MI130004, a Novel Antibody–Drug Conjugate Combining Trastuzumab with a Molecule of Marine Origin, Shows Outstanding In Vivo Activity against HER2-Expressing Tumors

Antibody–drug conjugates (ADC) are a relatively novel class of therapeutic agents that have drawn the attention of the pharmaceutical industry in the past decade. Pursuing the “magic bullet” concept coined by Paul Ehrlich in 1917 (1), it was not until 2000 that the first ADC (Mylotarg, gemtuzumab ozogamicin) was approved by the FDA for treating acute myeloid leukemia. Although Mylotarg was withdrawn from market in 2010, two new ADCs were launched soon thereafter: Adcetris (brentuximab vedotin) in 2011, for the treatment of Hodgkin lymphoma as well as systemic anaplastic large cell lymphoma (2), and Kadcyla (trastuzumab emtansine) in 2013 for treating HER2-positive metastatic breast cancer (3). These successful cases have boosted interest in the discovery and development of new ADCs. Interestingly, the vast majority of ADCs in clinical development or in early preclinical studies include cytotoxic payloads belonging to the same chemical series that those of Mylotarg, Adcetris, and Kadcyla, that is, calicheamicins, auristatins, or maytansinoids. Although new scaffolds acting through different mechanisms of action are also being explored, with the notable examples of pyrrolobenzodiazepines and camptothecin analogues, they only account for a minority of cases (4, 5). Consequently, there is a need for novel molecules to be used as payloads to increase the diversity of ADCs and thus to overcome possible resistance (6).

The wide multiplicity of life forms harbored by the sea and the different physical and chemical environment that marine organisms must face account for the much richer chemical diversity that can be found in the oceans as compared with that on earth, but such diversity still remains largely unexplored. Therefore, marine molecules constitute an attractive source of inspiration in the search for novel chemical scaffolds and successful cases of marine compounds with notable therapeutic potential are frequently reported. PM050489 is a good example of such success with its dechlorinated analogue PM060184 currently undergoing phase II clinical trials for the treatment of advanced, hormone receptor–positive, HER2-negative breast cancer (EU clinical trials database EudraCT Number: 2015-002395-24). The molecule, originally isolated from the sponge Lithoplocamia lithistoides (7), is an extremely potent interfacial microtubule inhibitor that binds to β-tubulin with single-digit nanomolar affinity at a new site (8, 9). On the basis of the interaction with tubulin dimers, these molecules inhibit microtubule assembly through a novel mechanism of action suppressing microtubule shortening and growing at a similar extent (10) and leading cells to death with subnanomolar in vitro antiproliferative activity against cancer cell lines (7). Such extremely high potency, an imperative requisite for ADC payloads (11), together with the novel mechanism of microtubule dynamics impairment, led us to consider the opportunity to use PM050489 as warhead for novel ADCs. Given the proven success of trastuzumab-based ADCs and the extensive validation for this antibody in the field, we decided to conjugate PM050489 with trastuzumab, and thus the resulting ADC, called MI130004, was prepared. This article summarizes the characterization of MI130004 and the results it has delivered both in vitro and in vivo.

PM050489, PM120160 (the result of adding a noncleavable linker to PM050489, see Fig. 1A), and MI130004 were prepared in PharmaMar S.A. Trastuzumab used in this study is the European Medicines Agency–approved version and formulation. Chromatography reagents and materials were from GE Healthcare. Immunoglobulin-degrading enzyme of Streptococcus pyogenes (IdeS) was purchased from Genovis. Mouse monoclonal anti–α-tubulin and rabbit anti–γ-tubulin antibodies were from Sigma-Aldrich. Alexa Fluor–conjugated goat anti-mouse or anti-rabbit antibodies were from Thermo Fisher Scientific. All other reagents were of the highest purity available. Throughout this work, trastuzumab concentration was determined spectrophotometrically by monitoring its absorbance at 280 nm using a molar extinction coefficient of 2.18E5 M⁻¹ cm⁻¹ and a molecular weight of 150 kDa.

