Purpose:

Next-generation sequencing studies and CRISPR-Cas9 screens have established mutations in the IFNγ-JAK-STAT pathway as an immune checkpoint inhibitor (ICI) resistance mechanism in a subset of patients with melanoma. We hypothesized ICI resistance mutations in the IFNγ pathway would simultaneously render melanomas susceptible to oncolytic virus (OV) therapy.

Experimental Design:

Cytotoxicity experiments were performed with a number of OVs on a matched melanoma cell line pair generated from a baseline biopsy and a progressing lesion with complete JAK2 loss from a patient that relapsed on anti-PD-1 therapy, in melanoma lines following JAK1/2 RNA interference (RNAi) and pharmacologic inhibition and in Jak2 knockout (KO) B16-F10 mouse melanomas. Furthermore, we estimated the frequency of genetic alterations in the IFNγ-JAK-STAT pathway in human melanomas.

Results:

The melanoma line from an anti-PD-1 progressing lesion was 7- and 22-fold more sensitive to the modified OVs, herpes simplex virus 1 (HSV1-dICP0) and vesicular stomatitis virus (VSV-Δ51), respectively, compared with the line from the baseline biopsy. RNAi, JAK1/2 inhibitor studies, and in vivo studies of Jak2 KOs B16-F10 melanomas revealed a significant increase in VSV-Δ51 sensitivity with JAK/STAT pathway inhibition. Our analysis of The Cancer Genome Atlas data estimated that approximately 11% of ICI-naïve cutaneous melanomas have alterations in IFNγ pathway genes that may confer OV susceptibility.

Conclusions:

We provide mechanistic support for the use of OVs as a precision medicine strategy for both salvage therapy in ICI-resistant and first-line treatment in melanomas with IFNγ-JAK-STAT pathway mutations. Our study also supports JAK inhibitor–OV combination therapy for treatment-naïve melanomas without IFN signaling defects.

See related commentary by Kaufman, p. 3278

Translational Relevance

Despite the remarkable success of immune checkpoint inhibitors (ICI) in providing durable response for patients with cutaneous metastatic melanoma, a large subset of patients do not benefit from ICI therapy. Mutations in the IFNγ response pathway is one of the best-established mechanisms of ICI resistance. Disablement of the antiviral IFNγ response should render melanomas sensitive to oncolytic viruses (OV), which are natural or genetically modified viruses generated to specifically target tumors. Our study provides functional evidence to support the use of OVs as a precision medicine strategy to treat ICI-resistant and treatment-naïve melanomas with defects in the IFNγ pathway, and also demonstrates the potential clinical utility of JAK inhibitor–OV combination as a melanoma therapy.

Major strides in the treatment of advanced melanoma have been achieved with the advent, approval and clinical implementation of immune checkpoint inhibitors (ICIs). Monoclonal antibodies (mAbs) targeting programmed cell death protein 1 (PD-1; nivolumab and pembrolizumab) and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4; ipilimumab) enhance T-cell antitumor response to produce durable, long-term survival benefits for patients with melanoma (1–3). The last published results from the phase III Checkmate 067 trial reported 5-year overall survival at 52% in the nivolumab-plus-ipilimumab group and 44% in the nivolumab single treatment group, as compared with 26% in the ipilimumab group (4). Although ICIs have been a breakthrough therapy for patients with metastatic melanoma, a large subset of patients still do not respond to ICI therapy or relapse upon continuous treatment (5). Currently, the melanoma and immune oncology research communities are investing significant resources to identify biomarkers predicting immunotherapy responses and resistance mechanisms (6). Mutations in IFNγ signaling pathway genes remains one of the best-established ICI resistance mechanisms reported to date. Whole-exome sequencing (WES) analyses of melanoma samples from patients treated with anti-CTLA-4 or anti-PD-1/PD-L1 have reported higher alteration frequencies of genes that regulate IFNγ signaling in nonresponders (7–9). Specifically, loss-of-function (LoF) mutations in Janus kinase 1 and 2 (JAK1 and JAK2) have been found in patients experiencing both primary and acquired resistance (7, 10, 11). For example, WES of tumor biopsies of 4 patients who demonstrated objective response to pembrolizumab (anti-PD-1) but later progressed, revealed that 2 of the 4 patients were found to have LoF mutation in either JAK1 or JAK2 (11). Furthermore, a number of CRISPR/Cas9 screens support these findings, confirming key components of the IFN response (JAK1/2, STAT1, IFNGR1/2) as central genes in immunotherapy response (12–15). Once IFNγ binds to IFN gamma receptor (IFNGR), it signals through the JAK/STAT cascade to upregulate a set of IFN-stimulated genes (ISG) that lead to effective antitumor response (16). IFNγ is also involved in cellular antiviral functions (17). While type I IFNs (α, β) principally regulate viral defense, IFNγ can exert antiviral properties, as it regulates an overlapping set of genes, and induces type I IFNs expression in an amplification loop (17).

Oncolytic viruses (OV) are a form of immunotherapy that makes use of natural or genetically altered viral strains to preferentially target cancer cells. While interest in this cancer therapy dates back to the early 20th century, OVs have more recently received significant attention due to our improved abilities to modify viral genomes, thus enhancing their oncoselectivity and safety (18). In 2015, talimogene laherparepvec (T-VEC) a modified strain of the herpes simplex virus-1 (HSV-1), became the first OV to be approved by the FDA for the treatment of melanoma (19). Vesicular stomatitis virus (VSV) is another OV that has been investigated extensively due to a number of favorable properties that makes this virus an attractive immunotherapy (20–23). OVs utilize a dual mechanism to target cancers by preferentially replicating in malignant cells resulting in tumor death (oncolysis) and inducing the host's antitumor immune response (19). Compared with normal cells, tumors harbor mutations that confer unlimited proliferation, defects in apoptosis, and immunosurveillance escape. While these mutations result in growth advantages, they also lead to defects in cellular antiviral function (18).

In this study, we use patient-derived cell lines, RNA interference (RNAi), and pharmacologic inhibition to demonstrate that defects in the IFNγ response pathway in melanomas that can occur during ICI resistance result in enhanced oncolysis effects of OVs. Furthermore, we performed copy-number analysis for melanoma cases from The Cancer Genome Atlas (TCGA) dataset to investigate the frequency of genomic alterations in the IFNγ-JAK-STAT pathway genes, estimating that approximately 11% of ICI treatment-naïve melanomas may be susceptible to OV therapy. In summary, our study provides mechanistic support for the use of OVs as a precision medicine strategy for both salvage therapy for ICI-resistant melanomas and first-line treatment in melanomas with defects in IFNγ signaling. Our study also supports the use of JAK inhibitors to further enhance OV's cytotoxic effect in melanomas with intact IFNγ signaling.

