Susceptibility of pediatric HIV-1 isolates to recombinant CD4-IgG₂ (PRO 542) and humanized mAb to the chemokine receptor CCR5 (PRO 140)

Article Outline

• Acknowledgment
• References
• Copyright

To the Editor:

Although remarkable advances have been made in treating HIV-1 infection with antiretroviral agents, including reverse transcriptase inhibitors (RTIs) and protease inhibitors (PI),¹ the use of immune-based HIV-1 entry inhibitor proteins remains an attractive alternate approach, because the HIV-1 virion is prevented from attaching to target cells.² These molecules act via unexploited mechanisms, are active against multidrug-resistant viruses in vitro, and offer new, potentially long-acting modes of therapy.³ In an earlier analysis, we reported on the ability of PRO 542, a recombinant tetrameric CD4-IgG₂ fusion protein that binds to HIV-1 gp120, to reduce HIV-1 viral load in children being treated with combination antiretroviral agents including PI in the Pediatric AIDS Clinical Trials Group (PACTG) protocol 351.⁴ Subsequently, we have had the opportunity to compare the coreceptor usage and in vitro susceptibility of pretreatment viral isolates of protocol 351 study subjects to PRO 542 and PRO 140, a humanized mAb to the chemokine receptor CCR5.⁵ PRO 140 exhibits a broad spectrum of inhibition of primary R5 HIV-1 (binds to CCR5 receptor) isolates obtained from adult patients from the United States and Africa.⁶ PRO 140 also compares favorably with several other mAb and HIV-1 entry inhibitor molecules in its ability to inhibit acute and chronic HIV-1 viruses from European adult patients.⁷ Prolonged coating of CCR5⁺ lymphocytes (>60 d) by PRO 140 (5 mg/kg) was reported in healthy adult male human subjects in a dose-finding safety study.⁸ To provide a rationale for designing a clinical trial of PRO 140 in pediatric patients, we compared PRO 140 and PRO 542 for inhibition of pretreatment HIV-1 isolates of patients enrolled in PACTG protocol 351.

Study subjects were HIV-1 infected by perinatal transmission, 2 to 12 years of age, receiving stable antiretroviral medications for at least 3 months, and exhibiting an HIV-1 RNA viral load of ≥10,000 copies/mL. Two separately recruited cohorts were studied: cohort I (1999; n = 6) and Cohort II (2000-2004; n = 13). Patient plasma and sera were preserved in liquid N₂ (−140°C). Five cohort I patients and 10 cohort II patients had adequate samples for study. Peripheral blood CD4⁺ T-cell counts (cells/μL) and percentages and plasma HIV-1 viral load (copies/mL) determination were performed as previously described.⁴ For the current study, only pretreatment HIV-1 viral isolates, HIV-1 RNA levels, and CD4⁺ T-cell values were used.

Tropism and neutralization data were obtained using a phenotypic assay (HIV PhenoSense Entry Assay; Monogram Biosciences, Inc, South San Francisco, Calif; formerly ViroLogic, Inc) on stored plasma samples collected before study treatment.³ In this assay, HIV-1 envelopes representing the circulating quasispecies (typically 500-5000 quasispecies) are amplified from plasma or sera, subcloned, and used to create single-round reporter viruses. The envelope-complemented reporter viruses are tested for infectivity on U87-CD4-CCR5 and U87-CD4-CXCR4 (chemokine receptor found on CD4⁺ T cells) cells in the presence and absence of inhibitors. IC₅₀ is the concentration of antibody (μg/mL) required to inhibit the virus by 50%. The IC₅₀ values for a given patient isolate were divided by that observed for the dual-tropic control virus (92HT594) on the respective cell type, resulting in a relative IC₅₀ (rIC₅₀). This normalization process controls for minor day-to-day variations in assay performance.

Descriptive statistics, overall and classified by cohort, were computed for the following baseline characteristics: sex, race/ethnicity, age (years), weight (kg), CD4⁺ T-cell count/percentage, and HIV-RNA viral load. Summary statistics of IC₅₀ values were calculated for PRO 140 and PRO 542 (overall and by cohort). Estimates of the prevalence of R5X4 (HIV-1 that binds to both CCR5 and CXCR4 receptors) tropism were also calculated and bounded by 95% exact confidence intervals.

