RAPID COMMUNICATION

Maternal Toxicity With Continuous Nevirapine in Pregnancy

Results From PACTG 1022
Jane Hitti, MD,* Lisa M. Frenkel, MD,* Alice M. Stek, MD,† Sharon A. Nachman, MD,‡
David Baker, MD,‡ Adolfo Gonzalez-Garcia, MD,§ Arthur Provisor, MD,㛳 Edwin M. Thorpe, MD,¶
Mary E. Paul, MD,# Marc Foca, MD,** Jorge Gandia, MD,†† Sharon Huang, MS,‡‡
Lee-Jen Wei, PhD,‡‡ Laura M. Stevens, BS,§§ D. Heather Watts, MD,¶¶, and James McNamara, MD¶¶
for the PACTG 1022 Study Team

Conclusions: Continuous nevirapine may be associated with in-

Objective: To compare the safety of nelfinavir and nevirapine-
based antiretroviral treatment in HIV–1-infected pregnant women.
Methods: In Pediatric AIDS Clinical Trials Group Protocol 1022,
38 antiretroviral-naive pregnant women at 10–30 weeks’ gestation
were randomized to nelfinavir or nevirapine with zidovudine plus la-
mivudine. The study was suspended because of greater than expected
toxicity and changes in nevirapine prescribing information. The inci-
dence of treatment-limiting hepatic or cutaneous toxicity was com-
pared between groups for all subjects and for the subset with CD4 cell
counts greater than 250 cells/µL at study entry.
creased toxicity among HIV-1-infected pregnant women with CD4
cell counts greater than 250 cells/µL, as has been observed in non-
pregnant women.
Key Words: pregnancy, hepatic toxicity, cutaneous toxicity, nelfina-
vir, nevirapine
(J Acquir Immune Defic Syndr 2004;36:772–776)

Results: Toxicity was seen in 1 (5%) of 21 subjects randomized

to nelfinavir and 5 (29%) of 17 subjects randomized to nevirapine
(P = 0.07). Within the nevirapine group, 1 subject developed fulmi-
nant hepatic failure and died, and another developed Stevens-Johnson
syndrome. The one adverse event associated with nelfinavir occurred
in a subject with a CD4 cell count less than 250 cells/µL. All 5 events
among subjects with a CD4 cell count greater than 250 cells/µL were
associated with nevirapine (P = 0.04).
Received for publication March 31, 2004; accepted May 14, 2004.
From the *University of Washington, Seattle, WA; †University of Southern
California, Los Angeles, CA; ‡State University of New York, Stony
Brook, NY; §University of Miami, FL; 㛳Columbus Regional Healthcare
System, Columbus, GA; ¶University of Tennessee Health Science Center,
Memphis, TN; #Baylor College of Medicine, TX; **Columbia University,
New York, NY; ††City Hospital of San Juan, Puerto Rico; ‡‡Harvard
School of Public Health, Boston, MA; §§Frontier Science and Technology
Research Foundation, Buffalo, NY; and ¶¶National Institutes of Health,
Bethesda, MD.
This study was supported in part by the Pediatric AIDS Clinical Trials Group
of the National Institute for Allergy and Infectious Diseases, the
Pediatric/Perinatal HIV-1 Clinical Trials Network of the National Institute
of Child Health and Development, and the General Clinical Research Cen-
ter Units funded by the National Center for Research Resources. Agouron
Pharmaceuticals: A Pfizer Company, Boehringer-Ingelheim Pharmaceu-
ticals and GlaxoSmithKline Company donated the study medications.
Reprints: Jane Hitti, MD, MPH Department of Obstetrics/Gynecology Uni-
versity of Washington Medical Center, Box 356460 Seattle, WA 98195
(e-mail: jhitti@u.washington.edu).
P otent antiretroviral therapy is commonly prescribed for
pregnant women to reduce maternal viral load and thereby
minimize the risk of mother-to-child transmission of HIV-1, as
well as to treat maternal HIV-1 disease when indicated. How-
ever, the relative safety and efficacy of specific antiretroviral
medications have not been well defined in pregnancy. We ini-
tiated a prospective randomized trial, Pediatric AIDS Clinical
Trials Group (PACTG) 1022, to evaluate the safety and effi-
cacy of two frequently prescribed antiretroviral regimens for
HIV-1–infected pregnant women. Enrollment into this trial
was suspended after greater than expected toxicity was ob-
served, and in response to changes in nevirapine prescribing
information. The objectives of this report are to compare the
protocol-defined treatment-limiting toxicities associated with
nelfinavir compared with nevirapine, and to explore the asso-
ciation of nevirapine toxicity with initiation of HAART among
women with higher CD4 cell counts.
METHODS
HIV–1 infected antiretroviral-naive pregnant women
between 10–30 weeks’ gestation were randomized to either
nelfinavir (1250 mg twice daily) or nevirapine (200 mg once
daily for two weeks and then 200 mg twice daily), which were
administered with zidovudine (300 mg twice daily) plus lam-
ivudine (150 mg twice daily) in an open-label trial design. Eli-
gibility criteria included an HIV-1 viral load greater than 1,000

