Phase I Study of P-Cadherin–Targeted Radioimmunotherapy with Y-FF

Published OnlineFirst August 18, 2020; DOI: 10.1158/1078-0432.CCR-20-0037

CLINICAL CANCER RESEARCH | CLINICAL TRIALS: TARGETED THERAPY

Phase I Study of P-cadherin–targeted Radioimmunotherapy with 90Y-FF-21101 Monoclonal Antibody in Solid Tumors A C Vivek Subbiah1, William Erwin2, Osama Mawlawi2, Asa McCoy2, David Wages3, Catherine Wheeler3, Carlos Gonzalez-Lepera4, Holly Liu1, Homer Macapinlac4, Funda Meric-Bernstam1, David S. Hong1, Shubham Pant1, Dao Le4, Elmer Santos4, Jose Gonzalez2, Jason Roszik5, Takeaki Suzuki3, Ruth Ann Subach3, Timothy Madden3, Mary Johansen3, Fumiko Nomura6, Hirokazu Satoh6, Tadashi Matsuura6, Masamichi Kajita7, Eri Nakamura7, Yuichi Funase7, Satoshi Matsushima7, and Gregory Ravizzini4

ABSTRACT ◥ Purpose: 90Y-FF-21101 is an Yttrium-90–conjugated, chimeric in the majority of patients. Organ dose estimates for all patients were mAb that is highly specific for binding to human placental (P)- below applicable limits. P-cadherin expression H-scores ranged cadherin, a cell-to-cell adhesion molecule overexpressed and associ- from 0 to 242 with 40% of samples exhibiting scores ≥100. FF-21101 ated with cancer invasion and metastatic dissemination in many protein pharmacokinetics were linear with increasing antibody cancer types. We report the clinical activity of 90Y-FF-21101 in a first- dose, and the mean half-life was 69.7 (12.1) hours. Radioactivity in-human phase I study in patients with advanced solid tumors. clearance paralleled antibody clearance. A complete clinical Patients and Methods: The safety and efficacy of 90Y-FF-21101 response was observed in a patient with clear cell ovarian carcino- were evaluated in a phase I 3þ3 dose-escalation study in patients ma, correlating with a high tumor P-cadherin expression. Stable with advanced solid tumors (n ¼ 15) over a dose range of disease was observed in a variety of other tumor types, without dose- 5–25 mCi/m2. Dosimetry using 111In-FF-21101 was performed limiting toxicity. 1 week prior to assess radiation doses to critical organs. Patients Conclusions: The favorable safety profile and initial antitumor who demonstrated clinical benefit received repeated 90Y-FF-21101 activity observed for 90Y-FF-21101 warrant further evaluation of administration every 4 months. this radioimmunotherapeutic (RIT) approach and provide initial Results: 111In-FF-21101 uptake was observed primarily in the clinical data supporting P-cadherin as a potential target for cancer spleen, kidneys, testes, lungs, and liver, with tumor uptake observed treatment.

Introduction forming extracellular cadherin (EC) domains. These EC domains provide the means for cadherins to bind to and interact with other Cadherins are cell-surface glycoproteins that mediate calcium- cells and extracellular matrix proteins. Five EC domains have been dependent cell–cell adhesion required for the formation of solid identiﬁed, EC1–EC5, with EC1 playing a critical role in mediated tissue (1). The so-called classic cadherins, epithelial (E), neural (N), cell–cell adhesion. and placental (P)-cadherin, were among the earliest studied and are Because cadherins have been shown to be overexpressed in a variety known to be associated with adherens junctions of cells. Their struc- of tumors, they are believed to play a role in effecting malignant cell tures are notable for having repeated 110 amino acid sequences, behavior (2) and, accordingly, may be an attractive target for tumor- speciﬁc therapies. Results from numerous preclinical models have now

1 shown that targeting or disrupting cadherins results in antitumor Department of Investigational Cancer Therapeutics, Division of Cancer Medi- – CDH3 cine, The University of Texas MD Anderson Cancer Center, Houston, Texas. activity (2 5). P-cadherin, encoded by the gene , represents a 2Department of Imaging Physics, Division of Diagnostic Imaging, The University particularly attractive target because it is overexpressed in a variety of of Texas MD Anderson Cancer Center, Houston, Texas. 3FUJIFILM Pharmaceu- tumors, with minimal expression in normal tissues of adult humans ticals U.S.A., Inc., Cambridge, Massachusetts. 4Department of Nuclear Medicine, (Supplementary Fig. S1; refs. 2, 4, 6–9). Park and colleagues and Zhang Division of Diagnostic Imaging, The University of Texas MD Anderson Cancer and colleagues previously characterized an unmodified, humanized 5 Center, Houston, Texas. Department of Melanoma Medical Oncology, Division mAb, PF-03732010, directed against P-cadherin (5, 10), which showed of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, Texas. 6Perseus Proteomics, Inc., Tokyo, Japan. 7FUJIFILM Toyama promising antitumor activity in preclinical models. However, although Chemical Co., Ltd., Chiba, Japan. PF-03732010 was well tolerated in initial human trials, it did not demonstrate clinical efficacy (9, 11), suggesting that the unmodified Note: Supplementary data for this article are available at Clinical Cancer – Research Online (http://clincancerres.aacrjournals.org/). anti P-cadherin antibody may not be suitable for the treatment of solid tumors. Corresponding Author: Vivek Subbiah, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd – Unit 455, Houston, TX 77030. Phone: 713- More recently, targeted antibodies or peptides have been conjugated 563-1930, Fax: 713-792-0334; E-mail: [email protected] with radionuclides to enhance their antitumor activity. This approach has been used successfully to treat well-differentiated neuroendocrine Clin Cancer Res 2020;XX:XX–XX tumors of the pancreas (12, 13), prostate cancer (14), and lympho- doi: 10.1158/1078-0432.CCR-20-0037 mas (15). In this paper, we describe the initial Phase 1 study results for a 2020 American Association for Cancer Research. chimeric human/mouse immunoglobulin G (IgG1) monoclonal

AACRJournals.org | OF1

Subbiah et al.

