Therapeutic Mechanism and Efficacy of the Antibody–Drug Conjugate BAY 79-4620 Targeting Human Carbonic Anhydrase 9

Carbonic anhydrase IX (CAIX) was first identified by Pastorekova as an endogenous HeLa cell antigen that was recognized by the antibody M75 raised against this cervix carcinoma cell line (1). Although initially referred to as MN antigen, it was shortly thereafter identified as a new carbonic anhydrase by the same group (2) and was confirmed to be identical to the tumor marker G250 (named after the antibody with which it was identified) published by Oosterwijk and colleagues (3).

CAIX is a cell surface glycoprotein that is expressed in carcinomas of several histologic types, including a strikingly high proportion of renal cell carcinomas (4, 5), carcinomas of the esophagus (4, 6), cervical carcinomas (4, 7), malignant colon carcinomas (4, 8), non–small cell lung carcinomas (NSCLC; refs. 4, 9), and, to a lesser degree, breast carcinomas (4, 10). By contrast, the expression of CAIX on normal tissues is largely restricted to the apical surface of cells of the stomach, bile duct mucosa (4, 11), and small intestine (4, 12). The extracellular domain of this type I transmembrane protein comprises both a proteoglycan domain implicated in cell adhesion through homotypic interaction (13) and the carbonic anhydrase domain that catalyzes the reversible hydration of carbon dioxide to bicarbonate and protons (14) and is involved in the regulation of the pH within the tumor environment (15).

CAIX gene expression is under the direct control of the transcription factor hypoxia-inducible factor-1 and is significantly upregulated by tumor hypoxia (16). CAIX expression was found to correlate with (i) a high mean vessel density, cancer stage, and degree of necrosis in head and neck carcinoma (17); (ii) poor survival in nasopharyngeal carcinoma (18); (iii) tumor grade, negative estrogen receptor status, higher relapse rate, and poor survival for invasive breast carcinoma (10). This association with tumor grade and overall survival, together with its cell surface distribution and restricted expression in normal tissues, implicates CAIX as an important therapeutic target for monoclonal antibody (mAb)-based therapy. In this article, we report the identification of a potent and selective internalizing human antibody directed against CAIX that, when conjugated to monomethylauristatin E (MMAE), yielded a highly efficacious antibody–drug conjugate BAY 79-4620, with activity against a variety of solid tumor types.

The hybridoma producing CAIX mAb M75 (1, 13) was obtained from American Type Culture Collection (ATCC) and used to express and purify mAb M75 using standard protocols.

The HuCAL Gold Fab-phage library was obtained from MorphoSys AG and was prepared as described elsewhere (19), comprising a highly diverse library of 10¹⁰ different monovalent phage encoded within phagemid vector pMORPH23 and allowing for monovalent CysDisplay of Fab fragments. Solid phase panning was carried out as described in the Supplementary Data.

Soluble Fab fragments were produced from the isolated phage clones as described in the Supplementary Data.

Isolated Fabs were tested for binding to the purified ectodomain of CAIX in an ELISA, positive Fabs were recloned in the IgG format and expressed in Chinese hamster ovary cells (see Supplementary Data). Anti-CAIX IgG antibodies were purified using protein A sepharose.

Surface plasmon resonance (SPR) was carried out on a BIAcore 3000 instrument (BIAcore; see Supplementary Data).

The complexes formed between the antibodies and biotin-labeled cell proteins were immunoprecipitated and visualized by immunoblots developed with enzyme-linked streptavidin using an adaptation of methods described elsewhere (20).

Adherent CAIX-expressing PC-3 mm2 cells, non-CAIX expressing MiaPaCa-2 cells, and MiaPaCa-2 cells transfected with full-length CAIX were detached from their flasks with 1:10 trypsin/Versene in PBS solution for 5 to 10 minutes. Cells were spun down (1,000 rpm, 5 minutes), washed once with ice-cold RPMI 10% FBS, and resuspended in ice-cold staining buffer [Ca⁺ Mg⁺-free PBS, 2% bovine serum albumin (BSA), and 0.05% sodium azide] to achieve 6 × 10⁶ cells/mL. Anti-CAIX IgG1 or control human IgG1 antibodies at 25 μg/mL were incubated with 6 × 10⁵ cells on ice for 1 hour. Unbound antibody was removed by washing the cells with the ice-cold staining buffer. The cells were fixed with 2% formaldehyde in PBS for 10 minutes, then washed twice with staining buffer. The cell pellet was resuspended in 100 μL ice-cold staining buffer containing a 1:200 dilution of Alexa Fluor 488–labeled secondary antibody (Molecular Probes/Invitrogen) and incubated on ice for 1 hour. The unbound antibody was washed from the cells 2 times with flow buffer (PBS containing 2% BSA), and the cells were resuspended in 1 mL flow buffer. Fluorescence-activated cell sorting (FACS) analysis of the resuspended cells was done on a Beckman FACS Calibur instrument. All cell lines were obtained from ATCC; MiaPaCa-2-CAIX was generated using standard transfection methods. Cell lines are regularly checked for authenticity by DNA fingerprinting at the DSMZ (German Collection of Microorganisms and Cell Cultures), Braunschweig, Germany.

