Antibody-based therapies designed for human use frequently fail to cross-react with the murine isoform of their target. Because of this problem, preclinical studies of antibody-based mesothelin (Msl)-targeted therapeutics in immunocompetent systems have been limited by the lack of suitable mouse models. Here, we describe two immunocompetent humanized mesothelin transgenic mouse lines that can act as tolerant hosts for C57Bl/6-syngeneic cell lines expressing the human isoform of mesothelin. Thyroid peroxidase (TPO) mice have thyroid-restricted human mesothelin expression. Mesothelin (Msl) mice express human mesothelin in the typical serosal membrane distribution and can additionally be utilized to assess on-target, off-tumor toxicity of human mesothelin–targeted therapeutics. Both transgenic strains shed human mesothelin into the serum like human mesothelioma and patients with ovarian cancer, and serum human mesothelin can be used as a blood-based surrogate of tumor burden. Using these models, we examined the on-target toxicity and antitumor activity of human mesothelin–targeted recombinant immunotoxins. We found that immunotoxin treatment causes acute and chronic histologic changes to serosal membranes in Msl mice, while human mesothelin-expressing thyroid follicular cells in TPO mice are resistant to immunotoxin despite excellent drug delivery. Furthermore, poor delivery of immunotoxin to syngeneic orthotopic human mesothelin-expressing pancreatic adenocarcinoma limits antitumor activity both alone and in combination with immune checkpoint inhibition. In summary, we have developed two high-fidelity, immunocompetent murine models for human cancer that allow for rigorous preclinical evaluation of human mesothelin–targeted therapeutics.

A limited number of transformation-specific, viable, cell surface antigens exist that are suitable as drug targets for antibody-based therapies against solid tumors. Mesothelin (MSLN) was discovered as a cancer surface antigen with a pattern of expression in normal tissue limited to the dispensable mesothelial linings that form the pleura, pericardium, and peritoneum (1). MSLN expression has not been detected in the parenchyma of any vital organs under normal resting conditions (2) but may be upregulated in activated portal fibroblasts in response to cholestasis (3). Many solid tumors have strong expression of MSLN including mesothelioma, pancreatic and biliary carcinomas, ovarian, lung, thymic, and gastric cancers (2, 4, 5). Multiple studies have concurred that MSLN is expressed in at least 85% of pancreatic ductal adenocarcinomas (PDAC), and three-quarters of patient tumors express MSLN in at least 25% of cancer cells (6). As a result of its strong differential expression, MSLN has become a popular target for many directed antineoplastic therapies including mAbs, antibody–drug conjugates, radioimmunotherapy, CAR-T cells, vaccines, and recombinant immunotoxins (iTox; refs. 7–13).

Preclinical evaluation of MSLN-targeted therapeutics, such as iTox, presents a challenge owing to the low conservation (∼71%) of the MSLN amino acid sequence between human and murine isoforms (14). Abundant expression of murine mesothelin (mMsln) has been demonstrated in several murine models of ovarian and pancreatic malignancies. However, mMsln does not cross-react with presently existing antibodies designed to target human mesothelin (hMSLN). This limitation has been overcome for other therapeutic antibodies by developing a new “tool” antibody specific for the protein target's murine isoforms. The “tool” therapeutic can then be used as a surrogate for the clinical therapeutic in preclinical testing. However, existing anti-mMsln antibodies have insufficient specificity and affinity to serve as useful surrogates for such experiments. While in vivo studies examining the activity of hMSLN-targeted therapeutics against tumors expressing hMSLN can be performed using human cell lines inoculated into tolerant immunocompromised mice, the evaluation of hMSLN-targeted therapies in combination with immune modulatory agents is not feasible in these immunocompromised systems. Furthermore, on-target/off-tumor toxicity effects of MSLN-targeted therapies and the changes they cause within the native tumor immune environment cannot be evaluated in existing immunocompromised preclinical models. This is a common problem when evaluating mAb-based therapies for which no murine analogue exists.

One solution is to ectopically express the recombinant human cDNA in murine cancer cells subsequently transplanting these cells into syngeneic mice. Unfortunately, this introduces a nonnative (“nonself”) protein into the cancer cells, which could form the basis for an artifactual antitumor immune response or even complete rejection of the allografted tumor cells, as seen previously (15). Overcoming this issue requires development of an immunologically tolerant, transgenic host despite presence of a fully proficient immune environment. Here, we report two new immunocompetent, transgenic murine models suitable for preclinical testing of hMSLN-targeted therapies. We describe the toxicities caused by MSLN-targeted therapy in these models and evaluate the combination of MSLN-targeted therapy with anti–PD-1 and anti-CTLA4 immune checkpoint inhibitors (ICI).

Transgenic mouse engineering

A plasmid construct was designed to express full-length human ORF for hMSLN and megakaryocyte potentiating factor (MPF) in the follicular subset of thyroid cells by using a rat thyroid-specific peroxidase (TPO) promoter to drive transgene expression. An approximately 8.5 kb transgenic fragment was isolated from plasmid backbone, purified, and injected into fertilized C57Bl/6 mouse oocytes by standard transgenic techniques. Founder mice were screened by human MSLN sequence-specific PCR to detect transgene integration. Msl mice were made by genOway, Inc. (France) customized mouse modeling service via contract arrangement and obtained through a Collaborative Research and Development Agreement with F. Hoffman LaRoche. Husbandry, breeding, and genotyping procedures have been established to manage the colonies of TPO171 and Msl mice, according to NIH/NCI guidelines for animal use and care and applicable IACUC regulations.

Cell culture and reagents

All cell culture reagents were purchased from Invitrogen. Cells were grown at 37°C with 5% CO2 in RPMI1640 medium supplemented with l-glutamine (2 mmol/L), penicillin (100 U), streptomycin (100 μg), and FBS (10% final volume, Hyclone, Thermo Fisher Scientific). Identity was confirmed by short tandem repeat analysis. LMB-100 iTox was manufactured by Roche and provided for these studies through a Collaborative Research and Development Agreement. Structural details were previously described by Bauss and colleagues (16). Fluorescent labeling of LMB-100 was performed using an Alexa Fluor 647 Labeling Kit (catalog no. A20173, Molecular Probes, Invitrogen, Thermo Fisher Scientific). The absorbance of the conjugate solution at 280 nm and 650 nm was measured to determine the protein concentration and labeling efficiency. Inactive LMB-100 was synthesized (15) as described previously.