all study-related procedures. Adult patients must have been ≥18 years Translational Relevance of age with histologically or cytologically confirmed advanced solid Cadherins are cell-surface glycoproteins that mediate calcium- tumor malignancy and refractory or relapsed from prior therapy with dependent cell–cell adhesion required for the formation of solid no available therapy likely to provide clinical benefit or for whom no tissue. Several cadherins, including placental (P)-cadherin, are alternative therapy is available. Patients had to be at least 3 weeks overexpressed in many tumor types, correlating with increased beyond the last chemotherapy (or ≥5 half-lives, whichever was tumor cell motility and invasiveness. P-cadherin is of particular shorter), radiotherapy, major surgery, or experimental treatment and interest clinically because, while highly expressed in many tumor recovered from all acute toxicities (≤ grade 1). Adequate performance types, it is minimally expressed in normal tissue, making it a status [Eastern Cooperative Oncology Group (ECOG) ≤ 2], hemato- selective antitumor target. Promising activity has been demon- logic parameters, renal and hepatic function, and at least one mea- strated in preclinical models using monoclonal antibodies targeting surable disease site that meets target lesion requirements by Response P-cadherin. Here, we report the clinical activity of 90Y-FF-21101, Evaluation Criteria in Solid Tumors (RECIST) v1.1, were required. an anti–P-cadherin antibody conjugated to a radionuclide to Dose escalation proceeded in a standard 3þ3 design with obser- enhance its antitumor activity. In this first-in-human phase I study, vation for dose-limiting toxicity (DLT) for 6 weeks post the initial 90Y-FF-21101 was well tolerated and demonstrated signs of anti- treatment. A minimum of 6 weeks elapsed between the last subject tumor activity in patients with advanced solid tumors, including a dosed in a cohort and the first subject in a new cohort. The highest complete response in a patient with ovarian cancer. These results 90Y-FF-21101 level below the level eliciting DLT would be declared the provide initial clinical data supporting the potential for P-cadherin maximum tolerated radioactivity. The dose-escalation phase was as a target for cancer treatment using a radioimmunotherapeutic followed by an expansion phase, which is currently ongoing. approach. Radiolabeled antibody preparation 111 Radiolabeling of FF-21101 with InCl3 for dosimetry or with 90 YCl3 for treatment was carried out by the clinical site radio- antibody directed against P-cadherin, conjugated to Yttrium-90 (90Y) pharmacy. The mAb conjugated to DOTA (FF-21101) was provided for the treatment of solid tumors (90Y-FF-21101). Because 90Y emits as a 5 mg/mL solution in 250 mmol/L sodium acetate, pH 5.5, by beta particles with an average penetration of 4 mm in soft tissue, the FUJIFILM Diosynth Biotechnologies U.S.A., Inc. A clinical formula- beta ray from a 90Y-labeled antibody renders DNA damage to tion buffer [25 mmol/L sodium acetate, 2 mg/mL ascorbic acid, the bound cells and also to the adjacent or more distant cells due 0.3 mg/mL diethylenetriaminepentaacetic acid (DTPA), and 0.72% to its longer range compared with other beta emitters (16). Our sodium chloride, pH 5.5] was provided for dilution. 111In chloride preclinical research has shown preferential uptake of 90Y-FF-21101 sterile solution (Mallinckrodt, Inc.) and 90Y chloride sterile solution in P-cadherin–expressing tumors, and antitumor activity in mouse (Eckert & Ziegler Radiopharma, GmbH) were used for radiolabeling. xenograft models of ovarian and lung adenocarcinoma (Supple- Solutions were prepared to maintain a specific radioactivity of 1 mCi/ mentary Figs. S1–S3; ref. 17, 18). On the basis of these results, the mg (10%) 111In-FF-21101 for dosimetry and 8 mCi/mg (10%) for first-in-human phase I trial of 90Y-FF-21101 was conducted to evalu- 90Y-FF-21101 at the time of injection. Validation procedures were ate the dosimetry, safety and tolerability, pharmacokinetics, and performed at each site prior to patient dose preparation. Three antitumor activity in patients with advanced solid tumors. consecutive batches of 40 mCi 90Y-FF-21101 were prepared and tested against acceptance criteria for appearance, radioactivity, radiochemical purity, pH, bacterial endotoxin, sterility, and stability. Quality Patients and Methods control testing of each patient dose was also performed prior to Trial design and patient eligibility administration including visual inspection (clear, colorless to pale The phase I, first-in-human, open-label, dose-escalation study was yellow, free of visible particular matter) and radiochemical purity by conducted at the University of Texas MD Anderson Cancer Center instant thin-layer chromatography (≥95%). Other quality attributes (MDACC, Houston, TX), in accordance with the Declarations of of prepared drug doses (purity by SEC, immunoreactivity by ELISA) Helsinki. The primary objective was to determine the safety and were optionally tested and documented in site-specific compounding tolerability in patients who receive 90Y-FF-21101 for the treatment validation records. of advanced solid tumors refractory to or relapsed from prior therapy (ClinicalTrials.gov Identifier: NCT02454010). Secondary endpoints 111In-FF-21101 dosimetry included determination of overall response rate (ORR), duration of 111In radioactivity in serum and urine was evaluated, and axial response, progression-free survival (PFS), overall survival (OS), and SPECT, CT, and fused (SPECT/CT) and whole-body anterior/ evaluation of serum antibody protein pharmacokinetics (PK) and posterior planar 111In scintigraphy was performed in all patients presence of human anti-CDH3 human/mouse chimeric antibody approximately 1 week prior to administration of the therapeutic dose (HACA). Baseline tumor P-cadherin expression was evaluated in all to assess the acceptability of radiation dose estimates to critical organs patients with a biopsy-accessible tumor to characterize and correlate and red marrow. The prepared 111In-FF-21101 dose was administered response to potential markers of activity. To allow patients to proceed intravenously through a 0.22-mm in-line filter by slow injection at a to treatment with 90Y-FF-21101, estimation of the radiation absorbed rate not to exceed 1 mL/minute (approximately 10 minutes). Anterior 90Y-FF-21101 dose to marrow and major organs was evaluated by and posterior whole-body planar scintigrams were acquired using a measurement of Indium-111 (111In)-FF-21101 biodistribution, and, as dual-head gamma camera (Symbia S, Siemens Medical Solutions USA, a secondary objective, selective uptake of 111In-FF-21101 in tumor Inc.) at 0.25, 4, 24, 72, and 144 hours after 111In-FF-21101 adminis- tissue was also assessed. tration; SPECT images were acquired 24 hours postadministration, The protocol was approved by the institutional review board at followed by a CT scan (750HD, GE Healthcare) for SPECT attenuation MDACC, and all patients provided written informed consent prior to correction and organ volume contouring. The images were used to

