Safety and Immunogenicity of an HIV-1 Recombinant Canarypox ...

BRIEF REPORT

Safety and Immunogenicity of an HIV-1 Recombinant Canarypox Vaccine in Newborns and Infants of HIV-1–Infected Women Daniel C. Johnson, Elizabeth J. McFarland, Petronella Muresan, Terence Fenton, James McNamara, Jennifer S. Read, Elizabeth Hawkins, Pamela L. Bouquin, Scharla G. Estep, Georgia D. Tomaras, Carol A. Vincent, Mobeen Rathore, Ann J. Melvin, Sanjay Gurunathan, and John Lamberta

Mother-to-child transmission of HIV-1 can occur during the antepartum, intrapartum, or postnatal period. Breast-feeding contributes to as much as 42% of the overall risk for motherto-child transmission of HIV-1 [1]. Active immunization of newborns, coupled with antiretroviral prophylaxis or passive immunization, is a potential intervention to prevent HIV-1 transmission via breast-feeding. The first study of HIV-1 vaccines in HIV-1–exposed infants evaluated 2 envelope glycoprotein gp120 recombinant subunit vaccines [2]. The vaccines were well tolerated, and most infants developed HIV-1–specific proliferative and antibody responses [3, 4]. Subunit vaccines fail to elicit cytotoxic T lymphocyte

Received 28 February 2005; accepted 8 July 2005; electronically published 9 November 2005. Reprints or correspondence: Dr. Daniel C. Johnson, Dept. of Pediatrics, Sinai Children’s Hospital, F444, 15th St. and California Ave., Chicago, IL 60608 ([email protected]). The Journal of Infectious Diseases 2005; 192:2129–33 2005 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2005/19212-0016$15.00

Presented in part: 5th Conference on Retroviruses and Opportunistic Infections, Chicago, 1–5 February 1998 (abstract 251); 6th Conference on Retroviruses and Opportunistic Infections, Chicago, 31 January–4 February 1999 (abstract 43); Second Conference on Global Strategies for the Prevention of HIV Transmission from Mother to Infants, Montreal, 31 August–2 September 1999 (abstract 261); 13th International AIDS Conference, Durban, South Africa, 9–14 July 2000 (abstract WeOrC582); 10th Conference on Retroviruses and Opportunistic Infections, Boston, 10–14 February 2003 (abstract C-36). Potential conflicts of interest: S.G. is employed by Sanofi Pasteur. All other authors: no conflicts reported. Financial support: Pediatric AIDS Clinical Trials Group, sponsored by the National Institute of Allergy and Infectious Diseases (cooperative agreement AI-41089); the National Institute of Child Health and Development (contracts HD-33162 and HD-33345); Children’s Hospital of Philadelphia and University of Colorado Health Sciences General Clinical Research Center, National Centers for Research Resources (RR-000240 and RR-00069); funding for the endpoint ELISAs was provided by the HIV Vaccine Trials Network, sponsored by the National Institute of Allergy and Infectious Diseases (AI-46725). a Author affiliations are listed after the text.

BRIEF REPORT • JID 2005:192 (15 December) • 2129

Downloaded from http://jid.oxfordjournals.org/ by guest on September 12, 2016

Pediatric AIDS Clinical Trials Group protocol 326 is a study of 2 formulations of recombinant canarypox ALVAC vaccine (vCP205) against human immunodeficiency virus type 1 (HIV-1). HIV-1–exposed infants were randomized to receive 1 of 2 formulations of vCP205 or placebo at birth and 4, 8, and 12 weeks. The vaccines were safe. Lymphoproliferative responses were detected at ⭓2 time points in 44%–56% of vaccinees and none of the placebo recipients. A cytotoxic T lymphocyte response on at least 1 occasion was detected in 62.5% of infants in cohort 1 (106.08 median tissue culture dose [TCID50] vaccine formulation) and 44% of infants in cohort 2 (106.33 TCID50 vaccine formulation). Rare mucosal immunoglobulin A responses and no measurable vaccineelicited serum antibodies were detected. In children, vCP205 appeared to be safe and immunogenic.

