Anaphylaxis caused by repetitive doses of a GITR agonist monoclonal antibody in mice

Abstract

Immunotherapy for cancer using antibodies to enhance T-cell function has been successful in recent clinical trials. Many molecules that improve activation and effector function of T cells have been investigated as potential new targets for immunomodulatory antibodies, including the tumor necrosis factor receptor superfamily members GITR and OX40. Antibodies engaging GITR or OX40 result in significant tumor protection in preclinical models. In this study, we observed that the GITR agonist antibody DTA-1 causes anaphylaxis in mice upon repeated intraperitoneal dosing. DTA-1-induced anaphylaxis requires GITR, CD4(+) T cells, B cells, and interleukin-4. Transfer of serum antibodies from DTA-1-treated mice, which contain high levels of DTA-1-specific immunoglobulin G1 (IgG1), can induce anaphylaxis in naive mice upon administration of an additional dose of DTA-1, suggesting that anaphylaxis results from anti-DTA-1 antibodies. Depletion of basophils and blockade of platelet-activating factor, the key components of the IgG1 pathway of anaphylaxis, rescues the mice from DTA-1-induced anaphylaxis. These results demonstrate a previously undescribed lethal side effect of repetitive doses of an agonist immunomodulatory antibody as well as insight into the mechanism of toxicity, which may offer a means of preventing adverse effects in future clinical trials using anti-GITR or other agonist antibodies as immunotherapies.

Figure 1

DTA-1 causes anaphylaxis in a GITR-dependent manner. C57BL/6J mice were treated with 1 mg of (A) DTA-1, (B) mDTA-1, or (C) isotype control (rat IgG2b, clone LTF-2) on days 0, 4, and 11. GITR^−/− mice were treated with 1 mg (D) DTA-1 or (E) OVA. Rectal temperatures of individual mice were monitored for 1 hour after each dose, represented by individual lines in the graphs. Data are representative of 3 independent experiments (n = 5 mice per group).

Figure 2

Repeated intraperitoneal doses of the TNFR superfamily agonist antibodies DTA-1 and OX86 cause anaphylaxis. C57BL/6J mice were treated with 1 mg of (A) DTA-1, (B) OX86, (C) anti-4-1BB (clone LOB.12), or (D) anti-CD40 (clone FGK45.5) on days 0, 4, and 11. Rectal temperatures of individual mice were monitored for 1 hour following the final injection, represented by individual lines in the graphs. Data are representative of 3 independent experiments (n = 5 mice per group).

Figure 3

CD4⁺ T cells, B cells, and IL-4 are required for DTA-1–induced anaphylaxis. (A) RAG^−/− mice, (B) TCRβ-deficient mice, (C) MHC class I–deficient mice, (D) MHC class II–deficient mice, and (E) µMT mice were injected with 1 mg DTA-1 on days 0, 4, and 11. Rectal temperatures of mice were monitored for 1 hour following the final injection. (F) C57BL/6J mice were treated with 1 mg DTA-1 on days 0, 4, and 11. Concurrently with the day 0 injection, mice also received 0.5 mg anti-IL-4 (clone 11B11) or isotype control (rat IgG1, clone HRPN). Rectal temperatures were monitored following the day –11 dose of DTA-1. Lines in the graphs represent the mean of each group of mice. The data are representative of 2 independent experiments (n = 5 mice per group). *P < .05; **P < .0001 by unpaired 2-tailed Student t test.