Previous antiretroviral use was quantified using the total number of drugs used in the following categories: PI, RTI (nucleoside and nonnucleoside), the fusion inhibitor enfuvirtide (T20), and total antiretroviral. Antiretroviral use was compared by cohort and by tropism using medians to summarize the results and the Kruskal-Wallis test to examine statistical significance. Associations between previous antiretroviral use and susceptibility to both PRO 140 and PRO 542 were tested using the Spearman rank correlation analysis.

Univariate correlations between viral tropism and various baseline factors such as viral load (HIV-RNA), CD4⁺ T-cell absolute count, and susceptibility to PRO 140 and PRO 542 using rIC₅₀ were assessed using Kruskal-Wallis test. Some dual-tropic patients had PRO 542 IC₅₀ and rIC₅₀ from both R5 and X4 (HIV-1 that binds to CXCR4 receptor) assays. Thus, in those cases, that is, PRO 542, the nonnormalized R5 and X4 IC₅₀ values were compared, and the higher IC₅₀ value and its corresponding rIC₅₀ value were used in the analyses, providing an overall measure of IC₅₀ and rIC₅₀, respectively. As expected, PRO 140 inhibited R5 virus entry but did not affect entry of viruses into U87-CD4-CXCR4 cells, and consequently R5 rIC₅₀ values were used exclusively for PRO 140.

The small sample size provides very limited statistical power; thus, many of the results are presented using descriptive statistics. Nonparametric analyses to test for significance were performed to avoid normality assumptions. The level of significance used was α = 0.05.

Overall baseline characteristics of children (n = 15) who received multiple doses in the PACTG 351 study are as follows: 60% were male, 60% were Hispanic, 33% were African American, and 7% were Caucasian. The median age was 8 years (range, 3-11). The median weight was 25.2 kg (range, 17.6-47.8). The median CD4⁺ T-cell count was 477 cells/μL (range, 13-1849; Table I). The median CD4⁺ T-cell percentage was 23 (range, 1-46). The median HIV-1 RNA was 34,079 copies/mL (range, 2590-167,025 copies/mL; Table I). Patients' baseline characteristics did not differ significantly across cohorts.

Table I. Viral tropism by cohort: cohort I versus cohort II

PATID, Patient identification number; NA, not available (insufficient level of viral replication to determine rIC₅₀).

Patients who were enrolled to cohort II were relatively more susceptible to both PRO 140 and PRO 542 based on their lower median rIC₅₀ compared with patients enrolled to cohort I (Table I). The median PRO 140 rIC₅₀ of patients enrolled to cohort I was 3.10, whereas for cohort II, the median was 1.59, and this 2-fold difference was statistically significant (P = .0196), reflecting in part the modest variation in viral susceptibility to PRO 140 observed in this study. The difference in rIC₅₀ medians of PRO 542 across cohorts was marginally significant (P = .0662; cohort I median, 7.00; cohort II median, 3.26). The overall median nonnormalized IC₅₀ values were 0.25 μg/mL (range, 0.043-0.48 μg/mL) for PRO 140 and 5.5 μg/mL (range, 0.39 to >76 μg/mL) for PRO 542 (data not shown).

Patients of cohort I and cohort II had a median use of 1 PI and 2 RTI. Previous RTI use did not differ significantly between patients with R5 and R5X4 virus (P = .7920). However, previous PI use marginally differed (P = .0952) between patients with R5X4 (median, 2; n = 5) and R5 (median, 1; n = 10) virus. The Spearman rank correlations did not identify any significant correlations between the number of PI or RTI previously used and viral susceptibility to either PRO 140 or PRO 542, indicating no obvious cross-resistance between antiretroviral and PRO 140 or PRO 542. Rather, there was a modest trend (r = −0.4073; P = .1483) toward increased susceptibility to PRO 140 with increased PI use.