(CTL) responses that may be critical to prevent or eliminate HIV-1 infection [5]. Priming of CD8+ CTLs can be successfully achieved with live attenuated virus or virus vector vaccines. One such vaccine is a recombinant canarypox virus (ALVAC) engineered to express foreign genes. In mammalian cells, canarypox undergoes an abortive cycle of replication and does not produce infectious progeny, suggesting that ALVAC will not disseminate or be transmitted to contacts [6, 7]. Subjects, materials, and methods. Pediatric AIDS Clinical Trials Group protocol 326 is a multicenter, randomized, double-blind, placebo-controlled study of the safety and immunogenicity of live recombinant ALVAC HIV-1 vaccines in HIV-1–exposed newborns and infants. Site institutional review board approvals and informed consent from enrolling parents or guardians were obtained. All HIV-1–infected pregnant women and their infants received prevention management of HIV-1 mother-to-child transmission in accordance with the site’s standard of care. Infants were excluded at study entry if they were of a gestational age of !37 weeks, had received HIV-1–specific immunotherapy, were documented to have or were suspected of having a serious illness, or had significantly abnormal laboratory values. Infants could not be breast-fed. Infants were randomized to receive 1 of 2 formulations of study vaccine or saline placebo. Doses were given at study entry (within 72 h of birth) and at weeks 4, 8, and 12. Production of the recombinant canarypox vector has been described elsewhere, as has a description of the study vaccine, vCP205 (Sanofi Pasteur) [8, 9]. Two formulations of the vaccine were used. Cohort 1 received 106.08 TCID50, and cohort 2 received 106.33 TCID50. Production of the latter formulation in-

2130 • JID 2005:192 (15 December) • BRIEF REPORT

Matthews). Titers (equivalent between serum/plasma pairs) are defined as the reciprocal dilution that yielded 50% maximum binding of the standard positive control replicated on each assay, as described elsewhere [12]. Salivary HIV-1–specific IgA antibodies to p24, gp160MN, gp120MN, gp120LAV, and baculovirus supernatant control (Protein Sciences) were determined by batched dot blot assay at weeks 0, 12, and 24 [13]. Assays included HIV-1–positive and –negative control saliva samples to establish intensity cutoffs. Reactivity to any of the HIV-1 antigens defined a positive response. For subjects who did not complete treatment, T cell and immunological data from time points subsequent to treatment discontinuation were not used in analyses. The difference across cohorts with respect to changes from baseline in CD4+ and CD8+ cell count and percentage were assessed using the Kruskal-Wallis test. The rates of lymphoproliferative responses were compared among the 3 cohorts, using Fisher’s exact test. Results. From March 1998 to July 1999, 28 newborns at 7 sites were randomized into 1 of 3 groups: 9 into cohort 1, 11 into cohort 2 (9 completed the study regimen), and 8 into the placebo group (5 completed the study regimen). Two subjects randomized into cohort 2 did not complete the regimen, at the parents’ request. One placebo recipient withdrew from the study before immunization, because of thrombocytopenia; 2 others withdrew at the parents’ request. The distribution of sexes and races among subjects was 46% male, 11% white, 75% black, 11% Hispanic, and 3% other. There were no significant differences in baseline characteristics across cohorts. Other than 1 placebo recipient, no child was known to have motherto-child transmission of HIV-1. Local reactions were infrequent and minor. No subject experienced systemic, vaccine-associated adverse events greater than grade 3, and no subject experienced vaccine-associated laboratory abnormalities greater than grade 3. CD4+ and CD8+ T cell counts and percentages from baseline to subsequent time points were not significantly different among the 3 cohorts (data not shown). At week 6, after 2 doses of vaccine, positive responses were observed for gp160 in 17% and 11% and for p24 in 40% and 33% of vaccinees in cohorts 1 and 2, respectively (figure 1). At weeks 24 and 52, a maximum of 1 placebo recipient was positive, and none had repeat positive lymphoproliferative responses. Differences among the groups with regard to cumulative lymphoproliferative responses to p24 through week 104 were statistically significant (P p .01 ) (figure 2). Because some placebo recipients had single positive lymphoproliferative responses, we applied a more stringent criterion, requiring at least 2 positive assay results to define a vaccine response. By use of this definition, 5 and 1 subjects in cohorts 1 and 2, respectively, responded to gp160, and 4 and 4 subjects in cohorts 1 and 2,


