Purpose: To assess the safety, tolerability, and preliminary antitumor activity of the investigational anti–guanylyl cyclase C (GCC) antibody–drug conjugate TAK-264 (formerly MLN0264) in adult patients with advanced gastrointestinal malignancies.
Experimental Design: Adult patients with GCC-expressing gastrointestinal malignancies (H-score ≥ 10) were eligible for inclusion. TAK-264 was administered as a 30-minute intravenous infusion once every 3 weeks for up to 17 cycles. Dose escalation proceeded using a Bayesian continual reassessment method. At the maximum tolerated dose (MTD), 25 patients with metastatic colorectal cancer were enrolled in a prespecified dose expansion cohort.
Results: Forty-one patients were enrolled, including 35 (85%) with metastatic colorectal cancer. During dose escalation (0.3–2.4 mg/kg), four of 19 patients experienced dose-limiting toxicities of grade 4 neutropenia; the MTD was determined as 1.8 mg/kg. Patients received a median of two cycles of TAK-264 (range, 1–12); nine received ≥four cycles. Common drug-related adverse events (AEs) included nausea and decreased appetite (each 41%), fatigue (32%), diarrhea, anemia, alopecia, and neutropenia (each 27%); grade ≥3 AEs included neutropenia (22%), hypokalemia, and febrile neutropenia (each 7%). Peripheral neuropathy was reported in four (10%) patients. Pharmacokinetic data showed approximately dose proportional systemic exposure and a mean plasma half-life of around 4 days, supporting the dosing schedule. Overall, 39 patients were response-evaluable; three experienced durable stable disease; and one with gastric adenocarcinoma had a partial response. GCC expression did not appear to correlate with treatment duration.
Conclusions: These findings suggest that TAK-264 has a manageable safety profile, with preliminary evidence of potential antitumor activity in specific gastrointestinal malignancies. Further investigation is underway. Clin Cancer Res; 22(20); 5049–57. ©2016 AACR.
This article is featured in Highlights of This Issue, p. 4961
The use of antibody–drug conjugates (ADCs) to target specific antigens preferentially expressed on cancer cells is a feasible and active treatment approach in patients with Hodgkin lymphoma, anaplastic large cell lymphoma, and HER2-positive breast cancer, with multiple ADCs under investigation in other lymphomas and solid tumors. ADCs comprise a monoclonal antibody, a linker, and a cytotoxic small-molecule drug. For gastrointestinal cancers, the transmembrane cell surface receptor guanylyl cyclase C (GCC) has been identified as a potential target for a monoclonal antibody, being expressed in 60% to 70% of pancreatic, gastric, and esophageal cancers and 95% of primary and metastatic colorectal cancer. This first-in-human phase I study investigated TAK-264, a novel investigational ADC targeting GCC, in adult patients with advanced, GCC-expressing gastrointestinal malignancies. Results demonstrated a manageable safety profile at the maximum tolerated dose and preliminary evidence of antitumor activity and early signs of clinical benefit in patients with pancreatic, esophageal, and gastric carcinoma.
Monoclonal antibodies are established as anticancer therapies in a number of different malignancies. Recently, antibody–drug conjugates (ADCs) have been approved, and several more are being investigated, for various types of cancer. ADCs are novel targeted agents composed of a monoclonal antibody, a linker, and a cytotoxic small-molecule drug (1, 2). The monoclonal antibody is targeted to a specific antigen preferentially expressed on cancer cells; the cytotoxic drug is released upon internalization (1). There are two ADCs currently approved: brentuximab vedotin for Hodgkin lymphoma and anaplastic large cell lymphoma (3–6) and ado-trastuzumab emtansine for HER2-positive breast cancer (7).
