Specific inhibitor to KRASG12C induces tumor-specific immunity and synergizes with oncolytic virus for enhanced cancer immunotherapy

Abstract

Background

Oncolytic virus (OV)-mediated immunotherapy has been shown limited efficacy. Small molecule inhibitors specific to the KRAS^G12C driver oncoprotein have recently been developed for cancer treatment. The combination of a potent OV with a KRAS^G12C inhibitor could be a potent combination strategy for treating KRAS^G12C cancer.

Methods

We explored combination therapies using KRAS^G12C inhibitor and OV in cancer cells in vitro and in two KRAS^G12C cancer models. We employed flow cytometry to evaluate the immune cell profiles, including dendritic cells, macrophages, myeloid-derived suppressor cells, natural killer (NK), subsets of CD4⁺ and CD8⁺ T cells, and exhaustion markers (CTLA-4, PD-1, TIM-3), activation markers (granzyme B, IFN-γ and 4-1BB) as well as enzyme-linked immunospot assay to identify tumor-antigen specific T cells. The importance of CD4⁺, CD8⁺ T and NK cells in the therapeutic effects was evaluated by antibody-mediated depletion in vivo.

Results

We confirmed that three inhibitors for KRAS^G12C, AMG510 (sotorasib), MRTX849 (adagrasib) and MRTX1257, all displayed potent cytotoxicity to cancer cells harboring KRAS^G12C, but not to cancer cells without this specific KRAS mutation in vitro. All three inhibitors exhibited potent antitumor activity in KRAS^G12C Lewis lung cancer, but not in MC38 colon cancer with wild-type KRAS. In two KRAS^G12C tumor models, either an IL-36γ-armed OV or orally delivered MRTX1257 inhibited tumor growth, but the combination worked much more efficiently, and efficacy was further improved with PD-1 blockade although with no statistical difference in survival, leading to complete tumor remission in a large fraction of the mice. Mechanistic studies revealed that MRTX1257, and other KRAS^G12C inhibitors as well, are potent inducers of antitumor immunity by themselves, and that it worked with OV to elicit potent innate and adaptive tumor-specific immunity. The combination therapeutic efficacy depended largely on increased tumor-specific CD8⁺ cytotoxic T cells, and to a smaller degree, on CD4⁺ T and NK cells.

Conclusions

Small molecule inhibitors of KRAS^G12C are novel inducers of tumor-specific immunity, and a unique triple combination regimen is highly efficacious through elicited potent antitumor immunity for KRAS^G12C cancers.

Correspondence to Dr Zong Sheng Guo; [email protected] ; Dr Kai Li; [email protected]

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Oncolytic virus (OV)-mediated cancer immunotherapy has shown limited efficacy. A specific inhibitor of KRAS^G12C has shown efficacy in lung cancer expressing KRAS^G12C oncoprotein.

WHAT THIS STUDY ADDS 

• The rational combination of a potent Th1-cytokine-armed OV with KRAS^G12C small molecule inhibitor greatly enhances antitumor immunity and leads to improved therapeutic efficacy. The addition of PD-1 blockade further improves the efficacy.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Our rationally designed triple combination strategy of OV, KRAS^G12C small molecule inhibitor and immune checkpoint blockade may provide the best efficacy for cancer driven by this particular type of KRAS mutated oncoprotein. Similar strategies could be developed for other KRAS-G12 mutated or overexpressed cancers.

Introduction

The oncoprotein KRAS has been one of the most challenging targets in cancer. Cancers driven by KRAS mutations are both common and deadly. KRAS^G12C is an oncogenic driver mutation in multiple cancer types, including lung, colorectal, and pancreatic cancers. About 45% of all non-small cell lung carcinoma (NSCLC) KRAS mutations in the USA are KRAS^G12C,¹ which represents 25,000 patients with NSCLC each year.² Researchers struggled to inhibit mutated forms of KRAS-which earned a reputation as “undruggable”. Due to persistent efforts by numerous investigators over the years, a breakthrough was finally achieved a few years ago. A few small molecule inhibitors specific to KRAS^G12C became available for preclinical and clinical studies.^3–5 Some of these inhibitors function by allosterically controlling GTP affinity and effector interactions.⁶ Two phase I clinical trials of the KRAS^G12C inhibitors AMG510 (sotorasib) and MRTX849 (adagrasib) have shown robust efficacy in human patients, including those with NSCLC.^{5 7} Two key features related to the current study have been noticed in the application of KRAS^G12C inhibitors. One is that the KRAS^G12C inhibitor has resulted in the downregulation of IL-36γ in the tumor.⁸ The second one is that AMG510, and possibly others as well, can induce inflammation and thus promote infiltration of immune cells into the tumor tissues.⁵ Based on the definition of immunogenic cell death (ICD), AMG510 induced a type of apoptosis that belongs to ICD.⁹ Intriguingly, the second-generation inhibitors MRTX-849 and MRTX1257 are potent, highly selective and can be delivered orally.¹⁰ Through painstaking preclinical and clinical studies, two small molecule inhibitors, sotorasib and adagrasib, have gained accelerated approval by Food and Drug Administration (FDA) to treat KRAS^G12C-mutant NSCLC after positive results in phase 2 trials.^{11 12}

Cancer immunotherapy is rapidly evolving, and treating cancer by activating the patient’s immune system presents an attractive therapeutic strategy. Oncolytic virus (OV) is a promising regimen of immunotherapy, as showcased by T-VEC for advanced melanoma approved in the USA, Europe and Japan.^{13 14} OV selectively infects and replicates in cancer cells and/or cancer-associated stromal cells in vivo while leaving normal cells unharmed. Oncolytic vaccinia virus (OVV) and other OVs induce oncolysis of cancer cells, usually in a mode of ICD,^15–17 leading to subsequently antitumor immunity that is associated with survival of patients with cancer.¹⁸ OVs have been shown in both preclinical studies and clinical trials to induce adaptive antitumor immunity and contribute to the overall efficacy.¹⁸ A randomized phase 3 PHOCUS trial of sequential treatment with Pexa-Vec and sorafenib in patients with advanced hepatocellular carcinoma did not achieve increased clinical benefit and fared worse compared with sorafenib alone.¹⁹ The most promising combination, an OV, T-VEC, with immune checkpoint blockade (ICB), showed highly improved efficacy in patients with advanced melanoma in early phase clinical studies yet failed in randomized, double-blinded, placebo-controlled phase III trial.²⁰ These results underscore the need to develop more rational and innovative combinations of OV-mediated cancer therapies.^21–23

We have focused our efforts on OVVs in the last two decades.²⁴ We and others have engineered OVVs to express tumor antigens, T-cell costimulatory molecules, and inflammatory chemokines and cytokines and studied their therapeutic efficacy and safety in preclinical tumor models.^25–31 Evidence has been accumulated that OVVs induce ICD, with release of extracellular ATP, HMGB1, and increased cell surface exposure of calreticulin and heat shock proteins, hallmarks of danger signals from ICD.^{16 32–34} This ICD lays the foundation for subsequent elicited antitumor immunity. We previously demonstrated that a new member of the IL-1 cytokines, IL-36γ, can modulate the tumor microenvironment (TME) and elicit potent type I antitumor immunity.^{35 36} Later we developed OVVs armed with the IL-36 gene and showed their efficacy in multiple syngeneic tumor models.³⁷ Indeed we learned that innovative combinations may be necessary to dramatically improve therapeutic efficacy in aggressive tumor models.^{22 23}

In the current study, we designed and studied a rational combination of a specific mutant KRAS-targeted inhibitor and an OV-mediated antitumor immunity for KRAS^G12C cancers. Because the small inhibitor to KRAS^G12C could downregulate IL-36γ in the tumor,⁸ we hypothesized that it is rational to combine a small molecule inhibitor to KRAS^G12C with an IL-36γ-expressing OV for improved antitumor immunity. We showed that this combination worked well in two KRAS^G12C tumor models in syngeneic mice, and the addition of ICB further improved its therapeutic efficacy. We envision that this combinatorial approach to targeting various aspects of cancer for synergistic cytotoxicity and potent antitumor immunity would lead to optimal efficacy for KRAS^G12C mutant cancer with the potential to become a long-term cure in human patients.