772 J Acquir Immune Defic Syndr • Volume 36, Number 3, July 1 2004
J Acquir Immune Defic Syndr • Volume 36, Number 3, July 1 2004 Maternal Toxicity With Continuous Nevirapine

copies/mL and the intention to continue antiretroviral therapy
after delivery. Prior zidovudine monotherapy, limited to <8
weeks, was allowed. There was no restriction as to CD4 cell
count at study entry. Women with a baseline alanine amino-
transferase (ALT) greater than 2.5 times the upper limit of nor-
mal, active hepatitis B or C or other serious concurrent illness
were not eligible to enroll in the trial. Participating centers in-
cluded University of Southern California, State University of
New York at Stony Brook, Columbus Regional Healthcare
System, University of Miami, Baylor College of Medicine,
Columbia University, City Hospital at San Juan, University of
Tennessee Health Science Center, University of Washington,
University of Cincinnati, University of California Los Ange-
les, Mount Sinai Hospital, University of California San Diego,
Yale University, Howard University, and the Bronx-Lebanon
Hospital. The Institutional Review Boards of all centers ap-
proved the study protocol, and all subjects provided written
informed consent prior to study entry.
The study protocol included stringent toxicity manage-
ment guidelines that were applied uniformly across all centers.
Abnormal ALT and aspartate aminotransferase (AST) values
were graded according to the Division of AIDS toxicity guide-
lines for adults: Grade 0 < = 1.25 times the upper limit of nor-
mal; Grade 1 = 1.25–2.5 times the upper limit of normal; Grade
2 = 2.5–5.0 times the upper limit of normal; Grade 3 = 5.0–10.0
times the upper limit of normal; and Grade 4 = >10 times
the upper limit of normal. Study treatment was stopped for a
confirmed Grade 3 or 4 ALT or AST, any symptoms possibly
consistent with clinical hepatitis at any grade ALT, or rash
accompanied by urticaria, mucous membrane involvement or
constitutional symptoms. Study personnel at each site were
responsible for deciding whether non-specific symptoms were
possibly consistent with clinical hepatitis and merited treat-
ment discontinuation.
Enrollment was suspended in August 2003 because of
greater than expected toxicity and changes in nevirapine pre-
scribing information that recommended caution for women
with CD4 cell counts greater than 250 cells/µL. These events
led to a previously unscheduled intent-to-treat interim analysis
comparing the incidence of treatment-limiting toxicity by arm,
for all subjects and also for the subset of women with an entry
CD4 cell count greater than 250 cells/µL. A 2-tailed Fisher
exact test was used to determine statistical significance.
RESULTS
Twenty-one subjects were randomized to the nelfinavir
arm and 18 subjects to the nevirapine arm for a total of 39
subjects. One subject in the nevirapine arm withdrew before
treatment was dispensed. This report includes follow-up data
on the remaining 38 subjects through January 1, 2004. Table 1
summarizes the baseline demographic and clinical character-
istics, which were similar for each treatment group. Of note,
74% of subjects had an entry CD4 cell count greater than
250 cells/µL, and all subjects were asymptomatic with respect
to HIV-1. The median length of follow-up as of January 1,
2004 was 38 weeks and was similar between treatment arms.

TABLE 1. Demographic and Clinical Characteristics of the Treatment Groups

Nelfinavir Nevirapine
(n = 21) (n = 17)
n (%) n (%)

Age (median) 25 28
Race/ethnicity
White 0 (0) 1 (6)
African American 8 (38) 9 (53)
Hispanic 12 (57) 7 (41)
Asian/Pacific Islander 1 (5) 0 (0)
Weeks’ gestation at entry (median, range) 22 (15–30) 20 (14–28)
Weeks of follow-up* (median, range) 37 (20–79) 41 (6–72)
Entry viral load (median and range, copies/mL) 7,762 (1,738–85,114) 9,772 (3,388–67,608)
Entry CD4 count (median and range, cells/µL) 324 (30–912) 359 (99–371)
Entry CD4 > 250 cells/µL 14 (67) 14 (82)
Entry CDC** Category A 21 (100) 17 (100)
Entry ALT*** 1.25–2.5 × upper limit of normal 0 (0) 3 (18)
History of Hepatitis B or C 0 (0) 0 (0)
*As of January 1, 2004.
**Centers for Disease Control and Prevention.
***Alanine aminotransferase.