Brentuximab vedotin is composed of an anti-CD30 antibody, a valine–citrulline protease-cleavable linker, and the potent microtubule-disrupting agent monomethyl auristatin E (MMAE; refs. 5, 6). It has demonstrated substantial efficacy in pivotal phase 2 studies, including overall response rates of 75% (34% complete remission) in relapsed/refractory Hodgkin lymphoma and 86% (57% complete remission) in relapsed/refractory systemic anaplastic large cell lymphoma (4, 6). Brentuximab vedotin received accelerated approval by the FDA in 2011 and conditional approval in Europe in 2012 for patients with relapsed or refractory Hodgkin lymphoma and anaplastic large cell lymphoma (8, 9). Other ADCs are in development based on the same linker/drug technology, including polatuzumab vedotin and pinatuzumab vedotin (targeting CD79b and CD22, respectively), both of which are in phase II development for the treatment of diffuse large B-cell lymphoma and follicular lymphoma (10). Ado-trastuzumab emtansine is composed of the anti-HER2 antibody trastuzumab connected via a stable thioether linker to the potent antimicrotubule agent derivative of maytansine 1, an inhibitor of microtubule dimerization (11, 12). It was approved in the United States and Europe in 2013 for patients with advanced or metastatic HER2-positive breast cancer (13, 14).
Exploring the therapeutic potential of ADCs in gastrointestinal cancers is of significant interest, as they are a leading cause of cancer-related deaths; colorectal, pancreatic, and gastric cancer had the second-, fourth-, and fourteenth-highest estimated rates of cancer-related deaths in the United States in 2015 (15). One potential target is the transmembrane cell surface receptor guanylyl cyclase C (GCC), which is normally expressed on intestinal epithelial cells but not on extragastrointestinal tissues (16, 17). In primary and metastatic tumors derived from intestinal epithelial cells, GCC expression is maintained during neoplastic progression (16, 18). GCC is expressed in 60% to 70% of pancreatic, gastric, and esophageal cancers (19–21) and 95% of primary and metastatic colorectal cancer (16, 18, 22–24). Systemically delivered GCC-targeting agents are expected to be preferentially delivered to GCC receptors in tumor tissue, while leaving normal tissues unaffected, as GCC is expressed on the apical side of epithelial tight junctions (10, 16, 23, 25, 26). Access to GCC receptors is enabled in tumor tissues as a result of disrupted cell polarity, altered tight junction architecture, and disruption of its apical localization (16, 18, 22, 24). TAK-264 (formerly MLN0264) is a novel ADC consisting of a fully human IgG1 monoclonal anti-GCC antibody conjugated via a protease-cleavable linker to MMAE. Following binding to GCC, TAK-264 is internalized and transported to lysosomes where MMAE is released and binds to microtubules, leading to cell-cycle arrest and apoptosis. TAK-264 has been investigated in vivo in animal models of gastrointestinal tumors, including GCC-expressing human colorectal cancer xenografts and pancreatic cancer xenograft models, demonstrating selective binding and internalization into GCC-expressing tumor cells and antitumor activity (27, 28).
This phase 1, first-in-human study assessed the safety and tolerability, dose-limiting toxicities (DLT), maximum tolerated dose (MTD), pharmacokinetics, and preliminary antitumor activity of TAK-264 in adult patients with advanced, GCC-expressing gastrointestinal malignancies.
Patients and Methods
Patients aged ≥18 years diagnosed with a GCC-expressing gastrointestinal malignancy (H-score ≥ 10, derivation described below), for whom standard treatment was no longer effective or did not offer curative or life-prolonging potential, were eligible. Eligible malignancies included, but were not limited to, metastatic colorectal cancer, gastric carcinoma, esophageal carcinoma, small intestine cancer, pancreatic cancer, and unknown primary malignancies. Patients should have progressed to standard-of-care therapies in all cases. For the prespecified expansion cohort, only patients with metastatic colorectal cancer were eligible. Patients required measurable disease per Response Evaluation Criteria in Solid Tumors (RECIST 1.1), Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1, life expectancy of ≥12 weeks, adequate bone marrow [absolute neutrophil count (ANC) ≥ 1,500 cells/mm3, platelet count ≥ 100,000/mm3], hepatic [total bilirubin ≤ 1.5 × upper limit of normal (ULN), serum alanine or aspartate aminotransferase (ALT/AST) ≤ 2 × ULN, serum albumin ≥ 3.0 g/dL], and renal (serum creatinine ≤ 1.5 × ULN and/or calculated creatinine clearance ≥60 mL/min) function. Patients had to have completed prior chemotherapy, immunotherapy, or radiotherapy ≥4 weeks prior to enrollment and to have available archived or fresh tumor tissue. Patients were excluded if they had any comorbidities that in the view of the treating physician rendered the patient at high risk from treatment complications, known infection or inflammatory bowel disease, history of another primary malignancy not in remission for at least 3 years, New York Heart Association class III or IV, or grade ≥2 peripheral neuropathy. All patients provided written informed consent.