Trastuzumab was initially reduced with Tris(2-carboxyethyl) phosphine hydrochloride (TCEP). Briefly, a 70 μmol/L (10.5 mg/mL) solution of the antibody in 50 mmol/L sodium phosphate pH 8.0 buffer was mixed with the appropriate amount of a 5 mmol/L solution of TCEP in water to keep the reducing agent in a 2.5-fold molar excess over the antibody. The mixture was incubated and stirred for 60 minutes at 20°C, and afterward, the appropriate volume of a 10 mmol/L PM120160 solution in DMSO was added to reach a fair molar excess of the compound over the antibody. DMSO was added if needed to keep its concentration at 5% (v/v), and the mixture was incubated for 30 minutes at 20°C. Afterward, N-acetyl-cysteine was added to quench the reaction, using the appropriate volume of a 10 mmol/L solution in water to match the concentration of PM120160. The resulting ADC was finally purified from the rest of reagents by gel filtration in Sephadex G-25 using PD-10 columns. The presence of aggregates was checked through analytic size exclusion chromatography using an Äkta FPLC system equipped with a Superdex-100 10/300 column running an isocratic method with PBS at 1 mL/minute: If the area of the peak corresponding to aggregates exceeded 10% of the total peak area, monomers were purified using the same chromatography system with a Superdex 200 16/600 preparative column running the same method described above. Final ADC concentration was determined spectrophotometrically by monitoring its absorbance at 280 nm using the molar extinction coefficient of trastuzumab; if the ADC concentration was below 2 mg/mL, it was concentrated using Vivaspin devices from GE Healthcare, and the new concentration was again determined as above.

Analysis of MI130004 by hydrophobic interaction chromatography (HIC) was performed on a Agilent 1100 HPLC system (Agilent) equipped with a 3.5 × 4.6 mm TSK-gel butyl-NPR column with 2.5 μm particle size (Tosoh Bioscience), using 1.5 mol/L ammonium sulfate in 25 mmol/L sodium phosphate, pH 7.5 as mobile phase A, and 25 % (v/v) isopropanol in 25 mmol/L sodium phosphate pH 7.5 as mobile phase B. Elution was run at a constant 0.8 mL/minutes flow rate using a 0% to 100% B gradient in 12 minutes, followed by a 5-minute isocratic elution in 100% B.

A total of 100 μg of trastuzumab or MI130004 was incubated with 100 U of enzyme in 100 μL of PBS at 37°C for 30 minutes. The sample was then diluted with 200 μL of 100 mmol/L Tris-HCl pH 7.5 with 6 mol/L guanidine-HCl, and 2 mmol/L EDTA, and afterward, 2 μL of 1 mol/L DL-dithiothreitol (DTT) was added and the mixture incubated at 56°C for 45 minutes. Reduction was stopped by adding 1 μL of glacial acetic acid, and samples were finally subjected to LC-MS analysis following the same procedure described by Wagner-Rousset and colleagues (12).

The following cell lines were obtained from ATCC: MDA-MB-231 (HTB-26) and BT-474 (HTB-20), breast adenocarcinoma; N87 (CRL-5822) and Hs 746T (HTB-135), gastric adenocarcinoma; and SK-OV-3 (HTB-77) ovarian adenocarcinoma. JIMT-1 breast adenocarcinoma cells (ACC589) were obtained from DSMZ. A2780cis ovarian adenocarcinoma cells (93112517) were obtained from ECACC. Gastric-008 gastric adenocarcinoma cell line was generated and generously given by Dr. Manuel Hidalgo (Beth Israel Deaconess Medical Center, Boston, MA). All cell lines were maintained in DMEM supplemented with 10% FBS, 2 mmol/L l-glutamine, and 100 U/mL of penicillin/streptomycin at 37°C and 5% CO₂. Cell lines were classified as HER2 positive or negative according to IHC analysis.