Cell culture reagents for growing and maintaining KPCdelMsln + hMSLN cells used for orthotopic injections were purchased from Thermo Fisher Scientific. Cells were grown in filter-sterilized, DF12/10 (DMEM/F-12, HEPES (11330032) supplemented with 10% FBS (10437028) and 1% penicillin–streptomycin (15140122)) at 37°C with 5%CO2. Cells were passaged at least once after thawing before use in experiments and each passage was performed once cells were at least 80% confluent. Serum-free pancreas ductal adenocarcinoma (PDA-sf) Media, used to isolate KPCdelMsln cells, included the following components: [final concentration]: DMEM/F-12, HEPES [1X](Thermo Fisher Scientific, 11330032), Nicotinamide [1.22 mg/mL](Sigma, N3376), Glucose [5 mg/mL](Sigma, G6152), ITS-A [0.5%](Thermo Fisher Scientific, 51300044), Gentamicin [100 μg/mL](Sigma, G1397), Amphotericin B [2.5 μg/mL](Sigma, A2942), Nu-Serum IV [5%](Corning, 355104), Soybean Trypsin Inhibitor [0.1 mg/mL] (Sigma, T6522), Penicillin–Streptomycin [1%](Thermo Fisher Scientific, 15140122). Filter sterilize then add Bovine Pituitary Extract [25 μg/mL](Corning, 356123), Epidermal Growth Factor [20 ng/mL] (Sigma, E4127], 3,3′,5-Triiodo-l-thyronine (T3) [50 nmol/L]Sigma, T2877), and Dexamethasone [1 μmol/L] (Sigma, D1756).

KPC/delMsln + hMSLN cell line generation

Two-thirds of PDAC tumors that arose in a KRASG12D/+Trp53R172H/+:p-48-Cre:Msln–/– animals were harvested into Dulbecco's Phosphate Buffered Saline (DPBS; 14190250) on ice, transferred to a biological safety cabinet (BSC), washed three additional times with fresh DPBS, and then minced into small approximately 2 × 2 mm pieces. The tumor pieces were dissociated in 10 mL of collagenase II (Worthington, LS004176) in DMEM/F-12, HEPES at 2 mg/mL (final concentration) for approximately 40 minutes at 37°C with stirring. Following dissociation, the sample was returned to the BSC, passed through a 100 μm cell strainer, pelleted by centrifugation, and washed twice with DMEM/F-12, HEPES. The final pellet was resuspended into PDA-sf and plated onto a collagen-coated plate. Following a 24-hour incubation (37°C, 5% CO2), the media was replaced with fresh PDA-sf and the plate was monitored daily for the outgrowth of epithelial tumor cells. To minimize fibroblast contamination, the plate was “stripped” once with 0.25% Trypsin-EDTA (Thermo Fisher Scientific, 25200056) for 2.5 minutes with gentle rocking. Loose cells were removed by aspiration and the reaction was stopped by rinsing twice with medium, before adding fresh PDA-sf. Cells were passaged at approximately 80% confluency and progressively adapted to serum-containing DF12/10 media onto standard tissue culture–treated plates to establish KPCdelMsln cell line. KPCdelMsln cells were modified genetically to express the orthologous human mesothelin protein using lentiviral transduction. The pLenti-PGK-Puro-hMSLN plasmid vector expressing human mesothelin under control of human phosphoglycerate kinase (PGK) 1 promoter was constructed by inserting untagged hMSLN open reading frame (Origene, NM_005823) into pLenti PGK Puro DEST lentiviral vector (Addgene, plasmid #19068) using the Gateway recombination cloning technology (Thermo Fisher Scientific), then purified using a plasmid purification kit utilizing a bacterial endotoxin removal step (Qiagen). Subsequent in vitro packaging of infectious lentiviral stocks was performed at the Viral Production Core Facility (NIH/NINDS). Upon transduction, expressing cells were selected in 0.5 μg/mL puromycin. Single-cell clones were sorted using a BD FACSAria following staining with purified anti-hMSLN MN antibody (BioXcell) followed by PE-labeled Goat anti-Mouse IgG (Jackson ImmunoResearch: 115–116–146).

Immunoblot

Mouse tissues were harvested, then dissociated and lysed with RIPA Buffer (Thermo Fisher Scientific, 89900). Lysates were heat-denatured in SDS-containing loading buffer and separated by SDS-PAGE gel electrophoresis before transfer to polyvinylidene membrane. Membranes were incubated with anti-hMSLN mouse Monoclonal Antibody (clone MN, Rockland) or GAPDH (Cell Signaling Technology, #5174) followed by appropriate HRP-labeled secondary antibody before ECL Detection (Pierce, SuperSign West Pico).

ELISA

Blood was collected from mice using submandibular puncture into serum separator tubes or K2EDTA plasma vacutainer tubes (BD Scientific). Serum was collected after centrifugation at 9,400 × g for 5 minutes. MSLN was detected using DMSLNO Kit (R&D Systems). Signal was detected using read buffer (Meso-Scale Discovery, R92TC-3) and luminescence measured with a QuickPlex SQ 120 (Meso-Scale Discovery).

qPCR

Mouse tissue samples (∼30 mg) were homogenized using a TissueLyser II (Qiagen) and tRNA isolated using a PureLink RNA Mini Kit (Life Technologies). One microliter of extracted tRNA was quantified using a NanoDrop spectrophotometer. Quality of the extracted tRNA was assessed by running 1μL of the extracted RNA on an Agilent Bioanalyzer. A total of 1 μg of extracted tRNA was converted to cDNA using a SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). The resulting cDNAs were diluted 1:100 with RNase free water and 18 μL of each diluted sample was analyzed in triplicate 20 μL reactions using StepOnePlus Real Time PCR Instrument (Applied Biosystems) using TaqMan Gene Expression Master Mix (Applied Biosystems, catalog no. 4444963). The target hMSLN mRNA was detected with an inventoried FAM-MGB TaqMan Gene Expression Assay (Thermo Fisher Scientific, catalog no. 4331182, assay hs00245879_m1). Normalization was provided by multiplexing using a primer limited endogenous murine GAPDH control VIC-MGB Probe (Thermo Fisher Scientific, catalog no. 4352339E).

Mouse tumor experiments

All animal experiments were performed in accordance with NIH guidelines and approved by the NCI Animal Care and Use Committee. Female and male 6- to 10-week-old mice were inoculated into the intraperitoneal (IP) cavity with 1 × 106 cells in RPMI1640-lacking additives. For orthotopic inoculations, 10- to 19-week-old mice were anesthetized with isoflurane, then injected subcutaneously with Buprenorphine-SR [0.5–1.0 mg/kg], followed by the application of artificial tears to prevent animal eye drying. A small incision was made on the left flank, and the spleen and pancreas were externalized. KPCdelMsln + hMSLN were injected into the pancreas using a Hamilton syringe. The peritoneum was sutured shut, 1–2 drops of Bupivacaine [0.25%] were applied to the incision site, and the skin was closed with 2–3 9 mm wound clips. For single-agent LMB-100 studies, 0.5–2 × 105 cells in 5–10 μL of 1% nitrocellulose (Corning) were inoculated and cohorts of tumor-bearing mice were euthanized when the first mouse became moribund (14–21 days after inoculation) unless indicated otherwise, then all visible tumor was dissected from the abdominal cavity and weighed. For studies that included immune checkpoint inhibitor drugs, 1 × 104 KPCdelMsln + hMSLN cells/3μL in a mixture of DMEM/F-12 serum-free media and 4% methyl cellulose (SIGMA M7027) were injected. Animals were palpated weekly, starting at 14 days postsurgery, then prescreened by abdominal ultrasound imaging technique (VisualSonics Vevo 2100) to determine the volume of nascent tumors. Mice were enrolled onto drug treatment study when tumors reached 50–100 mm3 volume. A rolling cohort and covariate adaptive randomization approach were applied to allocate animals to specific treatment arms.

