An antimesothelin-monomethyl auristatin e conjugate with potent antitumor activity in ovarian, pancreatic, and mesothelioma models.

Published OnlineFirst September 23, 2014; DOI: 10.1158/1535-7163.MCT-14-0487-T

Molecular Cancer Therapeutics

Large Molecule Therapeutics

An Antimesothelin-Monomethyl Auristatin E Conjugate with Potent Antitumor Activity in Ovarian, Pancreatic, and Mesothelioma Models Suzie J. Scales1, Nidhi Gupta1, Glenn Pacheco2, Ron Firestein3, Dorothy M. French3, Hartmut Koeppen3, Linda Rangell3, Vivian Barry-Hamilton1, Elizabeth Luis4, Josefa Chuh5, Yin Zhang6, Gladys S. Ingle1, Aimee Fourie-O'Donohue5, Katherine R. Kozak5, Sarajane Ross2, Mark S. Dennis6, and Susan D. Spencer2

Abstract Mesothelin (MSLN) is an attractive target for antibody–drug conjugate therapy because it is highly expressed in various epithelial cancers, with normal expression limited to nondividing mesothelia. We generated novel antimesothelin antibodies and conjugated an internalizing one (7D9) to the microtubuledisrupting drugs monomethyl auristatin E (MMAE) and MMAF, finding the most effective to be MMAE with a lysosomal protease-cleavable valine–citrulline linker. The humanized (h7D9.v3) version, aMSLNMMAE, specifically targeted mesothelin-expressing cells and inhibited their proliferation with an IC50 of 0.3 nmol/L. Because the antitumor activity of an antimesothelin immunotoxin (SS1P) in transfected mesothelin models did not translate to the clinic, we carefully selected in vivo efficacy models endogenously expressing clinically relevant levels of mesothelin, after scoring mesothelin levels in ovarian, pancreatic, and mesothelioma tumors by immunohistochemistry. We found that endogenous mesothelin in cancer cells is upregulated in vivo and identified two suitable xenograft models for each of these three indications. A single dose of aMSLN-MMAE profoundly inhibited or regressed tumor growth in a dosedependent manner in all six models, including two patient-derived tumor xenografts. The robust and durable efficacy of aMSLN-MMAE in preclinical models of ovarian, mesothelioma, and pancreatic cancers justifies the ongoing phase I clinical trial. Mol Cancer Ther; 13(11); 2630–40. 2014 AACR.

Introduction Mesothelin (MSLN) is a glycoprotein limited to the surface of the single mesothelial cell layer lining the pericardia, pleura, and peritoneum (1). It is expressed as an approximately 75 kDa fusion protein with the so-called megakaryocyte-promoting factor (MPF), which is removed by furin cleavage, leaving an approximately 50 kDa glycosylphosphatidylinositol-anchored mature mesothelin on the cell surface (2). Mesothelin-knockout mice exhibit no discernible phenotype, thus the normal biologic functions of MPF and mesothelin are unknown (3). Mesothelin is also overexpressed in a high percentage of various epi1 Department of Molecular Biology, Genentech, South San Francisco, California. 2Department of Translational Oncology, Genentech, South San Francisco, California. 3Department of Pathology, Genentech, South San Francisco, California. 4Department of Protein Chemistry, Genentech, South San Francisco, California. 5Department of Biochemical and Cellular Pharmacology, Genentech, South San Francisco, California. 6Department of Antibody Engineering, Genentech, South San Francisco, California.

Note: Supplementary data for this article are available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/). Corresponding Author: Suzie J. Scales, Genentech, 1 DNA Way, MS37, South San Francisco, CA 94080. Phone: 650-225-1510; Fax 650-4678882; E-mail: [email protected] doi: 10.1158/1535-7163.MCT-14-0487-T 2014 American Association for Cancer Research.

2630

thelial cancers, including ovarian, pancreatic, and mesothelioma (see references in Supplementary Table S1). The high tumor versus normal expression makes mesothelin an attractive target for cancer therapy. Indeed, besides vaccine (4) and chimeric antigen receptor T-cell immunotherapy (5), a murine dsFv to mesothelin has been fused to the protein synthesis inhibiting (PE38) domain of pseudomonas exotoxin to generate SS1P (CAT-5001; refs. 1, 6). This immunotoxin exhibits marked antitumor activity against xenografts of mesothelin-transfected A431 cells, but was not tested in endogenously expressing xenografts due to lack of suitable models. In a phase I trial, SS1P produced only 4 of 34 minor responses in mesothelin-selected cancers, and was dose limited by pleuritis (7), likely because PE38 inhibited protein synthesis in the mesothelin-positive pleura. Immunogenicity prevented further cycles of treatment even following lymphocyte depletion (8), so a deimmunized version (RG7787) is currently under investigation (9). Chimeric SS1 IgG without PE38 (MORAb-009, amatuximab) exerts modest preclinical activity via antibody-dependent cell-mediated cytotoxicity (10), but no clinical activity (11). Recent advances in antibody–drug conjugate (ADC) technology suggest that antimitotic ADCs with human or humanized antibodies to mesothelin could be a more promising approach. ADCs comprise whole IgGs

Mol Cancer Ther; 13(11) November 2014

Downloaded from mct.aacrjournals.org on June 25, 2015. © 2014 American Association for Cancer Research.


Preclinically Efficacious Antimesothelin–MMAE Conjugate

covalently linked to cytotoxic drugs that are orders of magnitude more potent than traditional chemotherapeutics (12, 13). ADCs are essentially prodrugs, optimally requiring internalization so the antibody and/or linker is digested to release the active drug (12, 14). The recent approval of two microtubule-inhibiting ADCs with different linker-drug platforms—anti-CD30–valine–citrulline–monomethyl auristatin E (SGN-35; ref. 15) and anti-HER2-MCC-DM1 (T-DM1; ref. 16)—offers hope for these technologies with other tumor antigens. Indeed, several other ADCs are at various stages of clinical development, including an antimesothelin–DM4 conjugate (BAY-94-9343) with a reducible SPDB linker (17). Antimesothelin ADCs have three major potential advantages over SS1P: antimitotic activity conferring inherent specificity for rapidly dividing cancerous over nondividing mesothelial cells; less immunogenicity, permitting multiple dosing cycles; and longer serum half-life to increase therapeutic exposure. Given the association of DM4 with ocular toxicity (18, 19), we chose to test auristatin-based antimesothelin ADCs using novel monoclonals to mesothelin. The most efficacious was an internalizing proteasesensitive valine–citrulline–linked monomethyl auristatin E (vcMMAE) conjugate. In addition, because SS1P preclinical activity in highly overexpressing MSLNA431 xenografts did not translate to the clinic, we conducted comprehensive IHC of mesothelin expression levels in human tumors and xenografts to identify the most representative models for efficacy studies. We used these to demonstrate that a single dose of humanized anti-MSLN-vcMMAE could regress ovarian, pancreatic, and mesothelioma xenografts.