Antibody internalization was assessed using a Cellomics Array Scan automated confocal microscope system (see Supplementary Data).

mAb conjugation with MMAE was done essentially as described elsewhere for conjugation with an anti-CD30 mAb (21).

Antigen-positive and antigen-negative cells were plated at 5,000 cells per well in 96-well plates in 100 μL media per well overnight at 37°C, in 5% CO₂ to adhere. The media was replaced with fresh media containing antibody or antibody-vcMMAE conjugate and the plates further incubated at 37°C, in 5% CO₂ for 72 hours. Cytotoxicity was assessed using the Alamar Blue cell viability assay. Alamar Blue was added at a final concentration of 10% for the last 4 hours of the incubation, and its transformation to fluorescent product by viable cells determined spectrofluorometrically by plate reader (544 nm excitation–590 nm emission).

Noninvasive in vivo imaging studies were conducted using the CRI Maestro in vivo imaging system with multispectral acquisition and analysis (CRI; see Supplementary Data).

All animal experiments were conducted in accordance with either the United Kingdom Coordinating Committee on Cancer Research regulations for the welfare of animals, the German animal welfare law, or the Institutional Animal Care and Use Committee, in addition to approval by local authorities. For xenograft experiments based on human tumor cell lines, immunocompromised mice (8–12 weeks of age) were subcutaneously implanted on day 0 with the respective tumor cells in 0.1 mL volume as detailed in Supplementary Table ST2. For xenograft experiments based on patient-derived NSCLC tumors, tumor fragments were subcutaneously passaged on naive NMRI nu/nu mice (Lu7506, Lu7298, Lu7913, Lu7406, Lu7747; refs. 22, 23). Treatment was initiated when tumors reached an average size of 60 to 160 mm³ depending on the model used (Supplementary Table ST2). Dosing of test compounds was carried out according to the dosing levels, schedules, and routes of administration described in Supplementary Table ST2. Carboplatin was obtained from Hexal, taxol (Paclitaxel) from Bristol Myers Squibb, cisplatin from Sigma Aldrich, and gemcitabine from Eli Lilly. A dose volume of 0.1 mL/10 g body weight was used throughout and intravenous administration was by tail vein injection. The health status of animals was monitored daily. The length and width of each tumor was measured by electronic calipers 2 to 3 times per week and tumor volumes (mm³) were calculated as [length (mm) × width (mm)²]/2. Percentage increases in tumor size during the study were calculated by the formula [100 × tumor volume at treatment end/tumor volume at treatment start] − 100 for every single animal. Final tumor volume was defined by the last time point in which the vehicle treated control remained within the experiment. Antitumor efficacy was also assessed from the incidence of regression in which a tumor regression is defined as a reduction in tumor size of more than 30% at study end relative to the initial size.

In pharmacodynamic experiments, tumors were collected at 4 hours, 1, 3, and 5 days following single administration of BAY 79-4620, formalin-fixed, paraffin-embedded (FFPE), sectioned at 5 μm, deparaffinized, and stained using standard protocol for fluorescent immunohistochemistry (IHC) of human tissues using mouse antibodies against -α/β-tubulin, phospho-Histone H3, and DNA. Slides were observed by fluorescent microscopy.

In parallel with efficacy studies, a group of animals carrying each tumor type was sacrificed, the tumors were preserved as FFPE and cryo (frozen in Tissue-Tek) tissues. Tumor tissue was cut in 3-μm slices and stained with the anti-CAIX mouse mAb M75 at 0.5 μg/mL (cryo sections) and 1 μg/mL (FFPE sections) final concentrations in conjunction with Animal Research Kit Peroxidase (Dako), mouse IgG block, peroxidase block and biotin block. Epitope demasking for FFPE tissue was done for 17 minutes in a vegetable steamer in Tris-buffered solution (pH 9).

