Original Article
A Multicenter Trial of Oral Zidovudine in Children with Advanced Human Immunodeficiency Virus Disease
N Engl J Med 1991; 324:1018-1025April 11, 1991
- Abstract
- Abstract
Background and Methods.
Zidovudine has been shown to be an effective antiretroviral treatment in adults with human immunodeficiency virus (HIV) infection. We examined the safety of zidovudine and the tolerance of and therapeutic response to the drug in 88 children with advanced HIV disease. During a 24-week outpatient trial, zidovudine (180 mg per square meter of body-surface area per dose) was given by mouth every six hours and serial measurements were made of clinical, immunologic, and virologic indexes. Children who completed 24 weeks of treatment were permitted to continue receiving zidovudine.
Results.
Of the 88 children (mean age, 3.9 years; range, 4 months to 11 years), 61 completed the initial 24-week trial, and 49 continued to receive zidovudine for up to 90 weeks (median follow-up, 56 weeks). The patients generally tolerated zidovudine well. One or more episodes of hematologic toxicity occurred in 54 children (61 percent) — anemia (hemoglobin level, <75 g per liter) in 23 children (26 percent) and neutropenia (neutrophil count, <0.75×109 per liter) in 42 (48 percent). Many of these abnormalities resolved spontaneously, but 30 children required transfusions or a modification of the dose of zidovudine. Only three children had to stop receiving the drug because of hematologic toxicity. Kaplan–Meier analysis demonstrated that the probability of survival was 0.89 after 24 weeks and 0.79 after 52 weeks. There was marked improvement in weight gain, cognitive function (mainly in children <3 years old), serum and cerebrospinal fluid concentrations of p24 antigen, and the proportion of cerebrospinal fluid cultures negative for HIV. CD4+ lymphocyte counts (mean at base line, 0.263×109 per liter) improved during the first 12 weeks, although the improvement was not sustained through the 24th week.
Conclusions.
Zidovudine in a dose of 180 mg per square meter every six hours can be safely administered to children with advanced HIV disease. The resultant clinical, immunologic, and virologic improvements in children are similar to those reported with zidovudine in adults. (N Engl J Med 1991; 324:1018–25.)
Media in This Article
Figure 1
Kaplan–Meier Curve for Survival during the 24-Week Trial and Follow-up.Figure 2
Percentile Values for Weights and Weight-Gain Rates in 57 Children.Article Activity
- Article
PLACEBO-controlled trials in adults have demonstrated that zidovudine (formerly known as azi-dothymidine, or AZT) is effective for the treatment of human immunodeficiency virus (HIV) disease, including asymptomatic infection.1 2 3 4 5 6 Phase I studies in a small cohort of children indicated that zidovudine was well tolerated and had pharmacokinetic properties similar to those in adults.7 8 9 10 We present data on the safety of zidovudine in 88 symptomatic children with HIV infection enrolled in a multicenter phase II study and on their tolerance of the drug. In addition, we present analyses of the clinical, immunologic, and virologic variables evaluated during the study.
Methods
Patient Population
This outpatient study was performed at nine centers under the auspices of the AIDS Clinical Trials Group and the Burroughs Wellcome Company. Children with HIV infection who were 3 months to 12 years of age were enrolled, and all had either the acquired immunodeficiency syndrome (AIDS), as indicated by conditions defined by the Centers for Disease Control,11 or evidence of advanced HIV disease, as indicated by two or more of the following conditions, or one of the following conditions and one of the secondary conditions: failure to thrive, persistent or recurrent oral candidiasis, or a CD4+ lymphocyte count ≤0.5 × 109 per liter. The secondary conditions were severe diarrhea, hepatomegaly, splenomegaly, cardiomyopathy, nephropathy, two or more episodes of herpes stomatitis within one year, or two or more episodes of recurrent varicella–zoster or chronic zoster infection. Children with lymphocytic interstitial pneumonitis were enrolled only if they had another AIDS-defining condition or met the clinical criteria for advanced HfV disease. HIV infection was documented by culture, the detection of p24 antigen in the serum, or the detection of HIV antibody with confirmation by Western blot testing. All participants had to have at least one of the following immunologic abnormalities: dysgammaglobulinemia, a CD4+ lymphocyte count less than or equal to 0.5×109 per liter, a CD4+:CD8+ ratio less than or equal to 1.0, decreased lymphocyte proliferative responses to at least one mitogen, or cutaneous anergy if the child was more than one year of age.
Children with acute opportunistic or bacterial infections requiring parenteral therapy were excluded from entry until their infections resolved. Children were also excluded if they had received any antiretroviral drug, including zidovudine, within two months, intravenous immune globulin within one month, or any experimental drug or agent within two weeks of entry that could cause prolonged neutropenia or severe renal dysfunction. Also excluded were children with any of the following abnormal laboratory values (after adjustment for age): a serum total bilirubin level more than three times the upper limit of normal, a serum aspartate aminotransferase concentration more than five times the upper limit of normal in the presence of an abnormal bilirubin level, an estimated creatinine clearance less than 50 ml per minute per 1.73 m2 of body-surface area, a total white-cell count less than 1.5 × 109 per liter, a neutrophil count less than 0.8× 109 per liter, a hematocrit less than 0.24, or a hemoglobin level less than 80 g per liter.
Patients were enrolled between April 1988 and January 1989. Informed consent was obtained from the parent or guardian of each participant; approval of the protocol and consent form by each center's institutional review board was required.
Evaluation and Follow-up
Participants were seen at least once before entry (during screening), at entry, weekly during the first month, biweekly for the next two months, and monthly thereafter. Vital signs and weight were recorded at each visit. A comprehensive physical examination, including documentation of growth and age-appropriate neurologic assessment, was performed at screening, at entry, and then every eight weeks. Age-appropriate neuropsychological and psychomotor evaluations were performed at entry and at week 24. All infections, concurrent medications, receipt of blood products, and adverse events were recorded at each scheduled visit.
