Next generation of tumor-activating type I IFN enhances anti-tumor immune responses to overcome therapy resistance

Abstract

Type I interferon is promising in treating different kinds of tumors, but has been limited by its toxicity, lack of tumor targeting, and very short half-life. To target tumors, reduce systemic toxicity, and increase half-life, here we engineer a masked type I IFN-Fc (ProIFN) with its natural receptor connected by a cleavable linker that can be targeted by tumor-associated proteases. ProIFN has a prolonged serum half-life and shows an improved tumor-targeting effect. Interestingly, ProIFN-treated mice show enhanced DC cross-priming and significant increased CD8⁺ infiltration and effector function in the tumor microenvironment. ProIFN is able to improve checkpoint blockade efficacy in established tumors, as well as radiation efficacy for both primary and metastatic tumors. ProIFN exhibits superior long-term pharmacokinetics with minimal toxicity in monkeys. Therefore, this study demonstrates an effective tumor-activating IFN that can increase targeted immunity against primary tumor or metastasis and reduce periphery toxicity to the host.

Fig. 1. Blockage of IFN-I activity by IFNAR1/2.

a Schematic depicting IFN-I binds to IFNAR1 and IFNAR2. b Design of IFN-I prodrug which IFNAR1/2 extracellular domain fused to IFN-Fc via a cleavable linker. c IFN activity of human IFNAR1/2 masked IFNa2b-Fc was assessed using a human IFN reporter cell line (n = 2). d IFN activity of mouse IFNAR1/2 masked IFNa4-Fc was assessed using mouse IFN reporter cell line (n = 2). e Reducing SDS-PAGE analysis of hProIFNa2b-Fc as well as MMP-14- or MMP-2-cleaved hProIFNa2b-Fc. The experiment was repeated two times independently with similar results. f IFN activity of MMP-activated hProIFNa2b-Fc, hIFNa2b-Fc, and hProIFNa2b-Fc was assessed via human IFN reporter cell line (n = 2). g Reducing SDS-PAGE analysis of mProIFNa4-Fc as well as MMP-14- or MMP-2-cleaved mProIFNa4-Fc. The experiment was repeated two times independently with similar results. h IFN activity of MMP-activated mProIFNa4-Fc, hIFNa2b-Fc, and mProIFNa4-Fc was assessed via mouse IFN reporter cell line (n = 2). Human IFN activity was assessed by 293T-Dual™ hSTING-R232 cells via secreted embryonic alkaline phosphatase activity. Mouse IFN activity was assessed by RAW-Lucia ISG cells via secreted coelenterazine-utilizing luciferase activity. Source data are provided as a Source Data file.

Fig. 2. In vivo assessment of toxicity.

a The pharmacokinetics of mProIFNa4-Fc and mIFNa4-Fc were compared in C57BL/6 mice following one intravenous (i.v.) dose at 1 nmol (n = 4 animals). b, c Healthy C57BL/6 J mice were intraperitoneally (i.p.) treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc, every 3 days for 6 times (n = 4 animals). Body weight (b) and survival curve (c) were shown. d–h Healthy C57BL/6 J mice were intraperitoneally (i.p.) treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc, every 3 days for three times (n = 9 animals). Plasma samples were collected 2 days after the last treatment. The levels of IFN-γ (d), IL-6 (e) and MCP-1 (f) were quantified by cytometric bead array (CBA). The levels of ALT (g), AST (h) were determined by UTSW Metabolic Phenotyping Core. i, j C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 cells and i.p. treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc on day 11 (n = 7 animals). Liver and tumor tissues were harvested 48 h after treatment. Intracellular RNA was extracted for RT-qPCR assay to determine expression levels of MX1 in B16 tumors (i) and livers (j). Data are reported as mean ± s.e.m. Two-tailed t-tests were performed to calculate p-values. Source data are provided as a Source Data file.