IHC slides were analyzed by a pathologist with an intensity scale of 0, 1+, 2+, or 3+.

IHC slides were quantified using Histoscore (H-score). H-score considers staining intensity per cell as well as the distribution of the staining in the tumor tissue and is determined as percentage of tumor cells showing 3+ staining intensity × 3 + (% cells staining 2+) × 2 + 1 × (% cells staining 1+); range 0–300. Percent tumor growth inhibition (TGI) at 1 mg/kg BAY 79-4620 was plotted against the average H-Score for each model.

CAIX content of several tumors was also determined by Western blot (Supplementary Data).

To compare TGI between treatment groups and their respective vehicle group, 2-sided Dunnett tests (24) were carried out. The Dunnett test is a one-step test procedure that is very powerful in the given situation of comparing several groups in a common control. Dunnett's procedure keeps the multiple significance level. In each study, a significance level of 0.05 was chosen. The log-transformed ratios of tumor volume at the end of experiment to baseline tumor volume at day 0 were defined as the response variable of interest.

All calculations were carried out using SAS 9.1.3.

Human antibody fragments (Fabs) directed at the ectodomain of CAIX were selected by in vitro “panning” of the MorphoSys HuCal Gold nonimmune biased Fab-phage display library with a purified preparation of the extracellular domain of human CAIX. More than 50 CAIX-binding Fabs with unique VH and/or VL sequences were identified that bound CAIX by ELISA with signal to noise ratios of more than 10. The top 10 Fabs disclosed in Tamburini and colleagues (25) with respect to selective binding and internalization by CAIX-expressing cancer cell lines were converted to full-length human IgG1 antibodies, expressed in Chinese hamster ovary cells, and purified by affinity chromatography. Seven of these antibodies with acceptable expression and solubility were further characterized and found to exhibit high affinity binding to the CAIX ectodomain in vitro, with several mAbs exhibiting single-digit nmol/L K_d values by SPR (Supplementary Table ST1). Next, the capacity of 6 of these antibodies to selectively immunoprecipitate antigen from whole-cell lysates derived from biotin-labeled PC-3mm2 whole cells that express CAIX (CAIX⁺) or DLD-1 cells that do not (CAIX⁻) was examined by immunoblots developed with streptavidin–horseradish peroxidase (Fig. 1A). mAbs 3ee9 and 1aa1 selectively coimmunoprecipitated from biotin-labeled CAIX⁺ cells—but not from the CAIX⁻ cells—a single protein that comigrated with recombinant biotinylated CAIX (Fig. 1A). mAbs 3ef2, 5aa3, and 5a6 were not selective and coprecipitated many additional cell surface proteins in addition to CAIX. Importantly, the conjugate of mAb 3ee9 with MMAE (BAY 79-4620; CAIX-ADC) retained the affinity of the parent antibody for CAIX (Supplementary Table ST1), exhibiting a kDa of 3.6 nmol/L by SPR. The conjugate did not bind to 13 other carbonic anhydrases tested (data not shown). Binding to mitochondria-associated CA5 was observed, but this isozyme would be inaccessible to the antibody in vivo. BAY 79-4620 selectively bound (Fig. 1B) and underwent internalization by (Supplementary Fig. S1) CAIX⁺ PC-3mm2 cells in culture but did not bind CAIX⁻ MiaPaca2 cells (Fig. 1B) or undergo internalization by CAIX⁻ DLD-1 cells (Supplementary Fig. S1). A similar selectivity was observed with the ADC comprising mAb1aa1 (data not shown), and both CAIX-specific mAb conjugates selectively killed CAIX⁺ cells in culture. For example, BAY 79-4620 killed CAIX-transfected MiaPaCa-2-CAIX cells, with an IC₉₀ of ≤10 nmol/L, whereas CAIX-negative MiaPaca2 wild-type cells were quite resistant (Fig. 2), with at least a 10-fold higher IC₉₀. The mechanism of cell killing by BAY 79-4620 involved targeted tubulin disruption. Thus although treatment of CAIX⁺ PC-3mm2 cells in culture resulted in fragmented fibers and prevention of normal spindle formation, as visualized by tubulin staining, CAIX⁻ H460 cells retained normal spindle formation in the presence of the conjugate (Supplementary Fig. S2).