Cell culture

HMVII (RRID:CVCL_1282), WM3629 (RRID:CVCL_C275), Meljuso (RRID:CVCL_1403), GAK (RRID:CVCL_1225), WM164 (RRID:CVCL_7928), WM793A (RRID:CVCL_8787), SKMel-19 (RRID:CVCL_6025), WM983A (RRID:CVCL_6808), WM983B (RRID:CVCL_6809), 451Lu (RRID:CVCL_6357), Malme-3M (RRID:CVCL_1438), A375 (RRID:CVCL_0132), UACC62 (RRID:CVCL_1780), WM266-4 (RRID:CVCL_2765), WM9 (RRID:CVCL_6806), WM3918F (RRID:CVCL_C279), MEWO (RRID:CVCL_0445), IGR1 (RRID:CVCL_1303), CHL-1 (RRID:CVCL_1122), Colo857 (RRID:CVCL_2004), Ma-Mel-54b (RRID:CVCL_C288), L-929 (RRID:CVCL_0462), and B16-F10 (RRID:CVCL_0159) were maintained in RPMI1640 medium (Wisent Bioproducts) supplemented with 5% heat-inactivated FBS (Gibco, Life Technologies) and 1% penicillin/streptomycin (Gibco, Life Technologies). Patient-derived M420, M464, HEK293FT (RRID:CVCL_6911), and Vero (RRID:CVCL_0059) cell lines were maintained in DMEM (Wisent Bioproducts) supplemented with 10% FBS and 1% penicillin/streptomycin. All cell lines were kept at 37°C in a humidified 5% CO2 incubator. M420 and M464 cell lines were kindly provided by Antoni Ribas (University of California Los Angeles, Lose Angeles, CA). Vero cell line was kindly provided by Martin Richer (McGill University, Montreal, Canada). Cell lines were authenticated using short-tandem repeat profiling and regularly tested for Mycoplasma using MycoAlert Mycoplasma Detection Kit (Lonza).

Plasmids and siRNA

pDONR223-JAK2 was a gift from William Hahn and David Root (Addgene plasmid #23915; http://n2t.net/addgene:23915; RRID:Addgene_23915). JAK2 open reading frames were subcloned into plenti6.3 expression plasmids (Invitrogen) according to Thermo Fisher Scientific's Gateway cloning protocol.

For lentiviral transduction, HEK293FT was transfected with 2 μg of plenti6.3-JAK2, 1.5 μg of psPAX2, 0.5 μg of pMD2 g using 15 μL each of Lipofectamine 3000 Transfection Reagent and P3000 Enhancer Reagent (Thermo Fisher Scientific) in Opti-MEM reduced serum media (Thermo Fisher Scientific). Viral media were collected at 48 and 72 hours after transfection and filtered through a 0.45 μm filter. Viral media were concentrated using Polyethylene glycol (Sigma-Aldrich) and stored at −80°C until use. Colo857 and Ma-Mel-54b were incubated with viral media and hexadimethrine bromide (Sigma-Aldrich) at 6 μg/mL for 24 hours. Infected cells were selected using Blasticidin S HCl (Invitrogen) for 7 days.

siRNAs targeting JAK1/2 and scramble negative control were purchased from Dharmacon. siRNA transfection was performed according to the manufacturer's protocol.

  • siRNA scramble negative control: UAGCGACUAAACACAUCAA

  • siRNA JAK1_1: CCACAUAGCUGAUCUGAAA

  • siRNA JAK1_2: UGAAAUCACUCACAUUGUA

  • siRNA JAK2_1: GAGCAAAGAUCCAAGACUA

  • siRNA JAK2_2: GCCAGAAACUUGAAACUUA

OVs and propagation

GFP-expressing VSV-Δ51, GFP-expressing HSV1-dICP0, vaccinia virus JX-594, and reovirus type 3 Dearing were described previously (24, 25). VSV-Δ51 contains a deletion at the amino acid 51 of the M protein. HSV1-dICP0 has the infected cell protein 0 gene deleted. JX-594 contains disruption of viral thymidine kinase gene and expression of human GM-CSF and β-galactosidase. Vero cells were infected with VSV-Δ51, HSV1-dICP0, or JX-594 at low multiplicities of infection (MOI) of 0.01 for 1–2 hours. L-929 cells were used for propagation of reovirus. Afterward, viral media were removed and replaced with normal media. Plates were incubated for 1–2 days until cytopathic effects were observed in 80% of the cell population. Cells and supernatants were harvested and underwent three cycles of freeze-thaw. Supernatants were clarified using low-speed centrifugation and filtration through a 0.45 μm filter. Supernatants were concentrated by ultracentrifugation at 23,000 RPM using a 36% sucrose cushion, and the pelleted viruses were resuspended in small amount of media and stored at −80°C. For in vivo studies, VSV-Δ51 was subjected to an additional round of density gradient purification using 10%–40% Optiprep gradient (STEMCELL Technologies).

Standard plaque assay was performed to measure viral titer. Vero cells seeded in 6-well plates were infected with OV at increasing MOIs for 1 hour. Viral media were then replaced, and wells were overlaid with RPMI media containing 0.5% agarose. Plates were kept in a 37°C incubator for 3–5 days until plaques could be seen under a light microscope. Agarose overlay was removed, and cells were fixed with formalin at room temperature for 30 minutes. Cells were dyed with 0.1% crystal violet solution, and number of plaques were counted. Viral titer was calculated using the following equation: Plaque-forming unit (PFU)/mL = Number of plaques/(Dilution factor × Volume of viral media).

IFN treatment

Human IFNs α-2b, β, and γ were purchased from PBL Assay Science. Mouse IFNγ were purchased from Shenandoah Biotechnology. For signaling studies, IFNs were used at either 200 or 500 IU/mL for 48 hours. For cytotoxicity assay, cells were treated with IFNγ at 200 IU/mL for 24 hours.