Fig. 3. In vivo assessment of efficacy.

a Dose-dependent anti-tumor effect of mProIFNa4-Fc. C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 tumor cells and i.p. treated with 0, 0.01, 0.1, or 1 nmol of fusion protein on day 7, 10, and 13 (n = 10 animals). Tumor volume was shown. b Cleavable linker dependent anti-tumor effect of mProIFNa4-Fc. C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 tumor cells and i.p. treated with 1 nmol of hIg, mProIFNa4-Fc without cleavable substrate (Non-sub), or mProIFNa4-Fc on day 8 and 11 (n = 8 animals). Tumor volume was shown. c, d C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 tumor cells and i.p. treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc, on day 8, 11, and 14 (n = 10 animals). Tumor growth (c) and body weight (d) were shown. e, f C57BL/6 J mice were s.c. inoculated with 1 × 10⁶ LLC tumor cells and i.p. treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc, on day 12, 15, and 18 (n = 9 animals). Tumor growth (e) and body weight (f) were shown. g, h C57BL/6 J mice were s.c. inoculated with 1 × 10⁶ MC38 tumor cells and i.p. treated with 1 nmol of hIg, mIFNa4-Fc, or mProIFNa4-Fc, on day 9, 12, and 15 (n = 9 animals). Tumor growth (g) and body weight (h) were shown. Data are reported as mean ± s.e.m. Two-way ANOVA was performed to calculate p-values. Source data are provided as a Source Data file.

Fig. 4. Dendritic cells and cytotoxic T cells are required for ProIFN-mediated tumor regression.

a C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16-OVA tumor cells and i.p. treated with 1 nmol of the fusion protein on day 8, 11, and 14 (n = 5 animals). 200 μg anti-CD8 was administrated on day 8, 11, and 18. Tumor growth was shown. b C57BL/6 J mice were s.c. inoculated with 1 × 10⁶ MC38 tumor cells and i.p. treated with 1 nmol of the fusion protein on day 9, 12, and 15 (n = 5 animals). 200 μg anti-CD8 was administrated on day 9, 12, and 19. Tumor growth was shown. c IFNAR1^f/f mice were s.c. inoculated with 5 × 10⁵ B16-OVA tumor cells and i.p. treated with 0.2 nmol of hIg or mProIFNa4-Fc on day 8, 11, and 14. Tumor growth was shown (n = 7 animals). d CD11c^creIFNAR1^f/f mice were s.c. inoculated with 5 × 10⁵ B16-OVA tumor cells and i.p. treated with 0.2 nmol of hIg, or mProIFNa4-Fc, on day 8, 11, and 14. Tumor growth was shown (n = 8 animals). e MC38-OVA tumor-bearing C57BL/6 mice (n = 5 animals) were i.p. treated with 1 nmol of hIg or mProIFNa4-Fc on day 14 and 17. Two days after the last treatment, DCs were isolated and co-cultured with isolated OT-I transgenic CD8⁺ T cells. Forty-eight hours later, IFN-γ-producing cells were enumerated by ELISPOT assay. f, g C57BL/6 J mice were s.c. inoculated with 1 × 10⁶ MC38 tumor cells and i.p. treated with 1 nmol of hIg or mProIFNa4-Fc on day 10 and 13. Tumor tissues were harvested 4 days after the last treatment. CD8⁺ IHC (f) and HE (g) staining were performed. Red arrow in the (g) showed the apoptosis area. The maximum unit of the scale bar is 100 µm. The images are representative of four mice samples in each group. The experiment was repeated two times independently with similar results. h To assess the IFN-γ producing ability of T cells, B16 tumor-bearing IFN-γ YFP reporter mice (n = 4 animals) were i.p. treated with 1 nmol of mProIFNa4-Fc on day 11, 14, and 17. Two days after the last treatment, tumor tissues were harvested and YFP ⁺ CD8 ⁺ T cells were determined by flow cytometry. i B16 tumor-bearing WT mice (n = 4 animals) were i.p. treated with 1 nmol of mProIFNa4-Fc on day 11, 14, and 17. Two days after the last treatment, tumor tissues were harvested and granzyme in CD8 ⁺ T cells was determined by flow cytometry. Data are reported as mean ± s.e.m. Two-way ANOVA (a–d) or two-tailed t-tests (e, h, i) were performed to calculate p-values. Source data are provided as a Source Data file.