A colorimetric assay based on the reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was used for quantitative measurement of cell viability as described elsewhere (13). Briefly, cells were seeded in 96-well trays. Serial dilutions of all compounds dissolved in DMSO were prepared and added to the cells in fresh medium, in triplicates. Exposure to the drug was maintained during 72 hours, and cellular viability was estimated from conversion of MTT (Sigma) to its colored reaction product, MTT-formazan, which was dissolved in DMSO so as to measure its absorbance at 540 nm with a POLARStar Omega Reader (BMG Labtech). Determination of IC₅₀ values was performed by iterative nonlinear regression to a 4-parameter logistic equation using the Prism 5.0 statistical software (GraphPad). The data presented here correspond to the geometric mean of three independent experiments performed in triplicate.

Tumor cells were treated for 24 hours with 1% (v/v) DMSO in the presence or absence of MI130004 at 0.1 μg/mL or of 1 nmol/L PM050489, then fixed with methanol for 10 minutes at −20°C and incubated with a blocking solution (5% BSA in PBS) for 30 minutes. Cells were then incubated with primary mouse anti-human α-tubulin or primary rabbit anti-human γ-tubulin antibodies for 1 hour at room temperature. After three washes with a 1% (w/v) BSA in PBS solution, cells were incubated with Alexa Fluor 488–conjugated goat anti-mouse IgG or Alexa Fluor 594-conjugated goat anti-rabbit IgG secondary antibodies at room temperature for 1 hour. Cells were finally counterstained with addition of Hoechst 33258 (1 μg/mL) for 5 minutes and mounted with Mowiol mounting medium. Pictures were taken with a Leica DM IRM fluorescence microscope equipped with a 100× oil immersion objective and a DFC 340 FX digital camera (Leica).

For cell-cycle experiments, cells were treated for 24 hours with 1% (v/v) DMSO in the presence or absence of MI130004 at 0.1 μg/mL and then stained with 0.4 μg/mL propidium iodide. Samples were analyzed with a BD Accuri C6 flow cytometer (Beckton Dickinson) and the FlowJo7 cytometry analysis software (FlowJo).

All animal protocols were reviewed and approved according to regional Institutional Animal Care and Use Committees. Design, randomization, and monitoring of experiments (including body weights and tumor measurements) were performed using NewLab Software v2.25.06.00 (NewLab Oncology). Female athymic Nude-Foxn-1 nu/nu or CB17/IcrHsd-PrKdc-SCID mice (Envigo, RMS Spain S.L.) between 4 to 6 weeks of age were subcutaneously xenografted with each cell line (JIMT-1, BT-474, N87, gastric-008, A2780Cis, and SK-OV-3 as HER2 positive; MDA-MB-231 and Hs 746T as HER2 negative) into their right flank with, depending on the model, either ca. 5-10E6 cells suspended in 0.05 mL of a solution consisting of 50% Matrigel (Corning Inc.) and 50% cell culture medium, without serum or antibiotics, or with tissue from serial transplanted donor mice. Tumors were removed from donor animals debrided of membrane, hemorrhagic and necrotic areas cut into fragments (3 mm³), placed in Matrigel, and subcutaneously implanted. When tumors reached ca. 150 mm³, mice (n = 8–10 animals/group) were randomly allocated (day 0) to receive MI130004 (1, 5, or 10 mg/kg), trastuzumab (30 or 10 mg/kg), or vehicle. Intravenous treatments were weekly administered for 5 consecutive weeks. The control animals received an equal volume of vehicle with the same schedule. Caliper measurements of the tumor diameters were made three times a week, and tumor volumes were calculated according to the following formula: volume = (a × b²)/2, where a and b were the longest and shortest tumor diameters, respectively. For survival evaluation, time to endpoint was defined as the time from day 0 to death as a result of tumor growth (larger than 2,000 mm³) or any other cause (e.g., tumor necrosis). Complete tumor regression (CR) was defined as tumor volume below 63 mm³ for 2 or more consecutive measurements, such value corresponding to the lowest measurable limit considering the contribution of the mass from fibrous material, scar tissue, etc. Statistical differences, in animal survival, were assessed by Kaplan–Meier curves with the log-rank test. Animals were humanely sacrificed when their tumors reached 2,500 mm³ or if significant toxicity (e.g., severe body weight reduction) was observed. Differences in tumor volumes between treated and control group were evaluated using the Mann–Whitney U test. Statistical analyses were performed by GraphPad Prism v5.03 (GraphPad Software Inc.).