Methods

Inhibitor compounds: The three inhibitors (AMG510, MRTX849 and MRTX1257) were obtained from commercial sources. AMG510 was purchased from AbMole BioScience (Houston, Texas, USA) while MRTX849 and MRTX1257 were obtained from Chemietek (Indianapolis, Indiana, USA). All three compounds were dissolved in DMSO solvent as stock solutions and stored at −80°C.

Mice and cell lines: Female C57BL/6J (B6) mice, aged 5–6 weeks old, were obtained from the Jackson Laboratory (Bar Harbor, Maine, USA) and housed in specific pathogen-free conditions in the UPMC Hillman Cancer Center Animal Facility. For the animal studies conducted at China Medical University, it was also approved by the Institutional Committee.

Murine colon cancer cell line MC38, melanoma B16, and Lewis lung cancer (LLC), were originally obtained from American Type Culture Collection (ATCC; Manassas, Virginia, USA). Mouse malignant mesothelioma AE17 cancer cell line was purchased from Millipore Sigma (Burlington, Massachusetts, USA). Mouse colon cancer cell line MC38-luc was described previously.³⁸ All cell lines were tested for mycoplasma once every 3 months or so to ensure they were free of the mycoplasma contamination. The human NSCLC cell lines were originally obtained from ATCC (Manassas, Virginia, USA). H2122 cells contain the homozygous KRAS^G12C mutation, while A549 cancer cells contain KRAS^G12S mutation.

OVVs: OVVs used in this study were derived from the WR strain. The first virus vvTD, previously named vSPT, is an OV with mutations of triple viral genes encoding thymidine kinase (tk) and antiapoptotic genes SPI-1 and SPI-2.^{32 39} The second one, vvTD-IL36γ, is derived from vvTD with insertion of murine IL-36γ complementary DNA at the tk locus and used in our previous study.^{32 35 39} They were amplified in HeLa cells and then purified and tittered as previously described.³⁸

Cell viability assay: The viability of cancer cells infected by OV or treated with MRTX1257 in vitro was measured at 48 hours after infection using CellTiter 96 Aqueous Non-radioactive Cell Proliferation Assay (Promega, Madison, Wisconsin, USA) or Cell Counting Kit-8 (Boster Bio, Pleasanton, California, USA), according to the manufacturers.

Tumor models and therapeutic treatments: For subcutaneous (s.c.) tumor models, B6 mice were subcutaneously inoculated with 5.0e5 LLC or MC38 colon cancer cells. For the s.c. AE17 tumor, 1.0e7 cancer cells were mixed with Matrigel for increased initiation and growth of the tumor. When the tumors reached the sizes 100–125 mm³, vvTD, vvTD-IL36γ, or phosphate-buffered saline (PBS) was intratumorally injected at a dose of 2.0e6 pfu/tumor and mice were treated daily with oral administration of 0, 60, 100, and 200 mg/kg of MRTX1257, or with the other two inhibitors at indicated concentrations. For combination therapy, the viral dose at 2.e6 pfu/tumor and MRTX1257 at 60 mg/kg daily were used unless indicated otherwise. The primary tumor size was measured using an electric caliper in two perpendicular diameters followed by measurement once every 3 days. For long-term survival of mice, the health and survival of treated mice was closely monitored. Mice died naturally due to the disease or were sacrificed when their s.c. tumor size exceeded 20 mm in diameter.

Flow cytometry: Tumor tissues were collected from mice, then minced and incubated in RPMI 1640 medium containing 2% fetal bovine serum (FBS), 1 mg/mL collagenase IV (Sigma; #C5138), 0.1 mg hyaluronidase (Sigma; #H6254), and 200 U DNase I (Sigma; #D5025) at 37°C for 1 hour to make single cells. Single-cell samples were processed for flow cytometry as described.⁴⁰ After staining with 100 µL Zombie Aqua Fixable Viability Kit cell dye (BioLegend, San Diego, California, USA), cells were stained in 100 µL total stain volume (50 µL BV stain buffer, 50 µl 2% FBS) with antibody at a dilution of 1:200 for 30 min on ice in the dark. The sources of antibodies are listed in online supplemental table 1.

Analysis of immune responses by reverse transcription-quantitative PCR (RT-qPCR): On day 6 and day 11 post-viral treatments, tumor tissues or spleens were harvested, and then single-cell suspension was made for cell separation and further analysis of immune cells either by flow cytometry or by RT-qPCR. The primers used are listed in online supplemental table 2.

Enzyme-linked immunospot (ELISpot) assay: Collected tumor tissues were cut into pieces and incubated at 37°C in digestion buffer (Miltenyi Biotec, San Diego, California, USA) before being mashed over a 100 µM tissue strainer. We performed the ELISpot assay as described previously.⁴⁰

Statistics

Statistical analyses were performed using unpaired Student’s t test for two-group comparison. The Student’s t-test was applied to compare tumor sizes between groups at the final measured time point of each experiment. For multiple group comparisons, one-way analysis of variance was used, where p value is adjusted for multiple tests by Dunnett method (GraphPad Prism V.5). Animal survival is presented using Kaplan-Meier survival curves and compared by using log-rank test (GraphPad Prism V.5). Data are presented as mean±SD unless indicated otherwise. We considered p values of<0.05 to be statistically significant, and all p values were two-sided. In the figures, standard symbols are used: *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; and NS: not significant.

Results

Three inhibitors to KRAS^G12C, alone or in combination with OV, inhibit the growth of KRAS^G12C cancer cells

We have selected three small molecule KRAS^G12C inhibitors, with two of them approved by the FDA for cancer treatment, to confirm that they indeed inhibit growth of cancer cells harboring the specific KRAS^G12C mutation. For this experiment, LLC (KRAS^G12C mutant) and MC38 (KRAS wild type) were selected. The cancer cells were plated in 96-well plates overnight and then treated with various concentrations of the inhibitors. At 48-hour post-treatment, CCK8 assays were conducted to measure cell viability. All three inhibitors, AMG510 (sotorasib), MRTX849 (adagrasib) and MRTX1257, inhibited the growth of LLC cells in a dose-dependent manner (figure 1A). Under the same concentration, MRTX1257 showed the highest activity, followed by MRTX849, and AMG510 being the least potent. When tested in non-KRAS^G12C mutant MC38 cells, they showed little activity (figure 1B). These results confirmed and extended the reported results. As MRTX1257 showed most potent activity, we used MRTX1257 for most experiments carried out later but also confirmed key findings with the other two inhibitors, AMG510 and MRTX849.