OF2 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

generate organ region of interest time–activity curves (TAC), which also have met the protocol-specified minimum duration from baseline were fitted to exponential functions using Prism 6 (GraphPad) to (8 weeks), including for assessment of clinical benefit. Progression-free obtain 111In-FF-21101 and, by converting from 111In to 90Y radioactive survival was calculated from the date of first dose of 90Y-FF-21101 to decay, estimated 90Y-FF-21101 organ residence times. Blood samples the date of first objective evidence of disease progression or death, were obtained at 0.5, 1, 2, 4, 24, 72, and 144 hours after 111In-FF-21101 whichever was earlier. Overall survival was calculated from the date of administration to compute 111In-FF-21101 and 90Y-FF-21101 red first dose of 90Y-FF-21101 to the date of death due to any cause. marrow TACs and residence times using the Sgouros formula (19). Residence times and CT-based lung, liver, kidney, and spleen organ Safety assessments mass estimates were entered into OLINDA/EXM 1.1 (Vanderbilt Safety and tolerability based on AEs and laboratory parameters were University, Nashville, TN) to generate estimated 111In-FF-21101 and assessed from the day of 111In-FF-21101 administration through the 90Y-FF-21101 organ doses (20). Serum and urine 111In radioactivity end of study (28 days after the last treatment). Adverse events were concentrations (% ID/L) were measured using a precalibrated gamma classified according to the Medical Dictionary for Regulatory Affairs scintillation counter. (MedDRA) and graded using NCI Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Patients were monitored 90Y-FF-21101 treatment during the conduct of the physical examination throughout the study The administered radioactivity in the first cohort was 5 mCi/m2 for signs of potential pulmonary toxicity (e.g., cough, wheezing, (185 MBq/m2; 10%) 90Y-FF-21101, which was then escalated in dyspnea with or without fever, radiation pneumonitis, etc.). Radiologic 5 mCi/m2 increments to 10, 15, 20, and 25 mCi/m2 (370, 555, 740, and assessment could be used as necessary to confirm findings, and 925 MBq/m2; 10%) in subsequent cohorts. 90Y-FF-21101 was symptoms treated with corticosteroids as required. Eye examinations administered intravenously through a 0.22-mm in-line filter on day consisted of visual acuity and field tests, slit lamp examination fundo- 1 by slow injection at a rate not to exceed 1 mL/minute (approximately scopy, and optical coherence tomography performed at cycle 1 and 10 minutes). Patients were monitored for 4 hours after each dose of every 30 days thereafter. Dose modification or delay was allowed on 90Y-FF 21101 for infusion reactions. At the highest dose level to be the basis of protocol-defined criteria. explored (25 mCi/m2), the maximum radioactivity allowed per infu- DLT was defined as grade 4 hematologic toxicity lasting ≥ 7 days; sion was limited to 60 mCi (2,220 MBq) as a safety precaution based on grade 4 neutropenia with a fever (single temperature > 38.3Cor dosimetry performed in animals and estimated radiation absorbed sustained temperature of ≥ 38C for more than 1 hour); ≥ grade 3 dose to the lungs that would exceed the maximum allowable limit thrombocytopenia with clinically significant bleeding; failure of plate- defined for external beam radiotherapy (21, 22). The delivered activity lets, ANC, or hemoglobin to recover to ≤grade 1 within 12 weeks of estimates were required to be below standard limits for external beam dosing with 90Y-FF-21101, despite allowance of supportive measures radiotherapy typically used for nonmyeloablative radioimmunother- to aid recovery (e.g., RBC, platelet transfusions, growth factors); grade apeutic (RIT; <3 Gy for red marrow, <30 Gy for liver, and <20 Gy for 4 nonhematologic toxicity of any duration; grade 3 nonhematologic kidney and lung). If an estimate exceeded the limit, option for toxicity lasting ≥ 3 days, despite the use of supportive care measures treatment at a lower radioactivity was allowed if within allowable (e.g., antiemetics, antidiarrheals); or any protocol defined ophthal- limits. mologic finding. A minimum of 6 weeks would elapse between the last For purposes of scheduling study procedures, cycles were defined as subject dosed in a cohort and the first subject in a new cohort for safety each 28-day period following the initial 90Y-FF-21101 administration. assessment. The highest 90Y-FF-21101 dose level below the dose level On the basis of estimates of organ radiation dose being well below the eliciting DLT, or the highest dose tested in the absence of DLT, was standard limits for external beam radiation, the lack of observed declared the maximum tolerated activity. toxicities, coupled with response observed including an objective Patients remained on study until they chose to discontinue partic- response observed at Cohort 3, the protocol was amended to allow ipation or developed PD, unacceptable AEs, or an intercurrent illness repeated administration at a frequency of every 4 months in respond- or change in medical status that prevented further study participation. ing patients or patients who demonstrated clinical benefit if the Long-term follow-up involved a clinic visit or telephone call every following criteria were fulfilled: hemoglobin ≥10 g/dL, absolute neu- 3 months for up to 12 months. trophil count ≥1.5 109 cells/L, platelets >100,000 106 cells/L, creatinine ≤1.5 upper limit of normal (ULN), or creatinine clearance Pharmacokinetic assessment ≥60 mL/minute 1.73 m2/BSA, bilirubin <1.5 ULN, ALT and AST Serum FF-21101 antibody concentrations were assessed during ≤2.5 ULN (<5 ULN if liver metastases), and ECOG Performance cycle 1 by electrochemiluminescence (ECL) analysis using a validated status ≤2. Additional therapeutic doses of 90Y-FF-21101 were allowed method with a dynamic range of 5–1,000 ng/mL. Data were collected until disease progression (PD), observation of unacceptable adverse using a MSD Sector Imager 6000 and pharmacokinetic parameters events (AE), intercurrent illness, or changes in the patient's condition estimated using a noncompartmental approach (Phoenix pharmaco- that prevented further study participation. kinetic software, Certara). Pharmacokinetic parameters (trough levels, t T C 1/2, max, max, AUC, CL) were assessed on the basis of individual and Efficacy assessment mean serum concentrations of FF-21101 on days 1, 8, and 15. Analysis Disease assessments based on CT or MRI were obtained at week 8 of anti-drug antibodies (ADA) was performed in samples obtained and every 8 weeks thereafter until documented PD and classified on days 1 (pretreatment), 8, 15, and 22 of all cycles using a bivalent according to RECIST v1.1 (23). The primary efficacy endpoint was the ECL immunoassay fully validated in human serum that binds proportion of subjects with objective response [complete response anti–FF-21101 to both biotin- and ruthenium-labeled FF-21101. (CR) þ partial response (PR)] achieved by week 8 (ORR). The best Qualitative ECL immunoassay was used to initially screen for the overall response was defined as the best response recorded across all presence of antibodies to FF-21101. Test samples that confirmed time points. Clinical benefit was defined as confirmed CR, PR, or stable positive for the presence of anti–FF-21101 antibodies were titrated disease (SD). When SD was believed to be the best response, it must to report the sample titer.