© 2004 Lippincott Williams & Wilkins 773

Hitti et al
Table 2 summarizes the observed toxicity leading to
treatment discontinuation by arm. Treatment-limiting toxicity
was seen in 1 subject on nelfinavir and 5 on nevirapine. The
incidence of treatment-limiting toxicity was 5% (1 of 21) in
the nelfinavir and 29% (5 of 17) in the nevirapine groups
(P = 0.07). Among the subset of participants with an entry CD4
cell count greater than 250 cells/µL, 0 of 14 in the nelfinavir
group and 5 (36%) of 14 in the nelfinavir arm had toxicity
leading to treatment discontinuation (P = 0.04). These adverse
events occurred at 6 weeks on nelfinavir and at 2–26 (median
5) weeks on nevirapine.
Nevirapine-associated treatment-limiting toxicity in-
cluded one subject with Stevens-Johnson syndrome (Table 2,
Subject 2), two subjects with an increase in ALT accompanied
by non-specific symptoms thought by the managing clinicians
to be possibly consistent with clinical hepatitis (Subjects 3 and
4), one subject with Grade 3 ALT without symptoms (Subject
5), and 1 subject with fulminant hepatic failure and death (Sub-
ject 6). The woman who died was a 33-year-old African
American woman, gravida 2 para 1, who began study medica-
tions at 29 weeks’ gestation. At study entry she had no labora-
tory evidence of hepatitis B or C, a normal ALT (59 U/L), an
AST just above normal limits (49 U/L), and an entry CD4 cell
count of 330 cells/µL (22%). Her other medications at study
entry included prenatal vitamins, iron and pseudoephedrine as
needed for sinus congestion. Two weeks later, her ALT and
AST were 34 and 20 U/L, respectively. Four weeks after entry,
she reported a transient facial rash that had resolved by the time
of her study visit. Her ALT and AST had increased to 177 and
244 U/L. Six days later she presented with nausea, malaise,
and jaundice, and she had an ALT of 1851 U/L and AST of
3288 U/L. She stopped all antiretroviral medications, was ad-
mitted to the hospital and 3 days later had a cesarean delivery
of a viable 34-week male infant. She subsequently developed
multi-system organ failure and died 3 days after delivery. A
post-mortem liver biopsy showed diffuse hepatic necrosis
without steatosis or fibrosis. A full autopsy was not performed.
TABLE 2. Adverse Events Leading to Treatment Discontinuation
Weeks’ Weeks on Entry
Gestation Treatment CD4 Cell
Subject Agent at Entry at Onset Count
1 Nelfinavir 22 6 239
2 Nevirapine 17 2 380
3 Nevirapine 21 4 606
4 Nevirapine 23 22 259
5 Nevirapine 19 26 510
6 Nevirapine 29 5 330
*Alanine aminotransferase, U/L
774
J Acquir Immune Defic Syndr • Volume 36, Number 3, July 1 2004
DISCUSSION
We observed greater than expected toxicity associated
with nevirapine during the first phase of this randomized trial.
All of the adverse events in the nevirapine group occurred
among women with an entry CD4 cell count greater than
250 cells/µL. The hepatic necrosis seen on liver biopsy from
the subject who died is consistent with drug-induced hepatic
toxicity, possibly direct hepatocellular injury or an immunoal-
lergic response.1 The absence of significant hepatic steatosis at
the time of death, and the onset occurring early after initiation
of HAART makes nucleoside analogue–associated hepatic
steatosis unlikely.
The maternal death observed in this trial is similar to
three other case reports of maternal death on nevirapine/zido-
vudine/lamivudine.2,3 Two additional cases of maternal mor-
tality associated with nevirapine in pregnancy have been
noted, beyond those cited above (Jorge Pinto, Escola de Me-
dicina, Universidade Federal de Minas Gerais, personal com-
munication; Patrick Robinson, Boehringer-Ingelheim Phar-
maceuticals, personal communication).
The common features of all 6 cases include fulminant
hepatic failure presenting 4–5 weeks after starting nevirapine-
based antiretroviral therapy, in antiretroviral-naive subjects
with previously normal ALT and no history of hepatitis B or C.
Four of these 6 cases occurred in women with a baseline CD4
cell count greater than 250 cells/µL. Furthermore, in these
cases hepatitis progressed rapidly to hepatic failure and death
despite discontinuation of nevirapine and close clinical and