A complete blood count was performed at every visit. Serum biochemical values and the results of liver-function tests were assessed at screening, at entry, and then every four weeks. Lymphocyte subsets were determined at screening, at entry, and at weeks 4, 8, 12, and 24. HIV culture, serologic testing for p24 antigen, and measurement of immunoglobulin concentrations were performed at entry and at weeks 12 and 24. A lumbar puncture to obtain cerebrospinal fluid for HIV culture and p24 antigen determination was done at entry and at week 24. Computed tomography of the head, chest radiography, electrocardiography, and echocardiography were also performed at entry and at week 24.
The planned duration of the trial was 24 weeks, although children were allowed to continue receiving therapy and were followed after this period, fn patients followed after week 24, survival was assessed and modifications of the dose of zidovudine, complete blood counts, serum chemical values, and vital signs were recorded. These data were collected from October 1988 to January 1990.
Drug Administration
Zidovudine was administered in a dose of 180 mg per square meter orally every six hours, as a sterile, filtered aqueous solution mixed with water or fruit juice, or as 100-mg capsules. It was recommended that the dose be reduced to 120 mg per square meter if severe anemia (hemoglobin, <80 g per liter), neutropenia (neutrophil count, <0.5× 109 per liter), or thrombocytopenia (thrombocyte count, <50×109 per liter) developed. Transfusions could be given as an alternative to dose reduction in cases of anemia. Zidovudine was temporarily discontinued if progressive bone marrow dysfunction occurred (dependence on transfusion for more than four weeks, neutropenia [<0.5×l09 neutrophils per liter] persisting more than seven days after dose reduction, or thrombocytopenia [<25×109 thrombocytes per liter] lasting more than seven days after dose reduction). Drug administration was resumed if hematologic values rose above predefined levels — i.e., the hemoglobin level to more than 80 g per liter, the neutrophil count to more than 0.75× 109 per liter, or the platelet count to more than 75×109 per liter.
A change in the mean corpuscular volume was used as a screening mechanism to identify possible noncompliance with drug administration. In addition, the parents or guardians of the children were required to return all dispensed vials or bottles at each clinic visit so that investigators could determine drug use.
The use of concurrent medications that could affect hepatic glucuronidation, the principal route of zidovudine metabolism, was discouraged. Children who had recovered from a first episode of Pneumocystis carinii pneumonia were allowed to continue receiving prophylactic treatment for this condition, but primary prophylaxis was prohibited, as was prophylaxis for candidiasis or otitis media. Intravenous immune globulin therapy could be initiated if a child had three or more serious bacterial infections during the course of the study.
Weight-Gain and Growth Velocity
The base-line weight was defined as the average of the values at screening and entry. Rates of weight gain, normalized to a 6-month interval (182 days), were calculated for the first and second 6-month periods if more than 120 days of data were available for that interval. The rates were compared with normal values for age and sex.12
Tests of Cognitive Function
The neuropsychological tests used were the Bayley Scales of Infant Development 13 (in children 3 to 30 months old), the McCarthy Scales of Children's Abilities14 (31 months to 6 years), and the Wechsler Intelligence Scale for Children, Revised15 (≥6 years). Cognitive function was reflected by the mental index of the Bayley Scales, the cognitive index of the McCarthy Scales, and the full-scale IQ of the Wechsler Scales. The scores from each test were transformed in order to be comparable with a standard score mean of 100 and a standard deviation of 15.
Immunologic and Virologic Testing
Lymphocyte subclasses were measured by standard flow-cyto-metric methods. Levels of p24 antigen were determined on a realtime basis with the use of commercially available reagents. A positive test was indicated by a level above 30 ng per liter. HIV cultures were performed on peripheral-blood mononuclear cells according to standard techniques.16
Statistical Analysis
The sign test (two-tailed) was used in the analysis of HIV-culture and p24 antigen measurements. The Wilcoxon signed-rank test (two-tailed) was used in the analysis of the CD4+ cell count and neuropsychological data. Kaplan–Meier survival estimates (±SE) were obtained with the PROC LIFETEST of the SAS System (SAS Institute, Research Triangle Park, N.C).
Results
Study Population
Eighty-eight children were enrolled in the study. All but five patients met the eligibility requirements: two had hemoglobin levels below 80 g per liter, one had a neutrophil count below 0.8 ×109 per liter, and two had received intravenous immune globulin within four weeks of entry. All 88 patients were included in the data analysis according to the intention-to-treat principle.
The characteristics of the study participants are shown in Table 1Table 1
Demographic Characteristics of 88 Children Treated with Zidovudine for HIV Infection.. The mean and median ages at entry were 3.9 and 3.1 years, respectively (range, 4 months to 11 years). There were 39 girls (44 percent) and 49 boys (56 percent). Forty-three children (49 percent) were American black or Haitian, 12 (14 percent) Hispanic, and 33 (38 percent) non-Hispanic white. Sixty-three children (72 percent) met the. criteria for AIDS defined by the Centers for Disease Control, and the remaining 25 had other manifestations of advanced HIV disease. Thirty-one (49 percent) of the 63 patients with AIDS had more than one AIDS-defining condition before enrollment. Most study participants also had laboratory evidence of severe immunodeficiency before enrollment. Ninety-two percent had CD4+:CD8+ ratios less than 1.0, 63 percent had a CD4+ count less than 0.5×X109 per liter, 92 percent had abnormal quantitative immunoglobulin concentrations, and 90 percent of children more than one year of age were anergic. Only three children routinely received intravenous immune globulin therapy during the course of study.At the time of cutoff (January 1990), the children had received zidovudine for a median of 390 days (range, 3 to 627). Sixty-one children (69 percent) completed the 24-week study. As of January 1990, 49 children remained in the extended study. Fourteen children died, 10 within the first 24 weeks. Twelve children were voluntarily withdrawn from the study, five of whom continued to receive zidovudine outside the study. Three other children were withdrawn because of hematologic abnormalities, one because of a violation of entry criteria (base-line neutrophil count, <0.8× 109 per liter), and one because of disease progression. Twelve parents or guardians were known or suspected to be noncompliant with the administration of zidovudine to their children exactly as prescribed; eight of the children were withdrawn from the study because they did not return for follow-up visits, three before the 24th week.