Figure 1. Cytotoxicity of three KRAS G12C inhibitors and combination of MRTX1257 and OVs in KRAS G12C mutant and non-mutant cancer cells in vitro. (A) Dose-dependent inhibition of LLC cancer cells harboring KRAS G12C cells; (B) Effects on KRAS wild-type MC38 cancer cells. Cancer cells were plated in 96-well plates overnight and then cultured in the presence of drug at the indicated concentrations of the inhibitors for 48 hours and cell viability was assessed using CCK8 assay. (C) The schematic presentation of vvTD and vvTD-IL36[gamma] viral constructs. TKr, the right portion of the tk gene; TKl, the left portion of the tk gene. (D) Cancer cell susceptibility towards MRTX1257. The KRAS G12C mutant cancer cell lines include murine lung cancer LLC, mesothelioma AE17, and human lung cancer H2122, while non-KRAS G12C mutant cancer cell lines include murine colon MC38 and human lung cancer A549. Cancer cells were plated in 96-well plates overnight and then cultured in the presence of drug at concentrations from 0, 0.016 up to 750 nM for a duration of 48 hours and cell viability was assessed using CCK8 assay. (E) Cytotoxicity induced by OVs for the KRAS G12C mutant and non-mutant cancer cells. Cancer cells were infected with vvTD-IL36[gamma] at various MOIs from 0, 0.001 to 100, and cell viability was measured at 48 hours postinfection. Data represented at least triplicates. The value for viable cells at MOI=0 was set at 100%. (F-J) Cell killing activity of vvTD and vvTD-IL36[gamma] combined with MRTX1257. Cancer cells were treated with vvTD or vvTD-IL36[gamma] at an MOI of 0.5 and/or 100 nmol MRTX1257 simultaneously. Cytotoxicity was measured at 24, 36, 48, and 72 hours. Data represented at least three replicates. (K) LLC cells were infected with vvTD or vvTD-IL36[gamma] at MOI of 1 with or without 10 nM MRTX1257 simultaneously, then harvested at varying time points, as virus titers were determined by plaque assay on CV-1 cells. (L) Production and secretion of IL-36[gamma] from infected CV1 cells. Cells were mock-infected or infected with vvTD or vvTD-IL36[gamma] at an MOI of 1.0, with MRTX1257 alone or in combination. Then, conditioned media were harvested at 48 hours postinfection. The amount of the cytokine protein secreted into the medium was quantified using ELISA. LLC, Lewis lung cancer; MOIs, multiplicities of infection; OV, oncolytic virus; PBS, phosphate-buffered saline; tk, thymidine kinase.

We have used two OVVs, one backbone virus, vvTD (without IL-36γ gene) and the other armed with IL-36γ (vvTD-IL-36γ) in this study (Figure 1C). To further evaluate KRAS^G12C inhibitors alone, we first picked MRTX1257, a covalent inhibitor specific to KRAS^G12C oncoprotein, and studied its cytotoxicity in a panel of human and murine cancer cell lines that harbor KRAS^G12C mutant protein or non-KRAS^G12C protein (Figure 1D). Among murine cancer cell lines, LLC (lung cancer) and AE17 (mesothelioma) are KRAS^G12C mutant, while MC38 (colon cancer) is not. Among human lung cancer cell lines, H2122 is KRAS^G12C mutant while A549 is not (in fact, KRAS^G12S mutant). The cancer cells were plated in 96-well plates overnight and then treated with various concentrations of MRTX1257 (figure 1D). The cancer cells were treated with MRTX1257 at doses ranging from 0.0, 0.016 up to 10,000 nM. The viable cells were measured at 48 h post-treatment using CCK8 assay. We observed that LLC and AE17 cells (both KRAS^G12C mutant) are highly sensitive to MRTX1257, while MC38 is much less sensitive. The difference in sensitivity is ∼1,000-fold. Human cancer cells are generally less sensitive to the inhibitor, however, H2122 cells (KRAS^G12C mutant) are indeed much more sensitive to the inhibitor when compared with A549 (non-KRAS^G12S mutant).

We also tested these cancer cells for cytotoxicity mediated by OVs. Cells were plated in 96-well plates overnight and then infected with various multiplicities of infection (MOIs) of vvTD-IL36γ, up to 100 MOIs (figure 1E). We observed that the sensitivity of these cancer cells varied greatly, with a difference of LD50 over 1,000-fold. In this case, MC38 cells were the most sensitive, while A549 cells were the least sensitive.

Then, we examined the cytotoxicity of combinations of OV with MRTX1257 in these cancer cell lines (figure 1F-J). For straightforward comparison, we set the MOI of OV at 0.5 and MRTX1257 at 100 nM and then compared the five cancer cell lines for their susceptibility to these two types of antitumoral agents (figure 1F-K). We made the following observations. First, we confirmed that KRAS^G12C mutant cancer cells (murine LLC, AE17 and human H2122) are more sensitive to MRTX1257 than non-KRAS^G12C mutant cancer cells (murine MC38 and human A549). Second, all cancer cell lines are sensitive to cytotoxicity induced by OVV (both vvTD and vvTD-IL36γ), even though degrees of sensitivity were different. We also examined the effect of MRTX1257 on the viral replication of vvTD and vvTD-IL36γ (figure 1K). It turned out that MRTX1257 had little, if any, inhibitory effect on viral replication and accumulation. To verify, cells infected with vvTD-IL36 alone or in combination with MRTX1257 led to production and secretion of high levels of IL-36γ (figure 1L). Finally, when combined, the effect was cooperative in KRAS^G12C cancer cells, such as LLC and H2122 (figure 1F and H).

In summary, all three KRAS^G12C inhibitors (AMG510, MRTX849, MRTX1257) showed significant antitumor efficacy in KRAS^G12C mutant cancer cells, yet little activity on KRAS^G12C wild-type cancer cells. The combination of OV with MRTX1257 further enhanced the killing of cancer cells. In addition, the inhibitor did not affect the replication of OVs and production of the cytokine IL-36γ from vvTD-IL36γ-infected cancer cells.

KRAS^G12C inhibitors and/or vvTD-IL36γ inhibited tumor growth

We then asked if three inhibitors could inhibit tumor growth in both KRAS^G12C mutant and non-mutant tumors (figure 2 and online supplemental table 3). MRTX1257 was tested first. In KRAS^G12C mutant LLC and AE17 tumor models, MRTX1257 was very efficacious at inhibiting tumor growth and worked in a dose-dependent manner (figure 2A and B). However, the inhibitor had no effect on KRAS wild-type MC38 colon tumor model (figure 2C). The antitumoral effects were also reflected in the extension of survival of mice in a dose-dependent manner in the LLC and AE17 tumor models, but not in the MC38 tumor model (figure 2D-F). In addition, we have observed that this inhibitor displayed similar quantitative inhibitory effects over time on both LLC and AE17 tumors.

Figure 2. Antitumor efficacy of three KRAS G12C inhibitors in three syngeneic subcutaneous tumor models. B6 mice were subcutaneously inoculated with LLC (1.0e6 cells), AE17 (1.0e7 cells), or MC38 (2.0e5 cells) cancer cells, respectively, on day 0 (D0). When tumor sizes reached 100-150 mm 3 in size (around day 10), mice were treated daily with 0, 60, 100, and 200 mg/kg oral administration of MRTX1257. (A, D) LLC tumor; (B, E) AE17 mesothelioma; (C, F) MC38 colon cancer. LLC and AE17 are KRAS G12C tumors while MC38 tumor is non-KRAS G12C . (E, F) For a direct comparison, antitumor efficacy of AMG510, MRTX849 and MRTX1257 was tested in the same doses in two subcutaneous tumor models. B6 mice were subcutaneously inoculated with LLC or MC38 tumor cells on day 0 (D0). When tumor sizes reached 100-150 mm 3 in size (around days 7-10), mice were treated daily with 60 mg/kg oral administration of AMG510, MRTX849 and MRTX1257. (G, H) When sizes reached 100-150 mm³ in size (around days 7-10), mice were treated daily with 60 mg/kg oral administration of AMG510, MRTX849 and MRTX1257. Student’s t-test was applied to compare tumor sizes between groups at the final measured time point. (G, H) When sizes reached 100-150 mm³ in size (around days 7-10), mice were treated daily with 60 mg/kg oral administration of AMG510, MRTX849 and MRTX1257. n=6; *p<0.05, ***p<0.001. LLC, Lewis lung cancer; PBS, phosphate-buffered saline.