Mice were treated with the following therapeutics: LMB-100 (intravenous, 3.5 mg/kg animal body mass, every 4 days × 3), anti-CTLA4 (BioXcell, clone UC10–4F10–11, catalog no. BP-0032, intraperitoneally, 25 μg/mouse, every 4 days × 3), anti-PD1 (BioXcell, clone J43, catalog no. BE-0033–2, intraperitoneally, 10 mg/kg, every 4 days × 3), or control IgG (BioXcell, catalog no. BE-0089 and BP-0091). In combination treatment arms, LMB-110 immunotoxin was coadministered with anti-CTLA-4 or anti-PD1 checkpoint inhibitor antibodies. For assays using fluorescently labeled LMB-100, mice received a single 3.5 mg/kg dose of drug intravenously, then were euthanized 4 hours later.

Histologic analyses

Tissue specimens were fixed in 10% neutral buffered formalin solution for 48–72 hours at room temperature (HT501128, Sigma-Aldrich), then sent to Molecular Histotechnology Laboratory (MHL) Core facility for all histologic studies. Analysis and quantification were performed by a trained veterinary pathologist affiliated with the MHL Core. For quantification of serositis H&E staining was performed on prepared slides. Immune infiltration was measured using the HALO Imaging Analysis Platform (Indica Labs).

MSLN staining was optimized by the MHL Core using MSLN-specific antibody YP158 developed by Mitchell Ho and colleagues using rabbit hybridoma technology (U.S. Patent #US9409992B2). Five-micron thick paraffin sections were deparaffinized, rehydrated with ethanol, and rinsed in distilled water. Antigen retrieval (20 minutes at 97°C) with Tris/EDTA pH 9 (Agilent #S2367) was performed, followed by an endogenous peroxidase quench in 3% hydrogen peroxide. Slides were incubated overnight at 4°C with YP158. ImmPRESS HRP Horse anti-Rabbit IgG Polymer Detection Kit (Vector Labs #MP-7401), and ImmPACT DAB (Vector Labs #SK-4105) were used for detection. After hematoxylin counterstain, slides were dehydrated and coverslips applied. The following additional antibodies were used for IHC staining of mouse tissues: Iba1 (Biocare, catalog no. CP290, 1:500), CD3 (BioRad, catalog no. MCA1477, 1:100), cleaved caspase-3 (Cell Signaling Technology, catalog no. 9961, 1:100).

Statistics

GraphPad Prism and Microsoft Excel were used for all graphing, statistical calculations, and curve fitting. Data are presented as mean ± SD. Comparisons of >2 groups were assessed by ANOVA. For post hoc analysis and two group comparisons, Mann–Whitney nonparametric test or two-tailed Student t-test with Welch's correction were used where appropriate. All experiments were confirmed by repeat. Data are presented in composite form (as noted) when appropriate. N.S. = no statistically significant difference (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Human mesothelin is integrated into the murine genome

To develop an immunocompetent mouse model tolerant of tumors expressing hMSLN, C57BL/6 mice were genetically modified to endogenously express hMSLN in nonessential tissue. Thyroid-specific expression of hMSLN was produced by cloning the full-length hMSLN gene into a vector under control of the Thyroperoxidase (TPO) promoter (Fig. 1A). The transgene is composed of the full human MSLN gene sequence, which includes human megakaryocyte-potentiating factor (MPF), the protease cleavage site, and hMSLN. The MSLN cDNA is flanked by HS4 insulator sequences to ensure tissue-specific expression. Multiple founder animals were bred for at least five generations and line #171 was ultimately selected for further development (TPO171). Of note, despite extensive efforts to establish a homozygous line carrying two TPO171 alleles, we failed to identify TPO171tg/tg mice suggesting that gene insertion may have disrupted an essential gene or a function at the integration site. Heterozygous TPO171 transgenic mice grew and bred normally without gross phenotypic abnormalities. FISH hybridization was conducted using full-length hMSLN cDNA on murine embryonic fibroblasts derived from a TPO171tg/+ heterozygous fetus (Fig. 1B). Karyotype evaluation localized single-locus transgene on band F3 of murine chromosome 7. qRT-PCR was performed on RNA isolated from TPO mouse tissues and confirmed thyroid-specific expression of the hMSLN transgene (Fig. 1C).

Figure 1.

Genetically engineered mouse models with endogenous expression of hMSLN. A, Graphical representation of transgenic construct used in TPO171 mice. B, FISH was conducted using full-length hMSLN cDNA on murine embryonic fibroblasts derived from a TPO171 tg/+ heterozygous fetus revealing a single integrated locus on chromosome 7. C, Relative levels of hMSLN mRNA in tissues from TPO171 mice (n = 2) as determined by qRT-PCR. Positive controls (C), HeLa cell RNA, negative controls (0), no cDNA. D, Schema depicting transgene integration in Msl mouse model. The Flp-FRT–mediated recombination was used to target hMSLN transgenic cDNA into the native murine Msln locus of C57BL/6 so that the insertion disrupts expression of mMsln. LoxP sites flanking the coding region result in removal of the hMSLN transgene upon exposure to Cre to produce a mouse lacking both transgenic hMSLN and native mMsln expression.

Figure 1.

Genetically engineered mouse models with endogenous expression of hMSLN. A, Graphical representation of transgenic construct used in TPO171 mice. B, FISH was conducted using full-length hMSLN cDNA on murine embryonic fibroblasts derived from a TPO171 tg/+ heterozygous fetus revealing a single integrated locus on chromosome 7. C, Relative levels of hMSLN mRNA in tissues from TPO171 mice (n = 2) as determined by qRT-PCR. Positive controls (C), HeLa cell RNA, negative controls (0), no cDNA. D, Schema depicting transgene integration in Msl mouse model. The Flp-FRT–mediated recombination was used to target hMSLN transgenic cDNA into the native murine Msln locus of C57BL/6 so that the insertion disrupts expression of mMsln. LoxP sites flanking the coding region result in removal of the hMSLN transgene upon exposure to Cre to produce a mouse lacking both transgenic hMSLN and native mMsln expression.