Materials and Methods Antibody production Two human mesothelin antigens were expressed in E. coli 58F3: unizyme His (HQ)-tagged MPF-MSLN propeptide (aa 34–580 of GenBank NM_005823) and HQ-mature MSLN extracellular domain (aa 296–580). Both were purified on a Ni-NTA column (Qiagen) followed by Superdex 200 gel filtration in 20 mmol/L MES pH 6.0, 6 mol/L Guanidine-HCl, and then dialyzed into 1 mmol/L HCl. Antibodies were produced as described (20), and screened by ELISA (21; see Supplementary), then by FACS on MSLN-transfected SV-T2 cells. Cell lines Cell lines originating from ATCC (acquisition year in brackets) included NCI-H226 (2009), NIH:OVCAR3 (2006), Caov-3 (2009), HPAC (2010), MSTO-211H (2010), PC3 (2006), 293 (2007), and SV-T2 (before 2002). All except SV-T2 were obtained through in-house repositories, which authenticated the lines by short tandem repeat profiling before banking. MSTO-211H was instead completely sequenced by next-generation sequencing. No further authentication by our laboratory was performed, but only

www.aacrjournals.org

SV-T2 were not authenticated and cells were not passaged beyond 2 months. gD-MSLN- PC3, SV-T2, HT1080, and 293 cells were generated by stable transfection with herpes simplex virus glycoprotein D (gD)–tagged MSLN or gDMPF-MSLN (GenBank NM_005823), using the gD signal sequence in a puromycin-resistant pCMV vector. For efficacy studies, NIH:OVCAR3 and NCI-H226 cells were adapted for in vivo growth by dissociating the best growing tumor (in the SCID beige mammary fat pad and subcutaneously in athymic nude mice, respectively) and cultivating the cells in vitro. These x1 cells were retransplanted into mice, and resultant tumor growth was improved but variable, so the entire procedure was repeated (except orthotopically in the lung of NCr nudes for H226-x1) to generate OVCAR3-x2.1 and H226-x2, which grew reproducible tumors. Humanization and conjugation 7D9 hybridoma was chimerized onto a human IgG1 backbone as reported (22), and then conjugated to mcMMAF (3.8 drugs/Ab), mc-vc-PAB-MMAE (3.8/Ab), and mc-vc-PAB-MMAF (3.9/Ab) via cysteines of the interchain disulfide bonds at Seattle Genetics as described (23, 24). Negative control ADCs were antiragweed (aRW) with 2.8, 3.2, and 3.0 drugs/Ab, respectively. See Supplementary Methods for h7D9.v3 humanization. h7D9.v3 was protein A-purified from Chinese Hamster Ovary cells as described (22) and conjugated to an average of 3.5 vcPAB-MMAE per antibody at Genentech. The isotype control was humanized anti-gD-tag 5B6 (Genentech; averaging 3.3 MMAE/Ab), which doubled as a positive control on gD-MSLN cells. Internalization Internalization was performed as described (20) except using 1 mg/mL antimesothelin antibodies and 0.4 mg/mL Alexa488-EGF (Molecular Probes). FACS FACS was as published (25). qRT-PCR qRT-PCR was as reported (26), using human mesothelin probe set Hs00245879_m1, spanning exons 16 to 17 (Applied Biosystems), and RPL19 as the reference gene. Western blotting Cells were lysed [in 1% NP-40, 150 mmol/L NaCl, 50 mmol/L Tris pH 8.0, 1 mmol/L EDTA, 1 mmol/L phenylmethylsulfonyl fluoride (PMSF), 1 mmol/L sodium fluoride, and 1 Complete Protease Inhibitor tablet (Roche)]. Lysates of flash-frozen xenografts were prepared using Tissue Lyzer (Qiagen). Thirty micrograms of lysates (measured by BCA; Pierce) were separated by SDS-PAGE, transferred to nitrocellulose, and blocked in 5% w/v nonfat milk, 0.1% Tween-20 in PBS. Primary antibodies were used at 1 to 10 mg/mL overnight, detected



2631


Scales et al.

with horseradish peroxidase–conjugated secondaries and ECL (enhanced chemiluminescence) or ECL prime (all from Amersham), and exposed to BioMaxMR film (Kodak). Films were scanned on an HP ScanJet 8200 and figures made with Adobe Photoshop (CS5.1).

IACUC guidelines, animals were euthanized before tumor volumes reached 3,000 mm3 or ulcerated, or body weight loss exceeded 20%. Primary OVXF1023 and PAXF1657 xenograft studies were conducted at Oncotest GmbH; refs. 30, 31).

Scatchard analysis Scatchard analysis was as described (27) except 125Ih7D9.v3 was incubated for 2 hours at room temperature.

Results

IHC Formalin-fixed paraffin-embedded (FFPE) human tumor microarrays (Supplementary Table S1A) were stained with 10 mg/mL 19C3 or IgG2b isotype control MPC-11 (BD Pharmingen) at room temperature for 60 minutes on a Dako autostainer following TARGET retrieval (Dako). Detection used horse antimouse biotinylated secondary (Vector), ABC complex (ABC Elite Kit; Vector Laboratories) and DAB visualization (Pierce). To avoid mouse-on-mouse issues with xenografts, rabbit monoclonal antihuman mesothelin SP74 (Spring Bioscience; M3744) or isotype control DA1E (Cell Signaling Technology; 3900S) were used at 1 mg/mL for 16 minutes at 37 C on the Ventana platform, with CC1 digestion, Omnimap detection, and DAB visualization. 19C3 and SP74 were first validated for IHC (Supplementary Figs. S1 and S2). Mesothelin staining was scored by a Board-certified pathologist, noting the distribution and intensity of staining, with a 10% cutoff as follows: 0 (negative) indicates very weak or no staining in >90% of tumor cells; 1þ indicates predominantly weak staining (i.e., >50% of the stained cells are weak); 2þ, predominantly moderate; and 3þ, predominantly strong. This differs from the published 5B2 (Novocastra) scoring system, which typically uses a 1% cutoff and scores percentages of cells only, ignoring intensity. In vitro efficacy In vitro efficacy was assayed with CellTiterGlo II (Promega) as described (27), except with 2,000 gD-MSLNHT1080 cells per well. In vivo efficacy All animal experiments were conducted according to the Guide for the Care and Use of Laboratory Animals (28). Cells or tumor fragments were implanted subcutaneously into female mice as summarized in Supplementary Table S2, except OVCAR3 were in the mammary fat pad (27). Mice were randomized into groups of 8 to 11 with tumors averaging approximately 150 mm3. Tumors were measured with UltraCal-IV calipers and volumes calculated as 0.5 (length width2). Growth curves were fitted by linear mixed effect modeling and analyzed using the nlme package of R software (R Foundation for Statistical Computing; ref. 29). Percentage tumor growth inhibition was calculated relative to the vehicle on day 21 or day 22 after dosing. As per