Immunoblot

Cells were lysed in Pierce RIPA buffer (Thermo Fisher Scientific) supplemented with Halt protease inhibitor cocktail (Thermo Fisher Scientific). Cell lysates were cleared using centrifugation and protein concentrations were measured using DC Protein Assay (Bio-Rad). Lysates were denatured in NuPAGE LDS sample buffer (Invitrogen) and run on 8% Tris-glycine SDS polyacrylamide gel. Gel was transferred to a nitrocellulose membrane using the Transplot Turbo Transfer System (Bio-Rad) before immunoblotting. Antibodies against JAK1 (1:1,000; catalog no. 3344, RRID:AB_2265054), JAK2 (1:1,000; catalog no. 3230, RRID:AB_2128522), phospho-STAT1(Y701; 1:1,000; catalog no. 7649, RRID:AB_10950970), STAT1 (1:1,000; catalog no. 9172, RRID:AB_2198300), PD-L1 (1:3,000; catalog no. 13684, RRID:AB_2687655), STING (1:1,000; catalog no. 13647, RRID:AB_2732796), HSP70 (1:3,000; catalog no. 4872, RRID:AB_2279841), and GAPDH (1:3,000; catalog no. 2118, RRID:AB_561053) were purchased from Cell Signaling Technology. Antibodies against HLA class I heavy chain (1:10,000; catalog no. MUB2037P) were obtained from Nordic-MUBio. Anti-mouse (1:2,000; catalog no. 7076, RRID:AB_330924) and anti-rabbit (1:2,000; catalog no. 7074, RRID:AB_2099233) IgG horseradish peroxidase (HRP)-linked secondary antibodies were purchased from Cell Signaling Technology. Antibodies against mouse PD-L1 (1:1,000; catalog no. AF1019, RRID:AB_354540) and anti-goat IgG HRP-linked secondary antibodies (1:2,000; catalog no. HAF017, RRID:AB_562588) were purchased from R&D Systems.

Dose–response cytotoxicity assay

A total of 5,000 cells were seeded in 96-well white plates (Corning) in quadruplicates. After pretreatment with IFNγ at 200 IU/mL for 24 hours, cells were then infected with OVs at increasing MOI for 3 hours, before viral media were removed and replaced with normal media. Plates were incubated for 72 hours, and cell viability was measured afterward using Cell Titer Glo Luminescent Assay (Promega). Luminescence was read using FLUOstar Omega Luminometer at gains of 2,000. Experiment was performed in quadruplicates. Log-response curves were plotted using GraphPad Prism (RRID:SCR_002798) to calculate the TCID50 (50% tissue culture infective dose). Readings from mock-treated cells were used to establish 100% survival point, and readings of medium alone with reagents was used as 0% point. Grubbs' test was performed to identify and remove outliers. Three independent experiments were performed, and an unpaired two-tailed Student t test was performed on the resulting TCID50.

In vivo studies

All animal studies and protocols were preapproved by the McGill Comparative Medicine and Animal Resources Centre. Seven-weeks-old male C57BL/6J mice (Jackson Laboratory, RRID:IMSR_JAX:000664) were injected subcutaneously in the right flank with 2.5 × 105 of parental B16 or two different clones of B16 Jak2 knockout (KO) cells suspended in a mixture of 1:1 PBS and Matrigel (Corning). Ten days after tumor injection, mice were randomly reassigned into treatment (VSV-Δ51) and control (PBS) groups. Mice were intratumorally injected with either VSVΔ51 (108 PFU) or PBS once every 3 days for three injections total. Tumor volume changes were measured once every 2 days, and tumor volumes were calculated using the formula [4/3 × (π) × (Length/2) × (Width/2)2]. Mice were euthanized once any diameter exceeds 1.5 cm or when ulceration exceeds 3 mm. Kaplan–Meier survival curves were plotted using GraphPad Prism and survival analysis was performed using the log-rank test.

TCGA analysis

Homozygous gene inactivation was determined using copy-number calls from ABSOLUTE (26) and single-nucleotide variant (SNV) calls acquired from our previous study (27). We considered two types of alteration as homozygous inactivation: a copy number of zero or LoF mutation in combination with loss of heterozygosity (LOH). Specifically, if a gene has a segment anywhere along its length where both alleles' copy number is equal to zero or if the gene has a LoF mutation and the number of mutated copies was equal to the total number of copies, a homozygous inactivation call was assigned. LoF mutations were defined as having the following: start_lost, stop_gained, splice_donor_variant, splice_acceptor_variant, frameshift_variant, or stop_lost. The IFNγ pathway genes (Supplementary Table S1) were defined according to Gao and colleagues (7).

JAK2 LoF mutations in tumors of previously treated patient sensitizes tumors to oncolysis by OVs

Clinical trial studies have estimated that in addition to patients exhibiting de novo resistance to ICIs, more than one-third of patients who had an initial response to anti-PD-1 or anti-CTLA-4 relapsed (4, 28). Genetic alterations in the IFNγ pathway, such as JAK2 LoF mutations, have been discovered in melanomas from patients experiencing both primary and acquired ICI resistance (7–11). To determine whether IFNγ pathway dysregulation leads to increased OV sensitivity, we obtained matched cell lines from a baseline biopsy (M420) and a progressing lesion (M464) from a published study of a patient who relapsed on anti PD-1 therapy (11). The M464 cell line was reported to have a JAK2 F547 LoF splice site mutation with LOH. As expected, we confirmed by immunoblot analysis that M420 responds to both type I (α, β) and II (γ) IFNs stimulation (500 IU/mL), as seen by the phosphorylation of STAT1 and induction of a downstream ISG, PD-L1 (Fig. 1A). In contrast, M464 only responded to type I IFNs, due to the absence of JAK2 expression (Fig. 1A). To assess the impact of JAK2 loss on the potency of OVs, we performed a dose–response cytotoxicity assay using VSV-Δ51. Cells were pretreated with 200 IU/mL of IFNγ for 24 hours prior to VSV-Δ51 infection at increasing MOI for 72 hours, after which cell viability was measured. The JAK2 LoF mutant M464 cell line obtained from the anti-PD1 progressing lesion was significantly more sensitive to the cytotoxicity effect of VSV-Δ51 than the JAK2 wild-type M420 obtained from the pretreatment biopsy, as seen by a 22-fold decrease in 50% tissue culture infective dose (TCID50; P = 0.0197; Fig. 1B). We repeated the same experiment with an additional OV, HSV1-dIPC0. The M464 cell line had a 7-fold decrease in TCID50 compared with the M420 line (P = 0.0201; Fig. 1C). Taken together, these results suggest JAK2 loss sensitizes the cells against the oncolysis effects of OVs.

