BIW-8962, an Anti-GM2 Ganglioside Monoclonal Antibody, in Previously Treated Advanced/Recurrent Lung Cancer: a Phase I/II Study

BIW-8962 is a monoclonal antibody to GM2 ganglioside that shows pre-clinical activity towards lung cancer cell lines and in an animal model bearing small-cell lung cancer (SCLC) xenografts. In phase I, patients (N=16) with advanced, recurrent lung cancer received BIW-8962 (1-10 mg/kg) intravenously every 3 weeks. There were no dose-limiting toxicities and the maximum tolerated dose was not established. The highest dose (10 mg/kg) was administered to patients with advanced, recurrent SCLC (N=21) in phase II. The phase II study was prematurely terminated due to lack of efficacy. Objective response rate was 5% (95% CI: 0.1-24.9%) in the efficacy evaluable population (N=20). One patient showed a durable partial response and there were a few with stable disease, which was generally not durable. No pattern of consistent toxicity was observed across the phases: there were no treatment-related adverse events (AEs) grade ≥ 3, serious AEs, AEs leading to discontinuation of BIW-8962, or deaths. Exploratory analysis of circulating tumor cells and other potentially predictive orpharmacodynamic markers did not reveal any results consistent with an effect from BIW-8962. Given the complete lack of response in the study population, further development of BIW8962 has been discontinued. The reason for the lack of clinical activity with BIW-8962 is unknown. It is hoped that the negative findings of this study will contribute to other investigations of GM2 as a therapeutic target, possibly by combination therapy, and of alternative tumor markers in patients with SCLC.


Introduction
Lung cancer is estimated as the most commonly diagnosed and the leading cause of cancer death globally [1]. Small-cell Lung Cancer (SCLC) comprises ~15% of all lung cancers and is the most aggressive form of the disease [2]: survival is inferior compared to Non-smallcell Lung Cancer (NSCLC) [3]. It is characterized by rapid growth and a tendency to metastasize early in the disease course. About one-third of patients with SCLC present with limited-stage disease, which is usually responsive to first-line therapy (usually with cisplatin and etoposide) although survival remains poor. The disease typically develops drug resistance during first-line therapy in almost all cases, resulting in recurrences that respond poorly to additional treatment [4]. Patients with limited-stage SCLC treated with etoposide, concurrent thoracic radiotherapy, and subsequent prophylactic cranial radiotherapy may expect a representative median Overall Survival (OS) of 14.5 months, and 2-and 5-year OS rates of 25% and 10%, respectively [5]. In extensive-stage SCLC, etoposide chemotherapy can palliate symptoms and prolong OS [4]: median OS for patients with extensive-stage disease has remained fairly constant at 9-10 months regardless of whether etoposide is dose intensified or whether irinotecan is substituted for etoposide [6,7]. Objective Response Rates (ORRs) to second-line chemotherapies tend to range between 12% and 25% and median OS from the start of second-line therapy is only 4-6 months [8][9][10]. There is consequently an unmet need for novel, more effective agents in the treatment of SCLC.
Cell membrane gangliosides comprise a carbohydrate chain with sialic acid at the cell surface and a hydrophobic ceramide in the lipid bilayer [11]; some are involved in cell-cell recognition [12] and cell-matrix attachment [13] that regulate cell growth and differentiation [14,15]. Gangliosides have been proposed as therapeutic targets for cancer therapy [16,17] as their quantitative and qualitative expression can increase during cellular oncogenic transformation [12]. GM2 ganglioside is a tumor-associated antigen that is over expressed in a high proportion of several malignancies, e.g., SCLC, NSCLC, mesothelioma, melanoma, neuroblastoma, multiple myeloma [18][19][20][21].
BIW-8962 is a recombinant, humanized, nonfucosylated immunoglobulin G1 monoclonal antibody (mAb) directed against GM2 ganglioside. BIW-8962 is produced in Chinese Hamster Ovary cells that lack the FUT8 gene, rendering the mAb devoid of fucose. Nonfucosylated mAbs have up to 100-fold higher antibodydependent cell-mediated cytotoxicity (ADCC) against tumor cells compared to conventional fucosylated mAbs [22]. Assessment of the binding activity of KM8969 (a precursor mAb of BIW-8962 with the same complementarity-determining regions) in an enzymelinked immunoassay using a series of immobilized gangliosides showed strong reaction with N-acetyl-GM2 and N-glycolyl-GM2 but weak reaction with GD2 [23]. BIW-8962 (and/or KM8969) demonstrated ADCC and Complement-dependent Cytotoxicity (CDC) against GM2-positive human SCLC [23,24] or mesothelioma [21] cell lines using human peripheral blood mononuclear cells as the effector cells or human serum as the complement source, respectively, in vitro assays. In vivo activity of BIW-8962 has been demonstrated in murine SCID models usingGM2-positive SCLC or mesothelioma cells [21,24]. The in vivo antitumor activity of BIW-8962 (3 mg/kg intravenously [IV] once weekly for 3 weeks) has been demonstrated in a nude rat xenotransplant model using a human SCLC cell line, NCI-N417, with no severe toxicity [data on file, Kyowa Kirin Pharmaceutical Development, Inc.]. BIW-8962 has been investigated in a phase I clinical study in patients with heavily pre-treated multiple myeloma: the maximum tolerated dose (MTD) was not reached at up to 3 mg/kg IV every 2 weeks and no clinical activity was demonstrated at this dose using monotherapy [25].
The aim of the current phase I/II study was to determine the safety, tolerability, MTD, pharmacokinetics, potential immunogenicity, and preliminary clinical efficacy of BIW-8962 administered as monotherapy in patients with previously treated advanced/recurrent lung cancer.