2632


Antibody, linker, and drug selection We began by generating novel antibodies to mesothelin. In contrast with published antimesothelin antibodies (10, 32–35), our mouse monoclonals were raised to unfolded, nonglycosylated mesothelin extracellular domain (with and without MPF). Hybridoma supernatants were screened by ELISA, followed by FACS and IHC analysis of cells stably transfected with mesothelin to select those capable of detecting native glycosylated and formalin-fixed mature mesothelin, respectively (data not shown). The two highest affinity antibodies, 19C3 (0.06 nmol/L by surface plasmon resonance analysis) and 7D9 (0.23 nmol/L), which bound noncompeting epitopes, were further characterized. Both were specific by FACS and Western blotting, only recognizing transfected and endogenously mesothelin-expressing cells, but 19C3 was more sensitive on low expressers (Fig. 1A–C), possibly reflecting its higher affinity. However, because internalizing ADCs are generally more efficacious due to greater drug release from lysosomes (12, 13, 36), we selected 7D9 for conjugation because it internalized to lysosomes to a greater extent in H226, OVCAR3, and MSLN-PC3 cells (Fig. 1D and data not shown), all known to express mesothelin (34, 37). Even so, internalization was relatively slow, with little uptake before 6 hours and significant surface signal remaining after overnight incubation. Chimeric 7D9 antibody was conjugated via the protease-sensitive linker maleimidocaproyl-valine-citrulline-p-aminobenzyloxycarbonyl (mc-vc-PAB) to the auristatin MMAE and the relatively impermeable MMAF, and also to uncleavable maleimidocaproyl-MMAF (mc-MMAF; refs. 23, 24). Conjugation did not affect 7D9 binding, and all three conjugates, but not the naked antibody, exhibited antiproliferative activity in vitro against OVCAR3 cells (not shown). By IHC, monoclonal 19C3 specifically stained mesothelin-positive cells, including normal mesothelia (Supplementary Fig. S1), and detected mesothelin on OVCAR3 cells at 2þ by both IHC (Fig. 2A) and FACS (Fig. 1B, red). Serially transplanted OVCAR3 tumors exhibited stronger (IHC 3þ) apical mesothelin staining (Fig. 2B), similar to many ovarian tumors. In this model, a single 3 mg/kg dose of cleavable 7D9-mc-vc-PAB-MMAE was clearly more efficacious than both the MMAF conjugates (Fig. 2C and Supplementary Fig. S3A), with robust tumor regressions (138% inhibition) and mean tumor doubling delayed from 6.5 (vehicle) to more than 91 days (Supplementary Table S2). The relapsed tumors retained their original levels of mesothelin expression (Fig. 2B0 , D–G),





19C3

PC3 vector

OVCAR3

Clone 3

OVCAR3

7D9

98 1

10

100

1,000

10,000

64 50

71

36

8 3 5

7D9

22

10

100

1,000

10,000

Mean fluorescence intensity Ovarian cells

B 19C3

1

10

100

10

100

1,000

10,000

7D9

Alexa488-EGF

1

19C3

αMesothelin

D

# Events

7D9

1

1,000

10,000

Merge + DAPI

Figure 1. 7D9 has lower affinity but internalizes better than 19C3. FACS of MPF-MSLN–transfected PC3 stable cells (numbered subclones in different colors, A) or endogenously expressing ovarian cells (B) OVCAR5 (blue), OVCAR4 (green), Caov-3 (purple), and OVCAR3 (red), with PC3 vector (gray) as a negative control in all panels, using 1 mg/mL 19C3 (top) or 7D9 (bottom) detected with Alexa488-antimouse. C, Western blot (enhanced chemiluminescence detection) analysis showing specificity of 19C3 (left) and 7D9 (right) for mesothelin. Arrowhead, 75 kDa MPF-MSLN; arrow, mature 50 kDa mesothelin. D, H226 cells after 20 hours of cointernalization of 1 mg/mL 19C3 or 7D9 (top and red channel in merge) and the lysosomal pathway marker Alexa488-EGF (middle and green channel). 7D9 internalizes to lysosomes (arrows). Insets (50% magnification) show lack of isotype control uptake. Scale bars, 20 mm.

# Events

19C3

Clone 5

Clone 3

71 8 3 5

Clone 5

C

MSLN-PC3 cells

PC3 vector

A

Mean fluorescence intensity

suggesting target that downregulation did not account for recurrence and raising the possibility of re-treatment. We speculate that the lower efficacy of the uncleavable mc-MMAF conjugate (74% inhibition) may be due to slower diffusion than free MMAE/F into the cytoplasm of its lysosomal metabolite, cys-mc-MMAF (cys being the conjugated cysteine of the antibody that remains following its lysosomal degradation; ref. 24). The intermediate activity of the mc-vc-PAB-MMAF conjugate (110% inhibition) probably reflects more efficient diffusion of free MMAF (than cys-mc-MMAF) following val-cit cleavage and PAB self-immolation, but lack of a bystander effect (the ability to intoxicate neighboring cells; ref. 38) compared with free MMAE; MMAF has poor membrane permeability unless protonated in lowpH environments such as lysosomes (24). Humanized 7D9 (h7D9.v3) was, therefore, conjugated to mc-vc-PABMMAE, henceforth called aMSLN-MMAE, and characterized in detail.


In vitro characterization of aMSLN-MMAE ADC h7D9.v3 (hereafter named aMSLN) retained high affinity for mesothelin, ranging from 0.4 to 3 nmol/L as measured by Scatchard analysis of endogenous mesothelin expressed from 0.5þ on MSTO-211H to 2þ on HPAC and OVCAR3-x2.1 [in vivo growthadapted OVCAR3 cells (see Materials and Methods), which express equivalent mesothelin levels to the parental line, Fig. 2H] and 3þ on H226-x2 cells (Fig. 3A). MMAE conjugation neither inhibited antigen binding, nor introduced nonspecific binding, as was also the case for the agD control ADC (Fig. 3A–C). To verify cytotoxic activity, gD-MSLN-HT1080 cells (expressing 130,000 copies of mesothelin; Kd, 0.97 nmol/L), which internalize both aMSLN and agD ADCs (Scales and colleagues, unpublished data), were assessed for viability following 3 days’ incubation with these ADCs in vitro. aMSLN-MMAE exhibited more potent killing [average IC50 of 0.32 0.1 nmol/L (n ¼ 3 independent



2633


Scales et al.

Untreated

A

B

B¢

C

αRagweed-vc-MMAE

7D9-vc-MMAE

D

F

E

G

3,000 Vehicle

Fitted tumor volume (mm 3)

2,500

7D9-mc-MMAF 7D9-mc-vc-PAB-MMAF

2,000

7D9-mc-vc-PAB-MMAE αRW-mc-MMAF

1,500

αRW-mc-vc-PAB-MMAF αRW-mc-vc-PAB-MMAE

1,000 500 0 0

H

20

40 60 Days since dosing

80

100

100

Figure 2. 7D9 is more efficacious conjugated as MMAE than MMAF conjugates in an OVCAR3 xenograft model. Antimesothelin 19C3 IHC staining of OVCAR3 cells (A) and untreated (B, 10; B0 2.5), 7D9-vc-MMAE-treated relapsed (day 91; D and E, tumors from two representative animals), or aRagweed-vc-MMAE–treated (day 74; F and G) OVCAR3 tumors. Scale bars, 200 mm. C, growth rate [calculated by linear mixed effects (LME) modeling] of OVCAR3 tumor transplants in the mammary fat pad following one intravenous dose of 3 mg/kg chimeric 7D9 (black) or negative control antiragweed (aRW, gray) conjugates. Raw mean tumor volumes are shown in Supplementary Fig. S3A. H, FACS with 0.5 mg/mL h7D9.v3 showing ex vivo–derived OVCAR3-x2.1 (blue) express similar levels of mesothelin to parental OVCAR3 (red) cells by FACS. Dashed lines, Alexa488 antihuman alone; gray, cells alone.