Restoration of JAK2 in null lines reduces sensitivity to OVs

To gain a general understanding of IFN response in melanoma cell lines, we examined a panel of 20 ICI-treatment-naïve lines. Most cell lines investigated had a functional JAK/STAT signaling pathway, as seen by phosphorylation of STAT1 upon type I or type II IFN stimulation and induction of the ISG, PD-L1 (Supplementary Fig. S1; Supplementary Table S2). One cell line, Colo857, did not respond to IFNγ and had loss of JAK2 expression (Fig. 2A). Reintroduction of JAK2 stable expression restored phosphorylation of STAT1 and induced expression of another ISG, HLA class I (Fig. 2A). Importantly, cytotoxic assays demonstrated JAK2 overexpression reduced OV sensitivity, leading to a 135-fold increase in TCID50 following VSV-Δ51 treatment (P = 0.0414; Fig. 2B). We noted a similar trend, but less impressive effect, when we infected Colo857 with HSV1-dIPC0 where JAK2 overexpression resulted in a 2-fold increase in TCID50 (P = 0.0321; Supplementary Fig. S2). We next examined whether there were differences in endogenous expression of STING, a viral DNA sensor, between our JAK2-intact and JAK2-deficient melanoma lines as a potential explanation for the differences in sensitivity to VSV (RNA-based) and HSV1 (DNA-based) OVs. Importantly, we did not observe any differences in STING protein expression in the M420, M464, and Colo857 with and without reconstituted JAK2 (Supplementary Fig. S3A). To confirm that the increased VSV sensitivity in the JAK2-deficient lines in comparison HSV-1 was not simply due the viral genome type, we performed our cytotoxic assays with the DNA OV, vaccinia virus JX-594, and the double-stranded RNA OV, reovirus. We observed with the DNA OV, JX-594, that JAK2 overexpression resulted in a 3.7-fold increase in TCID50 (P = 0.0152), similar to the 2-fold difference observed with HSV-1 (Supplementary Fig. S3B). However, with the RNA OV reovirus, JAK2 overexpression resulted in a marginal increase in TCID50 (1.49-fold difference, P = 0.1828) compared with the 135-fold increase observed with VSV (Fig. 2B; Supplementary Fig. S3C). Therefore, we reason that the increased sensitivity observed with VSV in melanomas with defects in IFNγ signaling is less likely due to the type of viral genetic material, but rather on the high potentiation of VSV upon impaired IFN-induced innate immune responses in cancer that has been previously reported for this virus (20–23).

To validate our VSV findings in an additional melanoma line, we obtained the Ma-Mel-54b line that has been previously reported to be deficient in JAK2 protein expression and possesses a homozygous mutation (29). Reintroduction of JAK2 in Ma-Mel-54b restores IFNγ response, as seen by phosphorylation of STAT1 and induction of PD-L1 (Fig. 2C). Furthermore, expression of JAK2 in Ma-Mel-54b led to a significant 746-fold increase in TCID50 following VSV-Δ51 infection (P = 0.0070; Fig. 2D). Thus, restoration of JAK2 reduced sensitivity of JAK2-null melanoma lines to OVs.

JAK1 or JAK2 knockdown sensitizes melanoma to VSV-Δ51

To genetically validate our findings, we used RNAi to individually target JAK1 or JAK2. We selected one of the melanoma lines with an intact functional IFN response, HMVII, from our screen of 20 lines described above (Supplementary Fig. S1; Supplementary Table S2). Knockdown using two independent siRNAs targeting either JAK1 or JAK2 abrogated STAT1 phosphorylation and PD-L1 induction (Fig. 3A and B). JAK1 knockdown significantly sensitized HMVII to VSV-Δ51 oncolysis, resulting in approximately 239-fold and 84-fold decrease in TCID50 for two independent siRNAs (P = 0.0400, P = 0.0410, respectively; Fig. 3C). JAK2 knockdown in the HMVII line similarly increased sensitivity to VSV-Δ51, resulting in 141-fold and 65-fold decrease in TCID50 for two independent siRNAs; however, although observing impressive fold change differences in TCID50, we did observe more experimental variability between our replicate experiments (P = 0.1030, P = 0.1050, respectively; Fig. 3D). These results demonstrated loss of either JAK1 or JAK2 expression sensitizes melanomas to OV-mediated lysis.

JAK1/2 inhibitor ruxolitinib improves oncolysis effects of VSV-Δ51 against melanomas

We next asked whether pharmacologic treatment of a FDA-approved JAK inhibitor, ruxolitinib, could replicate our siRNA results sensitizing melanomas to OV. Ruxolitinib is a JAK1/2 inhibitor that is used to treat rheumatoid arthritis and myelofibrosis (30). In the HMVII cell line with intact IFNs signaling, treatment with ruxolitinib for 24 hours inhibited STAT1 phosphorylation and PD-L1 induction in the presence of IFNγ at 500 IU/mL (Fig. 4A). Furthermore, ruxolitinib treatment did not affect the viability of HMVII cells, even at extremely high concentrations of 100 μmol/L (Fig. 4B). However, with pretreatment of ruxolitinib, we observed increased sensitivity to VSV-Δ51 by 60-fold (P = 0.0468) compared with DMSO control (Fig. 4C). We observed the ruxolitinib-mediated increased sensitivity to VSV-Δ51 in two additional IFN-responsive melanoma cell lines, CHL-1 and WM3629, whereby addition of ruxolitinib decreased the TCID50 by 15 to 143-fold compared with DMSO controls (Supplementary Fig. S4). Thus, addition of ruxolitinib inhibited the IFN response in melanoma lines with intact JAK/STAT signaling, sensitizing melanomas to VSV-Δ51 oncolysis.

B16-F10 Jak2 KO melanoma tumors are more responsive to VSV-Δ51 therapy in vivo

To examine the effects of JAK2 loss on the OV response in vivo, we utilized the B16-F10 murine melanoma model. We generated two CRISPR/Cas9-mediated Jak2 KO single-cell clones. In both clones, Jak2 KO led to a reduction in STAT1 phosphorylation and PD-L1 induction upon mouse IFNγ treatment (Fig. 5A). Remarkably, while the B16 mouse line was less sensitive to VSV-Δ51 in comparison with human melanoma lines (Figs. 1–3), Jak2 KO sensitized B16 lines against VSV-Δ51 oncolysis in vitro (Fig. 5B). Subsequently, C57BL/6J mice were subcutaneously injected with the parental B16 line, or either Jak2 KO clones. After 10 days, xenografts were intratumorally injected with VSV-Δ51 or PBS every 3 days for three injections total, and tumor growth was measured every 2 days (Fig. 5C; Supplementary Fig. S5). The Jak2 KO clone 1 grew similarly in vivo to the parental B16 control (no significant differences in survival with PBS injections). However, intratumoral injection of VSV-Δ51 more than doubled the median survival of the Jak2 KO clone 1 xenograft mice compared with the parental B16 mice (P = 0.0391). The Jak2 KO clone 2 grew slower in vivo compared with both the B16 parental and Jak2 KO clone 1. However, upon VSV-Δ51 intratumoral injection, the Jak2 KO clone 2 xenograft mice obtained a similar survival benefit to the Jak2 KO clone 1, more than doubling the median survival compared with the parental B16 xenografts (Fig. 5C). These in vivo studies support the increased VSV-Δ51 sensitivity of melanomas with defects in IFNγ signaling.