Univariate analysis identified trends in lower CD4⁺ T-cell counts and higher HIV-1 RNA levels in R5X4 patients compared with R5 patients (Table II). R5 viruses had a median PRO 542 rIC₅₀ of 5.40 (n = 10), whereas R5X4 viruses were slightly more susceptible to PRO 542, showing a relatively lower median rIC₅₀ of 3.30 (n = 5). However, based on the P value of .4624, there was insufficient evidence to say that the difference was significant. R5 viruses had a median PRO 140 rIC₅₀ of 2.75 (n = 10). R5X4 viruses were slightly more susceptible to PRO 140 (rIC₅₀ = 1.00; n = 4) when tested on U87-CD4-CCR5 cells, and this difference was significant (P = .0237).

Table II. Univariate association between viral tropism and other baseline factors

Viral tropism	n	Median	Minimum	Maximum	P value∗
Viral tropism and CD4+ T-cell count (cells/μL)
R5	10	768.5	36	1849	.0662
R5-X4 (dual)	5	121.0	13	612
Viral tropism and HIV-RNA viral load (copies/mL)
R5	10	17,887.5	2590	66,990	.0864
R5-X4 (dual)	5	48,562.0	22,482	167,025
Viral tropism and PRO 542 rIC50
R5	10	5.4	0.30	59.32	.4624
R5-X4 (dual)	5	3.3	1.18	19.00
Viral tropism and PRO 140 rIC50
R5	10	2.75	0.90	5.2	.0237
R5-X4 (dual)	4	1.00	0.52	2.3

mAbs and immunoglobulin fusion proteins are assuming an important role in the treatment of numerous diseases in medicine, such as cancer, autoimmune disease, and infectious disease.⁹ The CCR5 mAb and CD4-Ig fusion protein studied here offer alternate molecular modes of therapy for HIV-1–infected patients. The results of this study suggest that PRO 140 effectively inhibits primary pediatric HIV-1 isolates, both R5 viruses and the CCR5-dependent component of R5X4 isolates. Also, the data suggest that R5 monotropism of pediatric HIV-1 strains is associated with higher CD4⁺ T-cell counts and lower HIV-1 RNA copy numbers, whereas R5X4 dual-tropism is associated with lower CD4⁺ T-cell counts and higher HIV-1 RNA copy numbers, results consistent with recent findings in adult patients.¹⁰

Some caveats need to be made in the interpretation of the results of this small pediatric study. It is apparent that the enrollment of patients into this study took several years, and the use of antiretroviral drugs changed considerably during this time, from 1999 to 2004. Thus, the later cohort used different background regimens as reflected in greater PI use. Therapy with PI and RTI can affect viral tropism.¹¹ Despite the preliminary nature of this in vitro study, it is reasonable to conclude that both PRO 140 and PRO 542 inhibit the growth of R5 and R5X4 viruses from children in vitro. In terms of PRO 140 and future phase I/II clinical trials in children, our findings provide support for a study to explore the potential of this CCR5 mAb as a potentially long-acting HIV-1 therapy.

Back to Article Outline

We thank the Pediatric AIDS Clinical Trials Group Protocol 351 Study Team Cohort II Coordinators: Veronica Y. Amos, PhD, MS, RN, APRN, Children's National Medical Center, Washington, DC; Theresa Dunaway, RN, BSN, MBA, Children's Hospital of LA, Los Angeles, Calif; Maryanne Dillon, BSN, NP, University of California Medical Center, Los Angeles, Calif; Chivon D. Jackson, RN, BSN, ADN, Baylor College of Medicine, Houston, Tex; Stephanie Kujawa, RN, Children's Hospital Orange County, Orange, Calif, and Susan E. Marks, RN, Miller Children's Hospital, Long Beach Memorial Medical Center, Long Beach, Calif.