The high selectivity of the 3ee9 targeting antibody component of BAY 79-4620 was further shown using established tumors derived from the CAIX-negative MIAPaCa-2 pancreatic carcinoma cell line, and its CAIX-positive transfectant. Administration of an 3ee9-Alexa Fluor 750 conjugate, but not a control hIgG Alexa Fluor 750 conjugate, to mice with bilaterally implanted CAIX-negative and CAIX-positive MiaPaca tumors resulted in exclusive retention of the Alexa-750 conjugate in the CAIX-positive tumor following elimination of the bulk of the unbound fluorescence via the liver and bladder on or around day 4 (Supplementary Fig. S3). Furthermore, BAY 79-4620 exhibited minimal TGI (50%) when administered to mice implanted with established CAIX-negative MiaPaca2 tumors, even at the higher dose of 10 mg/kg, but exhibited significant antitumor activity when administered to mice implanted with established CAIX-positive tumors derived with MiaPaca2 cells transfected with CAIX (data not shown), in which doses of 2.5, 5, and 10 mg/kg yielded TGI of 63%, 82%, and 94%, respectively, and 100% shrinkage at the latter dose.

BAY 79-4620 was highly active against colorectal cancer HT-29 tumors using a range of dosing schedules, including, quite remarkably, single intravenous doses (Fig. 3A, left panel) in the range 0.625 to 10 mg/kg. To explore the in vivo mechanism of action, tumors were harvested following single BAY 79-4620 doses of 1.25 and 5 mg/kg and evaluated by IHC (Fig. 3B). Whereas little effect was observed at 4 hours, increased numbers of cells in G₂/M arrest, the appearance of multipolar spindles, and a decreased level of tubulin was clearly seen by day 1. This pattern intensified through to 5 days postdosing when almost all cells in the 5 mg/kg dose group were severely affected as judged by the presence of defective spindles or apoptosis. These data indicate a mechanism of tumor killing by BAY 79-4620 that is both consistent with the effects of the compound on CAIX-expressing cells (Supplementary Fig. S2) and in line with the known properties of MMAE as a tubulin polymerization inhibitor that induces G₂/M arrest through prevention of normal spindle formation (26).

Despite the impressive results obtained in this model, the tumor response to single dose (Fig. 3A, left panel) and the 1 dose every 7 days (not shown) schedules, was, unlike the effects of the 3 doses every 4 days schedule (Fig. 3A, right panel), transient in nature. Therefore, for all following animal experiments the dosing schedule of 3 doses every 4 days was used. At this dose scheduling, BAY 79-4620 was highly active against established tumors derived from the HT-29 cell line (Fig. 3A, right panel). Whereas tumors from untreated, PBS vehicle treated, and 10 mg/kg unconjugated CAIX mAb-treated control groups grew progressively, with mean doubling times of around 6.4 days, treatment with BAY 79-4620 on a 3 doses every 4 days schedule, intravenous schedule produced robust antitumor efficacy at all doses examined, with 0.625, 1.25, 2.5, 5, and 10 mg/kg doses yielding 54, 72, 97, 100, and 100% TGI, respectively. The percentages of the tumors that exhibited regression were 20, 90, 100, 100, respectively, at the 1.25, 2.5, 5, and 10 mg/kg doses, and BAY 79-4620 was well tolerated with no weight loss at all doses. Paclitaxel (Taxol) at a dose of 15 mg/kg showed efficacy very similar to 2.5 mg/kg of BAY 79-4620. Free MMAE toxophore administered at a dose of 0.2 mg/kg, equivalent to the amount of MMAE administered in 10 mg/kg of BAY 79-4620 was much less active than the antibody conjugate, producing 60% TGI and no tumor regression. Although this dose of free MMAE was well tolerated, a 1 mg/kg dose produced 50% lethality with severe weight loss across all survivors (not shown).

BAY 79-4620 was highly active against a variety of other established tumor types (Fig. 4), including those established using the following CAIX⁺ human cancer cell lines: cervix carcinoma HeLa-MaTu, colorectal Colo205, prostate PC3mm2, gastric NCI-N87, lung A549, gastric SNU16, and gastric MKN45. Tumors derived using HeLa-MaTu were the most sensitive wherein a minimal effective dose of 0.625 mg/kg was observed. The 60 mg/kg dose of BAY 79-4620 produced 10% lethality and 20% body weight loss establishing this dose of BAY 79-4620 as the MTD, whereas all other treatment levels were well tolerated. Eighty percent of the HeLa-MaTu tumors exhibited tumor shrinkage at the 1.25 mg/kg dose, whereas at the 2.5 mg/kg dose and higher, 100% of the tumors showed regression or complete regression. BAY 79-4620 was also active against established tumor xenografts derived from the human prostate PC-3mm2 cell line (Fig. 4). Thus, whereas PBS vehicle or unconjugated mAb (not shown) at 30 mg/kg exhibited no antitumor efficacy, BAY 79-4620 produced a dose-dependent TGI ranging from 45% at 1 mg/kg to 100% at doses of 10 mg/kg and higher and was well tolerated at all doses.