Institutional review boards/ethics committees at the participating investigational centers approved the study, which was conducted according to the principles set out in the Declaration of Helsinki, International Conference on Harmonisation Good Clinical Practice guidelines, and local regulatory requirements.
In this multicenter, open-label, dose-escalation study (NCT01577758), TAK-264 was administered once every 3 weeks as a 30-minute intravenous infusion (day 1 of 21-day cycles) for up to 17 cycles or until disease progression or occurrence of unacceptable TAK-264–related toxicity. Dose escalation was conducted using a Bayesian continual reassessment method (CRM) based on two-patient cohorts, according to observed DLTs during cycle 1. DLTs were defined as grade 4 neutropenia (ANC < 500 cells/mm3); grade ≥ 3 neutropenia with fever (oral temperature ≥ 38.5°C) and/or infection; grade 4 thrombocytopenia (platelets < 25,000/mm3); grade ≥ 3 thrombocytopenia with clinically meaningful bleeding at any time; grade ≥ 3 nausea and/or emesis that occurred despite anti-emetic prophylaxis; grade ≥ 3 diarrhea that occurred despite optimal supportive care measures; any other grade ≥ 3 nonhematologic toxicity with the exception of brief (<1 week) grade 3 fatigue; inability to start the next cycle of therapy due to >2 weeks treatment delay because of a lack of adequate recovery of TAK-264–related hematologic/nonhematologic toxicities; other TAK-264–related grade ≥ 2 nonhematologic toxicities that, in the opinion of the investigator, required a dose reduction or discontinuation of TAK-264 therapy.
The CRM algorithm is shown in Fig. 1. Dose escalation or de-escalation was based on the observed toxicities in all DLT-evaluable patients. After the first two patients had been dosed and had completed the first cycle of therapy, the CRM algorithm was updated on the basis of the observed DLTs, and the predicted MTD was calculated. Once at least six patients had been treated at a given dose without the algorithm suggesting escalation or de-escalation, that dose was considered the MTD. After completion of dose escalation and MTD determination, additional patients with metastatic colorectal cancer were enrolled to a prespecified dose-expansion cohort to achieve up to 20 response-evaluable patients. Among these patients, at least six were required to have high GCC expression.
Objectives and assessments
The primary objectives were to assess the safety profile of intravenous TAK-264 in patients with advanced GCC-expressing gastrointestinal malignancies, determine the MTD, and describe the pharmacokinetic (PK) profile of TAK-264, total antibody, and MMAE. Secondary objectives were to evaluate disease response and evidence of antitumor activity in TAK-264–treated patients and the immunogenicity of TAK-264 [antitherapeutic antibody (ATA) development].
An immunohistochemical assay was performed to assess the expression of GCC by utilizing a fully human antibody specific to GCC (Millennium Pharmaceuticals, Inc.). Tissue sections 5- to 10-μm thick were dewaxed through 4, 5-minute changes of xylene and then placed in a series of graded alcohol solutions diluted with distilled water. Steam heat–induced epitope recovery (SHIER) was conducted for 20 minutes in the capillary gap in the upper chamber of a Black and Decker Steamer. An automated TechMate 500 or TechMate 1000 (Roche Diagnostics) was used to perform the staining. Following an overnight primary incubation, the visualization was achieved using a non–biotin-based peroxidase detection kit (Ultra Vision). After completing the staining, slides were dehydrated, and glass coverslips and CytoSeal were used to permanently cover the slides. Positive staining was shown by the presence of a brown (DAB-HRP) reaction product. Hematoxylin counterstain was used to visualize cell and tissue morphologies. All slides were stained and assessed under microscope by two blinded pathologists at the central laboratory to assess quality of staining and evaluate the GCC levels. On the basis of this semiquantitative method, the H-score for GCC expression was calculated using the sum of the percentage of tumor cells with GCC staining and intensity of 1+, 2+, and 3+, up to a maximum score of 300 (100% at 3+; ref. 29). Preclinical data suggested that both cytoplasmic and apical GCC expression may play a role in TAK-264 efficacy (data on file). Therefore, H-scores were determined for both cytoplasmic and apically oriented staining and summed, giving a total maximum combined H-score of 600. The staining was detected on tumor cells; there was no staining in the stroma or infiltrating inflammatory cells.