Tumors (or skin around the former tumor site) were dissected, formalin fixed, and paraffin embedded for histopathology evaluations. Mitotic aberrations were revealed by nuclear staining with Hoechst 33258 after 24 hours of the first administration of MI130004.

MI130004 is an ADC resulting from the conjugation of PM050489 to Cys residues of trastuzumab via a maleimide-based noncleavable linker (structure in Fig. 1A). According to size exclusion chromatography analysis, the ADC preparation was composed nearly exclusively (99.8% purity) of single monomeric species (Supplementary Fig. S1A). HIC of MI130004 revealed the existence of one major peak, eluting at 6.23 minutes, corresponding to the conjugated species and accounting for 75% of the total peak area, together with the peak corresponding to the naked antibody (16%) and another minor peak (7%) of higher hydrophobicity, eluting at 8.06 minutes, likely corresponding to minority species with a higher drug load (Supplementary Fig. S1B).

To determine the number and position of the drug in the ADC molecule, limited proteolysis and reduction of both ADC and antibody were performed, followed by LC-MS analysis of the resulting fragments. Molecular mass assignments to these fragments were done using the values reported in ref. 12 as we are using the same source of trastuzumab; therefore, a similar glycosylation pattern may be expected. The UV chromatograms of the digestion products of trastuzumab and MI130004 showed three common peaks, eluting at circa 12.7, 15.2, and 19.0 minutes whose MS analysis allowed the assignment to the Fc/2, LC, and Fd fragments of the antibody, respectively (Supplementary Figs. S1C–S1D and S2A–S2C) with different degrees of glycosylation causing the multiplicity of masses identified in the first peak. Only the MI130004 chromatogram showed two additional peaks eluting at longer times: 20.82 and 23.88 minutes. MS deconvolution of the peak at 20.82 minutes (Supplementary Fig. S3A) allowed the identification of four species, the major one with a mass of 24,303 Da, consistent with that of the LC fragment (23,439 Da) conjugated to one drug molecule (856 Da). Similarly, MS deconvolution of the peak at 23.88 minutes (Supplementary Fig. S3B) yielded two species, the major one corresponding to a mass of 26,244 Da, consistent with that of the Fd fragment (25,384 Da) conjugated to one payload (856 Da). Likewise, analysis of the peak areas in these chromatograms suggests a DAR of 1.04, whereas the results of the HIC chromatography (Supplementary Fig. S1B) suggest a DAR of 1.8. These data, together with the aforementioned mass assignments (summarized in Supplementary Table S1), suggest that, in average, one molecule of MI130004 contains 1 to 1.8 molecules of PM050489 conjugated to cysteines involved in one of the disulfide bonds connecting the light and the heavy chains.

