Results of the PK portion, study design, dosing, and sociodemographic characteristics have been reported previously (1, 2). Adverse events were graded according to the National Institute of Allergy and Infectious Diseases Division of AIDS (NIAID/DAIDS) toxicity tables (3). Signs and symptoms were graded as mild (grade 1), moderate (grade 2), severe (grade 3), or life-threatening (grade 4).

The protocol-targeted, maternal adverse events included any grade 3 or 4 event as well as grade 2 adverse events for glucose, aspartate transaminase (AST), alanine aminotransferase (ALT), uric acid, proteinuria, and neurocerebellar or neurosensory symptoms. All grade 3 and 4 adverse events were managed by discontinuation of all study drugs, with rechallenge for grade 3 subjects if toxicity resolved within 72 h. In addition, grade 2 adverse events were managed as if they were grade 3. Protocol-targeted, infant adverse events included all grade 3 and 4 events, with drug discontinuation except for anemia and neutropenia and with repeat testing performed within 72 h and discontinued if toxicity persisted. All adverse events were reviewed by the study team and classified as definitely related, possibly related, or not related to study medications. Three mothers entering the study with <400 copies/ml of viral RNA were already taking SQV and had evidenced a clinical response. Infants were born between March 2001 and October 2002. The proportion of women with undetectable plasma HIV-1 RNA (<400 copies/ml) increased from 23% (3/13) at baseline to 100% (13/13) at delivery (P = 0.002, McNemar's exact test) and 69% (9/13) and 77% (10/13), respectively, at 6 and 12 weeks postpartum. The median absolute CD4 count increased throughout the treatment period (not shown), and the median CD4 count increased by 287 cells/m³ (P = 0.001, Wilcoxon signed-rank test), representing a significant increase between study entry and delivery (P < 0.001) but not between delivery and 12 weeks postpartum.

Nine adverse events were reported for mothers; two were grade 3, with elevated amylase (possibly related) and hyperglycemia (nonrelated), and the remaining were grade 2 (Table ​(Table1).1). One mother had a grade 2 elevation in liver enzymes (AST and ALT), resulting in (protocol-mandated) permanent drug discontinuation at 35 weeks of gestation after 11 weeks of treatment.

One infant never began treatment; the mother stopped therapy prior to delivery. One discontinued therapy at 4 weeks of age because of grade 3 anemia. Eleven infants completed 6 weeks of study treatment. Nineteen adverse events were reported among 10 infants (Table ​(Table2).2). Five infants had seven adverse events that were possibly treatment related: three had grade 3 anemia, three had grade 3 neutropenia, and one had grade 4 hyperbilirubinemia.

One infant was HIV infected, with a negative HIV-1 RNA PCR at 6 weeks of age but detectable HIV-1 RNA at 12 and 24 weeks and a positive HIV DNA PCR at 24 weeks. This child's mother discontinued all study drugs (for 1 week) at 27 weeks' gestation due to grade 2 hepatotoxicity and restarted 1 week later with normalization of liver enzymes. Hepatotoxicity returned, and the study drugs were permanently discontinued at 35 weeks' gestation after another week of discontinuation. The PI was switched to nelfinavir, while zidovudine (ZDV) and lamivudine (3TC) were maintained for the remainder of pregnancy. A scheduled Cesarean section was performed at 38 weeks, and the viral load was <400 copies/ml. The baby received ZDV for the first 6 weeks of life and did not breast-feed.

SQV in a soft-gel capsule formulation with low-dose RTV given in combination with ZDV and 3TC was well tolerated by women during pregnancy, during delivery, and postpartum, and minimal infant toxicity occurred. The ZDV and 3TC regimen was also well tolerated by the infants. The most common adverse events in the infants, anemia and neutropenia, likely resulted from ZDV therapy. No adverse events were judged as definitely related to the study treatment, and no life-threatening or serious adverse events for mothers or infants occurred during the study. The antiviral activity of this regimen was demonstrated by a reduction of the maternal HIV-1 RNA level from study entry to delivery and a consistent CD4 cell count increase during treatment. The increase in viral loads observed in three of the women between delivery and 12 weeks postpartum, although not statistically significant, suggests the possibility of reduced medication adherence during the postpartum period, yet all of the women reported adherence. An alternative explanation would be the development of drug-resistant virus. Since viral-sensitivity testing was not available, we cannot prove or disprove this possibility.

The SQV soft-gel capsule formulation is no longer available; however, the results from this study are important because the new hard-gel capsule (HGC) formulation has been shown to achieve the targeted plasma levels in pregnant women (6). A dose of 1,200 mg HGC SQV/100 mg RTV given daily to two HIV-infected pregnant women, with areas under the concentration-time curve above 10,000 ng·h/ml, has also been studied. The clinical safety of the new formulation has been reported as well (7). The clinical efficacy and safety parameters of SQV during pregnancy make it a good choice for HIV-infected pregnant women.