AACRJournals.org Clin Cancer Res; 2020 OF3

Subbiah et al.

Analysis of P-cadherin expression Table 1. Patient characteristics and demographics. P-cadherin expression by IHC in formalin-fixed paraffin-embedded archived or fresh tumor biopsy was correlated with clinical outcome as Dose-escalation phase Treated patients (n ¼ 15) fi m an exploratory endpoint. IHC staining of paraf n-embedded 4- m Median age, years (range) 57 (31–70) patient tumor tissue sections was performed using a Benchmark Ultra Male/female, n 6/9 automated immunostaining device (Ventana Medical Systems) with Disease, n heat-induced antigen retrieval and Optiview DAB detection. Murine Soft tissue sarcoma 5 monoclonal anti–P-cadherin antibody was purchased from BD Bio- Leiomyosarcoma 1 sciences (material number 610227/8) and applied at 2 mg/mL. Poly- Liposarcoma 1 mer-negative control serum (BioCare Medical) was used as a negative Clear cell sarcoma 2 control. Samples were processed at a single CLIA reference laboratory Desmoplastic small round cell tumor 1 Chondrosarcoma 1 (Covance). P-cadherin expression was scored on the basis of staining Gynecologic cancer 3 intensity and the percentage of positive tumor cells with membrane Vaginal 1 staining. Staining intensity was rated as 0, absent; 1, weak; 2, moderate; Ovarian 2 or 3, strong. The approximate percentage of tumor cells with mem- Gastrointestinal cancer 3 brane staining at each intensity was reported from 0% to 100%. An Colon 1 H-score for each tissue specimen was calculated as follows, based on Appendix 1 the percentage of cells at each intensity: H-score ¼ 3 (% of cells at 3þ Cholangiocarcinoma 1 intensity) þ 2 (% cells at 2þ intensity) þ 1 (% cells at 1þ Neuroendocrine tumor 2 intensity). The maximum H-score was 300, representing 100% of Lung 1 Pancreatic 1 tumor cells positive for P-cadherin with a staining intensity of 3. Other (papillary thyroid carcinoma) 1 ECOG performance status, n Statistical analyses 01 Response rates were summarized using the number and percentage 114 of patients with best overall response of CR, PR, or SD. For PFS and OS, 90Y-FF-21101 dose per single administration the Kaplan–Meier product-limit method was used to estimate the Minimum 7.7 mCi (285 MBq) median survival. Patients who did not experience disease progression Maximum 51.6 mCi (1909 MBq) or death were censored at the date of the last evaluable disease Median no. of prior treatment regimens (range) 6.5 (3–10) assessment. For patients who have not died, OS was censored at the date the patient was last known to be alive or the date of last contact, whichever was later. Descriptive statistics were used to summarize demographic data, baseline disease characteristics, and safety out- Although only limited assessment of tumor localization could be comes in the dose-escalation phase. All analyses were performed using performed, radiotracer activity was appreciable at the location of the SAS for Windows version 9.3 or higher (SAS Institute, Cary, NC). primary tumor and in known sites of metastatic disease in the majority of patients evaluated. Figure 1A (i–iii) shows axial SPECT, CT, and fused (SPECT/CT) images of the pelvis obtained 24 hours after the Results administration of 5 mCi/m2 (5.54 mCi) 111In-FF-21101 in a patient Patient characteristics with squamous cell carcinoma of the anus enrolled in the first cohort, From January 21, 2016 to July 1, 2019, 15 patients (6 males, 9 providing early radiographic evidence of P-cadherin targeted antibody females) with advanced primary tumors (stage IV) were treated in five uptake in tumor. Increased radiotracer uptake was shown correspond- cohorts of the 3þ3 dose-escalation phase study (Table 1). Patients ing to a 6.5 5.0 4.4 cm soft tissue metastasis in the right pelvis in with a variety of tumor types were enrolled including sarcomas this patient, with serial whole-body planar images demonstrating the [leiomyosarcoma, liposarcoma, clear cell sarcoma (n ¼ 2), desmo- temporal accumulation of radioactivity in the pelvic lesion and in plastic small round cell tumor], gynecologic cancers [vaginal, ovarian metastases to the right hepatic lobe and anterior abdominal wall and (n ¼ 2)], gastrointestinal cancers (colon, appendix, cholangiocarci- rectal regions, with maximal tumor uptake observed by 72 hours noma), and neuroendocrine tumors (lung, pancreatic). All patients postadministration [Fig. 1A (iv)]. Image (iv) also reveals gradually were heavily pretreated and had received prior surgery, radiation, increasing splenic uptake. Although appreciable tumor uptake was and/or chemotherapy with a median number of prior treatment observed, the patient unfortunately was withdrawn following dosim- regimens of 6.5 (range 3–10). etry due to rapidly progressing disease. Overall, evidence of radiotracer activity in tumor was observed in all but two patients evaluated, one Dosimetry with ovarian cancer who achieved a CR as further described below, and Images from whole-body scintigraphy and single-photon emission another with cholangiocarcinoma for which uptake in the periportal CT/X-ray CT (SPECT/CT) demonstrated 111In-FF-21101 uptake pri- region was indeterminate. The latter patient had received three treat- marily in the spleen, kidneys, testes, lungs, and liver (Table 2). Total ments approximately 4 months apart, with stable disease maintained 90Y Gy organ radiation dose estimates are shown by increasing through cycle 12 (48 weeks) before progressing. administered activity over cohorts 1–5. For all patients, estimates for the 90Y-FF-21101 administered activity were below the applicable 90Y-FF-21101 treatment and P-cadherin staining limits for external beam radiotherapy to red marrow (<3 Gy), liver Subsequent 90Y-FF-21101 treatment was initiated at an adminis- (<30 Gy), kidneys, and lungs (<20 Gy; ref. 22). The spleen had the tered radioactivity of 5 mCi/m2 (185 MBq/m2; 10%), similar to highest amount of uptake in all patients except one who had prior that used in previous clinical trials for 90Y-drug conjugates (24–28), splenectomy. at a specific activity of 8 mCi/mg antibody. Dose was escalated to

OF4 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

Table 2. Normalized 90Y-FF-21101 organ radiation–absorbed dose estimates based on 111In-FF-21101 dosimetric imaging.