The in vitro antiproliferative activity of MI130004 was determined using a panel of tumor cell lines of different histotypes (breast, gastric, and ovary) expressing or not HER2 (Table 1). MI130004 showed excellent in vitro potency with selectivity for HER2-positive tumor cells (Table 1; Fig. 1B; Supplementary Fig. S4). The HER2-positive panel IC₅₀ geometric mean value was 0.44 μg/mL, whereas it was >10 μg/mL for the HER2-negative tumor cells. Of note, the potency of MI130004 in the parental and P-gp overexpressing versions of the A2780 cell line are virtually identical (2.00 vs. 3.44 μg/mL), consistent with the results obtained with PM050489 and its derivatives and confirming the potential of these series to overcome resistance mechanisms related to drug efflux. On the other hand, PM050489 caused a clear antiproliferative effect in all the cell lines tested regardless of their HER2 expression (Supplementary Table S2). Trastuzumab alone did not cause any effect on cell growth in any of these lines even at the highest concentration tested (50 μg/mL).

We have previously described that, in tumor cell lines, PM050489 impaired tubulin polymerization, causing mitotic aberrations with massive disorganization and disappearance of the microtubule mass and cell-cycle arrest in G₂–M (8). Interestingly, and differently from PM050489 that affected both HER2-positive and HER2-negative tumor cells (Supplementary Fig. S5), MI130004 mirrored the cellular effects of the payload only in HER2-positive tumor cells. Indeed, as shown in Fig. 1C and Supplementary Fig. S6, MI130004-treated HER2-positive breast tumor cells of different histotypes exhibited a disordered microtubule distribution and less microtubule fibers compared with nontreated cells. Moreover, immunofluorescent analyses showed that MI130004 induced altered mitotic phenotypes with multiple mitotic poles. This coincided with an increase in the percentage of cells in G₂–M phase (Table 2). In contrast, HER2-negative breast cancer cells did not show alterations in tubulin cytoskeleton nor in cell cycle (Fig. 1C; Supplementary Fig. S6; Table 2). Trastuzumab did not induce any alterations on the microtubule network (14).

To determine whether the in vitro antiproliferative activity of MI130004 translated into in vivo antitumor activity, we evaluated the effect of the ADC in mice xenograft models. The efficacy of the molecule was first tested in xenografts with breast tumor cells. HER2 expression levels in the tumors were confirmed by IHC 24 hours after the first administration (Supplementary Fig. S7A–S7C). At the drug doses used in the experiment, no significant toxicity or body weight loss was observed in the treated animals. As shown in Fig. 2A, MI130004 presented antitumor activity with a statistically significant inhibition of tumor growth and an improvement of mice survival only in the HER2-positive tested models but not in the HER2-negative one (i.e., MDA-MB-231), thereby confirming the selectivity of the ADC for HER2-expressing cells. For instance, in animals bearing JIMT-1 (HER2 positive) xenografts, MI130004 induced a long-lasting antitumor effect with statistically significant reduction of tumor volume compared with both vehicle and trastuzumab treatments. Moreover, a dose-dependent disappearance of HER2-expressing cells was observed in JIMT-1 xenografts after MI130004 (Fig. 2B). Mitotic aberrations consistent with the mechanism of action of PM050489 (revealed by nuclear staining with Hoechst-33258) were also noticeable (Fig. 2B).

In addition to breast xenografts, gastric and ovarian xenograft models were also used to evaluate the efficacy of MI130004, and similar results were obtained. In HER2-positive gastric cell lines (gastric-008 and N87), MI130004 induced a reduction in tumor volume as well as an increase in survival compared with HER2-negative cells (Hs 746T, Fig. 3A), whereas trastuzumab did not show any effect on tumor growth. In gastric-008 cells extracted from the tumor, a reduction of HER2 expression as well as an increase in mitotic aberrations can be noticed as a consequence of MI130004 treatment (Fig. 3B). Alike, the compound also showed activity in HER2-positive ovarian cancer cell lines (SK-OV-3 and A2780cis; Fig. 4A). Again, in SK-OV-3, MI130004 was more active than trastuzumab and induced a dose-dependent reduction of HER2 expression in tumor xenografts (Fig. 4B).

Altogether, our results indicate that MI130004, an ADC-conjugating PM050489 to trastuzumab, demonstrated significant and selective in vivo antitumor activity in human-derived HER2-expressing xenografts corresponding to breast, gastric, and ovarian tumors.