% of max

80 60 40 20 0

11 10 100 1,000 10,000 Mean fluorescence intensity

experiments)] than the positive control agD-MMAE (IC50 2.2 0.9 nmol/L) despite similar drug loading (Fig. 3D), likely reflecting its greater binding capacity as their EC50s were similar (Fig. 3B). Cytotoxicity was specific as there was no detectable killing except at the highest concentration on untransfected HT1080 cells (with comparable sensitivity with free MMAE: IC50 0.17 0.01 nmol/L versus 0.13 0.02 nmol/L for MSLN-HT1080, Fig. 3E), nor with aTENB2-vc-MMAE, whose antigen is absent from these cells (Fig. 3D). As expected, on lower expressing (40,000 copies) OVCAR3x2.1 cells with similar MMAE sensitivity (0.2 nmol/L), the IC50 of aMSLN-MMAE was substantially higher (20.3 nmol/L, data not shown). Naked aMSLN had no

2634


cytotoxic activity up to 100 nmol/L on any cell line tested (Fig. 3E and data not shown). Mesothelin prevalence and intensity in human tumors It is important to demonstrate in vivo efficacy in models expressing mesothelin at levels representative of human tumors; however, typical levels are unknown as most of the antimesothelin IHC (5B2 antibody) literature scores only percentages (not intensities) of positive cells (summarized in Supplementary Table S1). We, therefore, scored mesothelin staining by 19C3 in ovarian, pancreatic, and mesothelioma tissue microarrays (TMA) using intensity thresholds of no staining (0), weak (1þ), moderate





A Mean fluorescence intensity

2,500

40 20

1,000

0

2,500 2,000

2,600 1.04 nmol/L

1,500 1,000 500

10

μg/mL

50

5

0.5

50

0.05

H226-x2

216,000 ± 34,400 3.24 ± 0.59 nmol/L

HT1080 αMSLN αMSLN-MMAE αgD αgD-MMAE

2,500 2,000 1,500 1,000 500

0

1

5

39,963 0.56 nmol/L

3,000

αMSLN αMSLN-MMAE αgD αgD-MMAE

0.1

OVCAR3-x2.1

C

gD-MSLN-HT1080

0.0001 0.001 0.01

0.5

50

5

HPAC



3,000

0.05

MSTO-211H

0.5

0.05

50

5

0.05

0

10 100 1,000 10,000 Mean fluorescence intensity

0.5

1

500

0 0.0001 0.001 0.01

100

μg/mL antibody

0.1

1

10

100

μg/mL antibody

E

D

120

100

100

% Viability

120

80 60

20 0 0.001

80 60 40

40

αMSLN-MMAE αgD-MMAE αTENB2-MMAE 0.01

0.1

(2þ), and strong (3þ; Fig. 4). We used a 10% cutoff (instead of the 1% cutoff used in the 5B2 literature), hypothesizing that tumors would need 10% positive cells for meaningful antitumor responses with our bystander-capable ADC, but the actual threshold and intensity remain to be determined. Of 86 unspecified ovarian adenocarcinomas 85% were mesothelin positive, with a membranous pattern and diffuse distribution. Staining intensity was mostly moderate (2þ, 35% of the total) or strong (3þ, 36%), with only 14% weak (1þ, Fig. 4A). In a separate TMA, mesothelin prevalence was 75%, including 98% (42 of 43) serous

αMSLN free MMAE

20

1

(nmol/L)


αgD-MMAE

60

MSLN/cell 136 ± 4 Kd 0.36 ± 0.03 nmol/L (nmol/L)

B

αgD

αMSLN-MMAE

80 % of max

1,500

αMSLN

100

2,000

% Viability

Figure 3. In vitro characterization of aMSLN-MMAE ADC. A, FACS binding of naked aMSLN (black bars) and aMSLN-MMAE (gray) incubated at 0.05–50 mg/mL with mesothelin 0.5þ MSTO-211H, 2þ HPAC, 2.5þ OVCAR3-x2.1, and 3þ H226-x2 cells (detected with Alexa647-antihuman). Negative control agD (white) and agDMMAE (striped) were only tested at 50 mg/mL, showing minimal background binding. Inset, FACS histograms of 0.5 mg/mL naked aMSLN (cells in same order); gray is secondary alone on MSTO211H. Scatchard results for mesothelin copy number and affinity are shown below (mean of two experiments or mean of three SD). B, gD-MSLN-HT1080 stable cells were incubated as in A with naked (closed symbols) or MMAE-conjugated (open symbols, dashed lines) aMSLN (black triangles), or agD control (grey squares) at concentrations spanning 5 logs. Mean fluorescence intensity versus concentration plots reveal sigmoidal dose-dependent binding with an EC50 of 0.46 0.11 mg/mL for aMSLN-MMAE binding versus 0.5 0.17 for naked h7D9.v3 aMSLN; and 0.54 0.13 for agDMMAE versus 0.24 0.04 for agD. C, there was no detectable binding to untransfected HT1080 cells. D, cell viability of gD-MSLN-HT1080 (closed symbols) or untransfected HT1080 (open symbols and dashed lines) with aMSLN-MMAE (triangles), agD-MMAE (gray squares), or aTenB2-MMAE (circles), normalized to untreated cells. E, as D, except with naked aMSLN (triangles) and free MMAE (diamonds). The means and SDs of three independent duplicate experiments are shown in A–E.

10

100

0 0.001

0.01

0.1

1

10

100

(nmol/L)

adenocarcinomas, 23% of which were 1þ, 44% 2þ, and 33% 3þ. Twenty-six of 33 (79%) endometrioid, 5 of 13 clear cell, and 1 of 10 mucinous ovarian tumors were also positive. These prevalence data are consistent with the 5B2 literature (Supplementary Table S1B) and further validate 19C3 for IHC. In contrast, only 75% of 196 pancreatic ductal adenocarcinomas were mesothelin positive. The staining was usually weaker than in ovarian tumors (36% were 1þ, 27% 2þ, and only 11% 3þ) and often restricted to the apical membrane of the ducts (Fig. 4B). This is lower than the reported 95% prevalence (Supplementary Table S1C),



2635


Scales et al.

0

A

14%

2+ 35%

3+ 36%

Ovarian

15%

1+

25%

36%

27%

11%

55%

10%

14%

21%

Pancreatic

B

Mesothelioma

C

Figure 4. Representative mesothelin expression levels in human tumors by IHC. Typical antimesothelin 19C3 signal intensities on 86 ovarian adenocarcinomas (A), 196 pancreatic ductal adenocarcinomas (B), and 29 mesotheliomas (C). The percentage of tumors with each IHC score is indicated (0þ is negative or less than 10% cells stained; 1þ is predominantly weak; 2þ is moderate; and 3þ is strong staining). Scale bars, 50 mm.

likely because staining is focal in up to 20% cases (39, 40) and would have missed our 10% cutoff and/or not been captured on the TMA. Unexpectedly, only 45% of 29 mesotheliomas (4 of 12 pleural, 6 of 13 peritoneal and 3 of 4 pericardial) were positive, far fewer than the 86% for pleural (95% for epithelioid pleural) in the literature (see Supplementary Table S1D and S1E). Staining was mostly diffuse and membranous, as reported, and was of variable intensity: 10% 1þ, 14% 2þ, and 21% 3þ (Fig. 4C). Our lower prevalence is probably not due to lower 19C3 sensitivity, because in a head-to-head comparison 19C3 detected more ovarian samples than 5B2 (not shown). It is possible that this mesothelioma TMA was simply not representative, as histologic subtypes were not specified and sarcomatous mesotheliomas usually lack mesothelin. These data suggested mesothelin 2-3þ ovarian, 1-2þ pancreatic, and 3þ mesothelioma efficacy models would be the most representative of human tumors. Identification of suitable efficacy models To identify such xenograft models, we screened cell lines for mesothelin expression by FACS and selected ovarian OVCAR3-x2.1 (2þ), pancreatic HPAC (2þ), and mesothelioma H226-x2 (3þ). Mesothelioma MSTO-211H (0.5þ, with only 136 copies per cell) was additionally chosen as a low-expressing comparator (Fig. 3A). IHC scores for cell pellets were all consistent with FACS, but unexpectedly, all cell lines tested substantially upregu-