Drug Tolerance
Zidovudine was generally well tolerated during the study period. Twenty adverse clinical events were reported in 15 children, including vomiting (5 percent), insomnia (3 percent), nonspecific changes on electrocardiograms (3 percent), echocardiograms consistent with cardiomyopathy (2 percent), nervousness (2 percent), and abdominal pain (2 percent). Acute abdominal syndrome, headache, nausea, and weight loss each occurred in one patient. None of these events led to the discontinuation of zidovudine.
Hematologic Abnormalities
Evaluation of hematologic abnormalities was complicated by the presence of anemia at entry in 50 children (57 percent) and a need for transfusions within four weeks of entry in 9 children. During the study, 54 children (61 percent) had one or more episodes of anemia (hemoglobin level, <75 g per liter) or neutropenia (neutrophil count, <0.75×109 per liter). The median hemoglobin level was 103 g per liter at base line, 109 g per liter at week 24, and 111 g per liter at week 52. Twenty-three children had anemia presumed to be drug-related (mean number of episodes, 1.9; range, 1 to 9); 12 of these patients (52 percent) had only one episode of anemia. The time to the onset of the first episode ranged from 1 to 56 weeks (median, 8).
The management of the 43 episodes of anemia in the 23 children included transfusion for 28 episodes, transfusion with dose modification for 8 episodes (reduction of the dose in 5, interruption of treatment in 2, and dose reduction and treatment interruption in 1), and dose reduction for 3; no intervention was initiated for 4 episodes. Treatment was permanently discontinued in one patient because of anemia and leukopenia. Overall, 19 of the 23 children required transfusions when their hemoglobin concentration fell below 75 g per liter. Ten additional children received red-cell transfusions although their hemoglobin concentration was above 75 g per liter. As has been observed among adults treated with zidovudine, there was a gradual increase in the erythrocyte mean corpuscular volume over time. The median mean corpuscular volume at entry of 80 fl (range, 62 to 107) increased by week 24 to 96 fl (range, 75 to 119) and by week 52 to 99 fl (range, 75 to 114).
The median neutrophil counts at base line, week 24, and week 52 were 2.66, 1.89, and 1.86×109 per liter, respectively. The 42 children with presumably drug-related neutropenia had a mean of 3.0 episodes of neutropenia (range, 1 to 12). Eighteen (43 percent) had only one episode. The time to the first episode of neutropenia ranged from 1 to 70 weeks (median, 21). Twelve of the 42 children had 13 episodes of prolonged neutropenia (≥30 days), with a mean duration of 12 weeks (range, 4 to 23). Management of the 108 episodes of neutropenia in the 42 children included dose reduction for 7 episodes, treatment interruption for 17, and permanent discontinuation of treatment for 2; there was no intervention for 82 episodes (76 percent). Among the children in whom treatment was interrupted, the mean interval was 13 days; in the majority (65 percent), treatment was resumed with the same dose. In summary, of the 54 children with hematologic toxicity, 30 (55 percent) required transfusion or modification in the dose of zidovudine.
In general, platelet counts remained constant or increased over the treatment period. The median platelet counts at base line, week 24, and week 52 were 286, 332, and 344×109 per liter, respectively.
Serologic Findings
Monitoring of serum biochemical levels during the study revealed no significant trends in the levels of creatinine, urea nitrogen, total bilirubin, aminotransferases, lactate dehydrogenase, amylase, creatine kinase, vitamin B12, or folate.
Survival
Fourteen of the 88 children ( 16 percent) died, 3 during the first six weeks after enrollment. Eleven of these 14 patients had a history of multiple AIDS-defining conditions (median, three conditions), including HIV encephalopathy in 11. Eight deaths were due to opportunistic infections, the most common being P. carinii pneumonia (in five children). Two children died of respiratory failure (suspected to be due to aspiration in one), and one child each died of dilated cardiomyopathy, cardiopulmonary arrest, pulmonary tuberculosis, and wasting syndrome.
A Kaplan–Meier product-limit survival analysis indicated that the probability of survival (±SE) was 0.89±0.04 at week 24 and 0.79±0.05 at week 52 (Fig. 1Figure 1
Kaplan–Meier Curve for Survival during the 24-Week Trial and Follow-up.).Intercurrent Infections
Sixteen opportunistic infections developed in 13 patients during the 24-week study. Two of these infections were diagnosed within the first two weeks and probably represented preexisting infection. P. carinii pneumonia occurred in seven children after they had received therapy for a median of 93 days (range, 68 to 140). Two had a history of pneumocystis pneumonia, and neither received prophylaxis. Other opportunistic infections included candida esophagitis (two patients), cryptosporidiosis (one patient), disseminated Mycobacterium avium–intracellulare (two patients), Listeria monocytogenes meningitis (one patient), M. avium–intracellulare pneumonia (one patient), cytomegalovirus colitis (one patient), and cytomegalovirus retinitis in a patient with preexisting cytomegalovirus infection (pneumonia and colitis).
Thirty-five other serious infections occurred in 24 children, including 19 episodes of pneumonia, 15 episodes of bacteremia or fungemia, and 1 hip abscess. Thirty-seven percent of these infections (13 of 35) developed within the first six weeks of therapy. Six of the infections were reported to be related to the use of indwelling central venous catheters. Twelve of the episodes of pneumonia were of unknown origin, six of them diagnosed by physical examination only. Seven episodes of pneumonia were documented as caused by bacteria. The most frequent pathogen isolated in the 15 cases of bacteremia was Streptococcus pneumoniae, in 5 cases. No other organism was isolated more than twice.
