Changing Therapy
The occurrence of certain clinical events while receiving antiretroviral therapy is evidence of HIV disease progression and/or a poor prognosis for infants and children. The following clinical criteria warrant consideration of a change in antiretroviral therapy:

• Progressive neurodevelopmental deterioration (i.e., persistence or progression of deterioration documented on repeated testing as demonstrated by the presence of two or more of the following findings: impairment in brain growth, decline of cognitive function documented by psychometric testing, or clinical motor dysfunction). In such cases, the new treatment regimen optimally should include at least one antiretroviral drug with substantial central nervous system penetration (e.g., ZDV or nevirapine, which have cerebrospinal fluid/plasma ratios >0.5).
• Growth failure (i.e., persistent decline in weight-growth velocity despite adequate nutritional support and without other explanation).
• Disease progression (i.e., advancement from one pediatric clinical category to another (Table 2)). Prognosis is poorer as patients progress to more advanced clinical categories (59). However, in patients with stable immunologic and virologic parameters, progression from one clinical category to another (e.g., from clinical category A to category B) may not represent an indication to change therapy. For example, development of new opportunistic infections, particularly in patients who had severe immunosuppression at the time therapy was initiated, may not reflect a failure of antiretroviral therapy but persistence of immunologic dysfunction despite adequate antiviral response. Thus, in patients whose disease progression is not associated with neurologic deterioration or growth failure, virologic and immunologic parameters should be considered when deciding whether to change therapy.

Choice of a New Antiretroviral Regimen
The choice of a new antiretroviral regimen is dictated by the indications that warranted the change in therapy and the limited available alternative antiretroviral agents. Although the efficacy of different combination antiretroviral regimens in children probably can be extrapolated from clinical trial data obtained for adults, data are limited regarding the pharmacokinetics, appropriate dosing, and short- and long-term safety of various combinations in infected children. New regimens should be chosen partly on the basis of the impact of the changes on future treatment options.

The following principles should be followed when choosing a new antiretroviral regimen in children who have received prior treatment.

• When therapy is changed because of toxicity or intolerance, agents with different toxicity and side-effect profiles should be chosen, when possible. Health-care providers should have comprehensive knowledge of the toxicity profile of each agent before selecting a new regimen. In the event of drug intolerance, change of a single drug in a multidrug regimen and, in certain circumstances, dose reduction are permissible options. owever, antiretroviral drugs should only be reduced to the lower end of the therapeutic range for those antiretrovirals for which an effective dosing range is known, and adequacy of antiretroviral activity should be confirmed by the monitoring of HIV RNA levels.
• When changing therapy because of treatment failure (Table 9), adherence to therapy should be assessed as a potential cause of failure.
• If the patient is adherent to the prescribed drug regimen, assume the development of drug resistance and, if possible, change at least two drugs to new antiretroviral agents.Change in one drug or addition of a drug to a failing regimen is suboptimal. The new regimen should include at least three drugs, if possible. The potential for cross-resistance between antiretroviral drugs should be considered in choosing new drugs.
• When considering changing to a new regimen, all other medications taken by the patient should be reviewed for possible drug interactions.
• A change to a new regimen, especially one containing protease inhibitors, must include a discussion of treatment adherence issues between the caregivers of the infected child and the health-care provider. The health-care provider must recognize that certain medications are difficult to take in combination because of exacting and often conflicting requirements with respect to whether they can be taken with food and other antiretrovirals.
• When changing therapy because of disease progression in a patient with advanced disease, the patient's quality of life must be considered.

Antiretroviral Drug Resistance Testing
It is important to distinguish between the need to change therapy due to drug failure versus drug toxicity or poor compliance. Viral resistance to antiretroviral drugs is important, but not the only reason for treatment failure. The goal of antiretroviral therapy should be to reduce plasma HIV RNA to below detection of the most sensitive assay available (<50 copies/mL). Accomplishing optimal viral suppression will reduce the likelihood that genetic/phenotypic resistance will emerge.

Genotypic assays are available for detecting specific HIV genetic variants (mutations). They are based on amplification procedures and can usually detect mutations in plasma samples with more than 1000 copies/mL of HIV RNA. Expert clinical interpretation is required to determine if genomic variations coincide with changes known to be associated with antiretroviral resistance. A compilation of the most common HIV-1 mutations selected by the three classes of antiretroviral agents is available on the Internet at http://hiv-web.lanl.gov.

Phenotypic assays measure the 50% or 90% inhibitory concentrations of a drug against the virus in vitro. These assays historically have been cumbersome and time-consuming. However, more rapid assays based on recombinant DNA technology are being developed. Although phenotypic assays provide important information regarding the sensitivity patterns of the dominant virus tested, minor species of resistant viruses may be missed.

HIV resistance assays may prove useful in guiding initial therapy and in changing failing regimens. However, the value of phenotypic or genotypic assays in guiding treatment has not been established in children. Moreover, standardization of such assays will be necessary before guidelines can be established for the incorporation of these assays into clinical care. Epidemiological surveys are needed to monitor the prevalence of resistant viruses in specific pediatric populations.

Therefore, specific recommendations cannot be made at this time regarding use of resistance assays for directing antiretroviral drug choices in children. However, if resistance testing is performed to determine its contribution to drug failure, these assays should be done while the child is still receiving the antiretroviral drugs. In the absence of antiretroviral drug pressure, wild type virus is likely to replace strains that could mask the presence of resistant virus.

It should be noted that the presence of viral resistance to a particular drug suggests that the specific drug(s) is unlikely to be successful in suppressing viral replication. However, the absence of resistance to a drug does not insure that its use will be successful, particularly if that drug or drugs share cross-resistance with drugs previously used. Although initial studies performed in adults suggest that drug-resistance genotyping modestly improves the response to antiretroviral therapy as reflected by a decline in HIV RNA levels below 200 copies/ml (78), there are no long-term data on the impact of such testing. Moreover, no controlled clinical trials, to date, have been performed in children that assess the benefits of genotypic or phenotypic resistance.

previous page | next page

return to table of contents