laboratory monitoring. There is one additional published case
report of severe but non-fatal hepatitis associated with nevi-
rapine in a pregnant woman with an initial CD4 cell count
greater than 600 cells/µL.4 These cases are also similar to re-
ports of severe hepatic toxicity among HIV-1–negative indi-
viduals receiving nevirapine for post-exposure prophylaxis.5
These observations of hepatic toxicity in pregnancy, in
our study and other published case reports,2–4 agree with a re-
cent meta-analysis suggesting that non-pregnant women with
ALT* (Grade)
Entry Onset Description
11 (0) 568 (4) Clinical hepatitis
11 (0) 62 (1) Stevens-Johnson syndrome
21 (0) 135 (2) Nausea and vomiting
12 (0) 41 (0) Right upper quadrant pain
60 (1) 311 (3) Asymptomatic
59 (0) 1851 (4) Fulminant hepatic failure and death
© 2004 Lippincott Williams & Wilkins
J Acquir Immune Defic Syndr • Volume 36, Number 3, July 1 2004
CD4 cell counts greater than 250 cells/µL who receive continu-
ous nevirapine are at increased risk for hepatic toxicity6 in-
cluding fulminant hepatic failure and death. Men with higher
CD4 cell counts are also at increased risk for hepatic toxicity,
but at a greater CD4 threshold (400 cells/µL) compared with
women. Thus, there appears to be an interaction between gen-
der and immunologic status as risk factors for rash-associated
hepatic toxicity with continuous nevirapine. While the exact
mechanism for nevirapine-associated hepatic toxicity is not
understood, an immune-mediated hypersensitivity component
is postulated.
It is difficult to estimate the true incidences of treatment-
limiting toxicity and fulminant hepatic failure associated with
continuous nevirapine use in pregnancy. At present, there are
no reliable mechanisms to consistently identify and tabulate
severe but non-fatal toxicity associated with antiretroviral
therapy in pregnancy occurring outside of the context of a
clinical trial. Thus, it is unknown whether pregnancy poses an
additional risk for nevirapine-associated hepatic toxicity be-
yond the risk associated with female gender. It is also not pos-
sible to determine the incidence of fulminant hepatic failure
and death associated with continuous nevirapine use in preg-
nancy, because the number of pregnant women who have re-
ceived nevirapine-containing antiretroviral therapy is unknown.
The data from this and other reports2–4,6 should not be
extrapolated to the use of single-dose intrapartum nevirapine
to prevent mother-to-child HIV-1 transmission. In three large
randomized trials, over 1,600 pregnant women received a
single intrapartum dose of nevirapine without any increase in
maternal toxicity.7–9 Similarly, this report does not imply that
women who initiate nevirapine and tolerate it well should then
discontinue nevirapine if they experience immune reconstitu-
tion. There is no evidence to suggest that immune reconstitu-
tion increases the risk of nevirapine-associated toxicity, re-
gardless of pregnancy status. It is also not known whether
antiretroviral-experienced women with immune reconstitution
who switch to a nevirapine-containing regimen would have an
increased risk of hepatic toxicity.
The major limitation of this study is its small sample
size, which could lead to an overestimate of the incidence of
hepatic toxicity associated with nevirapine during pregnancy.
It is possible that the increased incidence of nevirapine-
associated hepatic adverse events observed in this trial will not
be confirmed in larger data sets. In addition, since the majority
of subjects in both treatment arms were asymptomatic and had
CD4 cell counts greater than 250 cells/µL, this study does not
provide information about nevirapine-associated toxicity
among pregnant women with more advanced HIV-1 disease
and lower CD4 cell counts. However, the toxicities observed
thus far in this small randomized trial may be clinically rel-
evant even though the sample size is small. The single case
of fulminant hepatic toxicity leading to death in our study,
together with similar published case reports involving other
© 2004 Lippincott Williams & Wilkins
Maternal Toxicity With Continuous Nevirapine