Weight Gain
Growth was analyzed in terms of age- and sex-specific percentiles of weight-gain velocity instead of absolute changes in weight, because the latter would be expected to vary considerably according to age and sex (Fig. 2Figure 2
Percentile Values for Weights and Weight-Gain Rates in 57 Children.). Data on changes in weight during two six-month periods were available for 57 children. At base line, 30 children (53 percent) had weights below the 10th percentile for their age and sex, and only 6 (11 percent) had weights above the 50th percentile. During the first six months, 32 of these children (56 percent) gained weight at a rate above the 50th percentile for age and sex, 14 of them (25 percent) at a rate above the 90th percentile. Only 11 patients (19 percent) gained weight at a rate below the 10th percentile. Weight-growth velocities during the second six months were slower, with 23 patients (40 percent) gaining at a rate below the 10th percentile and 21 (37 percent) gaining at a rate above the 50th percentile.Cognitive Function
Cognitive function was tested at base line and week 24 in 55 children (Table 2Table 2
Changes in Cognitive-Test Scores from Base Line to Week 24.). At base line, 39 patients (71 percent) had function at least 1 SD below normal, with 25 (45 percent) at least 2 SD below normal (as compared with 2.3 percent of children in a normal distribution). Twenty-three of the 55 patients who could be evaluated had an increase of 8 points or more in cognitive-test scores, 10 of whom had increases of more than 15 points. Children less than 30 months old tested with the Bayley scales had the greatest change in their cognitive performance during the first six months of the study.Virologic Findings
p24 Antigen
Base-line serum p24 antigen levels were measured in 71 patients. At entry, 37 patients (52 percent) were positive for the antigen (level, ≥30 ng per liter), with a median concentration of 184 ng per liter, and 34 (48 percent) were negative. Declines in the entry levels of serum p24 antigen were observed after 12 weeks in 21 of the 24 patients who were positive at base line (88 percent) (P<0.0001) and in 14 of 18 patients after 24 weeks (78 percent) (P<0.01). Of the patients who were positive at base line, 10 of 24 were negative at week 12, and 9 of 18 at week 24.
Cerebrospinal fluid concentrations of the antigen were measured in 51 patients at entry. Eighteen (35 percent) were positive, with a median level of 114 ng per liter, and 33 (65 percent) were negative. All eight patients positive at base line who underwent follow-up lumbar puncture at week 24 had a decrease in the p24 antigen concentration. Of these eight patients, six were negative for the antigen at week 24. Of the 16 patients who were negative at entry and had a lumbar puncture at week 24, 1 became positive for p24 antigen (level, 65 ng per liter).
HIV Cultures
Peripheral-blood mononuclear cells from 77 children were cultured for HIV at entry; the cultures of 58 children (75 percent) were positive, and those of 19 (25 percent) were negative. Follow-up cultures were available for 65 children, 55 (85 percent) of whom subsequently had positive cultures during therapy. Cerebrospinal fluid from 63 patients was cultured for HIV at base line; the cultures of 15 (24 percent) were positive. Thirty children had a repeat lumbar puncture at week 24. Six of seven patients (86 percent) positive at base line became negative (P<0.05), and all patients negative at base line remained negative.
Immunologic Findings
Among the study participants, the median base-line CD4+ cell count was 0.263 × 109 per liter (range, 0 to 3.61 × 109). Sixty-three percent (52 of 83) had a count less than or equal to 0.50× 109 per liter at entry. In the total study population, there was a median intrapatient increase in the CD4+ count of 0.044×109 per liter by week 4 of therapy (P = 0.006). This increase was sustained at week 12, but a median decrease of 0.022×109 per liter was noted at week 24. Similar changes were observed among participants with baseline CD4+ cell counts less than or equal to 0.50× 109 per liter. However, a small median increase in CD4+ cell counts was maintained in this group through week 24 of zidovudine therapy (Table 3Table 3
Changes in CD4+ Cell Counts during Zidovudine Therapy.).Eighty-five percent of the study population (52 of 61 patients) for whom serum immunoglobulin values were available had base-line IgG levels above the age-adjusted normal range. Eight percent (five patients) had normal IgG values, and 7 percent (four patients) had hypogammaglobulinemia. Overall, there was a trend toward normalization of IgG levels during the study (Fig. 3Figure 3
Base-Line Serum IgG Concentration in 50 Children in Relation to Change in IgG Concentration at Week 24.). The children with the greatest elevations at base line had the largest decreases in IgG while receiving zidovudine therapy. Abnormal, usually elevated, levels of IgM and IgA were also noted at entry. Like the levels of IgG, the elevated levels of IgM and IgA tended to decrease from base line through the 24 weeks of therapy.Discussion
This study evaluated the safety and tolerance of zidovudine in children with advanced HIV disease who were at high risk for infection, disease progression, and death. The seventy of the HIV disease in these children was at least comparable to that in adults studied in the initial placebo-controlled trial of zidovudine.1 Sixty-three percent of the children had a base-line CD4+ cell count less than 0.50×lO9 per liter. Since CD4+ counts in normal infants are higher than in normal adults,17 this level in children represents an even greater degree of immunocompromise than it does in adults. In addition, 72 percent of the patients in this study had AIDS—half with multiple AIDS-defining conditions—as compared with the 57 percent of the adults in the initial placebo-controlled trial who had AIDS.1
To maximize compliance, a dosage schedule of administration every six hours was selected. Pharmacokinetic data derived from phase I studies in children indicated that giving 180 mg per square meter every 6 hours would provide drug exposure comparable to that in adults given 250 mg every 4 hours, as measured by the 24-hour time—concentration curve. Compliance with the dosage regimen was judged to be good, with only 14 percent of the study population known or suspected to be noncompliant. Monitoring of the erythrocyte mean corpuscular volume provided a fairly reliable screening mechanism by which to identify possible noncompliance.
The selected dosing regimen was generally tolerated. Fifty-two percent of the patients required no dose modification as of the time of data cutoff. Hematologic abnormalities, primarily neutropenia, were the only major reasons for dose modification, and in many cases the hematologic problems resolved without any change in zidovudine therapy. Dose modification was performed as intervention for 24 percent of the episodes of neutropenia (neutrophil count, <0.75×109 per liter). Interventions for neutropenia primarily involved an interruption of treatment with a median interval of 13 days. The management of anemia involved transfusion more often than dose modification. Dose modification and transfusion were generally effective in mitigating suspected drug-induced hematologic abnormalities; in only three patients was zidovudine treatment permanently discontinued because of recurrent or persistent hematologic problems. Neutropenia and anemia occurred throughout the study period, but only a single episode of either condition developed in approximately half the patients with these problems.
The primary purpose of this study was to assess drug safety and tolerance; however, several immunologic, virologic, and clinical variables were also monitored as potential indicators of a response to zidovudine therapy. Although no control group was available for direct comparisons, the improvement in the children in this study closely paralleled the observations in controlled studies of adults receiving zidovudine.