Also, we thank the collaborating physicians, Antonio Carlos Arrieta, MD, Children's Hospital of Orange County, Orange, Calif, Audra Deveikis, MD, Miller Children's Hospital, Long Beach, Calif, and Robert J. Israel, MD, Progenics Pharmaceuticals, Inc, Tarrytown, NY; the support staff, Courtney Ashton, BS, MT, Laboratory Data Coordinator, Elaine Ferguson, RPh, MS, Division of AIDS, National Institute of Allergy and Infectious Diseases, Bethesda, Md, Stephen Ramos, Progenics Pharmaceuticals Inc, Yolanda Lie, Monogram Biosciences, Inc, and Erin Smith, Westat, Rockville, Md; and Carolyn P. Jackson for assistance in manuscript preparation.

Informed consent was obtained from parents or caretakers, and assent was obtained from children ≥7 years old where required. Human experimentation guidelines of the US Department of Health and Human Services and of the authors' institutions were followed in the conduct of this research.

Back to Article Outline

References 

1. Hammer SM. Clinical practice. Management of newly diagnosed HIV infection. N Engl J Med. 2005;353:1702–1710
   • View In Article
   • CrossRef
2. Castagna A, Biswas P, Beretta A, Lazzarin A. The appealing story of HIV entry inhibitors: from discovery of biological mechanisms to drug development. Drugs. 2005;65:879–904
   • View In Article
   • MEDLINE
   • CrossRef
3. Jacobson JM, Israel RJ, Lowy I, Ostrow NA, Vassilatos LS, Barish M, et al. Treatment of advanced human immunodeficiency virus type 1 disease with the viral entry inhibitor PRO 542. Antimicrob Agents Chemother. 2004;48:423–429
   • View In Article
   • MEDLINE
   • CrossRef
4. Shearer WT, Israel RJ, Starr S, Fletcher CV, Wara D, Rathore M, et al. Recombinant CD4-IgG2 in human immunodeficiency virus type 1-infected children: phase 1/2 study. The Pediatric AIDS Clinical Trials Group Protocol 351 Study Team. J Infect Dis. 2000;182:1774–1779
   • View In Article
   • MEDLINE
   • CrossRef
5. Trkola A, Ketas TJ, Nagashima KA, Zhao L, Cilliers T, Morris L, et al. Potent, broad-spectrum inhibition of human immunodeficiency virus type 1 by the CCR5 monoclonal antibody PRO 140. J Virol. 2001;75:579–588
   • View In Article
   • MEDLINE
   • CrossRef
6. Cilliers T, Nhlapo J, Coetzer M, Orlovic D, Ketas T, Olson WC, et al. The CCR5 and CXCR4 coreceptors are both used by human immunodeficiency virus type 1 primary isolates from subtype C. J Virol. 2003;77:4449–4456
   • View In Article
   • MEDLINE
   • CrossRef
7. Rusert P, Kuster H, Joos B, Misselwitz B, Gujer C, Leemann C, et al. Virus isolates during acute and chronic human immunodeficiency virus type 1 infection show distinct patterns of sensitivity to entry inhibitors. J Virol. 2005;79:8454–8469
   • View In Article
   • MEDLINE
   • CrossRef
8. Olson WC, Doshan H, Zhan C, Mezzatesta J, Assumma A, Czarnecky R, et al. New antiretroviral agents and approaches: clinical studies. 13th Conference on Retroviruses and Opportunistic Infections, Denver, Colo, February 5-9, 2006. Available at: http://www.retroconference.org/2006/.
   • View In Article
9. Liossis SN, Tsokos GC. Monoclonal antibodies and fusion proteins in medicine. J Allergy Clin Immunol. 2005;116:721–729
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (213 KB)
   • CrossRef
10. Moyle GJ, Wildfire A, Mandalia S, Mayer H, Goodrich J, Whitcomb J, et al. Epidemiology and predictive factors for chemokine receptor use in HIV-1 infection. J Infect Dis. 2005;191:866–872
    • View In Article
    • MEDLINE
    • CrossRef
11. Kitchen CM, Philpott S, Burger H, Weiser B, Anastos K, Suchard MA. Evolution of human immunodeficiency virus type 1 coreceptor usage during antiretroviral Therapy: a Bayesian approach. J Virol. 2004;78:11296–11302
    • View In Article
    • MEDLINE
    • CrossRef