Adverse events (AEs) were graded using NCI-CTCAE version 4.03. Disease responses were assessed at the end of every second cycle by RECIST v1.1. Blood samples were collected for pharmacokinetic analysis pre-dose and at multiple time points post-dose. Serum levels of TAK-264 and total antibody were assayed using a quantitative sandwich enzyme immunoassay; plasma levels of free MMAE were determined using a liquid chromatography tandem mass spectrometry assay. Blood samples were taken from patients before dosing and at the end of the study to evaluate presence of ATA, using validated assays to detect TAK-264–binding antibodies and to verify whether these antibodies had neutralizing activity.
Analysis populations and statistical analyses
The safety population included all patients who received at least one dose of study drug. The DLT-evaluable population (used to determine MTD) included dose-escalation patients who either experienced a DLT during cycle 1 or received TAK-264 and completed all study procedures in cycle 1 without DLTs. The PK-evaluable population included patients with sufficient dosing and PK data to estimate pharmacokinetic parameters. The response-evaluable population included patients with measureable disease who received TAK-264 and had at least one post-baseline response assessment.
Statistical analyses were primarily descriptive and graphical in nature. Summary tabulations were used to present the number of observations, mean, standard deviation (SD), median, minimum, and maximum for continuous variables, and the number and percentage per category for categorical data. Summary statistics were calculated for baseline characteristics, dosing, safety (including DLTs), AEs, serious AEs, laboratory values, vital signs, efficacy (including disease response), biomarkers (including ATA), and pharmacokinetic parameters. Progression-free survival (PFS) was estimated using the Kaplan–Meier method.
Forty-one patients were enrolled at three sites in the United States and one site in Spain between June 11, 2012 and February 12, 2014 and received at least one dose of TAK-264 (Table 1). In 35 of 41 (85%) patients, the primary diagnosis was metastatic colorectal cancer, as this was the only tumor type included in the prespecified expansion cohort. The other patients had pancreatic, esophageal, gastric, or other type of adenocarcinoma. GCC H-score ranged from 10 to 550; 13 patients with high GCC expression levels were enrolled in the expansion cohort.
DLTs and MTD determination
The dose-escalation cohort included 19 patients: two patients each at 0.3, 0.6, 1.2, and 1.5 mg/kg, six patients at 1.8 mg/kg, four patients at 2.1 mg/kg, and one patient at 2.4 mg/kg. There were no DLTs in the low-dose groups (0.3–1.5 mg/kg). At higher doses, a total of four patients experienced DLTs of grade 4 neutropenia. Of the four patients with DLTs, three were treated at doses above 1.8 mg/kg. One patient in the 1.8 mg/kg dose group experienced a DLT of grade 4 febrile neutropenia on day 12 of cycle 1. The patient discontinued the study per protocol and the neutropenia resolved in 2 days. Two patients in the 2.1 mg/kg dose group experienced DLTs. One patient experienced grade 4 neutropenia and grade 3 QTc prolongation on days 10 and 14 of cycle 1, respectively. Per protocol, the patient discontinued the study, and the DLTs resolved in 4 and 7 days, respectively. The other patient treated at 2.1 mg/kg had grade 4 neutropenia on day 15 of cycle 1, which resolved 7 days later. At a dose of 2.4 mg/kg, one patient experienced a DLT of grade 4 neutropenia on day 12 of cycle 1. The patient discontinued the study per protocol, and the DLT resolved in 4 days.
The MTD was determined as 1.8 mg/kg according to the Bayesian CRM method. A total of 25 patients with metastatic colorectal cancer were enrolled at this dose level, which included the patients from the expansion cohort and three patients from the 1.8 mg/kg dose-escalation cohort.
Treatment exposure and safety
Patients received a median of two cycles of TAK-264 (range, 1–12). Overall, nine patients (22%) received ≥4 cycles and five (12%) patients received ≥6 cycles. Reasons for treatment discontinuation were disease progression (n = 28, 68%), symptomatic deterioration (n = 7, 17%), AEs, and subject withdrawal (each n = 3, 7%).