Close modal

A second transgenic mouse (Msl mouse model) was made using Flp-FRT–mediated recombination to target integration of the human MSLN transgenic cDNA into the native murine Msln locus of C57BL/6, so that insertion disrupts expression of mMsln (Fig. 1D). Both homozygous and heterozygous Msl mice grew and bred normally.

Appropriate tissue-specific expression of hMSLN is seen in TPO171 and Msl models

Tissues from TPO171 and Msl mice were harvested and protein lysates were immunoblotted to detect hMSLN expression (Fig. 2A). As expected, in TPO171 mice hMSLN was detected in the thyroid, but not in any other organ. IHC analysis of TPO171 tissues corroborated this finding (Fig. 2B). In Msl mice, we expected hMSLN expression exclusively on serosal surfaces, mimicking the native murine expression of mMsln. Immunoblotting demonstrated strong hMSLN expression in whole-tissue extracts of heart, lung, and spleen with detectable, but significantly less expression in liver and kidney (Fig. 2A). IHC localization demonstrated that hMSLN expression was limited to the serosal surfaces of these organs without any expression in parenchyma (Fig. 2B).

Figure 2.

Protein expression of hMSLN in TPO171 and Msl mice. A, Immunoblot showing hMSLN expression in indicated tissues harvested from TPO171 or Msl mice. GAPDH was used as a loading control. B, IHC assessing hMSLN expression was performed on Msl and TPO mouse tissues using the YP158 antibody. LN, lymph node. C and D, Serum hMSLN concentration as measured by hMSLN-specific ELISA in wild-type C57Bl/6 (+/+), heterozygous TPO (Tg/+), homozygous (Ki/Ki) Msl, or heterozygous (Ki/+) Msl mouse serum. E, Msl fetal tissues harvested midgestation were probed with anti-hMSLN antibody.

Figure 2.

Protein expression of hMSLN in TPO171 and Msl mice. A, Immunoblot showing hMSLN expression in indicated tissues harvested from TPO171 or Msl mice. GAPDH was used as a loading control. B, IHC assessing hMSLN expression was performed on Msl and TPO mouse tissues using the YP158 antibody. LN, lymph node. C and D, Serum hMSLN concentration as measured by hMSLN-specific ELISA in wild-type C57Bl/6 (+/+), heterozygous TPO (Tg/+), homozygous (Ki/Ki) Msl, or heterozygous (Ki/+) Msl mouse serum. E, Msl fetal tissues harvested midgestation were probed with anti-hMSLN antibody.

Close modal

In humans, hMSLN is shed from serosal surfaces through the action of cellular proteases and then secreted into the serum. To determine whether hMSLN shedding also occurs in our models, we used ELISA to measure the concentration of hMSLN in mouse serum. Shed hMSLN was not detectable in the serum of WT C57Bl/6 mice (Fig. 2C). Heterozygous TPO171 mice did have detectable concentrations of serum hMSLN (Fig. 2C). Elevated serum hMSLN concentrations were also found in Msl mice. The serum concentration of shed hMSLN in Msl mice was dependent on gene dose and had an absolute concentration more than a log-fold higher than that of TPO171 mice (Fig. 2D).

Because of the overall poor specificity of antibodies against the murine isoform, studies localizing mMsln protein expression in embryologic tissues have previously proved challenging. Because hMSLN expression is driven from the native Msln promoter, we presumed that embryologic expression, like adult expression of the hMSLN transgene, would mimic that of mMsln. Our data show that the transgene was expressed in several embryonic serosal tissues, similar to adults, but could also be localized to maternal placenta and to the transitional epithelium of the bladder. Bladder expression was detectable from midgestation through the perinatal stage of embryogenesis (Fig. 2E). Of note, no expression in bladder transitional epithelium has been detected in adult animals nor in the epithelial lining of the human urogenital tract. This suggests MSLN may have a yet to be understood function in late embryonic development and/or organogenesis of the bladder and/or urogenital tract.

Characterization of on-target off-tumor toxicity in TPO171 mice

Recombinant immunotoxins (iTox) are anticancer therapeutics with an antibody-based targeting domain linked to a toxin payload. LMB-100 is a hMSLN-targeted iTox that delivers a modified Pseudomonas exotoxin A payload to hMSLN-expressing cells. The payload is inactive unless internalized, making bystander effect unlikely. The high specificity of the drug for hMSLN-expressing cells has been demonstrated previously (17). We anticipated that treatment of TPO171 mice with LMB-100 would ablate or injure the thyroid follicular cells, which robustly express hMSLN resulting in hypothyroidism. To test this, TPO171 mice were treated with vehicle or LMB-100 (at a dose and schedule that is well tolerated by WT C57BL/6 mice) and surveyed for toxicity. Administration of LMB-100 to both WT and TPO171 mice produced an initial drop in weight, which recovered similarly after cessation of LMB-100 treatment (Fig. 3A). Ultimately, both strains returned to or exceeded their baseline weight, suggesting the reversible weight loss was due to acute LMB-100 toxicity. There was no indication of excessive weight gain in TPO171 mice as would be expected with hypothyroidism. Furthermore, no histopathologic changes to TPO171 mouse thyroid tissue was appreciated on H&E following LMB-100 treatment (Fig. 3B). Naïve TPO171 mice were further challenged with two 3.5 mg/kg doses (on days 1 and 3) of either functional LMB100 immunotoxin or its mutant variant carrying an inactive PE exotoxin moiety, then euthanized on day 5. Staining of thyroid tissues with either Iba1-detecting tissue infiltrating macrophages and activated microglia or cleaved caspase-3–specific antibodies revealed no differences in samples treated with active versus inactive LMB100 iTox (Fig. 3C and D; Supplementary Fig. S1). However, a statistically significant increase in CD3-expressing lymphocytes was detected following LMB100 treatment, (Fig. 3D; Supplementary Fig. S1). To assess for possible long-term toxicity, TPO171 mice were treated with vehicle versus LMB-100 and monitored for nearly 2 months. Excessive weight gain, a sign of hypothyroidism, was not observed in LMB-100–treated TPO171 mice (Fig. 3E), nor were changes in serum concentration of thyroid hormone observed after LMB-100 treatment (Fig. 3F). To summarize, we unexpectedly found no signs or symptoms of thyroid dysfunction in treated mice despite the small acute increase in acute CD3+ cell infiltration.

Figure 3.

A, Weight change over time in C57Bl/6 wild-type (WT) mice versus TPO mice treated with LMB-100. Arrows on x-axis indicate times of LMB-100 administration to both strains. B, H&E staining of thyroid tissue harvested from WT and TPO171 mice administered LMB-100 as in A. C, Comparative assessment of tissue-infiltrating macrophages (left panel), CD3-expressing lymphocytes (right panel), and cells undergoing apoptosis (middle panel) in thyroid samples isolated from TPO171 mice treated with either functional or inactive forms of LMB100 immunotoxin. D, Tissue-infiltrating macrophages are visualized in thyroid gland samples by anti-iba1 staining. E, Weight change over time in TPO171 mice given either LMB-100 or vehicle control (PBS) on days 1, 3, and 5. F, Serially measured serum thyroid hormone (T4) concentration in TPO171 mice treated as in C. G, TPO171 mice were intravenously injected with fluorescently labeled LMB-100, then euthanized 4 hours later. Left, H&E of thyroid tissue. Right, Corresponding fluorescence microscopy detecting Alexafluor-labeled LMB-100 (magenta) and DAPI (blue).