Fig. 5. Checkpoint blockage benefits from ProIFN-optimized tumor microenvironment.

a–e C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16-OVA tumor cells and i.p. treated with 1 nmol of hIg, or mProIFNa4-Fc, on day 9, 12, and 15 (n = 5 animals for hIg, n = 4 animals for mProIFNa4-Fc). 5 days after the last treatment, mice were euthanized. Tumors and draining lymph nodes (dLNs) were extracted, digested in collagenase/DNase, and resuspended as single cells. Tumor-infiltrating T cells were analyzed via flow cytometry for the frequency of total CD45⁺ cells (a), CD8⁺ T cells (b), Foxp3⁺ T cells (c), CD8⁺/Foxp3⁺ ratio (d), and Ki67⁺CD8⁺ T cells (e). f–i C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 cells and i.p. treated with 1 nmol of hIg or mProIFNa4-Fc on day 11 (n = 7 animals). Tumor tissues were harvested 48 h after treatment. Intracellular RNA was extracted for RT-qPCR assay to determine expression levels of CXCL-9 (f), CXCL-10 (g), CCL-17 (h), and CCL-22 (i). j, k PD-L1 expression in tumors and draining lymph nodes (dLNs) from tumor-bearing mice in a was measured by flow cytometry. Mean fluorescent intensities (MFIs) of PD-L1 staining in dLN (j) and tumor (k) were shown (n = 5 animals). l Combination of mProIFNa4-Fc and a-PD-L1 antibody therapy. C57BL/6 J mice were s.c. inoculated with 1 × 10⁶ MC38 tumor cells and i.p. treated with 1 nmol of mProIFNa4-Fc on day 12, 14, and 16 (n = 8 animals). 200 μg anti-PD-L1 was i.p. administrated on day 12, and 16. Tumor growth was shown. Data are reported as mean ± s.e.m. Two-tailed t-tests (a–k) or two-way ANOVA (l) were performed to calculate p-values. Source data are provided as a Source data file.

Fig. 6. ProIFN improves radiation efficacy.

a, b C57BL/6 J mice were s.c. inoculated with 5 × 10⁵ B16 tumor cells and i.p. treated with 1 nmol (8 mg/kg) of mProIFNa4-Fc on day 10, 13, and 16 (n = 9). Tumors were locally received a single 30-Gy dose on day 9. Tumor growth (a) and survival curve (b) were shown. c Western blot analysis of MMP-2 expression level from the culture supernatant of MDA-MB-231 cells. The experiment was repeated twice independently with similar results. d Flow cytometry analysis of MMP-14 expression level on MDA-MB-231 cells. e–g Human CD34⁺ hematopoietic stem cells transferred NSG-SGM3 mice were s.c. inoculated with 2 × 10⁶ MDA-MB-231 tumor cells and i.p. treated with 10 mg/kg hProIFNa2b-Fc on day 18, 21, 24, 31, 40, and 47 (n = 9 animals for Ctrl, n = 10 animals for hProIFNa2b-Fc, n = 9 animals for IR, n = 8 animals for hProIFNa2b-Fc + IR). Tumors were locally received a single 12-Gy dose on day 17. Mice were euthanized on day 57. Lung metastasis percentage (e), liver metastasis percentage (f), and number of liver metastasis nodules were shown (g). Data are reported as mean ± s.e.m. Two-way ANOVA (a), the logrank test, (b) or two-tailed t-tests (g) were performed to calculate p-values. Source data are provided as a Source Data file.