Alteration frequency of IFNγ pathway genes in ICI treatment-naïve melanomas

CDKN2A located on 9p21.3 is the most frequently silenced tumor suppressor gene in cutaneous melanoma (27, 31, 32). Previous studies have reported that JAK2 found on 9p24.1 and a cluster of 17 type I IFN genes located approximately 500 kb away from CDKN2A are frequently codeleted (29, 33, 34). Therefore, we next sought to determine the frequency of genetic alterations that are predicted to significantly disrupt the IFNγ signaling pathway in ICI treatment-naïve melanomas. To address this, we leveraged TCGA skin cutaneous melanoma (SKCM) dataset to identify inactivating alterations in key signaling components of IFNγ pathway (IFNGR1/2, JAK1/2, STAT1, and downstream ISGs; Supplementary Table S1). Inactivating alterations were defined as homozygous deletions, or LoF mutations with accompanying LOH. Using ABSOLUTE, we obtained genome-wide copy numbers and LOH events for 449 SKCM samples, taking into account tumor purity and ploidy (26). We identified 5 patient samples carrying homozygous LoF alterations in either IFNGR1, JAK1, or JAK2 (Fig. 6). We then looked at alterations in downstream ISGs and identified an additional 43 of 449 patient samples with homozygous inactivating alterations in at least one ISGs. The vast majority of these cases had homozygous segmental deletions overlapping the type I IFN gene cluster located on chromosome 9p21.3. (Fig. 6). Concomitant deletion of CDKN2A was observed for all such cases. In summary, we estimate that 11% of ICI-treatment naïve melanomas possess genetic alterations in IFNγ pathway genes that may predict sensitivity to OVs.

Dysregulation of the IFNγ response pathway is one of the most commonly reported ICI resistance mechanisms (10–15, 29). We postulated that disablement of this IFNγ response would render melanomas more sensitive to OVs. Here, we first observed in matched melanoma cell lines obtained from the same patient, that the anti-PD-1 progressing lesion line possessing homozygous JAK2 loss was significantly more sensitive to VSV and HSV-1 compared with the wild-type baseline biopsy line. Second, by restoring JAK2 expression in JAK2-null lines, we reduced the oncolysis effect of OVs. Third, we formally determined that JAK1 and JAK2 loss sensitized melanomas to VSV in RNAi experiments. Fourth, we showed that Jak2 loss improves the efficacy of VSV in a syngeneic mouse model. Finally, we revealed that a FDA-approved JAK1/2 inhibitor sensitized melanomas to VSV. Taken together, we demonstrate that melanomas with deregulated IFNγ response that have been reported to confer resistance to ICI therapy are significantly more sensitive to OV therapy.

Tumors can evade immunosurveillance through dysregulation of type I and II IFN response pathways (35). When examining a panel of 20 melanoma cell lines, we found one line with IFNγ signaling dysregulation. From analysis of TCGA data, we estimate that 11% of ICI-treatment naïve melanomas have LoF genetic events in the IFNγ pathway. Previous studies have reported a significantly higher frequency of alterations when analyzing a similar set of IFNγ pathway genes in a subset of the SKCM TCGA dataset (n = 287–367 cases; refs. 7, 29). We suspect our analysis of a larger cohort (n = 449) and use of copy-number data adjusted for tumor purity and ploidy, which can influence detection sensitivity (8, 26), likely explains this discrepancy. Our estimations are also consistent with a recent study that examined IFNγ response by RNAseq in melanoma cell lines and biopsies (36). Furthermore, although published longitudinal genetic analysis of matched pre- and post-melanoma biopsies from ICI-treated patients remains limited, one next-generation sequencing study discovered a JAK1 or JAK2 LoF mutation in two of the four anti-PD-1 acquired resistance melanomas that were sequenced (11). Ex vivo studies have also shown that patient-derived melanomas cell lines under selective pressure of an effective T-cell response and continuous IFNγ exposure evolve into IFNγ-resistant clones that are considerably less susceptible to T-cell effector mechanisms (29). Thus, we speculate that IFN signaling defects will be more common in melanomas that develop ICI resistance.

T-VEC, a modified strain of the HSV-1 approved in the United States for the treatment of melanoma, has already been shown to promote intratumor lymphocyte infiltration to turn immunologically “cold” tumors “hot” and improve anti-PD-1 therapy (ClinicalTrials.gov: NCT02263508; ref. 37). Once infected, tumors undergo immunogenic cell death (ICD), characterized by cell surface exposure of damage-associated molecular patterns such as HSPs, ATP, and DNA. Specifically, ICD activates dendritic cells, who then migrate to lymph nodes and activate T cells, shaping the adaptive antitumor immune response (18). Furthermore, the OV Newcastle disease virus (NDV) has been shown to induce T-cell infiltration, but also induction of PD-L1, in human and mouse tumor models whereby only combination of NDV with anti-PD-1 or PD-L1 ICIs resulted in complete tumor rejection of B16 melanomas in the majority of treated mice (38). Here, we provide mechanistic insight supporting OVs as an effective salvage therapy for a subset of melanomas that become ICI resistant by dysregulation of the IFN response. We should note that upon examination of 434 publicly available melanoma whole exomes from patients treated with ICIs (8, 9, 39–41), we identified 8 patients (1.8%) with LoF mutations (stop gain or frameshift indel) in either JAK1 or JAK2, of which 6 had a partial response to ICI. We postulate that OV/ICI combination therapy may lead to more frequent complete response for these patients.

Ongoing clinical trials are addressing the clinical utility of T-VEC in combination with pembrolizumab following progression on prior anti-PD-1 therapy (ClinicalTrials.gov: NCT04068181). However, we have seen more dramatic responses to VSV than HSV-1 in melanomas with JAK1/2 loss (Figs. 1 and 2; Supplementary Fig. S2). As T-VEC is a modified strain of HSV-1, we speculate alternative OVs such as VSV may produce better responses in ICI-resistant melanomas. Furthermore, recent studies have identified other approaches to overcome ICI resistance mediated by IFNγ dysregulated signaling (42–44). These strategies include B0-112, a potent nanoplexed version of poly I:C [an immunostimulant of Toll-like receptor 3 (TLR3) that mimics viral infection] to target JAK1 mutants, and a TLR9 agonist for JAK1/2-mutant cancers (42, 44). RIG-1, an immunoreceptor activated by viral RNA mimetics such as poly I:C, has also been identified as a target to restore HLA-I expression and ICI response in IFN-resistant melanomas (43). We demonstrate here that VSV has enhanced oncolysis in IFNγ dysregulated melanomas, and surmise VSV would also lead to an increase in antigen presentation via RIG-I, which we are currently investigating.