Mean (Total Gy) Mean Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5 (mGy/mCi) CV 5 mCi/m2 10 mCi/m2 15 mCi/m2 20 mCi/m2 25 mCi/m2 Organ (n ¼ 15) StdDev (%) (n ¼ 3) (n ¼ 3) (n ¼ 3) (n ¼ 3) (n ¼ 3)

Spleen 1,031.5 399.9 38.8 12.82 18.99 20.38 43.95 37.42 Testes 339.5 122.6 36.1 3.22 11.01 N/A 15.62 14.06 Kidneys 219.7 75.8 34.5 2.70 3.20 6.62 8.92 8.15 Lungs 168.9 56.9 33.7 1.88 2.88 4.51 6.65 8.05 Liver 160.6 58.5 36.4 1.85 2.73 4.08 6.39 7.32 Heart wall 89.7 18.1 20.1 0.67 1.99 2.43 3.84 5.08 Osteogenic cells 34.7 13.2 38.1 0.30 0.71 0.98 1.24 2.30 Red marrow 29.6 11.0 37.1 0.30 0.56 0.78 1.02 1.87 Total body 25.7 2.6 10.1 0.20 0.55 0.69 0.99 1.38 Ovaries 14.6 4.3 29.6 0.11 0.26 0.36 0.61 0.93 Uterus 14.6 4.3 29.6 0.11 0.26 0.36 0.61 0.93 All other organs 12.5 5.1 40.6 0.09 0.29 0.36 0.50 0.86

Note: All 16 (7 male, 9 female) patients were evaluated (1 male patient did not proceed to therapy). Total dose estimates for each patient were their individual normalized dose estimates multiplied by their administered 90Y radioactivity, and are shown by increasing administered activity over cohorts 1–5 (testes N/A for Cohort 3, as all were female). 111In-FF-21101 uptake was observed primarily in the spleen, testes, kidneys, lungs, and liver. For all patients, estimates for 90Y-FF-21101 administered activity were below the applicable limits for external beam radiotherapy (<3 Gy to red marrow, <30 Gy to the liver, and <20 Gy to the kidneys and lungs).

25 mCi/m2 over five cohorts, with 5 patients who received more than In both patients enrolled at Cohort 4 (20 mCi/m2), ADA titers ranged one treatment (Fig. 2A). The median (range) time on study following from low-intermediate to high and remained stable throughout the the first 90Y-FF-21101 administration was 24 (4–90) weeks. Tumor course of treatment. It is unknown at this time if the ADA is samples from 13 of 15 subjects were available for IHC analysis. P- neutralizing. However, these patients went on to receive subsequent cadherin staining was predominantly localized in the tumor cell doses of 90Y-FF-21101 and maintained stable disease, one for 4 months membrane, with some nontumor staining of occasional fibroblasts and one for 12 months. and endothelial cells (Fig. 2B, a–d). P-cadherin expression H-scores ranged from 0 to 242 with 40% of samples exhibiting scores ≥100. One Clinical activity subject with metastatic clear cell ovarian carcinoma had the most One patient achieved a CR; no other patients achieved a PR, so the marked P-cadherin staining (H-score of 242, Fig. 2A; Fig. 2B, d) and ORR was 6.7% (1/15). Median PFS was 24 weeks, and the median OS achieved a clinical CR by RECIST, as further described below. Strong was 90 weeks. The clinical benefit rate defined as confirmed CR, PR, or P-cadherin staining was also observed in patients with pancreatic SD for a minimum of 8 weeks was 73% (11/15 patients). neuroendocrine tumor, intrahepatic cholangiocarcinoma, and sarco- The CR was observed in a 60-year-old woman with metastatic clear ma, specifically in clear cell, smooth muscle tissue. cell ovarian carcinoma (Fig. 3). The patient was a nulligravida female who presented with a persistent cough for several months. A diagnostic Pharmacokinetics and ADA titers work-up at the local emergency room for worsening abdominal pain, Over the dose range administered of 0.625 to 3.2 mg/m2 FF-21101 nausea, and vomiting confirmed a 10-cm left pelvic mass and a 6-mm protein, FF-21101 antibody pharmacokinetics in serum demonstrated splenic lesion suspected to be metastatic disease. Her CA-125 was generally dose proportional increases in exposure with increasing dose, elevated at 83 U/mL. She underwent exploratory laparotomy, total as measured by maximum concentration (Cmax) and area-under-the- abdominal hysterectomy with bilateral salpingo-oophorectomy, rad- fi concentration-time curve from zero to in nity (AUC0-¥), with a mean ical debulking, left pelvic and para-aortic lymph node sampling, and (StdDev) half-life for the circulating antibody in serum of 69.7 partial omentectomy. Final pathology showed a stage IIC clear cell (12.1) hours [Fig. 1B (i)]. With a physical half-life for 90Yof carcinoma of the ovary. According to the operative report, the spleen 64.1 hours, the effective half-life for 90Y-FF-21101 was calculated as was unremarkable and she was optimally debulked. Comprehensive 33.4 hours. Mean clearance and steady-state volume of distribution next-generation DNA sequencing of the resected tumor (Foundation (Vss) for the FF-21101 antibody ranged from 10.2 to 43.7 mL/hour/kg Medicine) showed PIK3CA E545K and N1044S mutations. Surgery and from 1,390 and 3,940 mL, respectively. On the basis of a limited was followed by six cycles of carboplatin/paclitaxel. Evidence of number of patients per cohort, there was high variability in exposure progressive nodal disease was seen on CT scan approximately 1 year within each cohort. However, the serum pharmacokinetics of the later. Pelvic and para-aortic lymphadenectomy showed 10 of 13 FF-21101 protein appears to be linear and not dose dependent in positive lymph nodes and was followed by radiotherapy (56 Gy in humans. Mean radioactivity clearance in serum (111In-FF-21101 27 fractions to high-risk areas) and hepatic resection of metastasis percent injected dose per liter, %ID/L) paralleled antibody clearance approximately 7 months before study entry. Further analysis of the from 0 to 144 hours [Fig. 1B (ii)], with an initial %ID/L of 20% at patients’ tumor confirmed the previous PIK3CA mutation [Oncology 0.5 hours (CV, 33%) that was more rapidly eliminated to < 1% Hotspot Panel (50 genes), sema4], and the following profile: cMET, measurable at 144 hours postadministration. Evidence of the forma- PD-1, PTEN, TOP2A, and TUBB3-positive, PD-L1–negative (Caris tion of circulating antibodies against the FF-21101 antibody was ob- Life Sciences). A follow-up restaging scan showed recurrence in the left served in only 2 patients within 4 weeks following the initial treatment. axilla, and biopsy of the left axillary lymph node at the time of