BAY 79-4620 was also active in the gastric carcinoma model NCI-N87 (Fig. 4). This aggressive tumor model responded well to the higher dose of 60 mg/kg on a 3 doses every 4 days schedule, although neither cisplatin nor paclitaxel treatment reached statistical significance compared with the vehicle group on day 37, when the vehicle group was sacrificed.

Efficacy of BAY 79-4620 in heterogeneous tumors derived from patient tumor explants was shown in several lung carcinoma models (Lu series). As these tumors consist of both CAIX-positive and CAIX-negative cells (Fig. 5), the observed efficacy in these models suggested that the release of toxophore from the CAIX-positive cells kills the CAIX-negative cells through a bystander effect.

In contrast to the effects observed in the above tumor models, BAY 79-4620 was inactive against the MDR line HCT-15 that overexpresses P-glycoprotein (P-gp). The P-gp substrate paclitaxel was similarly inactive, whereas gemcitabine, a non-Pgp substrate, exhibited TGI in this tumor model (Fig. 4).

Anti-CAIX–antibody drug conjugates are expected to release the attached toxophore specifically in antigen-expressing tumors. To test whether the activity of BAY 79-4620 correlated with CAIX expression levels in tumors, an immunohistochemical staining protocol for CAIX was developed. In parallel to the above studies of the antitumor activity of BAY 79-4620 in the various xenograft models, the levels of CAIX expression within these tumors was determined. Groups of animals were sacrificed at the time when treatment was started and tested for CAIX expression by IHC. CAIX expression was semiquantified from the average pixel intensities per cell, the percentage of the tumor cells staining positive for CAIX and the location of CAIX within the tumor tissue (Fig. 5). Most of the tumors, including A431, A549, Lu7913, Lu7298, and Lu7506, expressed CAIX in the area around necrotic cores. In most cases, CAIX staining was absent from the immediate vicinity of blood vessels consistent with its upregulation by hypoxia. The intensity of CAIX expression across the various tumor types varied strongly. HeLa-MaTu showed a strong homogenous CAIX expression pattern, whereas Hs746T showed almost no staining. The efficacy of BAY 79-4620 in tumor models showed a strong positive association with tumor CAIX expression level and/or H-score (Fig. 6). HeLa-MaTu, HT29, and PC3mm2 models characterized by the highest expression of CAIX showed 56% to 92% TGI, when treated with 1 mg/kg BAY 79-4620. This dose did not show TGI in the Hs746T, A549, and A431 models—characterized by the lowest CAIX levels observed–but these models did respond to doses higher than 1 mg/kg.

Although some recently published small molecule inhibitors (27, 28) successfully target the enzymatic activity of CAIX, the work described here uses a different therapeutic approach in exploiting CAIX as an anchor for the selective delivery of a toxic payload to tumors via an antibody-based therapeutic. A rigorous discovery cascade was implemented to generate and select fully human antibodies that are highly selective for binding to, and internalization by, CAIX-expressing cancer cell lines. These antibodies exhibited undetectable binding to other proteins within complex target cell proteomes by immunoprecipitation, did not bind to other extracellular carbonic anhydrases, and in the case of the CAIX-targeting antibody 3ee9 (BAY 79-4682), showed specific retention in CAIX-positive tumors in vivo by whole animal imaging using a fluorescently labeled version. The binding selectivity of 3ee9 was fully retained in the corresponding antibody–drug conjugate BAY 79-4620, which proved to be both potent and highly selective in the killing of cultured CAIX-expressing cell lines through a mechanism that involved tubulin disruption. Consistent with the in vitro selectivity of the conjugates, conjugate BAY 79-4620 exhibited little or no antitumor activity against a human tumor xenograft with low or no CAIX expression but was highly potent against xenografts comprising the same cell line that had been stably transfected with CAIX. The selectivity of the conjugate BAY 79-4620 was further confirmed in human and primate normal tissue binding studies, wherein the binding pattern mimicked exactly the expected tissue distribution of CAIX with significant staining only in the gastrointestinal tract and where no off-target binding was observed (data not shown).