Figure 3.

A, Weight change over time in C57Bl/6 wild-type (WT) mice versus TPO mice treated with LMB-100. Arrows on x-axis indicate times of LMB-100 administration to both strains. B, H&E staining of thyroid tissue harvested from WT and TPO171 mice administered LMB-100 as in A. C, Comparative assessment of tissue-infiltrating macrophages (left panel), CD3-expressing lymphocytes (right panel), and cells undergoing apoptosis (middle panel) in thyroid samples isolated from TPO171 mice treated with either functional or inactive forms of LMB100 immunotoxin. D, Tissue-infiltrating macrophages are visualized in thyroid gland samples by anti-iba1 staining. E, Weight change over time in TPO171 mice given either LMB-100 or vehicle control (PBS) on days 1, 3, and 5. F, Serially measured serum thyroid hormone (T4) concentration in TPO171 mice treated as in C. G, TPO171 mice were intravenously injected with fluorescently labeled LMB-100, then euthanized 4 hours later. Left, H&E of thyroid tissue. Right, Corresponding fluorescence microscopy detecting Alexafluor-labeled LMB-100 (magenta) and DAPI (blue).

Close modal

This finding raised concern that the hMSLN transgene may be incorrectly processed in murine tissue and poorly recognized by LMB-100. To test this, we injected TPO171 mice with fluorescently labeled LMB-100 and examined thyroid uptake. Robust delivery of the fluorescently labeled LMB-100 was observed in thyroid follicular cells within 4 hours of drug injection (Fig. 3G). This data demonstrate that LMB-100 can bind transgenic hMSLN in TPO171 mice and suggest that thyroid follicular cells may be resistant to killing by the iTox.

Characterization of on-target off-tumor toxicity in Msl mice

Phase I clinical studies of LMB-100 and the predecessor hMSLN-targeted iTox SS1P demonstrated that both iTox caused on-target off-tumor serosal membrane toxicity in human patients (18). In fact, pleuritis was found to be the dose-limiting toxicity of SS1P (19). Given that Msl mice have similar tissue distribution of hMSLN as human patients, we anticipated that treatment of Msl mice with LMB-100 or SS1P would cause inflammation of the serosal membrane tissues. Msl mice receiving LMB-100 at a dose and schedule nonlethal to WT mice showed no overt signs of illness, nor respiratory or cardiac dysfunction. However, histopathologic analysis of treated Msl mice demonstrated acute, mild to moderate infiltration of pleura (Fig. 4A) and pericardium (Fig. 4B) by immune cells that was not observed in WT control mice. These changes were quantified by HALO analysis, which revealed a statistically significant increase in serositis in Msl mice treated with LMB-100 or SS1P compared with controls. The observed inflammation persisted long-term. Mild changes were still visible at 30 days following the last treatment (Fig. 4C and D). Furthermore, while both Msl and WT mice given LMB-100 experienced an initial weight drop during LMB-100 treatment (as noted with TPO171 mice), only the Msl mice were unable to regain this weight (Fig. 4E). Taken together, these data suggest that hMSLN-targeted iTox can induce acute and subacute serositis.

Figure 4.

Msl (Ki) mice were administered either LMB-100 (L, 3.5 mg/kg, every other day × 3) or SS1P (S, 0.4 mg/kg, every other day × 3) iTox. Wild-type (WT) mice treated with LMB-100 served as negative control. A and B, Mice were euthanized 48 hours after treatment. H&E staining was performed on lung (A) and heart (B) tissue harvested from treated mice. Inflammatory changes were quantified using HALO image analysis. C, Mice were euthanized 30 days after completion of treatment. H&E staining of heart and lungs was performed to assess inflammatory changes in serosal membranes. Inflammation was graded on a scale of 0 to 4 (0, normal; 1, minimal; 2, mild; 3, moderate; 4, severe) by a pathologist blinded to treatment group. D, Representative H&E staining of inflamed pericardium, parietal, and visceral pleura (respectively, from left to right). E, LMB-100 was administered to wild-type (WT) or Msl (Ki) mice and weight change tracked over 30 days. NS, not significant.

Figure 4.

Msl (Ki) mice were administered either LMB-100 (L, 3.5 mg/kg, every other day × 3) or SS1P (S, 0.4 mg/kg, every other day × 3) iTox. Wild-type (WT) mice treated with LMB-100 served as negative control. A and B, Mice were euthanized 48 hours after treatment. H&E staining was performed on lung (A) and heart (B) tissue harvested from treated mice. Inflammatory changes were quantified using HALO image analysis. C, Mice were euthanized 30 days after completion of treatment. H&E staining of heart and lungs was performed to assess inflammatory changes in serosal membranes. Inflammation was graded on a scale of 0 to 4 (0, normal; 1, minimal; 2, mild; 3, moderate; 4, severe) by a pathologist blinded to treatment group. D, Representative H&E staining of inflamed pericardium, parietal, and visceral pleura (respectively, from left to right). E, LMB-100 was administered to wild-type (WT) or Msl (Ki) mice and weight change tracked over 30 days. NS, not significant.

Close modal

Testing immunotherapy in Msl mice bearing syngeneic tumors

To test the antitumor efficacy of hMSLN-targeted therapeutics in our immunocompetent transgenic animals required development of an hMSLN-expressing cell line syngeneic to C57BL/6. Therefore, we engineered an hMSLN-expressing cell line from a variant KPC-mouse–derived cell line. The KPC genetically engineered mouse (GEM) model harbors pancreas-specific mutations of Kras and Tp53 and spontaneously develops autochthonous PDAC. In brief, we took mice with global knockout of mMsln (20) and bred them into the KPC transgenic system to produce mMsln–/– mice with pancreas-specific mutation of Kras and Tp53. A primary PDAC cell line (KPCdelMsln) derived from these mice was then transduced with a lentiviral vector to establish expression of hMSLN (KPCdelMsln + hMSLN cells). KPCdelMsln + hMSLN cells are exquisitely sensitive to LMB-100 with IC50 approximately 5 ng/mL (Supplementary Fig. S2A). Orthotopic tumors generated using KPCdelMsln + hMSLN cells in Msl mice were consistently detectable by palpation within 14 days of cell inoculation and lethal by 28 days due to local invasion (see Supplementary Fig. S2B for histology). Interestingly, we found that serum hMSLN concentration of mice bearing KPCdelMsln + hMSLN–derived tumors could serve as a surrogate marker to track tumor growth. Msl mice bearing IP tumors from KPCdelMsln + hMSLN cells demonstrated a correlation of tumor burden to serum hMSLN concentration (r2 = 0.9934, P < 0.001; Supplementary Fig. S2C). These data demonstrate that serum hMSLN concentration can be used as an accurate surrogate for tumor burden during antitumor experiments using KPCdelMsln + hMSLN cells.