cluded an additional step, in an attempt to increase its concentration. The difference in titer between the 2 formulations was minimal. Safety assessments performed until age 2 years included direct observation for 1 h after immunization, telephone contact at 24 and 48 h after each immunization, and solicitation of an interim history at each study visit. Local and systemic reactions were recorded. Hematologic and chemical tests and lymphocyte subset determinations were performed at baseline or week 2 and at weeks 6, 10, 14, 24, and 104. Lymphoproliferative and CTL responses were determined at weeks 6, 10, 14, 24, 52, and 104 in 5 Pediatric AIDS Clinical Trials Group laboratories, using standardized methods and single, shared lots of reagents. Lymphoproliferative responses to p24 and baculovirus supernatant control (2.5 and 5 mg/mL; Protein Sciences), gp160MN (1 and 2 mg/mL; Aventis Pasteur), pokeweed mitogen (10 mg/mL; Sigma), tetanus antigen (1.25 and 2.25 mg/mL; Sanofi Pasteur), and Candida antigen (10 and 50 mg/mL; Geer) were measured by tritiated thymidine incorporation assay [2]. When 2 concentrations were tested, the maximal stimulation index (counts per minute for antigen divided by counts per minute for control) for a given antigen was used for analysis. Six samples were considered unreliable and excluded because the stimulation index was !5 for pokeweed mitogen and !3 for both tetanus and Candida antigen. A positive lymphoproliferative response was defined as a stimulation index of ⭓3 for the specified HIV-1 antigen. CTL responses were determined for HIV-1 Env and Gag by 51 Cr release assay, using peripheral blood mononuclear cells stimulated in vitro with autologous Epstein-Barr virus–transformed B lymphoblastoid cell lines expressing HIV-1 antigens [10, 11]. Target cells were 51Cr-labeled B lymphoblastoid cells infected with vaccinia virus recombinants expressing HIV-1 Env (vp1174:gp160MN), HIV-1 Gag (vp1287:gagLAI), HIV-1 Env/Gag (vp1291:gp120MN;gag/proLAI), or parental vector (L-var) (provided by J. Tartaglia, Sanofi Pasteur). The epitopes expressed by the hybrid and single HIV-1 gene vectors differed slightly, as confirmed by analysis with epitope-specific cloned cell lines (data not shown). HIV-1–specific lysis of ⭓10% for at least 2 effector-to-target ratios defined a positive assay. Assays were considered uninterpretable and were excluded if the highest effector-to-target ratio was !25:1 or the spontaneous-to-maximum ratio was 130%. Samples from HIV-1–infected subjects were tested on a quarterly basis as positive controls. The presence of vaccine-specific antibodies was assessed in cryopreserved serum and/or plasma by quantitative ELISA at weeks 0 and 14. Six serial dilutions beginning at 1:50 were tested in duplicate in microtiter plates coated with either purified gp120MN (provided by VaxGen) or DP31, a synthetic gp41 AVERY epitope not present in the vaccine (provided by T.