pregnant women2,3 raise concern about the safety of continu-
ous nevirapine in pregnancy in women with CD4 cell counts
greater than 250 cells/µL. Since these observations agree with
much larger data sets from non-pregnant women, it would
seem prudent to be cautious in the use of continuous nevirapine
in pregnant women with CD4 greater than 250 cells/µL. The
safety of nevirapine-containing antiretroviral regimens
for pregnant women with lower CD4 cell counts deserves
additional investigation, especially since continuous nevirap-
ine-containing antiretroviral treatment is rapidly becoming a
first-line regimen for pregnant women in resource-limited
countries.
ACKNOWLEDGMENTS
The following individuals assisted with conducting
PACTG 1022: Ana Melendrez, RN, Michael Neely, MD and
James Homans, MD, University of Southern California,
Los Angeles; Jennifer Griffin, RN, Deborah Hattenback, RN
and Sylvia Muniz, State University of New York, Stony
Brook; Dawn Barnes, RN and Anita Whitten, RN, Columbus
Regional Healthcare System; Robert Pass, MD and Terry
Byars, MD, University of Alabama; Chivon D. Jackson, RN,
Christine M. Owen, RN and Valencia Y. Johnson, RN, Baylor
College of Medicine; Alice Higgins, RN and Philip LaRussa,
MD, Columbia University; Midnela Acevedo, MD, Elvia Pérez-
Hernández, MPH and Antonio Rodriguez-Mimoso, MD, City
Hospital at San Juan; Nina Sublette, RN, Pat Flynn, MD, Jill
Utech, RN and Mary Dillard, RN, University of Tennessee
Health Science Center; Deborah Goldman, ARNP, Michele
Acker, ARNP and Kathey Mohan, ARNP, University of Wash-
ington; Joseph Mrus, MD, Patricia Kohler, RN and Michelle
Saemann, RN, University of Cincinnati; Margaret A. Keller,
MD, Marie Beall, MD and Judy Hayes, RN, University of
California, Los Angeles; Kathy Kabot, Paula Karnick, Gloria
Seals and Charles Lampley, Mount Sinai Hospital; Andrew D.
Hull, MD, Patricia Franklin, FNP, Mary Caffery, RN and Ste-
phen A. Spector, MD, University of California, San Diego;
Patricia Strock, PA, Jhoanna Roa, MD, Patricia Houston-Yu,
MS and Sohail Rana, MD, Howard University; Saroj Bakshi,
MD, Murli Purswani, MD, Marilyn Crane, CNM and Mavis
Dummitt, RN, Bronx-Lebanon Hospital.
REFERENCES
1. Lee WM. Drug-induced hepatotoxicity. N Engl J Med. 2003;349:474–
485.
2. Langlet P, Guillaume MP, Devriendt J, et al. Fatal liver failure associated
with nevirapine in a pregnant HIV patient: the first reported case. Gastro-
enterology. 2000;118:A1461.
3. Lyons F, Hopkins S, McGeary A, et al. Nevirapine tolerability in HIV
infected women in pregnancy—a word of caution. 2003: Second Interna-
tional AIDS Society Conference on HIV Pathogenesis and Treatment,
LB27 (http://www.ias.se/abstract/show.asp?abstract_id=11081)
4. Knudtson E, Para M, Boswell H, et al. Drug rash with eosinophilia and
systemic symptoms syndrome and renal toxicity with a nevirapine-con-
775
Hitti et al
taining regimen in a pregnant patient with human immunodeficiency
virus. Obstet Gynecol. 2003;101:1094–1097.
5. Centers for Disease Control and Prevention. Serious adverse events at-
tributed to nevirapine regimens for postexposure prophylaxis after HIV
exposures worldwide, 1997-2000. MMWR. 2001;49:1153–1156.
6. Stern JO, Robinson PA, Love J, et al. A comprehensive safety analysis of
nevirapine in different populations of HIV-infected patients. J Acquir Im-
mune Defic Syndr. 2003;34:S21–S33.
7. Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-
dose nevirapine compared with zidovudine for prevention of mother-to-
776
J Acquir Immune Defic Syndr • Volume 36, Number 3, July 1 2004
child HIV transmission of HIV in Kampala, Uganda: HIVNET 012 ran-
domised trial. Lancet. 1999;354:795–802.
8. Dorenbaum A, Cunningham CK, Gelber RD, et al. Two-dose intra-
partum/newborn nevirapine and standard antiretroviral therapy to reduce
perinatal HIV transmission: a randomized trial. JAMA. 2002;288:189–198.
9. Moodley D, Moodley J, Coovadia H, et al. A multicenter randomized
controlled trial of nevirapine versus a combination of zidovudine and la-
mivudine to reduce intrapartum and early postpartum mother-to-child
transmission of human immunodeficiency virus type 1. J Infect Dis. 2003;
187:725–735.
© 2004 Lippincott Williams & Wilkins