Figure 4

Serum antibodies from DTA-1–treated animals transfer anaphylactic activity to naive mice. (A-E) C57BL/6J mice were treated with 1 mg DTA-1 or isotype control (rat IgG2b, clone LTF-2) on days −10 and −6. On day 0, mice were euthanized, and sera, spleens, and lymph nodes were removed. Either (B) 50 × 10⁶ cells from pooled spleens and lymph nodes or (C-E) 200 µL pooled sera were transferred by intravenous tail vein injection into naive (C,D) C57BL/6J or (E) GITR^−/− mice. Three hours later, a single dose of DTA-1 was administered by intraperitoneal injection, and rectal temperatures were monitored for 1 hour, represented by individual lines in the graphs. Data are representative of 3 independent experiments (n = 5 mice per group). (F-G) Donor mice were treated as in (A). After sera were harvested, antibodies were fractionated from the pooled sera by using a protein A/G column. Either (F) 1 mg of the antibody fraction or (G) 11 mg of the antibody-depleted fraction (termed “flowthrough” fraction) was transferred by intravenous tail vein injection into naive mice. Three hours later, 1 mg DTA-1 was administered by intraperitoneal injection, and rectal temperatures of individual mice were monitored, represented by individual lines in the graphs. Data are representative of 2 independent experiments (n = 3 to 5 mice per group).

Figure 5

Anaphylaxis caused by DTA-1 is mediated by PAF, basophils, and IgG1 antibodies. (A) C57BL/6J or Kit^w/Kit^W-v mice were treated with 1 mg DTA-1 on days 0, 4, and 11. Rectal temperatures of individual mice were monitored for 1 hour following the final injection, represented by individual lines in the graphs. Data are representative of 2 independent experiments (n = 5 mice per group). (B) C57BL/6J mice were treated with 1 mg DTA-1 on days 0, 4, and 11. Thirty minutes prior to the final DTA-1 injection, mice were injected intraperitoneally with either 125 µg of the histamine inhibitor triprolidine or 125 µg of the PAF inhibitor CV6209. Rectal temperatures of individual mice were monitored for 1 hour following the final dose of DTA-1, represented by individual lines in the graphs. Data are representative of 3 independent experiments (n = 5 mice per group). (C) C57BL/6J mice were treated with 1 mg DTA-1 on days 0, 4, and 11. One day before the final dose of DTA-1, mice were either injected intravenously with 25 µg of anti-CD200R3 (clone Ba103) or intraperitoneally with 0.5 mg of anti-Ly6G (clone 1A8). Rectal temperatures of individual mice were monitored for 1 hour following the final dose of DTA-1, represented by individual lines in the graphs. Data are representative of 3 independent experiments (n = 5 mice per group). (D) Sera from DTA-1–treated mice were collected as in Figure 4A. The pooled sera were heat inactivated at 56°C for 3 hours. Heat-inactivated sera was then transferred by intravenous injection into the tail vein of recipient mice. Three hours later, a single dose of DTA-1 was administered, and rectal temperatures were monitored, represented by individual lines in the graphs. Data are representative of 2 independent experiments (n = 5 mice per group). (E) C57BL/6J mice were treated with DTA-1 or isotype control (rat IgG2b, clone LTF-2) on days 0 and 4. On day 11, sera were collected from individual mice and assayed by enzyme-linked immunosorbent assay for IgG1 antibodies recognizing DTA-1 or LTF-2. Data are representative of 3 independent experiments.

Figure 6

Inhibition of PAF protects mice from anaphylaxis while maintaining tumor protection provided by DTA-1 as part of a combination therapy. C57BL/6J mice were challenged intradermally with 1 × 10⁵ B16 melanoma cells. Groups of mice were either left untreated (naive) or were treated with 200 µg TA99 on days 5 and 9 after tumor challenge, 1 mg DTA-1 on days 5, 9, and 16 after tumor challenge, or a combination of both TA99 and DTA-1 at the aforementioned doses and schedules. One group treated with TA99 plus DTA-1 received 125 µg CV6209 by intraperitoneal injection in PBS 30 minutes before the day 16 dose of DTA-1. (A) Rectal temperatures were monitored following the day 16 dose of DTA-1. (B) Tumor growth was measured every 2 to 3 days, represented by individual lines in the graphs. Mice were euthanized when tumor diameter reached 1 cm. Data are representative of 3 independent experiments (n = 15 mice per group). *P < .05, unpaired 2-tailed Student t test.