2636


lated mesothelin in vivo: the 2þ OVCAR3-x2.1 cells (Fig. 2H) increased approximately 6.5-fold at both mRNA and protein levels to 3þ expressers in vivo (Fig. 5A–D; as did their parental OVCAR3 cells, Fig. 2A, B and H). Unfortunately, four pancreatic cell lines with 2þ mesothelin by FACS/IHC also upregulated to 3þ in vivo, including HPAC cells (9, Fig. 5E–G). We were unable to identify a suitable 1þ cell line that remained 2þ in vivo, thus selected 2þ primary pancreatic PAXF1657 transplants as the most representative pancreatic model (Fig. 5H). Two mesothelioma models similarly increased mesothelin expression in xenografts: H226-x2 (Fig. 5I–K) and MSTO-211H (from 0.5 to 1þ, Fig. 5L–O). Ovarian OVXF1023 (3þ) primary human tumors were identified by IHC (Fig. 5P). aMSLN-MMAE is highly efficacious in multiple xenograft models To determine the preclinical activity of humanized aMSLN-MMAE ADC, we measured the antitumor effects of single doses from 1 to 20 mg/kg in the identified models compared with the nontargeted agD-MMAE ADC. Efficacy parameters for the 10 mg/kg dose used in all models are shown in Supplementary Table S2. The mesothelin 3þ ovarian OVCAR3-x2.1 model was the most sensitive to aMSLN-MMAE, with an approximated IC50 for tumor growth inhibition (TGI) of 2.7 mg/kg and almost complete regressions (108% TGI) at 10 mg/kg, which increased the mean tumor-doubling time from





A

OVCAR3-x2.1

E

I

HPAC

20

H226-x2

L

12

12

8

8

4

4

MSTO-211H

16 8

12 8

4

4 n.a.

Relative mesothelin

12

0

0

1°Xenografts or isotype

SP74 staining Xenografts Cells

Cells

X1

X2/3 Ave X

Cells X1 X2

X3 Ave X

0

Cells

X1

0

Cells

B 2+

F 2+

J 3+

M 0.5+

C 3+

G 3+

K 3+

N 1+

D0

H 2+

PAXF1657

P 3+

OVXF1023

X1

X2

Ave X

O0

50 μm

Figure 5. Efficacy model selection and upregulation of mesothelin in vivo. Top row, expression level of mesothelin in OVCAR3-x2.1 (A), HPAC (E), H226x2 (I), and MST0-211H (L) by qRT-PCR (black, normalized to RPL19) and Western blotting (gray; Supplementary Fig. S4) of the same samples with rabbit monoclonal antimesothelin SP74 in cells and xenograft tumors (X1, X2, and X3), with cells normalized to 1. Error bars, SD of triplicate PCR reactions for individual samples or SD of the mean of all xenograft samples for "Ave X." n.a., insufficient tumor for Western blotting. Second row, SP74 IHC of cell pellets of OVCAR3-x2.1 (B), HPAC (F), H226-x2 (J), and MST0-211H (M) showing IHC scores consistent with the respective FACS shifts (Fig. 3A). Third row, SP74 IHC of xenografts of OVCAR3-x2.1 (C), HPAC (G), H226-x2 (K), and MST0-211H (N), confirming upregulation of mesothelin compared with parental cell lines. Bottom row, adjacent section isotype control IHC of OVCAR3-x2.1 (D) and MSTO-211H (O; rest in Supplementary Fig. S2) showing no signal; or SP74 IHC of primary xenografts PAXF1657 (H, 3/4 were 2þ) and OVXF1023 (P, 7/7 were 3þ). Scale bar, 50 mm (20, all panels).

4.5 (vehicle) to 83 days (Fig. 6A and Supplementary Fig. S3B). Activity was specific because agD-MMAE had little activity at this dose, and naked aMSLN had no activity at 15 mg/kg in any of the models tested, indicating the importance of conjugating MMAE to a mesothelin-targeting antibody. Despite similar mesothelin levels, the primary ovarian model OVXF1023 was less sensitive, with 10 mg/kg ADC achieving only 53% TGI after 3 weeks, necessitating a second dose that achieved 78% TGI by 84 days and slowed tumor doubling from 8 to 45 days (Fig. 6B and Supplementary Fig. S3C). aMSLN-MMAE was also active in pancreatic models: HPAC xenografts strongly regressed at 10 to 15 mg/kg (Fig. 6C) with an IC50 of 3.4 mg/kg and a tumor-doubling delay to 74 days. The 2þ primary pancreatic PAXF1657 model was more sensitive than OVXF1023, with tumor shrinkage (104% TGI) at 10 mg/kg and doubling delayed to 37 days. Although 20 mg/kg was more effective (107%


TGI), with 9 of 10 complete responses and doubling delayed beyond 81 days (Fig. 6D and Supplementary Fig. S3E), significant nonspecific activity occurred at this dose. Similar efficacy was observed in the higher (IHC 3þ) expressing PAXF736 primary pancreatic model (data not shown). In addition, aMSLN-MMAE inhibited the growth of the 3þ mesothelioma H226-x2 model, albeit only resulting in stasis (102% TGI) for 3 weeks and delaying tumor doubling to 34 days at 10 mg/kg (Fig. 6E). Although aMSLN-MMAE also caused stasis (96% TGI) in the low (1þ) mesothelin-expressing MSTO-211H xenografts, the duration of the response was shorter and the high permeability of these tumors resulted in some activity with agD-MMAE as well (Fig. 6F and data not shown), suggesting that mesothelin targeting played less of a role in this model. Pharmacokinetic analysis confirmed similar exposure in all models (data not shown).



2637


Scales et al.