All 41 patients reported AEs. A total of 36 patients (88%) reported at least one drug-related AE (Table 2). The most common drug-related AEs included nausea and decreased appetite (each n = 17, 41%), fatigue (n = 13, 32%), diarrhea, anemia, alopecia, and neutropenia (each n = 11, 27%). Grade ≥3 AEs were experienced by 26 patients (63%) treated at all doses (Table 2). Among these, 17 (41%) were drug-related. The drug-related grade ≥3 AEs, across all doses, included neutropenia (n = 9, 22%), hypokalemia, febrile neutropenia (each n = 3, 7%), and QTc prolongation (n = 2, 5%). Serious AEs were experienced by 16 patients (39%), with nine (22%) of these patients experiencing drug-related serious AEs including small intestinal obstruction [three patients (7%)], febrile neutropenia [three patients (7%)], constipation [two patients (5%)], intestinal obstruction [two patients (5%)], and pyrexia [two patients (5%)]. Grade 1 or 2 peripheral neuropathy was reported by four patients (10%) over the course of the study; three patients without baseline peripheral neuropathy experienced grade 1 peripheral neuropathy; two patients reported peripheral neuropathy during cycles 1 and 4, respectively, which was reported as not resolved by the end-of-study visit (∼30 days after the last dose of study drug or prior to the start of subsequent antineoplastic therapy), and the third patient reported grade 1 peripheral neuropathy during cycle 5, which resolved in 27 days. The fourth patient entered the study with grade 1 peripheral neuropathy and experienced grade 2 peripheral neuropathy events during cycles 1 and 2, which resolved in 6 and 9 days, respectively.
A total of three patients (7%) had dose reductions due to AEs. There were three discontinuations (7%) due to AEs, and one on-study death considered to be related to disease progression. The three patients who discontinued because of AEs were all treated at doses above the MTD and experienced DLTs. Therefore, per protocol, they discontinued the study. Two of the three patients were treated at 2.1 mg/kg and one at 2.4 mg/kg. The patient who died of disease progression was a 51-year-old male with esophageal carcinoma who received TAK-264 at the MTD (1.8 mg/kg) and initially discontinued because of a DLT of grade 4 neutropenia. A re-staging CT scan several months later, without any further therapy, showed a response. The patient re-entered the study 6 months after the initial discontinuation at a dose reduction of 1.5 mg/kg TAK-264. Unfortunately, although the patient tolerated the infusion, his clinical status declined and he died of disease progression 8 days after the second administration of TAK-264. The treating physician assumed that his disease was the primary cause of death. In this study there, were no confirmed ATA-positive results detected in the total number of samples.
The PK-evaluable population included all 41 patients. PK samples were collected pre-dose, post-dose, and on days 2, 3, 4, 8, and 15. Increases in exposure to TAK-264 and free MMAE were approximately proportional to dose (Fig. 2). Summary statistics for TAK-264, total antibody, and MMAE PK parameters during cycle 1 for patients treated at the MTD are shown in Table 3. Median time to maximum concentration occurred immediately after infusion for TAK-264 and total antibody and approximately 3 days after infusion for MMAE. Median half-life was approximately 4 days for TAK-264 and 3 days for free MMAE.
Among 39 response-evaluable patients, one partial response was observed in a 53-year-old white male patient with gastric adenocarcinoma who had a low GCC expression level, had received three prior therapies, and had been treated with TAK-264 at the MTD. This response was observed after two treatment cycles and sustained for 81 days. The patient was progression-free for 121 days. In addition, stable disease was observed in 17 (44%) response-evaluable patients. One 57-year-old white female patient with pancreatic carcinoma who had a low GCC expression level received 12 cycles of treatment. This patient was treated at 2.1 mg/kg through cycle 3 and at 1.8 mg/kg for cycles 4 to 12, maintaining stable disease from the first response assessment (unscheduled) until the end of cycle 12. A 30-year-old white male patient with pancreatic carcinoma who had a low GCC expression level received six cycles of treatment at the MTD and had a best response of stable disease, which lasted for 6 cycles. Both of these patients had previously received multiple lines of treatment, including chemotherapy, radiotherapy, and surgery. A 51-year-old white male patient with esophageal carcinoma who had an intermediate GCC expression level was treated at the MTD but experienced a DLT of febrile neutropenia during cycle 1 and discontinued, per protocol. A follow-up scan showed a decrease in the size and number of his lung metastases, suggesting a delayed response to treatment.