There remains a high unmet need for safe and effective antitumor agents targeting different tumor indications, yet despite the large amount of information about the association of CAIX with various solid tumors, there is no approved therapeutic drug targeting this antigen. Our approach uses an anti-CAIX antibody with a linker toxophore system based on the tubulin inhibitor MMAE that offers potency at the site of action and a high level of stability in the periphery. Accordingly, the conjugate BAY 79-4620 proved to be highly effective and well tolerated when used against CAIX-expressing human xenografts representing cervical, prostate, colorectal, gastric, and lung tumors. In each case, a 100% complete response rate could be shown at the higher doses within a 3 doses every 4 days schedule. At lower doses of BAY 79-4620 on or around 1 mg/kg, differences in the antitumor activity across the various models was observed with higher activity associated with tumor types exhibiting high CAIX expression, such as HeLa-MaTu, and lower activity against xenograft models with less CAIX expression such as PC-3mm2. The data suggests that the antitumor activity of BAY 79-4620 at 1 mg/kg is driven mainly by targeted delivery of the MMAE toxophore to CAIX-expressing tumor cells. Targeting of a CAIX-expressing tumor by the anti-CAIX antibody was shown in in vivo imaging studies, in which the antibody was selectively retained in the antigen-positive, but not the antigen-negative, tumor implanted on the same animal. The therapeutic efficacy of targeted delivery of the toxophore was also shown in the patient-derived lung tumor model Lu7406, in which both carboplatin and paclitaxel show little efficacy, whereas BAY 79-4620—which delivers a much lower total dose of cytotoxic compound to the tumor—showed high efficacy. The efficacy against this and other patient-derived models with heterogeneous expression of CAIX is consistent with BAY 79-4620–dependent tumor cell killing both by direct and bystander mechanisms. The mechanistic IHC studies of HT-29 tumors showed that the response to a single dose of CAIX-ADC was rapid, being detectable at day 1, and seemed to affect the majority of tumor cells by day 5. The efficacy of BAY 79-4620 seems to be limited to tumors that are susceptible to tubulin inhibition, as was shown by the lack of response to the ADC in the highly chemoresistant HCT-15 colon carcinoma model.

The safety and toxicologic evaluation of the ADCs described presently has been facilitated by cross-reactivity with primate CAIX and BAY 79-4620 is currently in phase I of clinical development.

We anticipate that the high potency and selectivity of the ADCs described presently will translate into the clinic and that these compounds may possess single-agent efficacy, possibly rising to become first-line therapeutics. We note that these therapeutics are combinable preclinically with standard of care therapeutics such as capecitabine (Xeloda; data not shown), and it is conceivable that therapeutic use could be further augmented in combination.

The mechanism of action of BAY 79-4620 suggests clinical development in tumor indications characterized by a high expression of CAIX and a reported sensitivity to spindle poisons. NSCLC and gastric cancer are 2 of several tumor indications showing CAIX expression as well as sensitivity to taxanes (29–31). The expression of CAIX varies between tumor indications as well as between individual patients with a tumor of the same histologic origin. Clinical development of BAY 79-4620 should, therefore, be supported by biomarker measurements of CAIX expression. Clinical studies should explore cut-offs for CAIX expression associated with response to BAY 79-4620.

Noninvasive assays to identify patients likely to respond to BAY 79-4620 treatment and assays monitoring early response to treatment would aid clinical development of this drug. Potential approaches currently under evaluation include the quantification of circulating levels of CAIX ectodomain and imaging-based methods such as positron emission tomography.

All authors are current or former employees of Bayer Healthcare.

BAY 79-4620 uses the Seattle Genetics linker-toxophore chemistry; the antibody 3ee9 is derived from the HuCAL phage display library (Morphosys AG). The authors thank Silvia Pastorekova and Jaromir Pastorek, Institute of Virology, for scientific consultancy and also the following individuals for their important contributions to this work: Dana Wirak, Richard Altman, Tom Donaldson, Haren Vasavada (molecular biology); Dave Wunderlich, Jennifer Pendleton, Karla D' Agostino (protein expression); Carla Pellegrino, Robert Dreyer, Steven Fisk (protein chemistry), Elizabeth Bourret, Peggy Bourguillon, Marina Ichetovkin, Susan Gawlak, Marina Reinelt, Katrin Weidner (cellular assays), Elizabeth Bortolon, Arris Henderson, Dahai Xue, Bianka Timpner, Karola Henschel, Katrin Jänsch (preclinical animal studies), Sabine Jabusch (IHC), Anna Behnke (biochemical assays) and Tina Müller (statistics).

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.