Study design
The primary objectives of the phase I study were to determine the safety, tolerability, and MTD/ recommended phase II dose (RP2D) of BIW-8962 in patients with advanced/recurrent lung tumors. Secondary objectives in phase I was to evaluate preliminary efficacy, determine the pharmacokinetic profile of BIW-8962, and screen for potential development of antibodies against BIW-8962. The primary objective of the phase II study was to determine the preliminary efficacy of BIW-8962 in patients with advanced/recurrent SCLC. Secondary objectives in phase II were to further evaluate the safety and tolerability of BIW-8962, and to screen for potential development of antibodies against BIW-8962.
The starting dose and schedule for administration of BIW-8962 in phase I was 1 mg/kg IV every 3 weeks. The starting dose level and its escalation were based on the safety and pharmacokinetic data from the first-in-human study of BIW-8962 in patients with relapsed/recurrent multiple myeloma, which examined doses up to 3 mg/kg IV every 2 weeks without reaching the MTD [25].
Phase I was conducted at a single study center (Samsung Medical Center) employing a standard 3+3 dose-escalation design with increasing doses of BIW-8962 (1, 3, 6, and 10 mg/kg) to determine the MTD and RP2D. Patients not receiving at least two full doses of BIW-8962 in a dose cohort were replaced, except for those who experienced BIW-8962-related toxicity. Doselimiting toxicity (DLT) was defined as any of the following treatment-related toxicities: grade 3 or 4 infusion reaction; grade 4 anemia or thrombocytopenia; grade 4 neutropenia ≥5 days despite treatment with granulocyte colony stimulating factor; grade 3 or 4 neutropenia if accompanied by fever ≥ 38.5°C for ≥ 4 hours; grade ≥ 3 non-hematologic toxicity (except nausea/vomiting or diarrhea that can be reduced to grade ≤ 2 within 24 hours with medical management, or grade 3 laboratory adverse event (AE) that is asymptomatic or rapidly reversible); and any other toxicity considered a clinically significant hazard to the patient.
Phase II was conducted at all five study centers using the RP2D for BIW-8962 in a maximum of 40 efficacy evaluable patients. Phase II included a Bayesian stopping rule for futility (lack of efficacy), with assessment of futility initiated after objective response of the first 15 patients was confirmed.
BIW-8962 was administered after dilution in 250 mL normal saline by IV infusion through a 0.22 µm proteinsparing/low-protein binding in-line filter over ≥ 2 hours. Prophylactic pre-medication for potential infusion-related reactions was administered 30-60 min before BIW-8962 administration according to the standard practice of each study center. Patients were allowed to continue treatment until disease progression, unacceptable toxicity, grade 3/4 infusion reactions, or withdrawal of consent.