Since the approval of Mylotarg in 2000, a vast research effort has been made in the ADC field, resulting in two novel entities being approved in 2011 to 2013 and a full pipeline with a large number of conjugates undergoing clinical studies. Despite the evident need for novel payloads that might help to overcome future resistance, a close study to the current pipeline shows that the vast majority of the novel ADCs under clinical investigation includes cytotoxic warheads belonging to the same chemical families than those of Mylotarg, Adcetris and Kadcyla, that is, calicheamicins, auristatins, and maytansinoids, respectively. Moreover, the novel payloads not belonging to those classes act through similar mechanisms of action, affecting either DNA biology (cross-linkers like pyrrolobenzodiazepines, alkylating agents such as indolinobenzodiazepines and duocarmycins or inhibitors of topoisomerases like camptothecin derivatives or doxorubicin) or microtubule dynamics (amberstatin or tubulysin analogues; see ref. 15 for a complete review). The discovery of novel warheads acting on unprecedented targets is a difficult task given the requisites such targets must accomplish, namely intracellular location and critical involvement in essential processes for cell survival or division. Therefore, the search for novel scaffolds acting differently on microtubules or DNA seems a more realistic approach to face possible resistance to current ADC payloads.

PM050489 is a marine molecule with extremely high antiproliferative potency that has inspired a chemical series leading to its dechlorinated analogue, PM106184, with promising therapeutic projection in oncology. Despite being a tubulin-binding agent, the molecule looks appropriate to be considered for antibody conjugation even if it binds to the so-called maytansine site recently described in β-tubulin and, therefore, its potentiality to overcome the hypothetical appearance of future resistance to maytansinoids could be argued. However, there are structural and functional results that may rebut such argument. First, although PM050489 derivatives share a common pharmacophore pocket with maytansine, their interaction area expands beyond such pocket to reach other zones within the β-tubulin surface, and indeed, they are involved in additional steric clashes with α-tubulin units, thus contributing further to prevent the formation of longitudinal contacts between tubulin molecules needed for microtubule growth (9). Second, although the main resistance mechanism known so far to affect ADC payload consists on the overexpression and increased activity of drug efflux pumps (16), an otherwise classical resistance mechanism for tubulin-binding agents (17), PM050489 derivatives are highly potent on inhibiting the in vitro growth of P-gp–overexpressing tumor cell lines and have shown activity in animal tumor models regardless of the P-gp status of the cell line used (10). In fact, MI130004 displays identical potencies in the ovarian cell line A2780 and in the P-gp–overexpressing version A2780cis. In our view, these two facts invest PM050489 derivatives with significant differential attributes with respect to maytansinoids that lead to consider them as attractive warheads for overcoming possible resistance to ADC payloads of current use.

PM050489 was conjugated to trastuzumab using a noncleavable linker. The stability of the linker used was demonstrated by the complete lack of activity observed for MI130004 in HER-negative cells both in vitro and in vivo. Furthermore, the extraordinary potency observed in vitro with MI130004, consistent in molar units with that of PM050489, demonstrates that the presence of the linker is not impairing significantly the antiproliferative potential of the original molecule and indeed the in vitro biological effects observed in cells treated with MI130004 (mitotic aberrations, massive disorganization, and disappearance of the microtubule network and cell-cycle arrest in G₂–M) resemble those obtained with PM050489.