We have then directly compared the efficacy of three inhibitors under the same concentration (60 mg/kg body weight) in LCC and MC38 tumors in syngeneic B6 mice (figure 2G and H). Not surprisingly, all three inhibitors inhibited the growth of LLC, with MRTX1257 showing the largest effect and AMG510 the smallest inhibitory effect (p<0.001 for all three inhibitors compared with PBS) (figure 2G). None of the inhibitors had any inhibitory effects on the growth of MC38 tumor, which possesses wild-type KRAS gene. In summary, these results obtained from three syngeneic tumor models are consistent with the specificity and action of all three KRAS^G12C inhibitors.

Then we proceeded to test the potential of combining the inhibitor with a potent IL-36γ-armed OV in KRAS^G12C mutant tumor models (figure 3; online supplemental table 3). Looking for potential cooperation between the two agents, we used the two agents at suboptimal concentrations: MRTX1257 at 60 mg/kg body weight (oral delivery), and OVs at 2.0e6 pfu/mouse (intratumoral delivery). In the LLC tumor model (figure 3A and B), MRTX1257 showed a significant antitumoral effect (figure 3A. p<0.05 compared with PBS-treated group). vvTD, the parental OV, showed very little therapeutic effect at this dose (not significant (ns), compared with PBS), yet IL-36γ-armed OV displayed potent antitumor efficacy (p<0.0001, when compared with PBS). The combination of vvTD with MRTX1257 did not further improve the therapeutic efficacy. In contrast, the combination of MRTX1257 and vvTD-IL36γ led to greater therapeutic efficacy than individual treatment (p<0.05 compared with vvTD-IL36γ; p<0.0001 compared with MRTX1257). Similar patterns of antitumoral effects were also observed in the AE17 tumor, another KRAS^G12C mutant tumor model (figure 3C). Consistently, we found that dual treatment with MRTX1257 and vvTD-IL36γ resulted in the longest overall survival in both LLC and AE17 tumor models (figure 3B and D) (p≤0.05). These data indicated that KRAS^G12C inhibitor and IL36γ-armed OV work cooperatively to exert their antitumoral effect in vivo.

Figure 3. Antitumoral effects of an OV or in combination with MRTX1257 in murine KRAS G12C tumor models. B6 mice were inoculated subcutaneously with 1.0e6 LLC or 1.0e7 AE17 cancer cells on the right flank. When the tumor reached an average of 100-150 mm 3 (mostly on day 10), the mice were randomly divided into six groups (PBS; vvTD; MRTX1257; vvTD+MRTX1257; vvTD-IL36[gamma]; vvTD-IL36[gamma]+MRTX1257) (n=7~8). On day 11, mice were injected intratumorally with 1.0e6 PFU of vvTD, vvTD-IL36[gamma], or PBS, and then on day 13 were treated with the MRTX1257 (60 mg/kg, daily for 3 weeks). The growth curves of tumors and survival curves of mice are shown with LLC cancer model (A, B) and AE17 tumor model (C, D). The growth curves of tumor sizes are presented as mean+-SD. Data represent two independent experiments. Student’s t-test was applied to compare tumor sizes between groups at the final measured time point. n=5; *p<0.05, **p<0.01, ***p<0.001. LLC, Lewis lung cancer; ns, not significant; OV, oncolytic virus; PBS, phosphate-buffered saline.

Modulation of the tumor microenvironment by inhibitors and vvTD-IL36γ in murine tumors harboring KRAS^G12C

We then examined the immune profile dynamics in the tumor tissue on days 6 and 11 after treatments. We analyzed CD4⁺ and CD8⁺ T cell subsets from samples collected on days 6 and 11 post-therapy (figure 4). There was a reduction of CD4⁺ T cells in all treated groups compared with the group treated with PBS on days 6 and 11 (figure 4A,B). However, there were more CD8⁺ T cells in groups treated with either vvTD-IL36γ or MRTX1257 when compared with PBS control group (p<0.01) (figure 4A,B). In dual therapy, there was a further increase of CD8⁺ T cells compared with either one of the monotherapy (p<0.05 compared with monotherapies; p<0.001 when compared with PBS). We continued to analyze the key molecular markers for the status of immunity in the TME using RT-qPCR (online supplemental figure 1 A-F). NKG2D, IFN-γ, and GzmB were all enhanced by either monotherapy, and either maintained at a high level or further increased in dual therapy (online supplemental figure 1C–E). As for IL-36γ messenger RNA expression, there were high levels of expression in vvTD-IL36γ or dual therapy on day 6 (p value: ns), and the levels were reduced by day 11 (online supplemental figure 1F).

Figure 4. Changes in quantities of CD4 + and CD8 + T cell subsets and key functional molecules in the LLC tumor microenvironment after therapy with vvTD-IL36[gamma], MRTX1257, or the dual combination. B6 mice were inoculated s.c. with 1.0e6 LLC cells. On day 11, mice were randomly split into groups and treated with PBS, single dose of vvTD-IL36[gamma] (2.0e6 PFU/mouse) that day, MRTX1257 (60 mg/kg; oral gavage daily for 3 weeks), or combination (vvTD-IL36[gamma]+MRTX1257). Tumor-bearing mice were sacrificed on days 6 and 11 post-treatment, and tumor tissues were collected, and single cells were prepared and analyzed using flow cytometry (A) CD4 + and CD8 + T cells on days 6 and 11 post-therapy; (B) Four to five mice were used for each treatment group, and ANOVA (Tukey’s multiple comparisons) was used for respective comparisons. *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001. ns: not significant. ANOVA, analysis of variance; LLC, Lewis lung cancer; PBS, phosphate-buffered saline; s.c., subcutaneous.

We also performed a direct comparison of the immune profiling on the tumors treated with AMG510, MRTX849, MRTX1257 in relation to the PBS group (online supplemental figure 2). They showed very similar patterns of increases or decreases of certain subsets of immune cells. CD8⁺ T cells, natural killer (NK) cells, dendritic cell (DC), ratio of M1/M2 macrophages, IFN-γ are all increased, while regulatory T (Treg) and myeloid-derived suppressor cell (MDSC) quantities showed a tendency to decrease although statistically insignificant. In summary, all three inhibitors elicited similar pro-antitumor immune profiles in the TME.

Then we analyzed subsets of immune cells in the tumor tissue using flow cytometry at day 6 after treatments with vvTD-IL36γ, MRTX1257 or combination (figure 5; online supplemental figure 3). The data set on day 11 are presented in online supplemental figure 4 and 5. There were NK cells and DCs in the treated groups (figure 5A,B). There were more macrophages (figure 5C), and more importantly, more M1-like macrophages in all treated groups (p<0.05 compared with PBS) (figure 5D). We then analyzed the subsets of CD8⁺ T cells (figure 5E–L). In all treated groups, both naive T cells (Tn) and central memory T cells (Tcm) were reduced (figure 5E,F), accompanied by a concurrent increase of effector memory T cells (Tem) (figure 5G). IFN-γ⁺CD8⁺ and PD-1⁺CD8⁺ T cells were increased, indicating activated cytotoxic T cells (figure 5H,I). However, CTLA-4⁺ CD8⁺ T cells were increased in all treated groups (figure 5J). Tim-3⁺CD8⁺ T cells were not increased in the monotherapy groups but were enhanced in the dual therapy group (figure 5K). Interestingly, PD-1^hi TIM-3⁺ CD8⁺ T cells, as previously shown to be terminally exhausted,⁴¹ were reduced significantly in all treated groups (p<0.01 between PBS and MRTX1257; p<0.001, PBS vs vvTD-IL36γ or dual therapy) (figure 5L). Finally, we examined three key types of immunosuppressive cells (figure 5M–O). M2 macrophages were reduced in all treated groups (figure 5M). MDSCs and Treg cells were reduced in the vvTD-IL36γ and dual-treated groups, but not in the MRTX1257 group (figure 5N,O).