Kaplan–Meier analysis performed on an intention-to-treat basis (in which all patients were included whether or not they continued to receive zidovudine) demonstrated a probability of survival of 0.79±0.05 after 52 weeks (Fig. 1). Although there is no matched, untreated control group for this study population, there are published reports on the natural history and prognosis of HIV infection in children that can serve as a point of reference, even though formal statistical comparisons cannot be made. Two studies reported that one-year survival after the diagnosis of HIV infection was 75 percent.18 19 Another retrospective study estimated that the probability of survival among children with HIV infection not treated with an antiretroviral drug was 0.83 one year after the onset of HIV-related symptoms (Connor E: unpublished data). Since the children enrolled in our study had advanced HIV disease before the start of therapy, and many had been given a diagnosis of AIDS long before entry, the observations in our study participants suggest a beneficial effect of zidovudine treatment on survival. Survival was similar to that among adults treated with zidovudine. Among the adults given zidovudine in the original placebo-controlled trial,1 the probability of 12-month survival after the start of therapy was 0.85. Overall, these adult patients had less advanced HIV disease than the children in our study. The survival rate among 4805 adult patients with AIDS treated with zidovudine in an investigational-new-drug program was 73 percent after 44 weeks.20
Children treated with zidovudine continued to have bacterial and opportunistic infections. The effect of the drug on the frequency of these events could not be assessed because of the lack of a control group. Among previously described adult recipients of zidovudine, infections continued to occur but developed less frequently and appeared to be less severe than infections among the recipients of placebo.1 Thus, both children and adults continue to have secondary infections while receiving specific antiretroviral therapy. P. carinii pneumonia was the most frequent cause of death among children in our study. HIV-infected children with advanced disease, like their adult counterparts,21 should receive primary and secondary prophylaxis against pneumocystis infection.
A number of major events occurred among the children within the first six weeks after enrollment, including 3 of the 14 deaths (21 percent), 2 of the 16 opportunistic infections (13 percent), and 14 of the 35 other serious infections (40 percent). The placebo-controlled trial in adults with advanced HIV disease showed no difference between placebo and zidovudine recipients in the occurrence of such events during the first six weeks. Consequently, the full clinical benefit of zidovudine may not be observed when the duration of therapy is limited. There may, however, be evidence of improvement in immunologic or virologic indicators as early as four weeks after the start of therapy.
Zidovudine therapy was associated with a substantial improvement in weight gain. Although the weight-gain rates (velocities) before entry could not be calculated, the fact that the weights of 57 percent of the subjects were less than the 10th percentile for age indicates a retarded rate of growth for a substantial period. The rapid rate of weight gain in some patients during the first six months of zidovudine therapy suggests a period of catch-up growth. This may be attributable either to zidovudine or to the close medical attention received by these patients.
There also appeared to be some improvement in cognitive function as measured by neuropsychological testing. The positive changes in cognitive-test scores confirm the previous observations of improvement in neuropsychological testing among children enrolled in phase I studies of zidovudine. Pizzo et al. reported improvement in cognitive function (specifically, improvement in IQ scores) among children receiving zidovudine by continuous intravenous infusion.8 Comparable improvement was observed among adult zidovudine recipients with advanced HIV disease in a placebo-controlled trial.2 This improvement may be related to either the direct effect of zidovudine on central nervous system HIV infection or improvement in general physical well-being.
The reductions in serum and cerebrospinal fluid levels of p24 antigen presumably reflect inhibition of HIV replication in vivo. The overall decreases in p24 antigen levels in this study population are consistent with changes observed among adults receiving zidovudine in controlled trials3 , 22 , 23 and among children in a phase I study.7 As in adults, zidovudine therapy did not consistently affect the recovery of HIV from nonquantitative cultures of peripheral-blood lymphocytes. In contrast to the results of sequential blood cultures for HIV, cultures of cerebrospinal fluid became negative in most (6 of 7) patients who were positive at entry. These data, along with the quantitative reduction in the cerebrospinal fluid level of p24 antigen, suggest a direct antiretroviral effect of zidovudine on central nervous system HIV infection. The neurologic improvements observed among children treated with zidovudine may be associated with such antiviral activity.
The differences between the effects of zidovudine on the recovery of HIV from cerebrospinal fluid and its effects on the recovery of the virus from blood may be explained by the fact that recoverable virus in the cerebrospinal fluid consists largely of free virus shed into the cerebrospinal fluid. In contrast, cultures of peripheral-blood mononuclear cells for HIV detect reactivated virus in latently infected lymphocytes, which are stimulated by mitogen in vitro to induce virus production. Zidovudine would not be expected to affect the retention of latent HIV provirus in cells, although it has been shown to reduce plasma HIV titers among adult patients with advanced HIV disease.24
Changes in CD4+ cell counts were similar to those among adult zidovudine recipients, even though the children were more immunocompromised at entry than the previously studied adults. There was a small increase in median CD4+ cell counts during the first 12 weeks of therapy and a subsequent decline to a number that nevertheless remained at or above baseline values after 24 weeks of therapy. The rise in CD4+ cell counts was proportionately greater in the children with more severe lymphocyte depletion at entry. The clinical importance of the changes in CD4+ cell counts is unclear, since data are not available on the natural history of changes in CD4+ cell counts over time in children with untreated HIV infection. There was also an apparent tendency toward normalization of serum immunoglobulin levels, particularly the level of IgG, as previously reported in phase I studies.7 This tendency may be due to a direct effect on B-cell function or an indirect measure of improvement in T-cell function.
In conclusion, oral zidovudine can be safely administered to children 3 months to 13 years of age. A number of clinical, immunologic, and virologic characteristics improved during the course of therapy. These trends were similar to those reported among zidovudine-treated adults and suggest that children with advanced HIV disease may benefit from zidovudine therapy. Recently completed studies in adults with a spectrum of HIV disease5 , 6 have indicated that lower doses of zidovudine are associated with fewer side effects but with efficacy equivalent to that of standard doses. These findings have led to a reduction in the total daily dose recommended for adults. The studies in adults, however, were conducted in patients with less advanced HIV disease than the children enrolled in this study, in particular neurologic disease. It is unknown whether the use of lower doses of zidovudine in children would provide equivalent improvement in HIV-related central nervous system manifestations, a major problem in pediatric AIDS. An ongoing study (AIDS Clinical Trials Group protocol 128) is comparing lower doses of zidovudine with the standard dosing regimen. Until additional data are available about lower doses, children may safely receive zidovudine in a dose of 180 mg per square meter every six hours in a setting of careful clinical and laboratory monitoring.