A 2,500

B

OVCAR3-x2.1

OVXF1023

1,500

αMSLN-MMAE: αgD-MMAE

1,000

1,500

5 10 15 20

1.0 2.5 5.0 10 15 20 15 naked Vehicle

2,000

1,000 500 500 0 0

Fitted tumor volume (mm3)

C 2,500

20

30

40

50

0

60

D

HPAC

0

3,000

10

20

30

40

50

60

30

40

50

60

PAXF1657

2,500

2,000

2,000 1,500 1,500 1,000 1,000 500 0

E

10

500

0

1,500

10

20

30

40

50

0

60

F

H226-x2

0

10

20

MSTO-211H

2,000

1,500 1,000 1,000 500 500

0 0

10

20

30

40

50

0

0

10

20

30

40

Figure 6. aMSLN-MMAE is highly efficacious in multiple xenografts of each indication. Xenografts 3 (150 mm , Supplementary Table S2) of ovarian OVCAR3-x2.1 (A), primary ovarian papillary serous adenomatous carcinoma OVXF1023 (B), HPAC (C), primary pancreatic ductal adenocarcinoma PAXF1657 (D), and mesotheliomas H226-x2 (E), and MSTO-211H (F) with 8 to 11 mice per group were dosed on day 0 (and day 21 for OVXF1023 only, arrow). Tumor growth curves were fitted using LME modeling. All models were dosed with vehicle (dashed line), aMSLN-MMAE (aMSLN, black), and control agD-MMAE (agD, gray) at 5 mg/kg (circles) and 10 mg/kg (triangles, dashed line), and some models also received 1 mg/kg (crosses), 2.5 mg/kg (open diamonds, dashed line), 15 mg/kg (squares), and/or 20 mg/kg (inverted triangles); see B for legend. For clarity, inefficacious 1 and 2.5 mg/kg groups are omitted in F. Naked aMSLN (hatched squares) had no effect at 15 mg/kg (A, C, and F). Graphs in A to D are truncated at day 60 to better reveal the effects of lower doses. Later time points are shown in Supplementary Fig. S3 as mean tumor volumes (SEM), with efficacy measures in Supplementary Table S2.

50

Days since dosing

To summarize, potent dose-dependent antitumor activity by a single systemic injection of aMSLN-MMAE was achieved in each of the three indications.

Discussion We have demonstrated strong and sustained antitumor activity by aMSLN-MMAE ADC in one mesothelioma, two ovarian, and two pancreatic xenograft models, including two primary tumor explants. In these five models, significant activity was seen at 5 mg/kg and tumor regressions at 10 mg/kg following a single dose. Similar results were obtained in OVCAR3, PAXF736, and H226 models with 10 mg/kg antimesothelin–DM4 conjugate (17), except with 3 to 6 repeated doses. Importantly, the

2638


models in which aMSLN-MMAE was highly active expressed clinically relevant levels of mesothelin (2-3þ by tumor IHC). Modest activity was seen in the 1þ mesothelioma MSTO-211H model. Although further work is required to determine the minimum mesothelin expression level required for activity, the differential responses of the 3þ models indicate that other factors such as internalization rate, MMAE-sensitivity, tumor accessibility, and growth rate likely also play a role. We took pains to undertake systematic scoring of mesothelin expression intensities in clinical samples to identify relevant xenografts, with the goal of achieving better predictive power for clinical activity than overexpressing models. Notably, the A431-K5 cells used to demonstrate SS1P preclinical activity express 2.5 million





copies of mesothelin per cell (41), 10-fold more than the highest endogenous line (H226-x2) and likely more than the patients who failed to respond to SS1P (7). In addition, aMSLN-MMAE internalizes consistently better in transfected than endogenous lines, even with matching expression levels (data not shown), supporting the superiority of endogenous models for assessing ADC activity. Nonetheless, in all lines, aMSLN accumulation in lysosomes was far slower than most other ADC targets (25, 36, 42), likely explaining the higher activity with cleavable vcMMAE than mc- or vc-MMAF, and also the lower (by an order of magnitude) in vitro potency on OVCAR3 cells (20 nmol/L) compared with other vcMMAE conjugates to similarly expressed endogenous targets in other adherent cell lines. Our efficacy data in OVCAR3 cells and xenografts are similar to those of aMuc16-auristatin conjugates, which also internalize slowly (27, 43). Unexpectedly, endogenous (but not exogenous) mesothelin was upregulated up to 10-fold in vivo by an unknown mechanism in all the lines tested (including others not shown). This enabled robust ADC activity even in HPAC cells with only 2,600 copies of mesothelin, suggesting many other tumorigenic cell lines could be potential efficacy models. It is unclear if upregulation is mediated by the CanScript promoter, as this is reportedly less active in OVCAR3 and MSTO-211H cells (44), which upregulated as much as the others. Cell–cell interactions might play a role, as we observed higher mesothelin in confluent cells (which may mimic the in vivo situation), but are likely independent of mesothelin-Muc16 binding (45) because Muc16 is not expressed in MSTO-211H. This would be consistent with the converse finding that mesothelin is lost from patient-derived mesotheliomas cultured in vitro (46). Overall, aMSLN-MMAE is highly active at inhibiting or even shrinking tumor growth in mesothelin-positive mouse models. It is also well tolerated, and did not exhibit pleuritis in non-human primates (Scales and

colleagues, manuscript in preparation). Furthermore, in phase I trials (NCT01469793), it (DMOT4039A) is showing activity in both ovarian and pancreatic cancers (47), suggesting that our preclinical models do indeed have predictive value. Disclosure of Potential Conflicts of Interest No potential conflicts of interest were disclosed.

Authors' Contributions Conception and design: S.J. Scales, S. Ross, M.S. Dennis, S.D. Spencer Development of methodology: D.M. French, L. Rangell, Y. Zhang Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S.J. Scales, N. Gupta, G. Pacheco, R. Firestein, D.M. French, H. Koeppen, L. Rangell, V. Barry-Hamilton, E. Luis, J. Chuh, Y. Zhang, G.S. Ingle, A. Fourie-O’Donohue, K.R. Kozak, S. Ross Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S.J. Scales, N. Gupta, R. Firestein, D.M. French, H. Koeppen, V. Barry-Hamilton, E. Luis, J. Chuh, A. Fourie-O’Donohue, K.R. Kozak, S. Ross, M.S. Dennis Writing, review, and/or revision of the manuscript: S.J. Scales, L. Rangell, S. Ross, S.D. Spencer Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Rangell, A. Fourie-O’Donohue, K.R. Kozak Study supervision: S.J. Scales, M.S. Dennis Other (conducting animal research, and providing raw data and some graphed figures that may have been changed): G. Pacheco

Acknowledgments The authors thank Seattle Genetics for the ADC technology license and are indebted to Jennifer Batson, Suzy Clark, Eric Cheng, Kathryn ParsonsReponte, Lisa Crocker, Nika Polevaya, Randy Musterer, Gail Colbern, Klara Totpal, Allison Cordrey, and Inna Balter for xenograft development; Leanne Ross, Jenny Bostrom, and Oncotest GmbH for primary xenograft studies; Fred de Sauvage, Paul Polakis, Jay Tibbitts, Michel duPage, Brigitte deVaux, Yvonne Chen, Sreedevi Chalasani, Caroline MelotyKapella, Danielle Pasqualone, and Genentech core services for excellent support. 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. Received June 6, 2014; revised August 1, 2014; accepted September 5, 2014; published OnlineFirst September 23, 2014.

References 1. 2.

3. 4.

5.

6. 7.