Figure 5. Changes in immune status in the tumor microenvironment after monotherapies or dual therapy with vvTD-IL36[gamma] and MRTX1257. B6 mice were inoculated s.c. with 1.0e6 LLC cells and treated with PBS, vvTD-IL36[gamma] as described (2.0e6 PFU/mouse 11 days post-tumor inoculation), MRTX1257 (60 mg/kg per mouse 13 days postinoculation using a daily schedule by oral gavage for 3 weeks), or combination (vvTD-IL36[gamma]+MRTX1257 scheduled as before). Tumor-bearing mice were sacrificed 6 days post-treatment and primary tumors were collected and analyzed using flow cytometry. (A) NK (NKG2D + CD3 - ); (B) DC (CD11c + CD11b + ); (C) Macrophages (CD11b + F4/80 + ); (D) M1-like Macrophages (CD206 - of CD11c + F4/80 + cells); (E) Naive T cells (CD62L + CD44 - ); (F) Central memory T cells (CD62L + CD44 + ); (G) Effector memory T cells (CD62L - CD44 + ); (H) CD8 + IFN-[gamma] + T cells; (I) PD-1 + CD8 + T cells; (J) TIM-3 + CD8 + T cells; (K) CTLA-4 + CD8 + T cells; (L) Exhausted CD8 + T cells (PD1 hi TIM3 + CD8 + ); (M) M2-like macrophages (CD206 + of CD11b + F4/80 + cells); (N) Myeloid-derived suppressor cells; (O) Regulatory T cells (Foxp3 + among CD4 + cells). Four to five mice were used for each treatment group and ANOVA (Tukey’s multiple comparisons) was used for respective comparisons. *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001. ANOVA, analysis of variance; DC, dendritic cell; LLC, Lewis lung cancer; MDSC, myeloid-derived suppressor cell; NK, natural killer; ns, not significant; PBS, phosphate-buffered saline; Tcm, central memory T cells; Tem, effector memory T cells; Treg, regulatory T cell.

We have also analyzed the immune cell subsets and immune checkpoint markers on the cell surface in splenocytes on days 6 and 11 after treatments and observed similar patterns of changes as in TILs (online supplemental figure 6 and 7).

The efficacy of dual therapy depends largely on CD8⁺ T cells, and anti-PD-1 further enhanced the efficacy

Having demonstrated that both MRTX1257 and OV can elicit adaptive antitumor immunity, we asked if one or more types of immune cells play an essential role right after initiation of the treatments (figure 6). The schedule of the treatments, and antibodies to deplete CD4⁺, CD8⁺, and NK cells in LLC-bearing mice, is presented in figure 6A. We conducted in vivo experiments on an LLC tumor model treated with dual therapy and some groups with additional infusion of antibodies to deplete CD4⁺, CD8⁺, or NK1.1⁺ cells. Tumor growth is plotted in figure 6B, and survival of the mice is plotted in figure 6C. As shown, tumor sizes in mice in the PBS group reached over 2,000 mm³ by day 24. In contrast, the group with dual treatment (vvTD-IL36γ and MRTX1257) remained at baseline on day 42, the end of the plot. When depleted with either CD4⁺ T or NK cells, there were small but significant losses of efficacy (p<0.05 for NK, and p<0.001 for CD4⁺ cells), indicating that NK and CD4⁺ T cells play some roles in therapy. When CD8 was depleted in those mice, we observed a complete loss of efficacy (p<0.0001), indicating that CD8⁺ T cells play absolute key roles in the therapeutic efficacy.

Figure 6. vvTD-IL36[gamma] or/and MRTX 1257 stimulates CD8 + T cell-dependent antitumor immunity, and triple combination led to best therapeutic efficacy. B6 mice bearing subcutaneous LLC tumors, when tumors reached 100-150 mm 3 in size (around D11), were injected intratumorally (i.t.) with PBS or 2.0e6 PFU vvTD-IL36[gamma], and/or MRTX1257 (on day 13 postinoculation) daily by oral gavage for 3 weeks. For cell depletion of CD8 + , CD4 + , and NK cells, anti-CD8 Ab (250 [micro]g per injection), anti-CD4 Ab (150 [micro]g per injection), or PK136 (300 [micro]g per injection) were intraperitoneally injected into mice. (A) Antibody immune cell depletion was performed according to the schedule shown. (B) Tumor size progression. Student’s t-test was applied to compare tumor sizes between groups at the final measured time point. (C) Long-term survival of mice. Log-rank (Mantel-Cox) test was used to compare survival rates in different groups. *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001. (D, E) Dual therapy induced PD-1 expression in tumors and PD-1 blockade dramatically enhanced antitumor efficacy. In separate experiments, LLC tumors were established in B6 mice as described before. Mice then were injected i.t. with PBS or vvTD-IL36[gamma] (at 2.0e6 PFU) on D11, MRTX1257 on D13 for 3 weeks, with or without 200 mg of anti-PD-1 Ab (200 [micro]g per injection), which was given every 2 days for a total of four doses. (D) Tumor sizes and (E) survivals of mice are presented. Log-rank (Mantel-Cox) test was used to compare survival rates. *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001. Ab, antibody; LLC, Lewis lung cancer; NK, natural killer; ns, not significant; PBS, phosphate-buffered saline.

As we have shown earlier, dual therapy enhanced PD-1⁺ CD8⁺ T cells, especially PD-1^hi TIM-3⁺ CD8⁺ T cells in the TME (figure 5I,L). Therefore, it is logical to ask if PD-1 ICB would enhance the efficacy of dual therapy (figure 6D,E). Considering all expected results from numerous control groups, we focus on the final two groups with best therapeutic effects. Indeed, the triple therapy group treated with vvTD-IL36γ+MRTX1257+αPD-1 had better results (p<0.05, when compared with the “vvTD-IL36γ+MRTX1257” dual therapy group). This superior tumor growth inhibition seen in the triple therapy group also translated into the best survival, with 100% of the mice surviving through the duration of the experiment (more than 90 days) (figure 6E). In the dual therapy, only 80% of mice survived, even though there was no statistical significance between the two groups.

The dual therapy elicited potent tumor-specific adaptive immunity

We examined the adaptive antitumor immunity by conducting IFN-γ ELISpot assays and analyzing a key activation marker molecule, 4-1BB, on T cells isolated from LLC tumor-bearing mice. First, we conducted the ELISpot assay using tumor-infiltrating lymphocytes (TILs) from LLC tumor tissues, and T cells from splenocytes isolated by CD90.2 affinity columns were incubated with irradiated LLC cells or control irradiated MC38 colon cancer cells (figure 7A–D). For TILs, there was a significant number of spots (∼180) in TILs from the MRTX1257 group (p<0.01 compared with PBS). vvTD-IL36 elicited more spots (∼480). In the dual therapy group, the number of spots increased further, to ∼700 (p<0.001, compared with either monotherapy group). There were very few spots when incubated with MC38 cancer cells, indicating tumor specificity.