respectively, responded to p24. No placebo recipient had a repeat response. When the vaccine cohorts were combined, the median stimulation indexes for those with repeated responses to gp160 and p24, respectively, were 6.4 (range, 3.5–17.8) and 6.15 (range, 3.1–70.6) for the first positive assay and 7.1 (range, 3.2–30.7) and 6.2 (range, 3–15.3) for the second positive assay. HIV-1–specific CTL responses were detected as early as week 6 and as late as week 52. At week 6, responses to env, gag, or env/gag were detected in 33%–50% of vaccinees. One placebo recipient had a response to env detected at week 14. The cumulative percentages of subjects responding to the env/gag combination vector on at least 1 occasion were 50% (4/8), 38% (3/8), and 0% (0/4) for cohort 1, cohort 2, and placebo recipients, respectively. Among those responding, the median env/ gag-specific lysis was 13.3% (range, 10.2%–29.3%) and 22.9% (range, 15.1%–45.1%) (at an effector-to-target cell ratio of 25: 1) for cohort 1 and 2, respectively. There was a CTL response on at least one occasion in 62.5% and 44% of the evaluable subjects in cohorts 1 and 2, respectively. One subject from each vaccine cohort had repeatedly positive CTL responses to 2 of

Figure 2. Cumulative percentage of subjects with a positive lymphoproliferative response at week 104 (top panel) or with repeated (⭓2) positive lymphoproliferative responses at any week after baseline through week 104 (bottom panel). n, no. of subjects tested at time point in cohort 1 (106.08 TCID50), cohort 2 (106.33 TCID50), and the placebo group, respectively. Rates of responses were compared among groups, using Fisher’s exact test. BRIEF REPORT • JID 2005:192 (15 December) • 2131


Figure 1. Percentage of subjects with positive lymphoproliferative responses to gp160 (top panel) and p24 (bottom panel), determined by tritiated thymidine incorporation in peripheral blood lymphocytes. A positive response was defined as a stimulation index of ⭓3. Vaccination with study vaccine or placebo was given at weeks 0, 4, 8, and 12. n, no. of subjects tested at time point in cohort 1 (106.08 TCID50), cohort 2 (106.33 TCID50), and the placebo group, respectively. Rates of responses were compared among groups, using Fisher’s exact test.

the same vectors. The small sample size did not permit meaningful statistical testing. All subjects had positive results in serum for HIV-1–binding antibody to both gp120 and DP31 at week 0 of the study. Maternal antibody levels decreased in all subjects, as indicated by the decrease in DP31-binding antibody titer. There were no significant differences in the gp120-specific binding antibody titers between the 3 groups at weeks 0 and 14. Eleven percent (1/9) of subjects in cohort 1 and 33% (3/9) of subjects in cohort 2 had measurable salivary IgA directed at HIV-1 at 1 time point. HIV-1–specific salivary IgA was not detected in placebo recipients (n p 4). Discussion. This study is the first, to our knowledge, to test an ALVAC vector-based vaccine in newborns or infants. Our data suggest that vCP205 is safe in newborns and infants. HIV-1–specific proliferative responses provide evidence of vaccine-induced cell-mediated immune responses. Because placebo recipients also had a lymphoproliferative response at single time points, perhaps because of in utero antigen exposure or nonspecific cross-reactivity, we applied a more stringent criterion, requiring subjects have ⭓2 positive assay results to be defined as responding to vaccine. No placebo recipient had repeated positive lymphoproliferative responses, whereas 44%– 56% of vaccinees had repeated lymphoproliferative responses