Presented as an abstract at the meeting of the Society for Pediatric Research, Anaheim, Calif., May 8, 1990, and at the Sixth International Conference on AIDS, San Francisco, June 22, 1990.
Supported by the AIDS Clinical Trials Group, National Institute of Allergy and Infectious Diseases; the Pediatric Branch, National Cancer Institute; and the Burroughs Wellcome Co.
*Members of the Protocol 043 Study Group are listed in the Appendix.
Source Information
From the Department of Pediatrics, Duke University School of Medicine, Durham, N.C. (R.E.M., K.O., C.M.W.); Burroughs Wellcome Company, Research Triangle Park, N.C. (M.A.M., G.K.R., S.N.L.); New Jersey Children's Hospital, Newark (E.M.C.); Division of AIDS, National Institute of Allergy and Infectious Diseases, Bethesda, Md. (J.F.); the Department of Pediatrics, University of Miami School of Medicine, Miami (G.B.S.); the Department of Biostatistics, Harvard School of Public Health, Boston (M.W., R.D.G.); the Children's Hospital Medical Center, Boston (K.M.); the Department of Pediatrics, New York University School of Medicine, New York (W.B.); the Department of Pediatrics, Johns Hopkins School of Medicine, Baltimore (J.F.M.); the Department of Pediatrics, University of California, San Francisco (P.W.); and the Division of Biostatistics and Epidemiology, Dana–Farber Cancer Institute, Boston (R.D.G.). Address reprint requests to Dr. McKinney at Box 3461, Duke University Medical Center, Durham, NC 27710.
Appendix
The Protocol 043 Study Group included the following participants (ACTU denotes AIDS Clinical Trial Unit): Duke University Medical Center Pediatric ACTU — Barbara Lane, R.N., M.S.N., Emmanuel B. Walter, M.D., and Sharon B. Kupit, P.A.-C; Children's Hospital Medical Center Pediatric ACTU — Ho Wen Hsu, M.D., Nancy Karthas, R.N., Kerrie Liebendorfer, R.N., Andrea Rubin, R.N., M.P.H., and Katherine Koury; University of California, San Francisco, Pediatric ACTU — Donna Brandt, R.N., and Carol Rumsey, F.N.P.; University of Miami Pediatric ACTU —Janet Gourley, R.N., Charles Mitchell, M.D., and Carol Owens, R.N.; Children's Hospital of New Jersey Pediatric ACTU —James Oleske, M.D., Bessie Bryant, R.N., M.A., Carmen Ayala, R.N., Nina Colabelli, R.N., B.S.N., Thomas Denny, B.S., Mark Mintz, M.D., and George McSherry, M.D.; Children's Hospital of Los Angeles — Edward Gomperts, M.D., Lin Woods, R.N., M.N., and Debbie Morgan, R.N.; National Cancer Institute — Philip A. Pizzo, M.D., and Janie Eddy, R.N.; New York University Medical Center Pediatric ACTU — Keith Krasinski, M.D., Maryam Minter, R.N., Robert Lawrence, M.D., and Mona Riguad, M.D.; Johns Hopkins Hospital Pediatric ACTU — Andrea Ruff, M.D., and Yvette Benjamin, P.A.-C; Burroughs Wellcome Co. — M. Robert Blum, Ph.D., Stephen Weiler, Ph.D., and Katherine A.R. Walmer, B.S.N.; and the National Institutes of Health, AIDS Program Office.
- References
References
1
Fischl
Full Text | Web of Science | Medline2
Schmitt
Full Text | Web of Science | Medline3
Parks
Web of Science | Medline4
Fischl
CrossRef | Web of Science | Medline5
Fischl
Web of Science | Medline6
Volberding
Full Text | Web of Science | Medline7
McKinney
CrossRef | Web of Science | Medline8
Pizzo
Full Text | Web of Science | Medline9
Balis
Web of Science10
Balis
CrossRef | Web of Science | Medline11
Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome . MMWR 1987; 36:Suppl 1S:1S–15S.
Medline12
Rowe PC, ed. The Harriet Lane handbook: a manual for pediatric house officers. 11th ed. Chicago: Year Book Medical, 1987:325–9.
13
Bayley scales of infant development. Cleveland: Psychological Corporation, 1969.
14
McCarthy scales of children's abilities. Cleveland: Psychological Corporation, 1972.
15
Wechsler intelligence scale for children —revised. Cleveland: Psychological Corporation, 1974.
16
Michalski
Web of Science | Medline17
Yanase
CrossRef | Web of Science | Medline18
Scott
Full Text | Web of Science | Medline19
Krasinski
Web of Science | Medline20
Creagh-Kirk
CrossRef | Web of Science | Medline21
Guidelines for prophylaxis against Pneumocystis carinii pneumonia for persons infected with human immunodeficiency virus . MMWR 1989; 38(S-5):1–9.