We next tested the antitumor efficacy of single-agent LMB-100 in IP and orthotopic tumor models using Msl mice bearing KPCdelMsln + hMSLN tumors. LMB-100 treatment reduced end tumor burden in the IP disease model as compared with vehicle (Fig. 5A). Serum hMSLN measurements from the treated mice suggested that tumor growth was slowed but not abrogated by LMB-100. In the orthotopic model, mice receiving LMB-100 did not have a statistically significant reduction in tumor burden but did trend toward a decreased rate of tumor growth as compared with vehicle (Fig. 5B).

Figure 5.

Msl mice bearing PDAC tumors implanted intraperitoneally (A) or orthotopically into the pancreas (B) were treated with LMB-100 (iTox) as described in Materials and Methods. Mice were euthanized 48 hours after treatment, and total tumor was dissected and weighed (A and B, left panel). Serum hMSLN was serially measured throughout the experimental course (A and B, right panel). C, C57Bl/6 wild-type (WT) or Msl mice (Ki) were treated with either LMB-100 alone (L), CTLA-4 inhibitor alone (C), or LMB-100 + CTLA-4 inhibitor (L+C). Mice were euthanized posttreatment and their lungs harvested for assessment of pleuritis using HALO analysis. D and E, Kaplan–Meier survival curves of Msl mice bearing orthotopic PDAC tumors treated as indicated. F, G, and H, Microscopy analysis of tissues from mice bearing orthotopically implanted PDAC tumors treated with fluorescently labeled LMB-100. F, H&E (left tumor panel, top lung panel), Fluorescent labeling (mid and right tumor panels, mid and lower lung panels) shows, anti-hMSLN YP158 (gold), Alexa-LMB-100 (magenta), DAPI (blue) from the same mouse. Scale bars, 100 μm for tumor, 50 μm for lung. G, Tumor is marked by *, and peritoneal membrane is indicated by arrow. Scale bar, 200 μm. H, Magnified image of tumor identifies focal delivery of LMB-100 to tumor tissue immediately adjacent to blood vessels (*). Alexa-LMB-100 (magenta).

Figure 5.

Msl mice bearing PDAC tumors implanted intraperitoneally (A) or orthotopically into the pancreas (B) were treated with LMB-100 (iTox) as described in Materials and Methods. Mice were euthanized 48 hours after treatment, and total tumor was dissected and weighed (A and B, left panel). Serum hMSLN was serially measured throughout the experimental course (A and B, right panel). C, C57Bl/6 wild-type (WT) or Msl mice (Ki) were treated with either LMB-100 alone (L), CTLA-4 inhibitor alone (C), or LMB-100 + CTLA-4 inhibitor (L+C). Mice were euthanized posttreatment and their lungs harvested for assessment of pleuritis using HALO analysis. D and E, Kaplan–Meier survival curves of Msl mice bearing orthotopic PDAC tumors treated as indicated. F, G, and H, Microscopy analysis of tissues from mice bearing orthotopically implanted PDAC tumors treated with fluorescently labeled LMB-100. F, H&E (left tumor panel, top lung panel), Fluorescent labeling (mid and right tumor panels, mid and lower lung panels) shows, anti-hMSLN YP158 (gold), Alexa-LMB-100 (magenta), DAPI (blue) from the same mouse. Scale bars, 100 μm for tumor, 50 μm for lung. G, Tumor is marked by *, and peritoneal membrane is indicated by arrow. Scale bar, 200 μm. H, Magnified image of tumor identifies focal delivery of LMB-100 to tumor tissue immediately adjacent to blood vessels (*). Alexa-LMB-100 (magenta).

Close modal

Given the modest single-agent activity of LMB-100 in this model and previous reports that minimal cell killing by iTox could facilitate antitumor activity of anti-PD1 (21) and anti-CTLA415 in lung, mesothelioma, and breast cancer models, we decided to test whether LMB-100 could boost the activity of ICIs in our PDAC model. To investigate these combinations, Msl mice bearing orthotopic KPCdelMsln + hMSLN tumors were treated with anti-CTLA4, or anti-PD1 with and without LMB-100. Interestingly, addition of anti-CTLA4 to LMB-100 treatment did not further increase pleuritis beyond that observed with single-agent LMB-100 (Fig. 5C). However, mice treated with LMB-100 or anti-CTLA4 alone did show prolonged survival compared with those receiving control IgG (Fig. 5D). Nevertheless, combining the two agents did not further enhance survival in this model. Neither was improved survival noted in mice treated with anti-PD1 alone or in combination with LMB-100 (Fig. 5E). Histologic studies confirmed continued robust expression of the transgene in tumor cells (Supplementary Fig. S3A), but also revealed increased stromal density in the orthotopic model as compared with the IP model (Supplementary Fig. S3B). This raised concern that the dense PDAC stroma in orthotopic tumors was impeding delivery of LMB-100. To test this, orthotopic tumor-bearing Msl mice were injected with fluorescently labeled LMB-100 and then tissues were harvested 4 hours later to assess delivery. Robust delivery of labeled LMB-100 to mesothelial cells in the pleura (Fig. 5F) and peritoneum (Fig. 5G) was observed, yet delivery to hMSLN-expressing tumor cells was quite poor (Fig. 5F&G) within the same animal. Delivery of LMB-100 was detected in less than 5% of tumor cells in all tumor-bearing mice examined (n = 3). Qualitatively, most labeled tumor cells were proximal to tumor blood vessels (Fig. 5H).

Here, we report the development and characterization of two high-fidelity, immunocompetent murine models for human cancer that allow for rigorous preclinical evaluation of hMSLN-targeted therapeutics. Using the Msl model, we found that systemic administration of a hMSLN-targeted iTox is associated with acute and subacute histopathologic serositis and that this toxicity is not exacerbated by coadministration of immune-activating agents such as anti-CTLA4. In addition, we were able to show that MSLN-targeted iTox LMB-100 lacks synergistic activity with ICI in PDAC and thus more investigation is needed before considering this combination for clinical trial in patients with pancreatic cancer. The absence of antitumor activity of LMB-100 alone or in combination with ICI is likely due to dense tumor stroma and lack of drug delivery to hMSLN-expressing tumor cells.