Adult (≥ 19 years) patients of either gender with Eastern
Cooperative Oncology Group (ECOG) performance status 0 or 1 and a histopathological or cytological diagnosis of measurable or non-measurable, unresectable, locally advanced primary recurrent, metastatic SCLC, NSCLC, or pleural mesothelioma unresponsive to standard therapy or for which there was no standard therapy available were eligible for inclusion in phase I. This was restricted to patients with SCLC exclusively in phase II. Other inclusion criteria included: life expectancy > 3 months; adequate hematologic, hepatic, renal, and cardiac function; and preserved lung function (SpO 2 ≥ 92% on room air only in phase I and ≥ 94% by pulse oximeter with or without supplemental oxygen in phase II). Standard clinical trial advice was given to men and women to preclude and avoid conception.
Exclusion criteria included: use of chemotherapy, oral biologic agents, hormonal therapy, radiotherapy (unless limited to non-visceral structures), immunotherapy, or investigational agents < 3 weeks prior to the first dose of BIW-8962; use of mAb of any type or for any form of disease < 4 weeks prior to the first dose; major surgery < 4 weeks prior to the first dose; known symptomatic brain metastases; leptomeningeal disease; uncontrolled intercurrent illnesses; known HIV disease; psychiatric illness, disability, or social situation likely to compromise safety, ability to provide consent, or limit compliance with study requirements; known hypersensitivity to mAbs or other therapeutic proteins, or any component of the BIW-8962 formulation, and the reaction could not be subsequently controlled or prevented with standard treatment (e.g., antihistamines, 5-HT3 antagonists, corticosteroids); history of other malignancy except for non-melanomatous skin cancer, cervical carcinoma in situ, or other primary solid tumor treated with curative intent and no known active disease present and no treatment administered for > 3 years; neurological paraneoplastic syndrome; and women who were pregnant.

Clinical and safety assessments
Demographics and medical/cancer histories were recorded together with pregnancy testing if applicable, echocardiography or multigated acquisition scan, and contrast magnetic response brain imaging at screening. Physical examination including vital signs, laboratory tests (blood chemistry, hematology, coagulation parameters, and urinalysis), 12-lead Electrocardiogram (ECG), and determination of antibodies to BIW-8962 using a bridging electrochemiluminescence assay were undertaken at screening and regular intervals during treatment. All patients were followed after the last dose until confirmation of disease progression or start of alternative treatment.
Response was evaluated by imaging and physical examination every 6 weeks. Best response (ORR and Clinical Benefit Rate [CBR]) was determined using Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 criteria [26] in the efficacy evaluable population, which included those patients with baseline and at least one on-study assessment for response. Confirmation of response was required ≥ 4 weeks after initial response. OS, progression-free survival (PFS), and response duration were estimated using the Kaplan-Meier method. OS and PFS were defined as the time from the date of the first dose of study drug until the first date on which death due to any cause or progressive disease was documented; patients were censored at the date of the last evaluable tumor assessment for patients who were lost to followup or who were alive at the time of analysis.
AEs were recorded following observations by the investigator during clinic visits or in response to nonleading questions, spontaneous reporting by the patient, or on the basis of clinical or laboratory tests. They were graded by National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) v4.0. And classified by the investigator with respect to relationship to treatment with BIW-8962 (definitely, probably, possibly, unlikely or unrelated). Treatment-related AEs included those considered definitely, probably, or possibly related to BIW-8962. The safety analysis population included all patients who received at least one dose of BIW-8962. SAEs were reported in an expedited manner.

Pharmacokinetics
Blood samples were taken pre-dose, and at 0 (end of infusion), 2, 6-8, 24, 48, 96, 192, 288, and 360 hours following the first and third dose of BIW-8962 as well as pre-dose following the second and fourth dose in phase I. Serum samples were analyzed at a central laboratory using a validated sandwich electrochemiluminescence assay. Pharmacokinetic parameters including area under the serum concentration-time curve from time zero extrapolated to infinity (AUC ∞ ) and over the dose interval (AUC 21d ), maximum serum concentration (C max ), time to C max (T max ), total systemic clearance (CL), volume of distribution at steady state (V ss ), elimination half-life (t 1/2 ), and mean residence time extrapolated to infinity (MRT ∞ ) were calculated using non-compartmental methods with WinNonlin software (Pharsight-A Certara Company, Mountain View, CA).

Exploratory assessments
When pre-treatment tumor biopsy samples were available, GM2 positivity was determined by immunohistochemistry (IHC) staining using BIW-8962 as the detection agent. Exploratory objectives in phase I were to assay circulating tumor cells(CTCs) in blood using the Cell search™ immunomagnetic enrichment system and to detect CTC GM2 positivity by fluorescence imaging before and after exposure to BIW-8962, and, in phase II, to assay exploratory variables including GM2 synthase mRNA, GM2 activator protein, chemokines (e.g.,IP-10, Mip-1alpha), and complement consumption in blood before and after exposure to BIW-8962 as potential pharmacodynamic and/or predictive biomarkers.