22
Jackson
Web of Science | Medline23
Chaisson
CrossRef | Web of Science | Medline24
Ho
Full Text | Web of Science | Medline
- Citing Articles (40)
Citing Articles
1
Scott Dryden-Peterson, Roger L Shapiro, Michael D Hughes, Kathleen Powis, Anthony Ogwu, Claire Moffat, Sikhulile Moyo, Joseph Makhema, Max Essex, Shahin Lockman. (2011) Increased Risk of Severe Infant Anemia After Exposure to Maternal HAART, Botswana. JAIDS Journal of Acquired Immune Deficiency Syndromes 56:5, 428-436
CrossRef2
Gabriela Marón, Aditya H. Gaur, Patricia M. Flynn. (2010) Antiretroviral Therapy in HIV-Infected Infants and Children. The Pediatric Infectious Disease Journal 29:4, 360-363
CrossRef3
Rohan Hazra, George K. Siberry, Lynne M. Mofenson. (2010) Growing Up with HIV: Children, Adolescents, and Young Adults with Perinatally Acquired HIV Infection*. Annual Review of Medicine 61:1, 169-185
CrossRef4
Veena Venugopalan, Alice C Thornton, Douglass T Steinke, Robert P Rapp, Frank Romanelli, David J Feola. (2009) Trimethoprim-Sulfamethoxazole Exposure Alters Ex Vivo Function of B Lymphocytes Isolated from Human Immunodeficiency Virus–Infected Patients Receiving Zidovudine. Pharmacotherapy 29:4, 373-382
CrossRef5
Cornelia Feiterna-Sperling, Katharina Weizsaecker, Christoph B??hrer, Simone Casteleyn, Andrea Loui, Thomas Schmitz, Volker Wahn, Michael Obladen. (2007) Hematologic Effects of Maternal Antiretroviral Therapy and Transmission Prophylaxis in HIV-1-Exposed Uninfected Newborn Infants. JAIDS Journal of Acquired Immune Deficiency Syndromes 45:1, 43-51
CrossRef6
Annelies Van Rie, Patrick R. Harrington, Anna Dow, Kevin Robertson. (2007) Neurologic and neurodevelopmental manifestations of pediatric HIV/AIDS: A global perspective. European Journal of Paediatric Neurology 11:1, 1-9
CrossRef7
Sanneke Koekkoek, Laura Eggermont, Leo Sonneville, Thidakat Jupimai, Saijai Wicharuk, Wichitea Apateerapong, Theshinee Chuenyam, Joep Lange, Ferdinand Wit, Chitsanu Pancharoen, Praphan Phanuphak, Jintanat Ananworanich. (2006) Effects of highly active antiretroviral therapy (HAART) on psychomotor performance in children with HIV disease. Journal of Neurology 253:12, 1615-1624
CrossRef8
Salvador Resino, Rosa Resino, Jose Ma Bellon, Dariela Micheloud, Ma Dolores Gurbindo Gutierrez, Ma Isabel de Jose, Jose Tomas Ramos, Pablo Martin Fontelos, Luis Ciria, Ma Angeles Munoz‐Fernandez, . (2006) Clinical Outcomes Improve with Highly Active Antiretroviral Therapy in Vertically HIV Type‐1–Infected Children. Clinical Infectious Diseases 43:2, 243-252
CrossRef9
Taha E. Taha, Newton Kumwenda, George Kafulafula, Johnstone Kumwenda, Rohit Chitale, Chiwawa Nkhoma, Pauline Katundu, Joshua Mukiibi, Shu Chen, Donald Hoover, Robin Broadhead. (2004) Haematological changes in African children who received short-term prophylaxis with nevirapine and zidovudine at birth. Annals of Tropical Paediatrics: International Child Health 24:4, 301-309
CrossRef10
Lucy Civitello. (2003) Neurologic aspects of HIV infection in infants and children: Therapeutic approaches and outcome. Current Neurology and Neuroscience Reports 3:2, 120-128
CrossRef11
S Resino, JM Bellón, D Gurbindo, JT Ramos, JA León, MÁ Muñóz-Fernández. (2002) Dynamics of progression markers in a non-study population of human immunodeficiency virus-1 vertically infected infants with different antiretroviral treatments. Acta Paediatrica 91:7, 776-782
CrossRef12
Mary E. Temple, Katalin I. Koranyi, Milap C. Nahata. (2001) The Safety and Antiviral Effect of Protease Inhibitors in Children. Pharmacotherapy 21:3, 287-294
CrossRef13
F. Albano, M. I. Spagnuolo, R. Berni Canani, A. Guarino. (1999) Adherence to antiretroviral therapy in HIV-infected children in Italy. AIDS Care 11:6, 711-714
CrossRef14
NICOLA PRINCIPI, PAOLA MARCHISIO, SUSANNA ESPOSITO, PAOLO ROSSI, GUIDO CASTELLI GATTINARA, LUISA GALLI, CLARA GABIANO, GIAN VINCENZO ZUCCOTTI, PAOLA ORLANDI. (1998) Zidovudine Therapy and HIV Type 1 Mutations in Children with Symptomatic HIV Type 1 Infection: Effect of Switching to Didanosine or Zidovudine Plus Didanosine Therapy. AIDS Research and Human Retroviruses 14:18, 1653-1659
CrossRef15
UWE WINTERGERST, FLORIAN HOFFMANN, BRIGITTE SÖLDER, GUNDULA NOTHEIS, THEONI PETROPOULOU, JOSEF EBERLE, LUTZ GÜRTLER, BERND H. BELOHRADSKY. (1998) Comparison of two antiretroviral triple combinations including the protease inhibitor indinavir in children infected with human immunodeficiency virus. The Pediatric Infectious Disease Journal 17:6, 495-499
CrossRef16
Gerd Horneff, Ortwin Adams, Volker Wahn. (1998) Pilot study of zidovudine–lamivudine combination therapy in vertically HIV-infected antiretroviral-naive children. AIDS 12:5, 489-494
CrossRef17
J. Gray. (1997) HIV in the neonate. Journal of Hospital Infection 37:3, 181-198
CrossRef18
Pamela L. Wolters, Pim Brouwers, Lucy Civitello, Howard A. Moss. (1997) Receptive and expressive language function of children with symptomatic HIV infection and relationship with disease parameters. AIDS 11:9, 1135-1144
CrossRef19