2132 • JID 2005:192 (15 December) • BRIEF REPORT

Chicago, Illinois (D.C.J.); Department of Pediatric Infectious Diseases, University of Colorado Health Sciences Center, Denver (E.J.M.); Statistical and Data Analysis Center, Harvard School of Public Health, Boston, Massachusetts (P.M. and T.F.); Clinical Immunology Branch, Division of Allergy and Immunology Transplantation, National Institute of Allergy and Infectious Diseases (NIAID), National Institutes of Health (NIH) (J.M.), Pharmaceutical Affairs Branch, Division of AIDS, NIAID, NIH (S.G.E.), and Pediatric, Adolescent, and Maternal AIDS Branch, National Institute of Child Health and Human Development, NIH (J.S.R.), Bethesda, Maryland; Social & Scientific Systems, Inc., Silver Spring, Maryland (E.H.); University of Maryland, Institute of Human Virology, Baltimore (J.L.); Department of Surgery and Center for Virology, Duke University Medical Center, Durham, North Carolina (G.D.T.); Children’s Hospital of Philadelphia, Philadelphia, Pennyslvania (C.A.V.); Sanofi Pasteur, Swiftwater, Pennyslvania (S.G.); University of Florida Health Science Center, Jacksonville (M.R.); University of Washington and Children’s Hospital and Regional Medical Center, Seattle (A.J.M.); Institute of Child Health, London, United Kingdom (J.L.). P.L.B. is independently employed. Acknowledgments We acknowledge Sanofi Pasteur, the Pediatric AIDS Clinical Trials Group system, the Pediatric Immunology Core Laboratories for performance of immunogenicity assays, and the following sites and staff members (as well as the patients from the sites) who contributed to this study: Steven D. Douglas, University of Pennsylvania; Melissa Scites and Abeer Khayat, University of Florida, Jacksonville; Lisa M. Frenkel, Children’s Hospital of Seattle; Christine Elsen, University of Chicago Children’s Hospital; Janie Kappius and Myron Levin, University of Colorado; and the Bellevue and San Francisco General Hospital sites. We also thank Ruth Dickover, Paul Harding, Katherine Lurzuriaga, Diane Wara, Carmen White, and Michael L. Greenberg for their involvement in this study.

References 1. Breastfeeding and HIV International Transmission Study Group. Late postnatal transmission of HIV-1 in breast-fed children: an individual patient data meta-analysis. J Infect Dis 2004; 189:2154–66. 2. Borkowsky W, Wara D, Fenton T, et al. Lymphoproliferative responses to recombinant HIV-1 envelope antigens in neonates and infants receiving gp120 vaccines. J Infect Dis 2000; 181:890–6. 3. Cunningham CK, Wara DW, Kang M, et al. Safety of two recombinant HIV-1 envelope vaccines in neonates born to HIV-1 infected women. Clin Infect Dis 2001; 32:801–7. 4. McFarland EJ, Borkowsky W, Fenton T, et al. Serologic responses to HIV-1 envelope in neonates receiving a HIV-1 recombinant gp120 vaccine. J Infect Dis 2001; 184:1331–5. 5. Rowland-Jones SL, McMichael A. Immune responses in HIV-exposed seronegatives: have they repelled the virus? Curr Opin Immunol 1995;7: 448–55. 6. Plotkin SA, Cadox M, Meignier B, et al. The safety and use of canarypox vectored vaccines. Dev Biol Stand 1995; 84:165–70. 7. Baxby D, Paoletti E. Potential use of non-replicating vectors as recombinant vaccines. Vaccine 1992; 10:8–9. 8. Pialoux G, Excler JL, Riviere Y, et al. The AGIS Group and lAgence Nationale de Recherche sur le SIDA: a prime-boost approach to HIV