FDA-approved JAK inhibitors, such as ruxolitinib, are used to treat inflammatory and myeloproliferative diseases, and are also being examined in the treatment of solid tumors (30). Signaling through JAK/STAT pathway has been shown to promote survival and growth in a number of malignancies (45). Therefore, clinical studies are underway combining JAK inhibitors with anti-PD1 checkpoint blockade (NCT02646748 and NCT03012230; refs. 46, 47). As might be expected given the link between IFNγ-JAK-STAT pathway mutations and ICI resistance, early results have not been favorable. However, our finding that ruxolitinib treatment potentiates the oncolysis effects of OVs supports this combination to enhance OV efficacy in melanoma. Published studies have already demonstrated the effectiveness of JAK/STAT inhibitors and OV combination for several other malignancies, including head and neck, non–small cell lung, and ovarian cancers (48–50). We postulate that in combination with OVs, JAK inhibitors would not only dampen the antiviral immune response to improve the oncolysis effect of OVs but would also alleviate the challenge of T-cell exhaustion due to chronic inflammation. Importantly, this strategy could expand the melanoma patient population that may benefit from OV therapy beyond patients with LoF genetic events in the IFNγ pathway.

T.-T. Nguyen reports grants from Canadian Cancer Society, Terry Fox Research Institute, Canada Research Chairs Program, and Quebec Cancer Consortium and financial support from the Quebec Ministry of Economy and Innovation through the Fonds d'accélération des collaborations en santé (FACS) during the conduct of the study. L. Ramsay reports grants from Canadian Cancer Society (grant no.: 706905), Terry Fox Research Institute (TFRI grant no.1084), Canada Research Chairs Program, and Ministère de l'Économie et de l'Innovation du Québec through the Fonds d'accélérations des collaborations en santé during the conduct of the study. M. Ahanfeshar-Adams reports grants from Canadian Cancer Society (grant no.: 706905), Terry Fox Research Institute (TFRI grant no.1084), Canada Research Chairs Program, and Quebec Cancer Consortium and financial support from the Ministère de l'Économie et de l'Innovation du Québec through the Fonds d'accélérations des collaborations en santé during the conduct of the study. M. Lajoie reports grants from Canadian Cancer Society (grant no.706905), Terry Fox Research Institute (TFRI grant no.1084), Canada Research Chairs Program, and Quebec Cancer Consortium and financial support from the Ministère de l'Économie et de l'Innovation du Québec through the Fonds d'accélération des collaborations en santé during the conduct of the study; other from Kew, Inc. outside the submitted work. D. Schadendorf reports grants, personal fees, nonfinancial support, and other from BMS and Novartis; grants from Amgen; personal fees, nonfinancial support, and other from MSD, Pierre Fabre, Regeneron, Roche, Merck-Serono, and 4SC; personal fees from Array, Pfizer, Philogen, Sun Pharma, InFlarX, Ultimovacs, Neracare, Helsinn, Immunocore, Replimune, and Sandoz; personal fees and nonfinancial support from Sanofi and Nektar outside the submitted work. I.R. Watson reports grants from Canadian Cancer Society (grant no.: 706905), Terry Fox Research Institute (TFRI - grant no. 1084), Canada Research Chairs Program, Quebec Cancer Consortium and the financial support from the Ministère de l'Économie et de l'Innovation du Québec through the Fonds d'accélérations des collaborations en santé during the conduct of the study; collaboration with KEW, Inc. outside the submitted work. No disclosures were reported by the other author.

T.-T. Nguyen: Formal analysis, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. L. Ramsay: Resources, formal analysis, investigation, visualization, writing–original draft, writing–review and editing. M. Ahanfeshar-Adams: Resources, formal analysis, validation, investigation, visualization, writing–original draft, writing–review and editing. M. Lajoie: Resources, formal analysis, validation, writing–review and editing. D. Schadendorf: Resources, formal analysis, validation, writing–review and editing. T. Alain: Resources, writing–review and editing. I.R. Watson: Conceptualization, resources, supervision, validation, funding acquisition, writing–original draft, writing–review and editing.