Hassan R, Ho M. Mesothelin targeted cancer immunotherapy. Eur J Cancer 2008;44:46–53. Kojima T, Oh-eda M, Hattori K, Taniguchi Y, Tamura M, Ochi N, et al. Molecular cloning and expression of megakaryocyte potentiating factor cDNA. J Biol Chem 1995;270:21984–90. Bera TK, Pastan I. Mesothelin is not required for normal mouse development or reproduction. Mol Cell Biol 2000;20:2902–6. Le DT, Brockstedt DG, Nir-Paz R, Hampl J, Mathur S, Nemunaitis J, et al. A live-attenuated Listeria vaccine (ANZ-100) and a live-attenuated Listeria vaccine expressing mesothelin (CRS-207) for advanced cancers: phase I studies of safety and immune induction. Clin Cancer Res 2012;18:858–68. Beatty GL, Haas AR, Maus MV, Torigian DA, Soulen MC, Plesa G, et al. Mesothelin-specific chimeric antigen receptor mRNA-engineered T cells induce anti-tumor activity in solid malignancies. Cancer Immunol Res 2014;2:112–20. Pastan I. Immunotoxins containing Pseudomonas exotoxin A: a short history. Cancer Immunol Immunother 2003;52:338–41. Hassan R, Bullock S, Premkumar A, Kreitman RJ, Kindler H, Willingham MC, et al. Phase I study of SS1P, a recombinant anti-mesothe-


lin 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. 8. Hassan R, Miller AC, Sharon E, Thomas A, Reynolds JC, Ling A, et al. Major cancer regressions in mesothelioma after treatment with an antimesothelin immunotoxin and immune suppression. Sci Transl Med 2013;5:208ra147. 9. Niederfellner G, Bauss F, Imhof-Jung S, Hesse F, Kronenberg S, Staak R, et al. RG7787 - a novel de-immunized PE based fusion protein for therapy of mesothelin-positive solid tumors. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; 2014, (abstr 4510). 10. Hassan R, Ebel W, Routhier EL, Patel R, Kline JB, Zhang J, et al. Preclinical evaluation of MORAb-009, a chimeric antibody targeting tumor-associated mesothelin. Cancer Immun 2007;7:20. 11. Hassan R, Cohen SJ, Phillips M, Pastan I, Sharon E, Kelly RJ, et al. Phase I clinical trial of the chimeric anti-mesothelin monoclonal antibody MORAb-009 in patients with mesothelin-expressing cancers. Clin Cancer Res 2010;16:6132–8.



2639


Scales et al.

12. Kovtun YV, Goldmacher VS. Cell killing by antibody-drug conjugates. Cancer Lett 2007;255:232–40. 13. Carter PJ, Senter PD. Antibody-drug conjugates for cancer therapy. Cancer J 2008;14:154–69. 14. Erickson HK, Park PU, Widdison WC, Kovtun YV, Garrett LM, Hoffman K, et al. Antibody-maytansinoid conjugates are activated in targeted cancer cells by lysosomal degradation and linker-dependent intracellular processing. Cancer Res 2006;66:4426–33. 15. Senter PD, Sievers EL. The discovery and development of brentuximab vedotin for use in relapsed Hodgkin lymphoma and systemic anaplastic large cell lymphoma. Nat Biotechnol 2012;30:631–7. 16. Burris HA III, Tibbitts J, Holden SN, Sliwkowski MX, Lewis Phillips GD. Trastuzumab emtansine (T-DM1): a novel agent for targeting HER2þ breast cancer. Clin Breast Cancer 2011;11:275–82. 17. Golfier S, Kopitz C, Kahnert A, Heisler I, Schatz CA, Stelte-Ludwig B, et al. Anetumab ravtansine - a novel mesothelin-targeting antibodydrug conjugate cures tumors with heterogeneous target expression favored by bystander effect. Mol Cancer Ther 2014;13:1537–48. 18. Mita MM, Ricart AD, Mita AC, Patnaik A, Sarantopoulos J, Sankhala K, et al. A phase I study of a CanAg-targeted immunoconjugate, huC242DM4, in patients with Can Ag-expressing solid tumors. J Clin Oncol 2007;25:18s, (suppl; abstr 3062). 19. Younes A, Kim S, Romaguera J, Copeland A, Farial Sde C, Kwak LW, et al. Phase I multidose-escalation study of the anti-CD19 maytansinoid immunoconjugate SAR3419 administered by intravenous infusion every 3 weeks to patients with relapsed/refractory B-cell lymphoma. J Clin Oncol 2012;30:2776–82. 20. Shang Y, Mao Y, Batson J, Scales SJ, Phillips G, Lackner MR, et al. Antixenograft tumor activity of a humanized anti-insulin-like growth factor-I receptor monoclonal antibody is associated with decreased AKT activation and glucose uptake. Mol Cancer Ther 2008;7:2599–608. 21. Hongo JA, Tsai SP, Moffat B, Schroeder KA, Jung C, Chuntharapai A, et al. Characterization of novel neutralizing monoclonal antibodies specific to human neurturin. Hybridoma 2000;19:303–15. 22. Presta LG, Chen H, O'Connor SJ, Chisholm V, Meng YG, Krummen L, et al. Humanization of an anti-vascular endothelial growth factor monoclonal antibody for the therapy of solid tumors and other disorders. Cancer Res 1997;57:4593–9. 23. Doronina SO, Toki BE, Torgov MY, Mendelsohn BA, Cerveny CG, Chace DF, et al. Development of potent monoclonal antibody auristatin conjugates for cancer therapy. Nat Biotechnol 2003;21:778–84. 24. Doronina SO, Mendelsohn BA, Bovee TD, Cerveny CG, Alley SC, Meyer DL, et al. Enhanced activity of monomethylauristatin F through monoclonal antibody delivery: effects of linker technology on efficacy and toxicity. Bioconjug Chem 2006;17:114–24. 25. Asundi J, Reed C, Arca J, McCutcheon K, Ferrando R, Clark S, et al. An antibody-drug conjugate targeting the endothelin B receptor for the treatment of melanoma. Clin Cancer Res 2011;17:965–75. 26. Wen X, Lai CK, Evangelista M, Hongo JA, de Sauvage FJ, Scales SJ. Kinetics of hedgehog-dependent full-length Gli3 accumulation in primary cilia and subsequent degradation. Mol Cell Biol 2010;30: 1910–22. 27. Chen Y, Clark S, Wong T, Chen Y, Chen Y, Dennis MS, et al. Armed antibodies targeting the mucin repeats of the ovarian cancer antigen, MUC16, are highly efficacious in animal tumor models. Cancer Res 2007;67:4924–32. 28. Resources IoLA. Guide for the Care and Use of Laboratory Animals. Washington, DC: National Academy Press; 1996. 29. Pinheiro J, Bates D, DebRoy S, Sarkar D. The R Core team. nlme: Linear and Nonlinear Mixed Effects Models. R package version 2009;3:1–96. 30. Drees M, Dengler WM, Hendriks HR, Kelland LR, Fiebig HH. Cycloplatam: a novel platinum compound exhibiting a different spectrum of anti-tumour activity to cisplatin. Eur J Cancer 1995;31A:356–61. 31. Kuesters S, Maurer M, Burger AM, Metz T, Fiebig HH. Correlation of ErbB2 gene status, mRNA and protein expression in a panel of

