Adeno-Associated Virus Delivery of Anti-HIV Monoclonal Antibodies Can Drive Long-Term Virologic Suppression

Abstract

Long-term delivery of anti-HIV monoclonal antibodies (mAbs) using adeno-associated virus (AAV) vectors holds promise for the prevention and treatment of HIV infection. We describe a therapy trial in which four rhesus monkeys were infected with SHIV-AD8 for 86 weeks before receiving the AAV-encoded mAbs 3BNC117, 10-1074, and 10E8. Although anti-drug antibody (ADA) responses restricted mAb delivery, one monkey successfully maintained 50-150 μg/mL of 3BNC117 and 10-1074 for over 2 years. Delivery of these two mAbs to this monkey resulted in an abrupt decline in plasma viremia, which remained undetectable for 38 successive measurements over 3 years. We generated two more examples of virologic suppression using AAV delivery of a cocktail of four mAbs in a 12-monkey study. Our results provide proof of concept for AAV-delivered mAbs to produce a "functional cure." However, they also serve as a warning that ADAs may be a problem for practical application of this approach in humans.

Keywords: AAV vector; HIV broadly neutralizing antibodies; SHIV infection; adeno-associated virus; anti-drug antibodies; functional cure; gene therapy; immunotherapy; rhesus macaques; viral reservoir.

Fig. 1.. Persistent virological control of SHIV in one monkey after AAV-mAb therapy (four-monkey study).

(A) Four rhesus macaques were experimentally infected with SHIV-AD8 at week 0 and viral loads were quantitated in plasma over time. At week 86 post-SHIV infection, the monkeys received intramuscular inoculations of rAAV vectors coding for three different anti-HIV broadly neutralizing antibodies (10E8, 3BNC117 and 10–1074). (B) Viral loads in monkey rh2438 before and after AAV-mAb delivery. (C) Extended SHIV viral loads in plasma before and after AAV-mAb therapy. Note: the viral load limit of detection (15 copies of viral RNA per ml of plasma) is indicated by a dashed line. Samples below the limit of detection were assigned the values of 5 or 10 for graphing purposes and to avoid overlap. See also Fig. S1 and Fig. S2.

Fig. 2.. mAb levels in serum following rAAV administration are high when anti-antibodies (ADAs) are low or absent (four-monkey study).

Levels of anti-HIV broadly neutralizing antibodies (A) 10E8, (B) 3BNC117 and (C) 10–1074 in serum were quantified by ELISA in the four AAV-recipient monkeys. The levels of delivered antibody remained similar to those shown through 132 weeks of measurements. Levels of host anti-antibodies generated to each of the antibodies delivered in A-C: (D) 10E8, (E) 3BNC117 and (F) 10–1074 were quantified by ELISA using serum from the four AAV recipients. Note: macaque rh2443 had to be euthanized at week 23 post-AAV inoculation due to SHIV related complications. See also Fig. S1.

Fig. 3.. Sharp decline in host anti-p27 Gag antibody responses, host anti-gp120 responses and cell-associated SHIV RNA and DNA levels following AAV-mAb therapy in rh2438.

(A) Antibody reactivity to p27 protein was measured in serum from rh2438 by standard ELISA. Samples obtained before SHIV infection, during infection and after AAV administration were tested. (B) Same as in (A) but now samples were tested against denatured gp120 protein by standard ELISA; 3BN117 and 10–1074 did not react with denatured gp120. PBMCs isolated from rh2438 were tested for the presence of (C) viral RNA and (D) DNA. Results were plotted as RNA or DNA copies per million cells.

Fig. 4.. Monkeys r14097 and r14121 exhibited suppression of viral loads after AAV-mAb therapy.

(A) Twelve rhesus macaques were experimentally infected with SHIV-AD8 at week 0 and viral loads were quantitated in plasma over time. At week 36 post-SHIV infection the monkeys were separated into two groups and received intramuscular inoculations of rAAV8 vectors coding for different anti-HIV broadly neutralizing antibodies (the “bi-group”, on the left panel, received 3BNC117 and 10–1074; the “quad-group”, on the right panel, received PGT145, 35O22, N6 and PGT128). At week 60 the monkeys received a booster inoculation of rAAV1 vectors coding for the same corresponding antibodies. Note: the viral load limit of detection (15 copies of viral RNA per ml of plasma) is indicated by a dashed line. Samples below the limit of detection were assigned the values of 5, 7 or 10 for graphing purposes and to avoid overlap. Macaque r13029 had to be euthanized at week 76 post-infection due to SHIV related complications. (B) Extent of virological control in monkeys r14097 and r14121 from (A). (C) Levels of anti-HIV broadly neutralizing antibodies PGT145, 35O22, N6 and PGT128 in serum of monkeys r14097 and r14121 were quantified by ELISA. See also Fig. S3 and Fig. S4.