We are especially grateful to Antoni Ribas for providing the M420 and M464 cell lines. We thank Martin Richer for providing the Vero cell line used in this study. We would like to thank the Watson lab for their thoughtful discussions and critical comments on the article. This research is funded by the Canadian Cancer Society (grant no.: 706905), the Terry Fox Research Institute (TFRI - grant no. 1084) and from the Canada Research Chairs Program. The authors are also grateful for support from the Quebec Cancer Consortium and the financial support from the Ministère de l'Économie et de l'Innovation du Québec through the Fonds d'accélérations des collaborations en santé. TTN is a recipient of the Canderel Graduate Studentship.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Ishida
Y
,
Agata
Y
,
Shibahara
K
,
Honjo
T
. 
Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death
.
EMBO J
1992
;
11
:
3887
95
.
2.
Leach
DR
,
Krummel
MF
,
Allison
JP
. 
Enhancement of antitumor immunity by CTLA-4 blockade
.
Science
1996
;
271
:
1734
6
.
3.
Luke
JJ
,
Flaherty
KT
,
Ribas
A
,
Long
GV
. 
Targeted agents and immunotherapies: optimizing outcomes in melanoma
.
Nat Rev Clin Oncol
2017
;
14
:
463
82
.
4.
Larkin
J
,
Chiarion-Sileni
V
,
Gonzalez
R
,
Grob
J-J
,
Rutkowski
P
,
Lao
CD
, et al
Five-year survival with combined nivolumab and ipilimumab in advanced melanoma
.
N Engl J Med
2019
;
381
:
1535
46
.
5.
Sharma
P
,
Hu-Lieskovan
S
,
Wargo
JA
,
Ribas
A
. 
Primary, adaptive, and acquired resistance to cancer immunotherapy
.
Cell
2017
;
168
:
707
23
.
6.
Baker
RG
,
Hoos
AX
,
Adam
SJ
,
Wholley
D
,
Doroshow
JH
,
Lowy
DR
, et al
The partnership for accelerating cancer therapies
.
Cancer J
2018
;
24
:
111
4
.
7.
Gao
J
,
Shi
LZ
,
Zhao
H
,
Chen
J
,
Xiong
L
,
He
Q
, et al
Loss of IFN-gamma pathway genes in tumor cells as a mechanism of resistance to anti-CTLA-4 therapy
.
Cell
2016
;
167
:
397
404
.
8.
Miao
D
,
Margolis
CA
,
Vokes
NI
,
Liu
D
,
Taylor-Weiner
A
,
Wankowicz
SM
, et al
Genomic correlates of response to immune checkpoint blockade in microsatellite-stable solid tumors
.
Nat Genet
2018
;
50
:
1271
81
.
9.
Liu
D
,
Schilling
B
,
Liu
D
,
Sucker
A
,
Livingstone
E
,
Jerby-Arnon
L
, et al
Integrative molecular and clinical modeling of clinical outcomes to PD1 blockade in patients with metastatic melanoma
.
Nat Med
2019
;
25
:
1916
27
.
10.
Shin
DS
,
Zaretsky
JM
,
Escuin-Ordinas
H
,
Garcia-Diaz
A
,
Hu-Lieskovan
S
,
Kalbasi
A
, et al
Primary resistance to PD-1 blockade mediated by JAK1/2 mutations
.
Cancer Discov
2017
;
7
:
188
201
.
11.
Zaretsky
JM
,
Garcia-Diaz
A
,
Shin
DS
,
Escuin-Ordinas
H
,
Hugo
W
,
Hu-Lieskovan
S
, et al
Mutations associated with acquired resistance to PD-1 blockade in melanoma
.
N Engl J Med
2016
;
375
:
819
29
.
12.
Manguso
RT
,
Pope
HW
,
Zimmer
MD
,
Brown
FD
,
Yates
KB
,
Miller
BC
, et al
In vivo CRISPR screening identifies Ptpn2 as a cancer immunotherapy target
.
Nature
2017
;
547
:
413
8
.
13.
Pan
D
,
Kobayashi
A
,
Jiang
P
,
Ferrari de Andrade
L
,
Tay
RE
,
Luoma
AM
, et al
A major chromatin regulator determines resistance of tumor cells to T cell-mediated killing
.
Science
2018
;
359
:
770
5
.
14.
Patel
SJ
,
Sanjana
NE
,
Kishton
RJ
,
Eidizadeh
A
,
Vodnala
SK
,
Cam
M
, et al
Identification of essential genes for cancer immunotherapy
.
Nature
2017
;
548
:
537
42
.
15.
Kearney
CJ
,
Vervoort
SJ
,
Hogg
SJ
,
Ramsbottom
KM
,
Freeman
AJ
,
Lalaoui
N
, et al
Tumor immune evasion arises through loss of TNF sensitivity
.
Sci Immunol
2018
;
3
:
eaar3451
.
16.
Platanias
LC
. 
Mechanisms of type-I- and type-II-interferon-mediated signalling
.
Nat Rev Immunol
2005
;
5
:
375
86
.
17.
Stark
GR
,
Kerr
IM
,
Williams
BR
,
Silverman
RH
,
Schreiber
RD
. 
How cells respond to interferons
.
Annu Rev Biochem
1998
;
67
:
227
64
.
18.
Twumasi-Boateng
K
,
Pettigrew
JL
,
Kwok
YYE
,
Bell
JC
,
Nelson
BH
. 
Oncolytic viruses as engineering platforms for combination immunotherapy
.
Nat Rev Cancer
2018
;
18
:
419
32
.
19.
Kohlhapp
FJ
,
Kaufman
HL
. 
Molecular pathways: mechanism of action for talimogene laherparepvec, a new oncolytic virus immunotherapy
.
Clin Cancer Res
2016
;
22
:
1048
54
.
20.
Belkowski
LS
,
Sen
GC
. 
Inhibition of vesicular stomatitis viral mRNA synthesis by interferons
.
J Virol
1987
;
61
:
653
60
.
21.
Durbin
JE
,
Hackenmiller
R
,
Simon
MC
,
Levy
DE
. 
Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease
.
Cell
1996
;
84
:
443
50
.
22.
Stojdl
DF
,
Lichty
B
,
Knowles
S
,
Marius
R
,
Atkins
H
,
Sonenberg
N
, et al
Exploiting tumor-specific defects in the interferon pathway with a previously unknown oncolytic virus
.
Nat Med
2000
;
6
:
821
5
.
23.
Alain
T
,
Lun
X
,
Martineau
Y
,
Sean
P
,
Pulendran
B
,
Petroulakis
E
, et al
Vesicular stomatitis virus oncolysis is potentiated by impairing mTORC1-dependent type I IFN production
.
Proc Natl Acad Sci U S A
2010
;
107
:
1576
81
.
24.
Zakaria
C
,
Sean
P
,
Hoang
H-D
,
Leroux
L-P
,
Watson
M
,
Workenhe
ST
, et al
Active-site mTOR inhibitors augment HSV1-dICP0 infection in cancer cells via dysregulated eIF4E/4E-BP axis
.
PLoS Pathog
2018
;
14
:
e1007264
.
25.
Hoang
H-D
,
Graber
TE
,
Jia
J-J
,
Vaidya
N
,
Gilchrist
VH
,
Xiang
X
, et al
Induction of an alternative mRNA 5′ leader enhances translation of the ciliopathy gene Inpp5e and resistance to oncolytic virus infection
.
Cell Rep
2019
;
29
:
4010
23
.
26.
Carter
SL
,
Cibulskis
K
,
Helman
E
,
McKenna
A
,
Shen
H
,
Zack
T
, et al