AACRJournals.org Clin Cancer Res; 2020 OF5

Subbiah et al.

Figure 1. Dosimetry and pharmacokinetics in a phase I study of 90Y-FF-21101. A, Dosimetric imaging in a 33-year-old male with metastatic squamous cell carcinoma of the anus. Axial SPECT (i), CT (ii), and fused (SPECT/CT) images (iii) of the pelvis obtained 24 hours after the administration of 5.54 mCi of 111In-FF-21101 demonstrate increased radiotracer uptake corresponding to a 6.5 5.0 4.4 cm soft tissue metastasis in the right pelvis (white arrows). Serial whole-body planar images in the anterior view demonstrate the time course of 111In-FF-21101 uptake in this patient (iv). The patient had metastases to the right anterior abdominal wall, right pelvis, and rectal region, with maximal tumor uptake observed by 72 hours postadministration (green arrow heads). Red arrows depict sequential radiotracer uptake in the right pelvic metastasis shown on the axial images. In addition, there was heterogeneous radiotracer activity at the site of multifocal metastases in the right hepatic lobe, the largest with central necrosis [circled in whole body planar (iv) and SPECT/CT (v) images]. Gradually increasing uptake was also noted in the spleen. B, Mean (StdDev) serum FF-21101 antibody protein concentration versus time following a single dose of 90Y-FF-21101 (n ¼ 3/cohort except Cohort 1 where n ¼ 1; i). Following the administration of 5 to 25 mCi/m290Y-FF-21101 over Cohorts 1–5 (equal to 0.625 to 3.2 mg/m2 FF-21101 protein), antibody protein was measurable in serum up to 300 hours postadministration. High variability in exposure was observed within each cohort; however, the serum pharmacokinetics of the FF-21101 protein appeared to be linear and not dose dependent in humans. The mean (StdDev) biologic half-life for the circulating antibody in serum was 69.7 (12.1) hours, with an effective half-life for 90Y-labeled FF-21101 of 33.4 hours. Radioactivity clearance in serum (111In-FF-21101 percent injected dose per liter, %ID/L) is shown for all patients in the dosimetry phase (n ¼ 16) following administration of 5 mCi (10%) 111In-FF-21101, containing 5 mg FF-21101, approximately 1 week prior to 90Y-FF-21101 dosing (ii). Radioactivity clearance paralleled antibody clearance from 0 to 144 hours. The mean initial %ID/L was 20% at 0.5 hours (CV, 33%; see inset) and < 1% at 144 hours post 111In-FF-21101 administration.

OF6 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

Figure 2. P-cadherin staining and response to therapy in a phase I dose-escalation study of 90Y-FF-21101. A, Best response to therapy in the phase I dose-escalation study by P-cadherin expression H-score and 90Y-FF-21101 dose. Dose administration is shown by circles. NE, not evaluable. B, Representative IHC images of P-cadherin staining in human tumor samples showing: a, weak staining of cholangiocarcinoma (H ¼ 108) with minimal background; b, moderate staining of ovarian carcinoma (H ¼ 180); c, focal strong membranous staining of clear cell sarcoma (H ¼ 115); and d, strong membranous and cytoplasmic staining of ovarian carcinoma (H ¼ 242) in a subject with a RECIST complete response. C, Percent change from baseline in tumor burden for all patients enrolled in the phase I dose-escalation study of 90Y-FF-21101 by P-cadherin expression H-score.

evaluation for study entry was positive for metastatic clear cell carcino- the target axillary nodes was not clearly appreciable in serial whole- ma. In addition, mildly enlarged abdominal adenopathy was observed body planar scintigrams nor the SPECT image acquired at 24 hours in the para-aortic region in addition to numerous enlarged nodes in the following 111In-FF-21101 administration. However, following a single retroperitoneum surrounding the abdominal aorta and a prominent 90Y-FF-21101 treatment, a PR in target lesions was observed at the gastro-hepatic node. The patient was then referred for study enroll- initial restaging at cycle 2 (week 8), which was conﬁrmed at week 16 ment in the third dose-escalation cohort to receive treatment with 90Y- (59.9% decrease in sum of diameters of target lesions). Given the FF-21101 at a dose of 15 mCi/m2 (24.2 mCi). Visible tumor uptake in excellent response, the protocol was amended to allow administration

AACRJournals.org Clin Cancer Res; 2020 OF7

Subbiah et al.

