EGFRvIII-targeted immunotoxin induces antitumor immunity that is inhibited in the absence of CD4+ and CD8+ T cells

Abstract

Purpose: Immunotoxins as anti-cancer therapeutics have several potential advantages over conventional agents including a high specificity, extraordinary potency, and a lack of an identified mechanism for resistance. It has been clearly demonstrated that Pseudomonas-based immunotoxins have a direct cytotoxic effect. However, delayed and often dramatic antitumor responses seen in human studies with targeted toxins led us to hypothesize that immunologic responses may be a secondary mechanism that enhances the therapeutic efficacy of these novel drugs.

Experimental design: This hypothesis was tested in a murine system using an immunotoxin, MR1-1 [MR1-1(dsFv)-PE38KDEL], that targets a syngeneic murine homologue of the tumor-specific human epidermal growth factor mutation, EGFRvIII, expressed on a murine cell line.

Results: Intratumoral treatment with MR1-1 eliminated EGFRvIII-expressing tumors (P < 0.0001). The antitumor activity of MR1-1 was dependent on the expression of EGFRvIII on some, but not all tumors cells, and was significantly inhibited in the absence of CD4+ (P = 0.0193) and CD8+ (P = 0.0193) T cells. MR1-1 induced EGFRvIII-specific immunity (P < 0.0005) and produced long lasting immunity against tumors expressing EGFRvIII as well as EGFRvIII-negative tumors.

Conclusions: These data suggest that immunotoxins may not be strictly dependent on direct cytotoxicity for their efficacy, but may also be potent inducers of antitumor immunity active even against cells that do not express the targeted antigen.

Fig. 1

a Tumor volume (mm³) of subcutaneous SMA560msEGFRvIII tumors in VM/Dk mice treated with MR1-1. Mice are challenged with 1.5 × 10⁵ SMA560msEGFRvIII cells subcutaneously and treated with intratumoral injection of MR1-1 (1 or 3 μg) or PBS on the day of tumor challenge. The time to detection of tumor growth differed significantly among the three treatment groups (P < 0.0001). Treatment with PBS failed to protect against tumor growth. Treatment with 1 or 3 μg of MR1-1 protected 100% of the mice from tumor growth and all mice treated with MR1-1 survived for >55 days without evidence of tumor (P < 0.0001). b At day 65 after the initial tumor challenge, mice were re-challenged with EGFRvIII-positive (SMA560msEGFRvIII) tumors. Initial treatment with MR1-1 protected mice from subsequent challenge with EGFRvIII-positive (SMA560msEGFRvIII) tumors as well (P < 0.0001)

Fig. 2

Proliferation of splenocytes to PEPvIII stimulation in vitro. Splenocytes were harvested from tumor-bearing mice 7 days after being treated with 3 μg MR1-1 and plated into 96-well plates (5 × 10⁵ cells/100 μl per well), and incubated with various concentrations (20, 10, and 1 μg/ml) of PEPvIII (LEEKKGNYVVTDHC), a peptide that spans the deletion mutation of EGFRvIII, for 72 h. About 1 μCi/25 μl of methyl-[³H]-thymidine was then added to each well. After 18 h, the splenocytes were harvested and radioactivity counted. Splenocytes from mice receiving intratumoral treatment with MR1-1 proliferate significantly in response to PEPvIII compared to spenocytes from untreated mice (*P < 0.0005)

Fig. 3

Tumor volume (mm³) of subcutaneous mixed EGFRvIII-positive (SMA560msEGFRvIII) and EGFRvIII-negative (SMA560) tumors in VM/Dk mice treated with MR1-1. Mice were challenged subcutaneously with a equal dose of cells (1.5 × 10⁵) of each type universally lethal in this mouse model for either tumor alone, and then treated with intratumoral injection of MR1-1 (1 or 3 μg) or PBS on the day of tumor challenge. Treatment with PBS failed to protect against tumor growth. Treatment with 1 or 3 μg of MR1-1 protected the mice from growth of either tumor type, and all mice treated with MR1-1 survived for >55 days (P < 0.0001)

Fig. 4

Subcutaneous mixed EGFRvIII-positive (SMA560msEGFRvIII) and EGFRvIII-negative (SMA560) tumors in VM/Dk mice treated with MR1-1 or anti-Tac(dsFv)-PE38. Mice were challenged with a mixture of a lethal dose of SMA560msEGFRvIII cells (1.5 × 10⁵) in combination with a lethal dose of SMA560 cells (1.5 × 10⁵) subcutaneously and treated with intratumoral injection of MR1-1 (3 μg) or an otherwise identical toxin targeted to the human CD25 molecule, anti-Tac(dsFv)-PE38 (3 μg) on the day of tumor challenge. Time for mice to develop tumors >1,000 mm³ is plotted. Treatment anti-Tac(dsFv)-PE38 failed to protect against tumor growth. Treatment with MR1-1 inhibited tumor growth, and all mice treated with MR1-1 survived for 55 days without evidence of tumor formation (P = 0.0079)

Fig. 5

VM/dK mice were challenged 1.5 × 10⁵ of SMA560msEGFRIII cells subcutaneously and treated with intratumoral injection of MR1-1 on the day of tumor challenge. The antitumor effects of intratumoral MR1-1 were evaluated after depletion of CD4+ or CD8+ T cells. Efficacy of MR1-1 was significantly reduced in the absence of CD4+ (P = 0.0193) or CD8+ (P = 0.0193) T cells. (Statistical analysis represents results of individual experiments, but for clarity, MR1-1 treated mice are pooled from repeated experiments.)