Three agonist antibodies in combination with high-dose IL-2 eradicate orthotopic kidney cancer in mice

Abstract

Background: Combination immunotherapies can be effective against subcutaneous tumors in mice but the effect against orthotopic malignant disease is less well characterized. In particular, a combination of three agonist antibodies, termed Tri-mAb, consisting of anti-DR5, anti-CD40 and anti-CD137 has previously been demonstrated to eradicate a large proportion of subcutaneous renal cell carcinoma (Renca) tumors (75% long-term survival), but the effect against orthotopic disease is not known.

Purpose: To determine the relative response of orthotopic tumors, we inoculated Renca into the kidney followed by treatment with Tri-mAb.

Results: We found that orthotopic tumors responded much less to treatment (approximately 13% survival), but a significant improvement in survival was achieved through the addition of IL-2 to the treatment regimen (55% survival). All three agonist antibodies and high dose IL-2, 100,000 IU for up to six doses, were required. CD8+ T cells were also required for optimal anti-tumor responses. Coadministration of IL-2 led to enhanced T cell activity as demonstrated by an increased frequency of IFN-gamma-producing T cells in tumor-draining lymph nodes, which may have contributed to the observed improvement of therapy against kidney tumors.

Implications: Responses of subcutaneous tumors to immunotherapy do not necessarily reflect how orthotopic tumors respond. The use of combination immunotherapy stimulating multiple facets of immunity and including cytokine support for T cells can induce effective anti-tumor responses against orthotopic and metastatic tumors.

Figure 1

Tri-mAb is less effective against orthotopic kidney cancer than it is against subcutaneous tumors. Mice were injected with Renca cells either subcutaneously (s.c.) (2 × 10⁵) or intrakidney (i.k.) (1 × 10⁵). On day 10 after tumor injection, some groups of mice received Tri-mAb treatment consisting of 100 μg each of anti-DR5, anti-CD40 and anti-CD137 on days 10, 13 and 17. Other groups of mice received control treatment consisting of isotype control antibodies (MAC4, rat IgG2a; UC8-1B9, hamster IgG) or were left non-treated (NT), and survival monitored. Two representative experiments of three are shown A and B. Control refers to non-treated in panel A and MAC4 rat IgG2a and UC8-1B9 isotype controls for panel B. Survival of s.c. tumor-bearing mice greater than mice with kidney tumors, P2 = 0.034 for pooled data from A and B, Log-rank Test. (s.c. = subcutaneous tumor; i.k. = intrakidney tumor).

Figure 2

Coadministration of IL-2 enhances the survival of mice bearing orthotopic kidney cancer. Groups of mice bearing 10-day established kidney tumors were treated with either Tri-mAb alone (days 10, 14 and 18), IL-2 alone (10⁵ IU i.p. daily for 4 to 6 days) or the combination of Tri-mAb + IL-2. Other groups were non-treated (NT). Numbers in parentheses depict the number of mice per group. Pooled results of 4 experiments are shown. (P2 < 0.0001 for Tri-mAb + HD IL-2 vs Tri-mAb alone, Log-rank Test). (NT = non-treated, HD = high dose).

Figure 3

Large doses of IL-2 are necessary for optimal anti-tumor activity. Mice with established orthotopic disease were treated with Tri-mAb alone or in combination with various dose levels of IL-2 delivered in five doses as indicated. HD: high dose (10⁵ IU); MD: medium dose (5 × 10⁴ IU); LD: low dose (10⁴ IU). Numbers in parentheses depict number of mice per group. Experiment performed once, P < 0.018 for high-dose IL-2 vs lower doses (Log-rank test).

Figure 4

All three agonist antibodies are necessary in combination with IL-2 for optimal anti-tumor effects. Following intrakidney injection of Renca cells, mice received the treatments listed as indicated by arrows. Some mice received the complete treatment of Tri-mAb + IL-2, whereas other groups received various pairs of antibodies as listed either alone or in combination with IL-2. Some mice received no treatment to act as controls. Numbers in parentheses depict number of mice per group. Results from a single experiment, P < 0.05 for three antibodies + IL-2 vs two antibodies + IL-2 (Log-rank test), except for the comparison between Tri-mAb + IL-2 and MD5.1/CD40 + IL-2 which tended towards significance at P = 0.079.

Figure 5

Tri-mAb in combination with IL-2 induced anti-tumor immunological memory. Long-term surviving mice were rechallenged with a subcutaneous injection of Renca (2 × 10⁵ cells) and survival monitored. Mice were derived from groups of kidney tumor-bearing mice that received either 3 doses of Tri-mAb alone or in combination with IL-2. Some mice came from groups receiving only one dose of Tri-mAb with or without IL-2. A group of naive mice was also injected subcutaneously with Renca as controls for tumor growth in the absence of immunological memory. Numbers in parentheses depict the number of mice per group. (P2 = 0.012, Log-rank test; Tri-mAb + IL-2 vs Tri-mAb alone).

Figure 6

Coadministration of IL-2 increases the frequency of IFN-γ-producing T cells. Mice were injected subcutaneously with 2 × 10⁵ Renca s.c. into the upper foot to enable isolation of popliteal lymph nodes as distinct tumor-draining nodes. Some groups received Tri-mAb i.p. on days 10 and 13, and/or IL-2 at the dose of 10⁵ IU i.p. on days 10, 11 and 13. Control mice received MAC4 (rat IgG2a) plus UC8-1B9 (hamster IgG) i.p. On day 14, mice were culled and popliteal lymph node cells stimulated on anti-CD3 antibody-coated plates with GolgiStop overnight. Each lymph node analyzed individually. Cells were harvested the next day and stained and assessed by flow cytometry for intracellular IFNγ. Standard error of the mean is shown for 6 mice per group.

Figure 7

CD8⁺ T cells are necessary for optimal anti-tumor effects. Mice were injected with 1 × 10⁵ Renca i.k. and then treated with Tri-mAb on day 11 only, and IL-2 at a dose of 10⁵ IU i.p. on days 11 and 12. Some mice received MAC 4 and UC8-1B9 antibodies as control treatment. In addition, some groups received CD8-depleting mAb as described in Materials and Methods. Results from a single experiment with 8 mice per group, P = 0.035 for Tri-mAb + IL-2 vs CD8-depleted Tri-mAb + IL-2 group (Log-rank test).