Overall, progressive disease was observed as the best response in 21 response-evaluable patients (54%). The number of treatment cycles received by each patient according to GCC expression level is shown in Fig. 3. GCC expression, as measured by H-score, did not appear to correlate with treatment duration.
Median PFS for the response-evaluable population was 44 days [95% confidence interval (CI), 39–83]. Using a Cox proportional hazard model, there was no association between GCC expression and PFS (HR, 1.002; P = 0.6052). There was also no difference in PFS when dichotomized by median GCC expression (HR, 0.806; P = 0.6545).
TAK-264 appeared generally well tolerated with a manageable safety profile in patients with advanced, GCC-positive gastrointestinal malignancies. The MTD for TAK-264, determined using a Bayesian CRM (30–32), was 1.8 mg/kg. A total of 19 patients were treated during the dose-escalation phase to establish the MTD, with eight of these patients receiving a dose below the MTD. An advantage of the CRM algorithm used in this study was that it allowed an accurate MTD to be determined with a minimal number of patients treated at suboptimal levels, while utilizing information from all treated patients to guide the dose modifications (33). It also allowed rapid dose escalation, was completed over approximately 12 months, through seven increasing doses.
DLTs were reported in four of 19 patients in the dose-escalation phase of the study. Three of the four patients with DLTs of grade 4 neutropenia were treated at doses above 1.8 mg/kg (MTD), whereas one patient who experienced grade 4 febrile neutropenia was treated at the MTD. Of the three patients treated at doses above the MTD, one patient also experienced grade 3 QTc prolongation. The DLTs occurred during days 10 to 15 of cycle 1, approximately 7 to 12 days after the MMAE Tmax of approximately 3 days; MMAE has a plasma half-life of approximately 3 days, and neutrophils have a circulation life of approximately 5 days (34). The mechanism by which TAK-264 causes neutropenia is not yet understood, but free MMAE is likely the cause, as GCC has not been found to be expressed in the bone marrow compartment; therefore, a role for this receptor in triggering neutrophil depletion seems unlikely. In other studies using microtubule-targeting agents such as taxanes, hematologic AEs were relatively common (35). It appears that neutrophils may be more susceptible to this toxicity than other bone marrow–derived cells.
Other common drug-related AEs included nausea, fatigue, decreased appetite, and diarrhea, which are often reported in patients with gastrointestinal malignancies. Interestingly, patients receiving brentuximab vedotin, who did not suffer from gastrointestinal malignancies, experienced a spectrum of AEs which included nausea, fatigue, neutropenia, diarrhea, and pyrexia (3–6), which are to some extent comparable with the AEs reported for TAK-264. The ADCs polatuzumab vedotin and pinatuzumab vedotin were also associated with similar toxicities, including neutropenia, peripheral neuropathy, and diarrhea (36). These ADCs incorporate the same cytotoxic agent as TAK-264, MMAE; it is assumed that the cytotoxicity of MMAE is related to its ability to inhibit cell division by binding tubulin, which arrests the target cell in the G2–M stage of the cell cycle and results in apoptosis (3). In contrast to these studies of other ADCs, peripheral neuropathy was not a commonly reported AE with TAK-264, occurring at grade 1 or 2 in 10% of patients. It should be noted, however, that the majority of patients participating in this study received two cycles of therapy. No ATA-positive samples were detected during this study, and no infusion site reactions occurred; this was consistent with expectations as TAK-264 utilizes a fully humanized antibody.
PK data, including a median half-life of approximately 4 days for TAK-264, suggest that steady-state PK for both the ADC and MMAE occurred by approximately 21 days, supporting the dosing schedule employed. Concentration–time profiles showed that TAK-264, total antibody, and MMAE concentrations peaked following infusion then decreased to low preinfusion concentrations prior to the next cycle, suggesting a lack of substantial accumulation in plasma or serum.