Statistics
The sample size for phase II was based on the onesided upper-tailed test for a single one sample proportion at a significance level of 0.025 and 20% beta (80% power). The threshold value of ORR under the null hypothesis and target ORR and the value under the alternative hypothesis were set to 5% and 20%, respectively. This test was equivalent to constructing a 95% confidence interval (CI) on the ORR and rejected the null hypothesis if the lower limit of the CI exceeded 5%. Under these assumptions, a sample size of 40 efficacy evaluable subjects provided ≥ 80% probability to demonstrate that the lower limit of the exact 95% CI for ORR exceeded the threshold ORR of 5%. Safety, efficacy, and pharmacokinetics were summarized by descriptive statistics. Response data were reported with the exact 95% CI.

Patient characteristics
The study is complete and was conducted between 10June 2013 and 23June 2016. The baseline clinical and demographic characteristics of the patients enrolled in phases I and II are summarized in table 1.Patients in phase I had SCLC (n=8) or NSCLC (n=8) and those in phase II exclusively had SCLC (n=21). All 16 patients in phase I were included in safety and efficacy populations. The safety and efficacy populations in phase II included 21 and 20 patients, respectively: one patient was excluded from efficacy evaluation because of a lack of postbaseline tumor assessment. Patient disposition and drug exposure are summarized in table 2. In both phase I and II, all patients received the total planned dose. Infusion of BIW-8962 was temporarily interrupted in one patient (6.3%) due to grade 2 infusion-related reaction in the cohort receiving 6 mg/kg in phase I and in two patients (9.5%) due to grade 2infusion-related reaction and grade 2 hypoglycemia, respectively. It was only possible to obtain pre-study biopsy samples for two patients with SCLC (1 each in phase I and phase II), both of which showed cell surface GM2 over expression of moderate intensity (50-75% staining) on IHC testing.

Dose-limiting toxicity
None of the patients in phase I developed DLT. The MTD was not reached. The RP2D was therefore the highest tested dose of 10 mg/kg every 3 weeks in the phase I trial.

Safety
AEs are summarized in table 3. Of particular note, there were no treatment-related AEs grade ≥ 3, serious AEs (SAEs), AEs leading to discontinuation of BIW-8962, or deaths in either phase. In phase I, overall treatmentrelated AEs (grade 1 or 2) did not appear related to dose across the cohorts and, by preferred term, were reported in individual patients except for infusion-related reaction (n=2). In phase II, overall treatment-related AEs (grade 1 or 2) occurred in seven of 21 (33.3%) of patients and, by preferred term, were reported in individual patients except for fatigue (n=3), and decreased appetite, increased Alanine Aminotransferase (ALT), and increased Aspartate Aminotransferase (AST) (each n=2). The only SAEs occurred in one patient receiving 10 mg/kg in phase I (atelectasis) and three patients receiving 10 mg/kg in phase II (2 with fatigue and 1 with hypercalcemia), all of which were considered unrelated to BIW-8962.
No unexpected trends or safety concerns were identified from laboratory parameter, vital sign, or ECG assessments. Anti-BIW-8962 antibodies were not detected in serum of any patient before or following treatment.
Among the 20 efficacy-evaluable patients in phase II, ORR was 5% (95% CI: 0.1-24.9%) and CBR was 15% (95% CI: 3.    BMI body mass index, ECOG Eastern Cooperative Oncology Group, NSCLS, non-small cell lung cancer, SCLC, small cell lung cancer Note: Percentages may not equal 100% exactly due to rounding. 8962 1-10 mg/kg are shown in figure 2. Given the limited numbers of patients evaluable following repeated cycles of BIW-8962 (n=1 for 3 mg/kg and n=1-3 for 10 mg/kg), any results should be interpreted cautiously. In Cycle 3, t 1/2 ranged from 211 to 352 hours for the 3 and 10 mg/ kg doses. Accumulation for BIW-8962 based on C max , trough serum concentration (C min ), and AUC 21d in Cycle 3 was modest for the 3 and 10 mg/kg doses ranging from 1.27 to 1.80. Visual inspection of BIW-8962 C min values following repeated administration of BIW-8962 indicated that steady state was reached at the end of Cycle 3.