to at least 1 HIV-1 antigen. CTL responses on 11 occasion were seen in 150% of vaccine recipients, compared with only 1 placebo recipient. The failure to identify more subjects who responded to the vaccine may be due to the low frequency of responding cells among circulating lymphocytes. Alternatively, the timing of samples might not be optimal, since the ideal time for measuring the immune response to vaccination has not been formally evaluated in infants. Finally, there might be biological variation in responses, with fluctuations related to external stimulation or sequestration of responder cells in lymphoid tissue. Vaccination failed to produce a measurable serum antibody response. The vaccine may not have been able to generate HIV1 antibody, or the response may have been too small and/or transient to be measured in the presence of maternal antibody. Studies of adults, using similar ALVAC constructs, have been shown to be safe and have shown humoral and cellular responses similar to or greater than those found in our study. Several ALVAC trials in adults have boosted vaccinees with a subunit vaccine. This usually generates antibody responses more rapidly and with greater intensity than does either vaccine alone [14]. In developing countries, breast-feeding recommendations are complex because of the health risk and social consequences associated with not breast-feeding; therefore, breast-feeding remains a significant cause of mother-to-child transmission of HIV-1 [1]. Vaccination of newborns could help to reduce transmission of HIV-1 via breast milk. Vaccination of newborn macaques with poxvirus simian immunodeficiency virus (SIV) vaccines (ALVAC-SIVgpe and modified vaccinia Ankara-SIVgpe) has protected against repeated oral exposure to SIV, suggesting that a parenteral vaccine could protect breast-fed infants [15]. Lymphoproliferative and CTL responses as early as week 6 in some of our study subjects suggests that an option worth examining to prevent transmission via breast-feeding is a combination of passive immunization and/or antiretroviral prophylaxis in the first weeks after birth, followed by vaccination to induce active immunity for the subsequent period of exposure to breast milk. Our study demonstrates some of the challenges associated with studying vaccine immunogenicity in newborns: small blood volumes and inherent difficulty of performing phlebotomy in young infants limits the quantity of blood for testing; repeated venipuncture, zidovudine prophylaxis, and physiological aspects of the newborn result in anemia; and, when this study began, relatively large blood volumes were needed to perform immunological studies, further complicating testing. Fortunately, assays developed more recently, such as enzymelinked immunospot or flow-based assays, measure responses with fewer cells and should facilitate future studies. Author affiliations. Sinai Children’s Hospital, Chicago, and Rosalind Franklin University of Medicine and Science, North

9. 10.

11.

12.

preventive vaccine using a recombinant canarypox virus expressing glycoprotein 160 (MN) followed by a recombinant glycoprotein 160 (MN/ LAI). AIDS Res Hum Retroviruses 1995; 11:373–81. Cadoz M, Strady A, Meignier B, et al. Immunization with canarypox virus expressing rabies glycoprotein. Lancet 1992; 339:1429–32. McFarland EJ, Harding PA, Schooley RT, Kuritzkes DR. In vitro effects of interleukin-12 on human immunodeficiency virus type 1 (HIV-1) specific cytotoxic T-lymphocytes from HIV-1 infected children. J Immunol 1998; 161:513–9. Belshe RB, Gorse GJ, Mulligan MJ, et al. Induction of immune responses to HIV-1 by canarypox virus (ALVAC) HIV-1 and gp120 SF-2 recombinant vaccines in uninfected volunteers. AIDS 1998; 12:2407–15. Goepfert PA, Horton H, McElrath MJ, et al. High-dose recombinant

canarypox vaccine expressing HIV-1 protein, in seronegative human subjects. J Infect Dis 2005; 192:1249–59. 13. Martin NL, Rautonen J, Crobleholme W, Rautonen N, Wara DW. A screening test for the detection of anti-HIV-1 IgA in young infants. Immunol Invest 1992; 21:65–70. 14. Gorse GJ, Patel GB, Belshe RB. HIV type 1 vaccine induced T cell memory and cytotoxic T lymphocyte responses in HIV type 1 uninfected volunteers. AIDS Res Hum Retroviruses 2001; 17:1175–89. 15. Van Rompay KKA, Abel K, Lawson JR, et al. Attenuated poxvirus-based simian immunodeficiency virus (SIV) vaccines given in infancy partially protect infant and juvenile macaques against repeated oral challenge with virulent SIV. J Acquir Immune Defic Syndr 2005; 38:124–34.


BRIEF REPORT • JID 2005:192 (15 December) • 2133