Absolute quantification of somatic DNA alterations in human cancer
.
Nat Biotechnol
2012
;
30
:
413
21
.
27.
Alkallas
R
,
Lajoie
M
,
Moldoveanu
D
,
Hoang
KV
,
Lefrançois
P
,
Lingrand
M
, et al
Multi-omic analysis reveals significantly mutated genes and DDX3X as a sex-specific tumor suppressor in cutaneous melanoma
.
Nature Cancer
2020
;
1
:
635
52
.
28.
Schachter
J
,
Ribas
A
,
Long
GV
,
Arance
A
,
Grob
J-J
,
Mortier
L
, et al
Pembrolizumab versus ipilimumab for advanced melanoma: final overall survival results of a multicentre, randomised, open-label phase 3 study (KEYNOTE-006)
.
Lancet
2017
;
390
:
1853
62
.
29.
Sucker
A
,
Zhao
F
,
Pieper
N
,
Heeke
C
,
Maltaner
R
,
Stadtler
N
, et al
Acquired IFNγ resistance impairs anti-tumor immunity and gives rise to T-cell-resistant melanoma lesions
.
Nat Commun
2017
;
8
:
15440
.
30.
Qureshy
Z
,
Johnson
DE
,
Grandis
JR
. 
Targeting the JAK/STAT pathway in solid tumors
.
J Cancer Metastasis Treat
2020
;
6
:
27
.
31.
Cancer Genome Atlas Network
. 
Genomic classification of cutaneous melanoma
.
Cell
2015
;
161
:
1681
96
.
32.
Curtin
JA
,
Fridlyand
J
,
Kageshita
T
,
Patel
HN
,
Busam
KJ
,
Kutzner
H
, et al
Distinct sets of genetic alterations in melanoma
.
N Engl J Med
2005
;
353
:
2135
47
.
33.
Horn
S
,
Leonardelli
S
,
Sucker
A
,
Schadendorf
D
,
Griewank
KG
,
Paschen
A
. 
Tumor CDKN2A-associated JAK2 loss and susceptibility to immunotherapy resistance
.
J Natl Cancer Inst
2018
;
110
:
677
81
.
34.
Linsley
PS
,
Speake
C
,
Whalen
E
,
Chaussabel
D
. 
Copy number loss of the interferon gene cluster in melanomas is linked to reduced T cell infiltrate and poor patient prognosis
.
PLoS One
2014
;
9
:
e109760
.
35.
Kaplan
DH
,
Shankaran
V
,
Dighe
AS
,
Stockert
E
,
Aguet
M
,
Old
LJ
, et al
Demonstration of an interferon γ-dependent tumor surveillance system in immunocompetent mice
.
Proc Natl Acad Sci U S A
1998
;
95
:
7556
61
.
36.
Grasso
CS
,
Tsoi
J
,
Onyshchenko
M
,
Abril-Rodriguez
G
,
Ross-Macdonald
P
,
Wind-Rotolo
M
, et al
Conserved interferon-gamma signaling drives clinical response to immune checkpoint blockade therapy in melanoma
.
Cancer Cell
2020
;
38
:
500
15
.
37.
Ribas
A
,
Dummer
R
,
Puzanov
I
,
VanderWalde
A
,
Andtbacka
RHI
,
Michielin
O
, et al
Oncolytic virotherapy promotes intratumoral T cell infiltration and improves anti-PD-1 immunotherapy
.
Cell
2017
;
170
:
1109
19
.
38.
Zamarin
D
,
Ricca
JM
,
Sadekova
S
,
Oseledchyk
A
,
Yu
Y
,
Blumenschein
WM
, et al
PD-L1 in tumor microenvironment mediates resistance to oncolytic immunotherapy
.
J Clin Invest
2018
;
128
:
1413
28
.
39.
Hugo
W
,
Zaretsky
JM
,
Sun
L
,
Song
C
,
Moreno
BH
,
Hu-Lieskovan
S
, et al
Genomic and transcriptomic features of response to Anti-PD-1 therapy in metastatic melanoma
.
Cell
2016
;
165
:
35
44
.
40.
Riaz
N
,
Havel
JJ
,
Makarov
V
,
Desrichard
A
,
Urba
WJ
,
Sims
JS
, et al
Tumor and microenvironment evolution during immunotherapy with nivolumab
.
Cell
2017
;
171
:
934
49
.
41.
Roh
W
,
Chen
PL
,
Reuben
A
,
Spencer
CN
,
Prieto
PA
,
Miller
JP
, et al
Integrated molecular analysis of tumor biopsies on sequential CTLA-4 and PD-1 blockade reveals markers of response and resistance
.
Sci Transl Med
2017
;
9
:
eaah3560
.
42.
Kalbasi
A
,
Tariveranmoshabad
M
,
Hakimi
K
,
Kremer
S
,
Campbell
KM
,
Funes
JM
, et al
Uncoupling interferon signaling and antigen presentation to overcome immunotherapy resistance due to JAK1 loss in melanoma
.
Sci Transl Med
2020
;
12
:
eabb0152
.
43.
Such
L
,
Zhao
F
,
Liu
D
,
Thier
B
,
Le-Trilling
VTK
,
Sucker
A
, et al
Targeting the innate immunoreceptor RIG-I overcomes melanoma-intrinsic resistance to T cell immunotherapy
.
J Clin Invest
2020
;
130
:
4266
81
.
44.
Torrejon
DY
,
Abril-Rodriguez
G
,
Champhekar
AS
,
Tsoi
J
,
Campbell
KM
,
Kalbasi
A
, et al
Overcoming genetically based resistance mechanisms to PD-1 blockade
.
Cancer Discov
2020
;
10
:
1140
57
.
45.
Buchert
M
,
Burns
CJ
,
Ernst
M
. 
Targeting JAK kinase in solid tumors: emerging opportunities and challenges
.
Oncogene
2016
;
35
:
939
51
.
46.
Koblish
HK
,
Hansbury
M
,
Wang
L-CS
,
Yang
G
,
Huang
T
,
Xue
C-B
, et al
Novel immunotherapeutic activity of JAK and PI3Kδ inhibitors in a model of pancreatic cancer [abstract]
. In:
Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18–22; Philadelphia, PA
.
Philadelphia (PA)
:
AACR
;
Cancer Res 2015;75(15 Suppl):Abstract nr 1336
.
47.
Kirkwood
JM
,
Iannotti
N
,
Cho
D
,
O'Day
S
,
Gibney
G
,
Hodi
FS
, et al
Effect of JAK/STAT or PI3Kδ plus PD-1 inhibition on the tumor microenvironment: Biomarker results from a phase Ib study in patients with advanced solid tumors [abstract]
. In:
Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14–18; Chicago, IL
.
Philadelphia (PA)
:
AACR
;
Cancer Res 2018;78(13 Suppl):Abstract nr CT176
.
48.
Escobar-Zarate
D
,
Liu
YP
,
Suksanpaisan
L
,
Russell
SJ
,
Peng
KW
. 
Overcoming cancer cell resistance to VSV oncolysis with JAK1/2 inhibitors
.
Cancer Gene Ther
2013
;
20
:
582
9
.
49.
Patel
MR
,
Dash
A
,
Jacobson
BA
,
Ji
Y
,
Baumann
D
,
Ismail
K
, et al
JAK/STAT inhibition with ruxolitinib enhances oncolytic virotherapy in non-small cell lung cancer models
.
Cancer Gene Ther
2019
;
26
:
411
8
.
50.
Dold
C
,
Rodriguez Urbiola
C
,
Wollmann
G
,
Egerer
L
,
Muik
A
,
Bellmann
L
, et al
Application of interferon modulators to overcome partial resistance of human ovarian cancers to VSV-GP oncolytic viral therapy
.
Mol Ther Oncolytics
2016
;
3
:
16021
.