2640


32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

>100 human tumor xenografts of different origin. Onkologie 2006; 29:249–56. Chowdhury PS, Viner JL, Beers R, Pastan I. Isolation of a highaffinity stable single-chain Fv specific for mesothelin from DNAimmunized mice by phage display and construction of a recombinant immunotoxin with anti-tumor activity. Proc Natl Acad Sci U S A 1998;95:669–74. Scholler N, Fu N, Yang Y, Ye Z, Goodman GE, Hellstrom KE, et al. Soluble member(s) of the mesothelin/megakaryocyte potentiating factor family are detectable in sera from patients with ovarian carcinoma. Proc Natl Acad Sci U S A 1999;96:11531–6. Chang K, Pastan I. Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesotheliomas, and ovarian cancers. Proc Natl Acad Sci U S A 1996;93:136–40. Onda M, Willingham M, Nagata S, Bera TK, Beers R, Ho M, et al. New monoclonal antibodies to mesothelin useful for immunohistochemistry, fluorescence-activated cell sorting, Western blotting, and ELISA. Clin Cancer Res 2005;11:5840–6. Polson AG, Calemine-Fenaux J, Chan P, Chang W, Christensen E, Clark S, et al. Antibody-drug conjugates for the treatment of nonHodgkin's lymphoma: target and linker-drug selection. Cancer Res 2009;69:2358–64. Ho M, Bera TK, Willingham MC, Onda M, Hassan R, FitzGerald D, et al. Mesothelin expression in human lung cancer. Clin Cancer Res 2007; 13:1571–5. Okeley NM, Miyamoto JB, Zhang X, Sanderson RJ, Benjamin DR, Sievers EL, et al. Intracellular activation of SGN-35, a potent anti-CD30 antibody-drug conjugate. Clin Cancer Res 2010;16:888–97. Argani P, Iacobuzio-Donahue C, Ryu B, Rosty C, Goggins M, Wilentz RE, et al. Mesothelin is overexpressed in the vast majority of ductal adenocarcinomas of the pancreas: identification of a new pancreatic cancer marker by serial analysis of gene expression (SAGE). Clin Cancer Res 2001;7:3862–8. Swierczynski SL, Maitra A, Abraham SC, Iacobuzio-Donahue CA, Ashfaq R, Cameron JL, et al. Analysis of novel tumor markers in pancreatic and biliary carcinomas using tissue microarrays. Hum Pathol 2004;35:357–66. Hassan R, Wu C, Brechbiel MW, Margulies I, Kreitman RJ, Pastan I. 111Indium-labeled monoclonal antibody K1: biodistribution study in nude mice bearing a human carcinoma xenograft expressing mesothelin. Int J Cancer 1999;80:559–63. Sutherland MS, Sanderson RJ, Gordon KA, Andreyka J, Cerveny CG, Yu C, et al. Lysosomal trafficking and cysteine protease metabolism confer target-specific cytotoxicity by peptide-linked anti-CD30-auristatin conjugates. J Biol Chem 2006;281:10540–7. Leipold D, Mallet WG. Case Study: An Antibody–Drug Conjugate Targeting MUC16 for Ovarian Cancer. Antibody-Drug Conjugates and Immunotoxins: From Pre-Clinical Development to Therapeutic Applications. Springer New York, New York; 2013. p. 221–39. Hucl T, Brody JR, Gallmeier E, Iacobuzio-Donahue CA, Farrance IK, Kern SE. High cancer-specific expression of mesothelin (MSLN) is attributable to an upstream enhancer containing a transcription enhancer factor dependent MCAT motif. Cancer Res 2007;67: 9055–65. Rump A, Morikawa Y, Tanaka M, Minami S, Umesaki N, Takeuchi M, et al. Binding of ovarian cancer antigen CA125/MUC16 to mesothelin mediates cell adhesion. J Biol Chem 2004;279:9190–8. Zhang J, Qiu S, Zhang Y, Merino M, Fetsch P, Avital I, et al. Loss of mesothelin expression by mesothelioma cells grown in vitro determines sensitivity to anti-mesothelin immunotoxin SS1P. Anticancer Res 2012;32:5151–8. Weekes CD, Lamberts LE, Borad MJ, Voortman J, McWilliams RR, Robinson Diamond J, et al. A phase I study of DMOT4039A, an antibody-drug conjugate (ADC) targeting mesothelin, in patients with unresectable pancreatic or platinum-resistant ovarian cancer. J Clin Oncol 2014;32:5s, (suppl; abstr 529).




An Antimesothelin-Monomethyl Auristatin E Conjugate with Potent Antitumor Activity in Ovarian, Pancreatic, and Mesothelioma Models Suzie J. Scales, Nidhi Gupta, Glenn Pacheco, et al. Mol Cancer Ther 2014;13:2630-2640. Published OnlineFirst September 23, 2014.

Updated version Supplementary Material

Cited articles Citing articles

E-mail alerts Reprints and Subscriptions Permissions

Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-14-0487-T Access the most recent supplemental material at: http://mct.aacrjournals.org/content/suppl/2014/09/24/1535-7163.MCT-14-0487-T.DC1.html

This article cites 43 articles, 27 of which you can access for free at: http://mct.aacrjournals.org/content/13/11/2630.full.html#ref-list-1 This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/13/11/2630.full.html#related-urls

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].


AMG 595, an Anti-EGFRvIII Antibody-Drug Conjugate, Induces Potent Antitumor Activity against EGFRvIII-Expressing Glioblastoma.

A chemically defined trifunctional antibody-cytokine-drug conjugate with potent antitumor activity.

Ibrutinib exerts potent antifibrotic and antitumor activities in mouse models of pancreatic adenocarcinoma.

A Conjugate Based on Anti-HER2 Diaffibody and Auristatin E Targets HER2-Positive Cancer Cells.

Selective ALK inhibitor alectinib with potent antitumor activity in models of crizotinib resistance.

A Novel Time-Dependent CENP-E Inhibitor with Potent Antitumor Activity.

Development and Properties of Valine-Alanine based Antibody-Drug Conjugates with Monomethyl Auristatin E as the Potent Payload.

Comparison of a mouse and a novel human scFv-SNAP-auristatin F drug conjugate with potent activity against EGFR-overexpressing human solid tumor cells.

Monoclonal Antibodies Targeting LecLex-Related Glycans with Potent Antitumor Activity.

Generation of a novel Antibody-Drug Conjugate targeting endosialin: potent and durable antitumor response in sarcoma.

Structural Basis of Microtubule Destabilization by Potent Auristatin Anti-Mitotics.

Postsurgical adjuvant or metastatic renal cell carcinoma therapy models reveal potent antitumor activity of metronomic oral topotecan with pazopanib.

Antitumor activity and clinical pharmacology of sulofenur in ovarian cancer.

The Aminosteroid Derivative RM-133 Shows In Vitro and In Vivo Antitumor Activity in Human Ovarian and Pancreatic Cancers.

6 inhibitors have potent activity in combination with pathway selective therapeutic agents in models of pancreatic cancer.

CXCR2 Inhibition Combined with Sorafenib Improved Antitumor and Antiangiogenic Response in Preclinical Models of Ovarian Cancer.

A priori prediction of tumor payload concentrations: preclinical case study with an auristatin-based anti-5T4 antibody-drug conjugate.

An antimicrobial peptide containing NGR motif has potent antitumor activity against CD13+ and CD13- tumor cells.

Monomethyl Auristatin E Phosphate Inhibits Human Prostate Cancer Growth.

Site-Specific Conjugation of Monomethyl Auristatin E to Anti-CD30 Antibodies Improves Their Pharmacokinetics and Therapeutic Index in Rodent Models.

Humanized anti-VEGFR-2 ScFv-As2O3-stealth nanoparticles, an antibody conjugate with potent and selective anti-hepatocellular carcinoma activity.

Antitumor effect of antitissue factor antibody-MMAE conjugate in human pancreatic tumor xenografts.

Novel quinonyl derivatives of muramyl dipeptide possessing potent antitumor activity.

LY2109761 enhances cisplatin antitumor activity in ovarian cancer cells.