Figure 7. vvTD-IL36[gamma] and MRTX1257 induced LLC tumor-specific CD8+T cells in LLC tumor model. (A, B) Representative image and quantified data of IFN-[gamma] ELISpot assay of 2.0e4 CD90.2+T cells from tumor tissues (TILs) on day 7 post oncolytic virotherapy or drug treatment and co-cultured 1:1 with specific (LLC) and unspecific (MC38) target cells. (C, D) Representative image and quantified data of IFN-[gamma] ELISpots of 1.0e5 CD90.2+T cells isolated from the spleen on day 7 post oncolytic virotherapy or drug treatment, and co-cultured 1:1 with specific (LLC) and unspecific (MC38 tumor cells, medium, or splenocytes) target cells. (E-H) Flow cytometric analysis of percentages of 4-1BB + CD8 + and 4-1BB + CD4 + T cells in the tumor tissues at day 11 post oncolytic virus, MRTX1257 or dual therapy. (E, F) 4-1BB + CD8 + T cells. (G, H) 4-1BB + CD4 + T cells. CD137 is another name for 4-1BB. *p<0.05; **p<0.01; ***p<0.001; and ****p<0.0001. ELISpot, enzyme-linked immunospot; LLC, Lewis lung cancer; ns, not significant; PBS, phosphate-buffered saline; TILs, tumor-infiltrating lymphocytes.

We also conducted ELISpot assays with T cells isolated from splenocytes. The number of spots was at least 10-fold lower, even though the patterns looked similar (figure 7C,D). These results indicated that tumor-specific T cells were enriched in the tumor tissues.

4-1BB expressed on the surface of T cells is an activation marker for tumor-reactive T cells.⁴² Therefore, we used flow cytometry to evaluate the quantities of 4-1BB⁺ CD4⁺ and 4-1BB⁺ CD8⁺ T cells (figure 7E–H). We observed a statistically significant increase of 4-1BB⁺ CD8⁺ T cells in both monotherapy groups and a trend to increase further in the dual therapy group (p<0.01 compared with PBS). 4-1BB⁺ CD4⁺ T cells also increased in all three treated groups (p<0.05 compared with PBS). Together, these results strongly indicate that, just like the potent OV, MRTX1257 can also elicit potent tumor-specific antitumor immunity. The combination works synergistically to elicit more optimal tumor-specific immunity for improved therapeutic efficacy.

Discussion

Cancer cells with mutant KRAS can convert a TME into an immunosuppressive state by actively recruiting immunosuppressive immune cells like MDSCs, Tregs, while simultaneously downregulating the activity of cytotoxic T cells, effectively shielding the tumor from immune attack.⁴³ Therefore, an effective cancer immunotherapy strategy would need to consider the factors of not only killing the tumor cells, but also improving the activation and proliferation of tumor-specific T cells and trafficking of those T cells to the tumor tissues. Proinflammatory cytokine-armed OVs have been shown to promote not only the trafficking, but also the activation of tumor-specific T cells, thus enhancing adaptive antitumor immunity.^{37 44} However, due to the fact that the highly immunosuppressive TME paralyzes the functions of T cells, often the OV itself is not potent enough to turn the corner, making the TME towards pro-antitumoral.⁴⁵ Therefore, it is strategic to develop combinations of OV with an antitumor agent that can target the driver oncoprotein KRAS^G12C.

In this study, we hypothesized that it is rational to combine a small molecule inhibitor to the driver oncoprotein KRAS^G12C with a potent OV for improved therapeutic efficacy on KRAS^G12C cancers. Two such small molecule inhibitors (AMG510 and MRTX849) have been shown recently to induce inflammation in cancer tissue.⁵ In our study, such an inhibitor by itself can induce potent innate and adaptive antitumor immunity. For innate immune cells, it increases the quantities of NK, DC, and M1 macrophages while decreasing M2 macrophages. As for the subsets of CD8⁺ T cells, it promotes the conversion from Tn, Tcm to Tem and effector T cells (Teff) cells. Even though it increases the expression of CTLA-4, PD-1, and TIM-3 on CD8+T cells, it reduces PD-1^hi TIM-3⁺ CD8⁺ T cells, the type of terminally exhausted cells, in the TME. Further analyses showed that it can increase the quantities of CD8⁺ T cells in tumor tissues, and NKG2D, granzyme B, and IFN-γ activation markers on those cytotoxic cells. Finally, ELISpot assays clearly demonstrated that MRTX1257 can induce tumor-specific T cells in the KRAS^G12C mutant tumor. In summary, our data showed convincingly that MRTX1257 induces tumor-specific T cells.

We previously characterized the properties of vvTD-IL36γ in other tumor models.³⁷ In the current study, we performed further studies with this OV on KRAS^G12C cancer. For the profile of innate immune cells, just like the small molecule inhibitors of KRAS^G12C protein, the OV increases the quantities of NK, DC, and M1 macrophages. More dramatically, the OV decreased M2 macrophages, MDSC, and Treg, all three major type immunoinhibitory cells in the TME. It can increase the quantities of CD8⁺ T cells in tumor tissues, and NKG2D, granzyme B, and IFN-γ activation markers on those cytotoxic cells. For CD8⁺ T cell subsets, the OV promotes the conversion from Tn, Tcm to Tem and Teff. Even though it enhances CTLA-4, PD-1, and TIM-3 on CD8⁺ T cells, it reduces PD-1^hiTIM-3⁺CD8⁺ T cells in the TME. Finally, the induction of tumor-specific T cells has been shown in ELISpot assays with both TILs and splenocytes, and for 4-1BB⁺ CD4⁺ and 4-1BB⁺ CD8⁺ T cells by flow cytometry.

It is rational to combine two or more components with functions that can work synergistically or at least additively to fight cancer. As stated earlier, a specific inhibitor to KRAS^G12C can result in the downregulation of IL-36γ,⁸ thus, it is most rational to combine an inhibitor to KRAS^G12C with an OV expressing this cytokine to elicit potent antitumor immunity.

Indeed, the combination of the OV and MRTX1257 led to more potent proimmune properties against cancer. Yet the expression levels of PD-1, CTLA-4, and TIM-3 were mostly higher than either agent alone, even though PD-1^hi TIM-3⁺ CD8⁺ T cells were reduced to the lowest level. Among these three coinhibitory molecules, PD-1 was the highest. Therefore, it is rational to add anti-PD-1 blockade to the regimen for improved antitumor immunity and hopefully therapeutic efficacy. Indeed, analyses using flow cytometry and quantitative RT-PCR indicated further improved levels of innate immune cells (NK, DC, and M1 macrophages), CD8⁺ T cells, and activation markers (NKG2D, granzyme B, IFN-γ, and 4-1BB), and further reduced levels of M2 macrophages, MDSC, and Treg. These changes led to further enhanced antitumor T cells intratumorally and systemically, as shown by ELISpot assays with TILs and splenocytes. Finally, the improved antitumor potency converts to the best therapeutic efficacy, leading to complete remission of LLC tumors in the majority of mice. This therapeutic efficacy depends heavily on CD8+T cells, and to a smaller degree on CD4⁺ T and NK cells.

In this study, we demonstrated that the triple combination led to complete regression of tumors in the KRAS^G12C mutant LLC tumor. One major issue in cancer monotherapy is the development of resistance to therapy. Cancer may possess innate or acquired resistance to immunotherapy.⁴⁶ In this case, KRAS^G12C-mutant cancer cells treated with KRAS^G12C inhibitors could quickly become either quiescent or resistant.⁸ The triple combined regimen developed in this study may overcome the resistance to each individual therapy and thus greatly enhance the overall therapeutic efficacy.