Preliminary response data indicated antitumor activity in this population, especially in patients with non–metastatic colorectal cancer. The efficacy of ADCs using MMAE in gastrointestinal malignancies is unproven, although they have been used successfully in Hodgkin lymphoma and anaplastic large cell lymphoma (3–6). Microtubule-disrupting agents have been shown to have varying effects in different tumor cell lines and may work by different mechanisms depending on cell type (26, 37). In this study, duration of disease stabilization and antitumor activity did not appear to correlate with GCC expression level. In total, 52% of patients in the expansion cohort had high levels of GCC expression. It is intriguing that 4 patients for whom there were early signals of potential clinical benefit, all non–metastatic colorectal cancer, had low levels of GCC expression relative to the overall population. Further investigation is needed to assess the role of GCC expression in clinical outcomes with TAK-264.
Several factors are involved in the ADC treatment response including the toxin selection, binding affinity of the antibody, delivering the ADC and cytotoxic agent to the tumor cells, and linker technology. While the data from this study are preliminary, there was less than expected clinical efficacy. The selection of GCC as a target remains experimental, and more data are needed to better assess the potential of targeting GCC to develop a successful treatment strategy for gastrointestinal malignancies. Various characteristics need to be better defined (such as heterogeneity and specificity of target expression, internalization rate, intracellular trafficking) to interpret the limitation of clinical efficacy.
Tissues tested for GCC expression were mostly archived tissues, including some from primary tumor during initial resection.
The relationship between primary and metastatic sites, and the effects of multiple lines of therapy, on GCC staining characteristics is currently unknown. Further insight into the optimal design characteristics of effective ADCs will be best gained as additional clinical data become available. Several ADCs are currently in clinical development, and TAK-264 is undergoing further investigation in two single-arm, phase II clinical studies in patients with gastric (NCT02202759) and pancreatic cancer (NCT02202785).
In conclusion, TAK-264 is a first-in-class ADC with a novel target, the GCC receptor. The data from this first-in-human phase I study suggest that TAK-264 has a manageable safety profile at the MTD of 1.8 mg/kg. Preliminary data suggest antitumor activity and early signs of clinical benefit in patients with pancreatic, esophageal, and gastric carcinoma. Combination strategies with active chemotherapy agents could be considered in the future. Antitumor activity did not appear to correlate with GCC expression levels. Evolving clinical data from the ongoing phase II clinical trials of TAK-264 will provide critical insight into the design of next-generation ADCs.
Disclosure of Potential Conflicts of Interest
J.E. Faris is an employee of Novartis, and is a consultant/advisory board member for N-of-One Therapeutics. D.P. Ryan is an employee of MPM Capital. W.A. Messersmith reports receiving commercial research grants from Millenium/Takeda. No potential conflicts of interest were disclosed by the other authors.
Conception and design: K. Almhanna, T. Kalebic, T. Wyant, A. Fasanmade, W.A. Messersmith, J. Rodon
Development of methodology: K. Almhanna, T. Kalebic, T. Wyant, A. Fasanmade
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T. Kalebic, C. Cruz, J.E. Faris, D.P. Ryan, T. Wyant, A. Fasanmade, W.A. Messersmith, J. Rodon
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): K. Almhanna, T. Kalebic, D.P. Ryan, J.A. Jung, T. Wyant, A. Fasanmade, W.A. Messersmith, J. Rodon
Writing, review, and/or revision of the manuscript: K. Almhanna, T. Kalebic, C. Cruz, D.P. Ryan, J.A. Jung, T. Wyant, A. Fasanmade, W.A. Messersmith, J. Rodon
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T. Kalebic, A. Fasanmade
Study supervision: T. Kalebic, A. Fasanmade, W.A. Messersmith
This study was funded by Millennium Pharmaceuticals Inc., a wholly owned subsidiary of Takeda Pharmaceutical Company Limited.
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.
The authors thank Helen Johns of FireKite, an Ashfield company, part of UDG Healthcare plc, for writing support during the development of this article, which was funded by Millennium Pharmaceuticals, Inc. ADC technology was licensed from Seattle Genetics, Inc.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).
W. Messersmith and J. Rodon share senior authorship.
- Received October 12, 2015.
- Revision received April 14, 2016.
- Accepted April 18, 2016.
- ©2016 American Association for Cancer Research.