Englund, Janet A., Baker, Carol J., Raskino, Claire, McKinney, Ross E., Petrie, Barbara, Fowler, Mary Glenn, Pearson, Deborah, Gershon, Anne, McSherry, George D., Abrams, Elaine J., Schliozberg, Jenny, Sullivan, John L., Behrman, RachelConnor, James C.Hetherington, SethLifschitz, Marta H.McLaren, ColinMendez, HermanMillison, KarenMoye, JackNozyce, MollyO'Donnell, KarenPurdue, LynetteSchoenfeld, DavidScott, GwendolynnSpector, Stephen A.Wara, Diane W.. (1997) Zidovudine, Didanosine, or Both as the Initial Treatment for Symptomatic HIV-Infected Children. New England Journal of Medicine 336:24, 1704-1712
Full Text20
D. Moodley, R. A. Bobat, H. M. Coovadia, T. Doorasamy, S. Munsamy, E. Gouws. (1997) Lymphocyte subset changes between 3 and 15 months of age in infants born to HIV-seropositive women in South Africa. Tropical Medicine & International Health 2:5, 415-421
CrossRef21
JANET A. ENGLUND, CAROL J. BAKER, CLAIRE RASKINO, ROSS E. MCKINNEY, MARTA H. LIFSCHITZ, BARBARA PETRIE, MARY GLENN FOWLER, JAMES D. CONNOR, HERMANN MENDEZ, KAREN O'DONNELL, DIANE W. WARA. (1996) Clinical and Laboratory Characteristics of a Large Cohort of Symptomatic, Human Immunodeficiency Virus-infected Infants and Children. The Pediatric Infectious Disease Journal 15:11, 1025-1036
CrossRef22
C. Exhenry, D. Nadal. (1996) Vertical human immunodeficiency virus-1 infection: Involvement of the central nervous system and treatment. European Journal of Pediatrics 155:10, 839-850
CrossRef23
Ay[dot below] Ünal, Eric Lorenzo, Michael Brown, Laurette Smith, Suzanne Matsuura, Gwendolyn Scott, Walter Scott. (1996) Reverse Transcriptase Mutations in HIV-1-Infected Children Treated with Zidovudine. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology 13:2, 140-145
CrossRef24
Barbara A. Styrt, Toni D. Piazza-Hepp, Gary K. Chikami. (1996) Clinical toxicity of antiretroviral nucleoside analogs. Antiviral Research 31:3, 121-135
CrossRef25
PHILIPPE LEPAGE, PHILIPPE MSELLATI, DEO-GRATIAS HITIMANA, ANATHOLIE BAZUBAGIRA, CHRISTIAAN VAN GOETHEM, ARLETTE SIMONON, ETIENNE KARITA, LAURENCE DEQUAE-MERCHADOU, PHILIPPE VAN DE PERRE, FRANÇOIS DABIS. (1996) Growth of human immunodeficiency type 1-infected and uninfected children: a prospective cohort study in Kigali, Rwanda, 1988 to 1993. The Pediatric Infectious Disease Journal 15:6, 479-485
CrossRef26
STEVEN R. NESHEIM. (1996) The Diagnosis and Management of Perinatal HIV Infection. Clinical Obstetrics and Gynecology 39:2, 396-410
CrossRef27
CHARLENE E. BUSH, RICHARD M. DONOVAN, ODETTE MANZOR, DWAYNE BAXA, ELLEN MOORE, FLOSSIE COHEN, LOUIS D. SARAVOLATZ. (1996) Comparison of HIV Type 1 RNA Plasma Viremia, p24 Antigenemia, and Unintegrated DNA as Viral Load Markers in Pediatric Patients. AIDS Research and Human Retroviruses 12:1, 11-15
CrossRef28
U. Wintergerst, B. Rolinski, Gundula Notheis, B. H. Belohradsky, F. -D. Goebel, Ilse Grosch-Wörner, Mechtild Vocks-Hauck, V. Wahn, K. -M. Debatin, . (1995) Pharmacokinetics of orally administered zidovudine in HIV-infected children and adults. Infection 23:6, 344-348
CrossRef29
R Manfredi, T Cariani, F Latini, F Chiodo. (1995) Recombinant granulocyte-macrophage colony-stimulating factor in the treatment of HIV-related leucopenia. Acta Paediatrica 84:8, 943-944
CrossRef30
Ross E. McKinney, Catherine Wilfert. (1994) Growth as a prognostic indicator in children with human immunodeficiency virus infection treated with zidovudine. The Journal of Pediatrics 125:5, 728-733
CrossRef31
Brigitta U. Mueller, Philip A. Pizzo, Maureen Farley, Robert N. Husson, Jonathan Goldsmith, Andrea Kovacs, Lin Woods, Janice Ono, Joseph A. Church, Pim Brouwers, Paul Jarosinski, David Venzon, Frank M. Balis. (1994) Pharmacokinetic evaluation of the combination of zidovudine and didanosine in children with human immunodeficiency virus infection. The Journal of Pediatrics 125:1, 142-146
CrossRef32
ANNE A. GERSHON. (1993) Antiviral Therapy for HIV Infection in Infants and Children. Annals of the New York Academy of Sciences 693:1 Pediatric AID, 166-177
CrossRef33
ANITA L. BELMAN. (1993) Neurologic Syndromes. Annals of the New York Academy of Sciences 693:1 Pediatric AID, 107-122
CrossRef34
T. Güngör, M. Funk, R. Linde, I. Kynast, A. Allendorf, C. Lotz, S. Ehrenforth, D. Hofmann, B. Kornhuber, W. Kreuz. (1993) Combined therapy in human immunodeficiency virus-infected children —a 4-year experience. European Journal of Pediatrics 152:8, 650-654
CrossRef35
Wood, Alastair J.J., , Hirsch, Martin S.D'Aquila, Richard T.. (1993) Therapy for Human Immunodeficiency Virus Infection. New England Journal of Medicine 328:23, 1686-1695
Full Text36
James Coplan. (1993) Child development. Current Problems in Pediatrics 23:2, 44-49
CrossRef37
Lipshultz, Steven E., Orav, E. John, Sanders, Stephen P., Hale, Andrea Rubin, McIntosh, Kenneth, Colan, Steven D., . (1992) Cardiac Structure and Function in Children with Human Immunodeficiency Virus Infection Treated with Zidovudine. New England Journal of Medicine 327:18, 1260-1265
Full Text38
ROBERT N. HUSSON, PHILIP A. PIZZO. (1992) The Use of Nucleoside Analogues in the Treatment of HIV-infected Children. AIDS Research and Human Retroviruses 8:6, 1059-1064
CrossRef39
M. C. Nahata. (1992) Advances in paediatric pharmacotherapy. Journal of Clinical Pharmacy and Therapeutics 17:3, 141-146
CrossRef40
(1991) HIV in children. The Lancet 337:8748, 1030
CrossRef