Analytic studies performed on MI130004 reveal the presence of one major species (75%, with naked trastuzumab being the most abundant species of the remaining 30%) likely bearing two molecules of payload bound per molecule of MI130004. To interpret the mass spectra data obtained, we have considered that, since we have used in this study the European Medicines Agency–approved version and formulation of trastuzumab, the glycosylation pattern of the antibody should be identical to that published for this version of the molecule; hence, the molecular mass assignations of the fragments obtained with the naked antibody should readily match those previously published, for example, in ref. 12. Our results are consistent with that assumption and enable us to propose that the two cytotoxic molecules are bound to both sides of the disulfide bond connecting one light and one heavy chain of the antibody as it has been possible to identify fragments whose masses correspond to the addition of only one payload of each chain (LC and Fd) located at each side of the disulfide bridge. Further attempts to increase the drug-to-antibody ratio (DAR) with alternative synthetic strategies were abandoned as they failed to deliver the desired product as most of the material was lost during the process due to poor solubility. Although it was commonly accepted that low DAR values correlate with higher bioavailability and reduced toxicity (18, 19), novel findings suggest that higher DAR values can be obtained without jeopardizing the ADC properties provided that the linker is endowed with properties that compensate for the higher hydrophobicity introduced by the payload (20). In the case of MI130004, it is highly remarkable that, despite its low DAR, the conjugate showed an extraordinary potency both in vitro and in vivo. As already reported by Strop and colleagues (21) for an auristatin-based ADC with engineered attachment sites in a Trop-2 antibody, a likely explanation for the high potency obtained at such low substitution degree may reside in the combination of an extremely potent payload with a high and homogeneous expression of the antigen (HER2 in the case of MI130004) within tumor cells and also with a rapid and efficient internalization of the antibody–receptor complex. Likewise, it sounds reasonable that such low DAR may also contribute to the outstanding in vivo efficacy observed due to the likely high bioavailability mentioned above.

The in vivo efficacy of MI130004 has been observed with several xenografts of HER2-positive tumor cells regardless of their histotype. Breast, gastric, and ovarian xenografts have experienced a significant reduction in tumor growth together with improved animal survival in a dose-dependent manner, and these effects were long lasting for the highest dose of MI130004 used. Histopathologic analysis showed that the biological effect caused in vivo was identical to that in vitro and is consistent with the mechanism of action of PM050489. Therefore, this marine molecule has demonstrated to be an excellent payload for novel ADCs endowed with promising features and reinforces the potential of marine natural products as a source of inspiration for novel chemical scaffolds.

J.F. Martínez-Leal is a department manager at and has ownership interest (including patents) in PharmaMar SA. No potential conflicts of interest were disclosed by the other authors.

Conception and design: P. Avilés, J.M. Domínguez, M.J. Guillén, J.F. Martínez-Leal, C.M. Galmarini, C. Cuevas

Development of methodology: J.M. Domínguez, M.J. Muñoz-Alonso, C. Mateo, R. Rodriguez-Acebes, A. Francesch

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M.J. Muñoz-Alonso, C. Mateo, R. Rodriguez-Acebes, J.M. Molina-Guijarro

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P. Avilés, J.M. Domínguez, M.J. Guillén, J.M. Molina-Guijarro, J.F. Martínez-Leal, C.M. Galmarini

Writing, review, and/or revision of the manuscript: P. Avilés, J.M. Domínguez, M.J. Guillén, J.M. Molina-Guijarro, J.F. Martínez-Leal, C.M. Galmarini, C. Cuevas

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.F. Martínez-Leal, S. Munt, C.M. Galmarini

Study supervision: P. Avilés, J.M. Domínguez, M.J. Guillén, J.F. Martínez-Leal, S. Munt

This work was partially supported by grant IPT-2012-0198-090000 (“MARINMAB” project) from Ministerio de Economía, Industria y Competitividad. The excellent technical skills of the in vivo Pharmacology team in PharmaMar and their critical contribution to perform the xenograft experiments is greatly appreciated. We want to thank Patricia Rupérez, Pablo Cobos, and Carles Celma at KYMOS for their work with the proteolytic digestion and LC-MS analysis of MI130004. We would also like to acknowledge the IHC work performed by the Oncogenesis and Antitumor Drug Group at the Hospital de Sant Pau in Barcelona.

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.