A number of MSLN-targeted therapies have undergone clinical investigation. Two iTox agents, mAbs, adoptive cell therapy, vaccines and antibody-drug conjugates (ADC) have all been tested in the clinic (1, 9, 10, 22). In addition, bicyclic and even trispecific antibodies are currently entering the clinical space (23). We believe the models described here provide an exciting opportunity to study these agents in a complex system that closely mimics the human patient with cancer. These new models provide a number of advantages in the study of MSLN-targeted therapeutics. Notably, they possess an intact immune system that can offer a complete picture of immune-mediated antitumor activity induced by a therapeutic regimen. Furthermore, the Msl model provides an accurate physiologic representation of both tissue and circulating hMSLN, which may influence both efficacy and toxicity. Finally, vital to a preclinical model of human cancer is the ability to measure tumor response. Pancreatic orthotopic and intraperitoneal tumors provide complex tumor microenvironments closely mimicking naturally occurring tumors but are not easily measurable before animal euthanization or death. Experimenter palpation is unreliable and not quantifiable. Fluorescently labeled cells are often used but introduce a confounding variable as immunocompetent mice can mount responses against the GFP antigen. Another option is ultrasound imaging, which is technically demanding and difficult to reproduce. Our Msl model allows us to overcome this barrier by measuring hMSLN in the serum. In this case, serum hMSLN concentration serves as a tumor marker that directly correlates to tumor burden. This surrogate for tumor measurement, unique to our model, gives us the ability to observe tumor growth over time with an easily repeatable assay.

There are several limitations to the use of our model in preclinical studies. Tumors that are grown by syngeneic cell inoculation, either intraperitoneally or orthotopically, may not provide a true representation of the PDAC microenvironment. Genetically engineered mice able to grow autocthonous tumors expressing hMSLN is an area of ongoing investigation by our group. In addition, hMSLN expression in KPCdelMsln + hMSLN cells is artificially driven by a lentiviral vector and may over-represent levels of cell surface and circulating hMSLN thus potentially altering treatment effect. Furthermore, it is unclear that the murine cellular machinery can process hMSLN in the same way as murine Msln. While we know hMSLN is expressed on the murine cell surface and shed like in humans, there may be variations in which parts of the molecule end up on the surface versus in the circulation or discarded. These differences may be important in iTox recognition and uptake by the cell.

Our finding that serositis develops in Msl mice treated with iTox is an important milestone in the study of these therapeutics. This is direct evidence that iTox can induce target-specific inflammatory response in normal host tissue. Developing methods to abrogate such a response may prove vital to advancing the clinical use of iTox. Investigation into specific immune components involved and molecular mediators of these changes is beyond the scope of this article but warrants further study. We hypothesized that serositis would be further exacerbated by addition of ICI. In fact, there was no statistically significant change in serositis detected supporting the safety of concurrently administering such a combination for tumor types such as lung cancer and mesothelioma where evidence for enhanced activity of the combination exists (24, 25).

Interestingly, there was no evidence of thyroid follicular cell killing by iTox and TPO mice did not develop thyroiditis or hypothyroidism after treatment with LMB-100 despite robust transgenic expression of the hMSLN target in thyroid cells and a small increase in CD3+ cell infiltrate. This data demonstrate that other factors besides target expression are required to facilitate iTox-mediated killing. One reason for target cell resistance may be that thyroid follicular cells lack the machinery to efficiently endocytose the iTox–MSLN complex as iTox must reach the cell cytosol to have killing activity. The experiments we have performed cannot rule out the possibility that iTox has reached the thyroid follicular cell surface, but failed to efficiently traffic into the cell. Alternatively, thyroid follicular cells may be intrinsically resistant to LMB-100. LMB-100 contains a modified Pseudomonas Exotoxin A payload that lacks most of Domain 2. Several cell lines highly sensitive to Domain 2–containing payloads lose this sensitivity when this region of unknown function is removed from the payload (26). Relative resistance to iTox-mediated killing was also observed in pancreatic tumor cells grown orthotopically, but our data demonstrate that poor delivery of the therapeutic is most likely responsible. Tumors expressing hMSLN also do not develop the level of immune infiltration observed in serosal membranes. Investigation into what factors may play a role in iTox-mediated inflammation may provide insight into how best to leverage potent MSLN-targeted iToxs into effective anticancer therapies.

In conclusion, we have engineered and characterized two novel mouse models of human pancreatic cancer. These models have the potential for a variety of applications in the field of MSLN-targeted immunotherapies and MSLN biology.

The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government.

C. Alewine reports nonfinancial support from Roche and grants from NIH Bench to Bedside Award during the conduct of the study as well as nonfinancial support from Minneamrita, ProDa, and Astra Zeneca outside the submitted work. No disclosures were reported by the other authors.

B. Hagerty: Investigation, data curation, writing–original draft, writing–review and editing. T. Norene O'Sullivan: Investigation. X. Zhang: Validation, investigation. N.K. Collins: Data curation. W. Custer Lawrence: Investigation. L.L. Bassel: Investigation. N. Pate: Investigation. J. Xu: Validation. T.M. Guerin: Investigation. S. Kozlov: Investigation, writing–original draft, writing–review and editing. C. Alewine: Investigation, writing–original draft, writing–review and editing.

The authors thank the MHL Core especially Elijah Edmondson, Baktiar Karim, and Donna Butcher. This research was supported by the Intramural Research Program of the NIH, NCI, Center for Cancer Research (C.A., Project No. ZIA BC 011652 and ZIA BC 011886). The study was funded in part (W. Custer Lawerence, L.L. Bassel, N. Pate, T.M. Guerin, and S. Kozlov) by federal funds from the NCI, NIH, under contract 75N91019D00024.