It is worth pointing out that all cancer cell lines contain multiple genetic mutations that may affect multiple signaling pathways. In addition to the KRAS^G12C mutation, another mutation, NRAS^Q61H mutation is also shared in both Lewis lung and AE17 cancers, the two models used in this study.⁴⁷ This mutation enhances NRAS’s ability to bind GTP, a molecular switch that activates downstream signaling pathways involved in cell growth and survival. However, it is not clear if NRAS^Q61H mutation affects the sensitivity to KRAS^G12C inhibitors if they exist as heterodimers (KRAS–NRAS dimer), and how it affects the RAS downstream signaling pathway and biological effects. It should be interesting to further study these inhibitors alone or in combination with OV in novel murine KRAS^G12C lung cancer cell line models as they are available now.⁴⁸

In summary, we found that a small molecule inhibitor to KRAS^G12C can elicit both innate immunity and tumor-specific T cell-mediated immunity. Moreover, the triple regimen we rationally developed acts mainly through potent antitumor immunity to exert profound antitumor activity. Recently, there have been exciting developments in new small molecule inhibitors targeting KRAS^G12D,⁴⁹ and pan-inhibitors to KRAS.⁵⁰ We reason that this combination strategy may be applicable to a variety of cancers with either other G12 mutations or elevated KRAS activity. Our impressive efficacy and safety using this triple combination in tumor models warrant translation in clinical studies in human patients with KRAS^G12C cancer.

We thank Ms Christine Burr for expert editing on the manuscript.

Data availability statement

Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information. Not applicable.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Animal studies were approved by the University of Pittsburgh Institutional Animal Care and Use Committee (IBC #201700279).

Footnote

ZZ and HC contributed equally.

Contributors Conception: ZZ, HC, ZSG. Acquisition, analysis and interpretation of data: ZZ, HC, CF, LC, CM, ZL, BL, KL, ZSG. Resources: ZL, ZQ, DLB, KL and ZSG. Statistical analysis: ZZ, ZSG. Drafting of the manuscript: ZZ, ZSG. Critical review of the manuscript: ZZ, HC, BL, KL, ZSG. Guarantor: ZZ, KL, ZSG.

Funding The research project was mostly conducted at the University of Pittsburgh, with a portion performed at China Medical University. ZZ, HC, and LC were supported in part by fellowships from their respective hospitals in China. This project has been partially supported by funds from NIH R21CA205727-01, internal funds from the University of Pittsburgh (grant #: N/A), and a grant from the Ministry of Sciences and Technology in China (grant #111002319320238000). This study used University of Pittsburgh shared facilities that are supported in part by the NIH award P30CA047904.

Competing interests ZZ, HC, BL, and ZSG have filed a patent application partly based on this study. The other authors declare no conflicts of interest.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Arbour KC, Jordan E, Kim HR, et al. Effects of Co-occurring Genomic Alterations on Outcomes in Patients with KRAS-Mutant Non-Small Cell Lung Cancer. Clin Cancer Res 2018; 24: 334–40. doi:10.1158/1078-0432.CCR-17-1841

² Biernacka A, Tsongalis PD, Peterson JD, et al. The potential utility of re-mining results of somatic mutation testing: KRAS status in lung adenocarcinoma. Cancer Genet 2016; 209: 195–8. doi:10.1016/j.cancergen.2016.03.001

³ Janes MR, Zhang J, Li L-S, et al. Targeting KRAS Mutant Cancers with a Covalent G12C-Specific Inhibitor. Cell 2018; 172: 578–89. doi:10.1016/j.cell.2018.01.006

⁴ Zeng M, Lu J, Li L, et al. Potent and Selective Covalent Quinazoline Inhibitors of KRAS G12C. Cell Chem Biol 2017; 24: 1005–16. doi:10.1016/j.chembiol.2017.06.017

⁵ Canon J, Rex K, Saiki AY, et al. The clinical KRAS(G12C) inhibitor AMG 510 drives anti-tumour immunity. Nature 2019; 575: 217–23. doi:10.1038/s41586-019-1694-1

⁶ Ostrem JM, Peters U, Sos ML, et al. K-Ras(G12C) inhibitors allosterically control GTP affinity and effector interactions. Nature 2013; 503: 548–51. doi:10.1038/nature12796

⁷ Hallin J, Engstrom LD, Hargis L, et al. The KRAS^G12C Inhibitor MRTX849 Provides Insight toward Therapeutic Susceptibility of KRAS-Mutant Cancers in Mouse Models and Patients. Cancer Discov 2020; 10: 54–71. doi:10.1158/2159-8290.CD-19-1167

⁸ Xue JY, Zhao Y, Aronowitz J, et al. Rapid non-uniform adaptation to conformation-specific KRAS(G12C) inhibition. Nature 2020; 577: 421–5. doi:10.1038/s41586-019-1884-x

⁹ Galluzzi L. Targeting Mutant KRAS for Immunogenic Cell Death Induction. Trends Pharmacol Sci 2020; 41: 1–3. doi:10.1016/j.tips.2019.11.004

¹⁰ Fell JB, Fischer JP, Baer BR, et al. Identification of the Clinical Development Candidate MRTX849, a Covalent KRAS(G12C). Inhibitor for the Treatment of Cancer J Med Chem 2020; 63: 6679–93. doi:10.1021/acs.jmedchem.9b02052

¹¹ Skoulidis F, Li BT, Dy GK, et al. Sotorasib for Lung Cancers with KRAS p. G12C Mutation N Engl J Med 2021; 384: 2371–81.

¹² Janne PA, Riely GJ, Gadgeel SM, et al. Adagrasib in Non-Small-Cell Lung Cancer Harboring a KRAS(G12C) Mutation. N Engl J Med 2022; 387: 120–31.

¹³ Shalhout SZ, Miller DM, Emerick KS, et al. Therapy with oncolytic viruses: progress and challenges. Nat Rev Clin Oncol 2023; 20: 160–77. doi:10.1038/s41571-022-00719-w

¹⁴ Ma R, Li Z, Chiocca EA, et al. The emerging field of oncolytic virus-based cancer immunotherapy. Trends Cancer 2023; 9: 122–39. doi:10.1016/j.trecan.2022.10.003

¹⁵ Guo ZS, Liu Z, Bartlett DL. Oncolytic Immunotherapy: Dying the Right Way is a Key to Eliciting Potent Antitumor Immunity. Front Oncol 2014; 4: 74. doi:10.3389/fonc.2014.00074

¹⁶ Ma J, Ramachandran M, Jin C, et al. Characterization of virus-mediated immunogenic cancer cell death and the consequences for oncolytic virus-based immunotherapy of cancer. Cell Death Dis 2020; 11: 48. doi:10.1038/s41419-020-2236-3

¹⁷ Panzarini E, Inguscio V, Dini L. Immunogenic cell death: can it be exploited in PhotoDynamic Therapy for cancer? Biomed Res Int 2013; 2013: 482160. doi:10.1155/2013/482160

¹⁸ Ling AL, Solomon IH, Landivar AM, et al. Clinical trial links oncolytic immunoactivation to survival in glioblastoma. Nature New Biol 2023; 623: 157–66. doi:10.1038/s41586-023-06623-2

¹⁹ Abou-Alfa GK, Galle PR, Chao Y, et al. PHOCUS: A Phase 3, Randomized, Open-Label Study of Sequential Treatment with Pexa-Vec (JX-594) and Sorafenib in Patients with Advanced Hepatocellular Carcinoma. Liver Cancer 2024; 13: 248–64. doi:10.1159/000533650

²⁰ Chesney JA, Ribas A, Long GV, et al. Randomized, Double-Blind, Placebo-Controlled, Global Phase III Trial of Talimogene Laherparepvec Combined With Pembrolizumab for Advanced Melanoma. J Clin Oncol 2023; 41: 528–40. doi:10.1200/JCO.22.00343

²¹ Bommareddy PK, Shettigar M, Kaufman HL. Integrating oncolytic viruses in combination cancer immunotherapy. Nat Rev Immunol 2018; 18: 498–513. doi:10.1038/s41577-018-0014-6