Figure 3. Complete response achieved with 90Y-FF-21101 treatment in a patient with metastatic clear cell ovarian carcinoma. A, Therapy timeline of a 60-year-old female with metastatic clear cell ovarian carcinoma and response to 90Y-FF-21101 therapy based on tumor marker (CA-125, U/mL) and target lesion size (longest diameter, mm). B, Serial CT images after 90Y-FF-21101 treatment showing left axillary adenopathy; target node reduced from 1.6 cm in the longest diameter at screening to 0.6 0.6 cm (red circle and arrows) following administration of four 15 mCi/m2 (24.2 mCi) 90Y-FF-21101 treatments administered 4 months apart; an additional ﬁfth dose was administered approximately 7 months later. A complete response was achieved by cycle 24 and has been maintained over 1 year to date without further need for therapy.

of additional doses in responding patients and her treatment was involving the rectum and abdominal wall and was progressing prior to continued every 4 months for three additional doses with continued study entry in the 2nd cohort, with a P-cadherin expression H-score of disease response (Fig. 3B), followed by an additional ﬁfth dose 180. Although disease was stable following 90Y-FF-21101 treatment, approximately 7 months later (cycle 20). Restaging at cycle 24 sub- the patient subsequently came off study after 16 weeks due to personal sequently showed achievement of a CR of her target lesion (axillary logistical (nonmedical) reasons. Stable disease responses were main- lymph node), and no evidence of metastatic abdominopelvic disease. tained for approximately 50 weeks on study in an additional 4 patients, The patient was thereafter followed by her local physician every other one with a pancreatic neuroendocrine tumor (NET) treated with a cycle beginning with cycle 25 and has maintained a CR with no further single dose at the lowest dose level of 5 mCi/m2 (9.8 mCi), one with treatment. CA-125 values decreased from 30 to 15 U/mL over the metastatic intrahepatic cholangiocarcinoma treated in the third cohort course of 90Y-FF-21101 treatment and, as noted above, tumor response with 15 mCi/m2 (25.4 mCi) for 3 doses, one with a lung NET treated in this patient correlated with a high P-cadherin expression H-score > with 2 doses of 20 mCi/m2 (36.3 mCi) administered 4 months apart, 200 (242), the highest documented on study (Fig. 2A–C). and one with a clear cell sarcoma who received a total of 4 doses An additional patient with metastatic low-grade serous ovarian administered 4 months apart at the highest dose level of 25 mCi/m2 carcinoma had stable disease following a single 90Y-FF-21101 treat- (49.5 mCi; Fig. 2A). P-cadherin expression H-scores were ≥ 100 in ment at a dose of 10 mCi/m2 (21.9 mCi). This patient had disease two of these patients (clear cell sarcoma and pancreatic NET), but

OF8 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

Table 3. 90Y-FF-21101–related adverse events in ≥ 2 patients.

Cohort 1 Cohort 2 Cohort 3 Cohort 4 Cohort 5 5 mCi/m2 10 mCi/m2 15 mCi/m2 20 mCi/m2 25 mCi/m2 Adverse event Total (n ¼ 16)a (n ¼ 4)a (n ¼ 3) (n ¼ 3) (n ¼ 3) (n ¼ 3)

All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 All Gr 3/4 Lymphopenia 11 (68.8%) 4 (25.0%) 0 0 3 1 3 1 3 2 2 0 Leukopenia 5 (31.3%) 1 (6.3%) 0 0 0 0 2 0 2 1 1 0 Thrombocytopenia 4 (25.0%) 1 (6.3%) 0 0 0 0 1 0 2 1 1 0 Dysgeusia 3 (18.8%) 0 0 0 0 0 0 0 1 0 2 0 AST, increased 2 (12.5%) 0 2 0 0 0 0 0 0 0 0 0 Rash 2 (12.5%) 0 2 0 0 0 0 0 0 0 0 0

Note: None of the grade 3 or 4 adverse events were DLTs. Abbreviation: AST, aspartate aminotransferase. aOne patient received the 111In-FF-21101 dose only. notably absent in the patient with lung NET (H-score of 3). Patients binding, and has recently undergone a surge of interest as more with an H-score ≥ 100 remained on study for a median (range) of sophisticated drugs and targeted therapies have found success (30). 54 (47–64) weeks compared with 8 (4–46) weeks in patients with an There have been successes with RIT such as 90Y ibritumomab tiuxetan H-score < 100. (Zevalin; Acrotech Biopharma, LLC), as well as radiolabeled peptide therapies such as lutetium 177Lu dotatate (Lutathera; Advanced Accel- Safety and treatment-related adverse events erator Applications USA, Inc.; ref. 31). However, it is a challenging DLT was not observed in any patient on study and dose reductions development path; there are also failures, such as RIT directed against were not required over the course of the dose escalation. Patients did CEA and MUC1, where early promising results have not yielded not exceed the allowed per treatment amount of calculated radiation effective therapies (28, 29, 32). Success requires the right target, exposure (Table 2). Treatment-emergent AEs (all events related and delivery vehicle, and therapeutic effector. unrelated) were typical of a phase I solid tumor trial, with the most Cadherins represent a very attractive target due to their frequent common being lymphopenia (75%), increased AST (56%), fatigue overexpression in a wide variety of tumors. Overexpression of (56%), anemia (44%), diarrhea (44%), dysgeusia (38%), leukopenia P-cadherin has been measured by both IHC and mRNA-based meth- (38%), increased ALT (25%), anorexia (25%), constipation (25%), ods in many tumor types, but because the literature on this subject dizziness (25%), dyspnea (25%), and hyperkalemia (25%). describes varying definitions and assay techniques, cross-study com- Similar to other 90Y-conjugated therapeutic antibodies, there were parisons of P-cadherin expression are difficult to perform and reversible myeloid effects (28, 29). The most common drug-related AE require cautious interpretation. Although the precise quantification was lymphopenia, which occurred in 11 (68.8%) of 16 patients [grade 3 of P-cadherin in tumor tissues may not have universal agreement, (n ¼ 3) or 4 (n ¼ 1) severity in four patients, 25%; Table 3]. In cohort 4, P-cadherin clearly appears to be expressed in greater amounts than in one subject experienced grade 4 lymphopenia within 1 week of the normal tissue, particularly in epithelial tumors. It should be appre- therapeutic dose but had normal overall white blood cell counts; this ciated that nonepithelial tumors also overexpress cadherins, including event did not meet DLT criteria. Transient grade 3 lymphopenia and P-cadherin (9). leukopenia was observed in one additional patient at cohort 4. There Strengthening the idea that P-cadherin is an important target, it is were no SAEs reported at Cohort 4, and no grade 3 or 4 lymphopenia notable that the majority of P-cadherin–directed therapies seem to observed in cohort 5 (25 mCi/m2). Drug-related leukopenia and have good activity in preclinical models. Furthermore, because cad- thrombocytopenia occurred in 31.3% and 25.0% of patients, respec- herins interact with and make up part of the tumor microenvironment tively, and were of grade 3 or 4 severity in only one patient each. Grade (TME), P-cadherin–directed therapy may complement other therapies 1 or 2 dysgeusia, increased AST, and rash were other common drug- that require TME alteration for greater effect. The elegant preclinical related AEs observed in more than 10% of patients. All AEs were work reported by Zhang and colleagues also supports the role of the reversible. No evidence of pulmonary function impairment or ocular TME in tumor viability and progression (10). Despite these potential effects was observed. There were no grade 3 or 4 AEs observed at the advantages, it should be noted that downregulation of P-cadherin may highest dose level (25 mCi/m2), and no drug-related serious AEs, stimulate some malignant cell properties in certain settings (33). discontinuations, or deaths were observed on study. A maximum Therefore, although P-cadherin is an attractive target, as previously tolerated administered activity was therefore not reached in dose mentioned, RIT therapy against the similarly attractive CEA and escalation based on the lack of DLT in patients receiving up to MUC1 led to disappointing results. 60 mCi/dose and cumulative doses up to 100 mCi administered on Here, we report the dose-escalation results of a first-in-human an every 4-month schedule. clinical trial of an RIT targeting P-cadherin. These data represent the first reported clinical trial results of a P-cadherin–directed therapy. Descriptions of several antibody therapies directed against a cadherin Discussion target have been previously published (9), but there have been no Antigens that are unique or virtually unique to tumors offer the definitive results reported on their activity or safety in humans. Phase I promise of targets that will yield drugs with both increased efficacy and studies have been conducted with PF-06671008, a bispecific antibody decreased toxicity. RIT offers some advantages over other target- against both P-cadherin and CD3 (NCT02659631), as well as with two directed therapeutic modalities such as potent toxins or direct target additional cadherin-directed antibody–drug conjugates, PCA-062