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.
Hassan
RBT
,
Pastan
I
. 
Mesothelin: a new target for immunotherapy
.
Clin Cancer Res
2004
;
10
:
3937
42
.
2.
Chang
K
,
Pastan
I
. 
Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesothiomas, and ovarian cancers
.
Proc Natl Acad Sci U S A
1996
;
93
:
136
40
.
3.
Koyama
Y
,
Wang
P
,
Liang
S
,
Iwaisako
K
,
Liu
X
,
Xu
J
, et al
Mesothelin/mucin 16 signaling in activated portal fibroblasts regulates cholestatic liver fibrosis
.
J Clin Invest
2017
;
127
:
1254
70
.
4.
Yen
M
,
Hsu
C
,
Mao
T
,
Wu
T
,
Roden
R
,
Wang
T
, et al
Diffuse mesothelin expression correlates with prolonged patient survival in ovarian serous carcinoma
.
Clin Cancer Res
2006
;
12
:
827
31
.
5.
Yu
L
,
Feng
M
,
Kim
H
,
Phung
Y
,
Kleiner
DE
,
Gores
GJ
, et al
Mesothelin as a potential therapeutic target in human cholangiocarcinoma
.
J Cancer
2010
;
1
:
141
9
.
6.
Argani
P
,
Iacobuzio-Donahue
C
,
Ryu
B
,
Rosty
C
,
Goggins
M
,
Wilentz
RE
, et al
Mesothelin is overexpressed in the vast majority of ductal adenocarinomas of the pancreas: identification of a new pancreatic cancer marker by serial analysis of gene expression (SAGE)
.
Clin Cancer Res
2001
;
7
:
3862
8
.
7.
Alewine
CHR
,
Pastan
I
. 
Advances in anticancer immunotoxin therapy
.
Oncologist
2015
;
20
:
176
85
.
8.
Hassan
R
,
Thomas
A
,
Alewine
C
,
Le
DT
,
Jaffee
EM
,
Pastan
I
. 
Mesothelin immunotherapy for cancer: ready for prime time?
J Clin Oncol
2016
;
34
:
4171
9
.
9.
Morello
A
,
Sadelain
M
,
Adusumilli
P
. 
Mesothelin-targeted CARs: driving T-cells to solid tumors
.
Cancer Discov
2016
;
6
:
133
46
.
10.
Hassan
R
,
Jennens
R
,
Meerbeeck
JPV
,
Nemunaitis
JJ
,
Blumenschein
GR
,
Fennell
DA
, et al
A pivotal randomized phase II study of anetumab ravtansine or vinorelbine in patients with advanced or metastatic pleural mesothelioma after progression on platinum/pemetrexed-based chemotherapy (NCT02610140)
.
J Clin Oncol
34
:
15s
, 
2016
(
suppl; TPS8576–TPS
).
11.
Hassan
R
,
Kindler
HL
,
Jahan
T
,
Bazhenova
L
,
Reck
M
,
Thomas
A
, et al
Phase II clinical trial of amatuximab, a chimeric antimesothelin antibody with pemetrexed and cisplatin in advanced unresectable pleural mesothelioma
.
Clin Cancer Res
2014
;
20
:
5927
36
.
12.
Le
DT
,
Wang-Gillam
A
,
Picozzi
V
,
Greten
TF
,
Crocenzi
T
,
Springett
G
, et al
Safety and survival with GVAX pancreas prime and Listeria Monocytogenes-expressing mesothelin (CRS-207) boost vaccines for metastatic pancreatic cancer
.
J Clin Oncol
2015
;
33
:
1325
33
.
13.
Li
M
,
Bharadwaj
U
,
Zhang
R
,
Zhang
S
,
Mu
H
,
Fisher
WE
, et al
Mesothelin is a malignant factor and therapeutic vaccine target for pancreatic cancer
.
Mol Cancer Ther
2008
;
7
:
286
96
.
14.
Zervos
EAS
,
Freistaedter
AG
,
Jones
GJB
,
Roper
RL
. 
Murine mesothelin: characterization, expression, and inhibition of tumor growth in a murine model of pancreatic cancer
.
J Exp Clin Cancer Res
2016
;
35
:
39
.
15.
Leshem
Y
,
O'Brien
J
,
Liu
X
,
Bera
TK
,
Terabe
M
,
Berzofsky
JA
, et al
Combining local immunotoxins targeting mesothelin with CTLA-4 blockade synergistically eradicates murine cancer by promoting anti-cancer immunity
.
Cancer Immunol Res
2017
;
5
:
685
94
.
16.
Bauss
F
,
Imhof-Jung
S
,
Hesse
F
,
Kronenberg
S
,
Lechmann
M
,
Krippendorff
B
, et al
RG7787 - a novel de-immunized PE based fusion protein for therapy of mesothelin-positive solid tumors [abstract]
. In:
Proceedings of the American Association for Cancer Research Annual Meeting; 2014; 2014 Apr 5-9; San Diego, CA
.
Philadelphia (PA)
:
AACR
; 
2014
. Abstract nr 4510.
17.
Alewine
C
,
Xiang
L
,
Yamori
T
,
Niederfellner
G
,
Bosslet
K
,
Pastan
I
. 
Efficacy of RG7787, a next-generation mesothelin-targeted immunotoxin, against triple-negative breast and gastric cancer
.
Mol Cancer Ther
2014
;
13
:
2653
61
.
18.
Hagerty
B
,
Pegna
G
,
Xu
J
,
Tai
C-H
,
Alewine
C
. 
Mesothelin-targeted recombinant immunotoxins for solid tumors
.
Biomolecules
2020
;
10
:
973
.
19.
Hassan
R
,
Bullock
S
,
Premkumar
A
,
Kreitman
RJ
,
Kindler
H
,
Willingham
MC
, et al
Phase 1 study of SS1P, a recombinant anti-mesothelin immunotoxin given as a bolus I.V. infusion to patients with mesothelin-expressing mesothelioma, ovarian and pancreatic cancers
.
Clin Cancer Res
2007
;
13
:
5144
9
.
20.
Bera
T
,
Pastan
I
. 
Mesothelin is not required for normal mouse development or reproduction
.
Mol Cell Biol
2000
;
20
:
2902
6
.
21.
Jiang
Q
,
Ghafoor
A
,
Mian
I
,
Rathkey
D
,
Thomas
A
,
Alewine
C
, et al
Enhanced efficacy of mesothelin-targeted immunotoxin LMB-100 and anti-PD-1 antibody in patients with mesothelioma and mouse tumor models
.
Sci Transl Med
2020
;
12
:
eaaz7252
.
22.
Kelly
R
,
Sharon
E
,
Pastan
I
,
Hassan
R
. 
Mesothelin-targeted agents in clinical trials and in pre-clinical development
.
Mol Cancer Ther
2012
;
11
:
517
25
.
23.
Molloy
ME
,
Austin
RJ
,
Lemon
BD
,
Aaron
WH
,
Ganti
V
,
Jones
A
, et al
Preclinical characterization of HPN536, a trispecific, T-cell–activating protein construct for the treatment of mesothelin-expressing solid tumors
.
Clin Cancer Res
2021
;
27
:
1452
62
.
24.
Leshem
Y
,
King
EM
,
Mazor
R
,
Reiter
Y
,
Pastan
I
. 
SS1P Immunotoxin induces markers of immunogenic cell death and enhances the effect of the CTLA-4 blockade in AE17M mouse mesothelioma tumors
.
Toxins
2018
;
10
:
470
.
25.
Mazor
R
,
King
E
,
Pastan
I
. 
Anti-drug antibodies to LMB-100 are enhanced by mAbs targeting OX40 and CTLA4 but not by mAbs targeting PD1 or PDL-1
.
Cell Immunol
2018
;
334
:
38
41
.
26.
Weldon
JE
,
Skarzynski
M
,
Therres
JA
,
Ostovitz
JR
,
Zhou
H
,
Kreitman
RJ
, et al
Designing the furin-cleavable linker in recombinant immunotoxins based on Pseudomonas exotoxin A
.
Bioconjug Chem
2015
;
26
:
1120
8
.

Supplementary data