²² Twumasi-Boateng K, Pettigrew JL, Kwok YYE, et al. Oncolytic viruses as engineering platforms for combination immunotherapy. Nat Rev Cancer 2018; 18: 419–32. doi:10.1038/s41568-018-0009-4

²³ Zhu Z, McGray AJR, Jiang W, et al. Improving cancer immunotherapy by rationally combining oncolytic virus with modulators targeting key signaling pathways. Mol Cancer 2022; 21: 196. doi:10.1186/s12943-022-01664-z

²⁴ Guo ZS, Lu B, Guo Z, et al. Vaccinia virus-mediated cancer immunotherapy: cancer vaccines and oncolytics. J Immunother Cancer 2019; 7: 6. doi:10.1186/s40425-018-0495-7

²⁵ Liu Z, Ravindranathan R, Kalinski P, et al. Rational combination of oncolytic vaccinia virus and PD-L1 blockade works synergistically to enhance therapeutic efficacy. Nat Commun 2017; 8: 14754. doi:10.1038/ncomms14754

²⁶ Kowalsky SJ, Liu Z, Feist M, et al. Superagonist IL-15-Armed Oncolytic Virus Elicits Potent Antitumor Immunity and Therapy That Are Enhanced with PD-1 Blockade. Mol Ther 2018; 26: 2476–86. doi:10.1016/j.ymthe.2018.07.013

²⁷ Ge Y, Wang H, Ren J, et al. Oncolytic vaccinia virus delivering tethered IL-12 enhances antitumor effects with improved safety. J Immunother Cancer 2020; 8: e000710. doi:10.1136/jitc-2020-000710

²⁸ Ahmed J, Chard LS, Yuan M, et al. A new oncolytic Vaccinia virus augments antitumor immune responses to prevent tumor recurrence and metastasis after surgery. J Immunother Cancer 2020; 8: e000415.

²⁹ Sun Y, Zhang Z, Zhang C, et al. An effective therapeutic regime for treatment of glioma using oncolytic vaccinia virus expressing IL-21 in combination with immune checkpoint inhibition. Mol Ther Oncolytics 2022; 26: 105–19. doi:10.1016/j.omto.2022.05.008

³⁰ Azar F, Deforges J, Demeusoit C, et al. TG6050, an oncolytic vaccinia virus encoding interleukin-12 and anti-CTLA-4 antibody, favors tumor regression via profound immune remodeling of the tumor microenvironment. J Immunother Cancer 2024; 12: e009302. doi:10.1136/jitc-2024-009302

³¹ Sun F, Guo ZS, Gregory AD, et al. Dual but not single PD-1 or TIM-3 blockade enhances oncolytic virotherapy in refractory lung cancer. J Immunother Cancer 2020; 8: e000294. doi:10.1136/jitc-2019-000294

³² Guo ZS, Naik A, O’Malley ME, et al. The enhanced tumor selectivity of an oncolytic vaccinia lacking the host range and antiapoptosis genes SPI-1 and SPI-2. Cancer Res 2005; 65: 9991–8. doi:10.1158/0008-5472.CAN-05-1630

³³ John LB, Howland LJ, Flynn JK, et al. Oncolytic virus and anti-4-1BB combination therapy elicits strong antitumor immunity against established cancer. Cancer Res 2012; 72: 1651–60. doi:10.1158/0008-5472.CAN-11-2788

³⁴ Whilding LM, Archibald KM, Kulbe H, et al. Vaccinia virus induces programmed necrosis in ovarian cancer cells. Mol Ther 2013; 21: 2074–86. doi:10.1038/mt.2013.195

³⁵ Wang X, Zhao X, Feng C, et al. IL-36γ Transforms the Tumor Microenvironment and Promotes Type 1 Lymphocyte-Mediated Antitumor Immune Responses. Cancer Cell 2015; 28: 296–306. doi:10.1016/j.ccell.2015.07.014

³⁶ Sun R, Gao DS, Shoush J, et al. The IL-1 family in tumorigenesis and antitumor immunity. Semin Cancer Biol 2022; 86: 280–95. doi:10.1016/j.semcancer.2022.05.002

³⁷ Yang M, Giehl E, Feng C, et al. IL-36γ-armed oncolytic virus exerts superior efficacy through induction of potent adaptive antitumor immunity. Cancer Immunol Immunother 2021; 70: 2467–81. doi:10.1007/s00262-021-02860-4

³⁸ Guo ZS, Parimi V, O’Malley ME, et al. The combination of immunosuppression and carrier cells significantly enhances the efficacy of oncolytic poxvirus in the pre-immunized host. Gene Ther 2010; 17: 1465–75. doi:10.1038/gt.2010.104

³⁹ Yang S, Guo ZS, O’Malley ME, et al. A new recombinant vaccinia with targeted deletion of three viral genes: its safety and efficacy as an oncolytic virus. Gene Ther 2007; 14: 638–47. doi:10.1038/sj.gt.3302914

⁴⁰ Feist M, Zhu Z, Dai E, et al. Oncolytic virus promotes tumor-reactive infiltrating lymphocytes for adoptive cell therapy. Cancer Gene Ther 2021; 28: 98–111. doi:10.1038/s41417-020-0189-4

⁴¹ Miller BC, Sen DR, Al Abosy R, et al. Subsets of exhausted CD8⁺ T cells differentially mediate tumor control and respond to checkpoint blockade. Nat Immunol 2019; 20: 326–36. doi:10.1038/s41590-019-0312-6

⁴² Ye Q, Song D-G, Poussin M, et al. CD137 accurately identifies and enriches for naturally occurring tumor-reactive T cells in tumor. Clin Cancer Res 2014; 20: 44–55. doi:10.1158/1078-0432.CCR-13-0945

⁴³ Yang L, Li A, Lei Q, et al. Tumor-intrinsic signaling pathways: key roles in the regulation of the immunosuppressive tumor microenvironment. J Hematol Oncol 2019; 12: 125. doi:10.1186/s13045-019-0804-8

⁴⁴ Pearl TM, Markert JM, Cassady KA, et al. Oncolytic Virus-Based Cytokine Expression to Improve Immune Activity in Brain and Solid Tumors. Mol Ther Oncolytics 2019; 13: 14–21. doi:10.1016/j.omto.2019.03.001

⁴⁵ Bejarano L, Jordāo MJC, Joyce JA. Therapeutic Targeting of the Tumor Microenvironment. Cancer Discov 2021; 11: 933–59. doi:10.1158/2159-8290.CD-20-1808

⁴⁶ Restifo NP, Smyth MJ, Snyder A. Acquired resistance to immunotherapy and future challenges. Nat Rev Cancer 2016; 16: 121–6. doi:10.1038/nrc.2016.2

⁴⁷ Giannou AD, Marazioti A, Kanellakis NI, et al. NRAS destines tumor cells to the lungs. EMBO Mol Med 2017; 9: 672–86.

⁴⁸ Sisler DJ, Hinz TK, Le AT, et al. Evaluation of KRASG12C inhibitor responses in novel murine KRASG12C lung cancer cell line models. Front Oncol 2023; 13: 1094123. doi:10.3389/fonc.2023.1094123

⁴⁹ Hallin J, Bowcut V, Calinisan A, et al. Anti-tumor efficacy of a potent and selective non-covalent KRAS^G12D inhibitor. Nat Med 2022; 28: 2171–82. doi:10.1038/s41591-022-02007-7

⁵⁰ Kim D, Herdeis L, Rudolph D, et al. Pan-KRAS inhibitor disables oncogenic signalling and tumour growth. Nature 2023; 619: 160–6. doi:10.1038/s41586-023-06123-3