Exploratory analyses
Determination of CTCs in phase I revealed measurable CTC counts (median 5 [range, 2-19] per 7.5 mL of whole blood) at baseline prior to BIW-8962 in six of 16 patients. After dosing with BIW-8962, CTCs were detected in 9 patients (median 3 [range, 1-40] per 7.5 mL of whole blood). Only one patient had a CTC sample that was GM2 positive and that was in a post-dose sample. None of the other pharmacodynamic biomarker assessments revealed predictive changes in relation to BIW-8962 administration or data collected were insufficient for meaningful analysis.     Note: n=3, 4, 3, and 6 for 1, 3, 6, and 10 mg/kg, respectively, in Cycle 1; n=1 and 3 for 3 and 10 mg/kg, respectively, in Cycle 3; n=2 for 10 mg/kg in Cycles 4, 5, and 6; and n=1 for 10 mg/kg in Cycle 7.

Discussion
This study was prematurely terminated due to lack of efficacy. ORR was 5% (95% CI: 0.1-24.9%) in phase II, which did not meet the efficacy hypothesis of the study, i.e., the lower limit of the exact 95% CI for ORR was 0.1%. Clinical development of BIW-8962 has been discontinued.
In phase I, the MTD of BIW-8962 was not reached within the dose range selected for study, so the RP2D was set at 10 mg/kg every 3 weeks, which was the highest tested dose. No DLTs were detected during phase I and no consistent pattern or severity of treatment-induced toxicity was noted in phase II. It might therefore be considered that the upper limit of the dose range studied in phase I might have been insufficient and a maximally effective dose was not reached for phase II study. However, C min values of BIW-8962 were within the range which would have been expected to cause cytotoxicity if in vitro data against SCLC cell lines and in the preclinical animal SCLC model that showed activity for BIW-8962 (see Introduction) were extrapolated to patients. The reason for the lack of in vivo activity is unknown. It may be that pre-clinical activity in vitro and in vivo for BIW-8962 does not translate in patients. Similar results concerning lack of efficacy for BIW-8962 have been found during the first-in-human phase I study in patients with relapsed/refractory multiple myeloma [25]. The potential synergism of BIW-8962 with other agents has not been investigated pre-clinically in animal lung tumor models and remains a possible option for investigation. None of the predictive biomarkers that were investigated showed pharmacodynamic-like changes following treatment with BIW-8962 that may have been indicative of a therapeutic effect, which may also be reflective of a lack of efficacy by BIW-8962. A limitation of the current study was that GM2 positivity was only able to be determined in two patients from biopsy samples, both of whom had SCLC GM2-positive tumors: one of whom had SD and the other was not evaluable for efficacy. The unlikely recruitment of predominance of patients with GM2-negative lung tumors would bias against detection of a therapeutic effect by BIW-8962.
BIW-8962 exposure increased in a generally doseproportional manner across the dose range of 1-10 mg/kg following a single IV infusion. In general, the pharmacokinetic properties of BIW-8962 were similar following single and multiple administrations with mean t 1/2 at approximately 235-324 hours. Accumulation of BIW-8962 was less than 2-fold and it appeared that stable BIW-8962 serum levels were reached as early as the end of Cycle 3 following multiple dose administration. The pharmacokinetic properties of BIW-8962 were similar to those previous reported following single-dose IV infusion in patients with relapsed/refractory multiple myeloma [25].
Only one patient demonstrated a response (PR) across phases I and II. Few patients demonstrated stable disease in phase I (n=4, including 3 at 10 mg/kg) or phase II (n=3), with little evidence of durable stable disease. The one patient with PR in phase II achieved response throughout the study and subsequently received BIW-8962 as part of a compassionate use program, attaining PFS of 463 days and duration of response of 382 days. This patient was a 46-year-old woman who had never smoked and was diagnosed with stage IV SCLC (T4N0M1b); she had received radiation therapy to the right lung and whole brain with no surgery; and she had received four lines of prior chemotherapy prior to BIW-8962 administration. The GM2 over expression status of her tumor was not determined as pre-study biopsy samples were not available.
In conclusion, further development of BIW-8962 has been discontinued given the complete lack of response in the study population. It is hoped that the negative findings of this study will contribute to other investigations of GM2 as a therapeutic target, possibly by combination therapy, and of alternative tumor markers in patients with SCLC.