AACRJournals.org Clin Cancer Res; 2020 OF9

Subbiah et al.

(NCT02375958) and HKT288 (NCT02947152; ref. 34), directed relative immunogenicity, the degree of glycosylation, and susceptibil- against P-cadherin and Cadherin 6, respectively. Yoshioka and col- ity to proteolysis. leagues reported preclinical data obtained from administration of a Localization of 111In-FF-21101 in tumor tissue in this study showed radioimmune-coupled mAb directed against P-cadherin, Mab-6 (35). uptake in lesions in the majority of patients evaluated; however, by Their preclinical data are strikingly similar to those reported for design, there were limitations for detailed evaluation of tumor target- 90Y-FF-21101; however, no further information has been made avail- ing. While the whole-body anterior/posterior planar 111In scintigraphy able on the progression of the compound to human trials. was conducted on multiple days up to 144 hours post 111In-FF-21101 Coupled to a highly energetic (0.934 MeV mean and 2.28 MeV administration, the SPECT/CT scan covering the lung, liver, and maximum energies) beta emitter, 90Y-FF-21101 may offer advan- kidney regions was acquired only at one time point on day 2 (approx- tages for targeting and affecting multiple tumor cells simultaneous- imately 24 hours postinfusion), primarily to correct organ time– ly. With a mean emission penetration of only 4 mm, it requires only activity curves for attenuation and scatter, and mass-correct organ modest external radiation precautions. Preclinical data have also radiation-absorbed doses. As illustrated by the lack of appreciable previously demonstrated that the FF-21101 anti–P-cadherin anti- uptake observed in target axillary nodes in the ovarian cancer patient body appears to be preferentially taken up by tumors, without who achieved a complete clinical response, assessment of tumor evidence of off-target toxicity or the formation of anti-drug targeting can be challenging and complex, requiring three- antibodies (refs. 17, 18; Supplementary Figs S1–S3). In this clinical dimensional imaging at multiple late time points for adequate contrast study, 90Y-FF-21101 doses up to 25 mCi/m2 were found to be very to assess smaller lesions, which was not performed in this study. In well tolerated in a heavily pretreated population of patients with addition, the inherent sensitivity of gamma camera imaging for advanced solid tumors. Because P-cadherin is expressed in the detection of radioactivity in very small lesions and micrometastases, retina, one safety concern was ophthalmologic effects; however, which are not uncommon, is low. And while further assessment will be repeated ophthalmology evaluations revealed no ocular findings in warranted throughout the course of development, evidence of tumor any patient on study. targeting may not always be used as a future prerequisite for treatment Dosimetric imaging revealed the organ with the highest radiotracer to proceed. As an example, the FDA eventually approved removal uptake was the spleen. As P-cadherin is a myo-epithelial antigen, this of pretreatment assessment of tumor uptake during the course may not have been anticipated. However, analysis of P-cadherin of development of Zevalin, due to its benefit in patients who were expression in tumors and normal tissues demonstrates a moderate known to have disease but lacked visual evidence of uptake in amount of P-cadherin expression in normal splenic tissue (mean 111In-radioimmunoconjugate imaging. StdDev CDH3 mRNA transcripts per million, 15.8 10.2; Supple- In this initial phase I study enrolling multiple types of solid tumors, a mentary Fig. S1). Nonspecific binding as well as macrophage uptake, as dose response was not observed. However, as a first-in-human dose- previously observed with some mAbs (36, 37), may also play a role in escalation phase trial with a novel radioimmunoconjugate, this study is the observed accumulation in the spleen, in addition to possible considered a signal-seeking study in patients with advanced and aggregation/complexation. Variable accumulation of radioactivity in refractory tumors, where lack of response may have been a conse- the spleen as a consequence of variable aggregation could possibly quence of multiple prior treatments and the refractoriness of advanced explain the non-dose–related lymphopenia observed. Although the cancers. The response to therapy in our patient with clear cell ovarian analytic methodology used in this study was not capable of measuring cancer despite reactivity was intriguing, and the high tumor aggregation, this, and complexation with ADA, remain potential P-cadherin expression in this tumor (H-score ¼ 242), one not contributing factors. Our clinical data to date demonstrates posttreat- commonly associated with P-cadherin overexpression (38, 39), under- ment ADA titers in only 2 patients, with one showing a relatively high scores the importance of further evaluation of target expression as a titer. Lack of measurable serum ADA could also be a function of potential correlate for activity. Although some patients with low complexation of the circulating antibody with low titer ADA, therefore expression maintained stable disease, the majority of patients additional evaluation is needed to further define this aspect of antibody with longer term disease control had relatively high expression fate. On the basis of the serum protein pharmacokinetics, mean (H-score > 100). clearance and volume of distribution for the FF-21101 antibody were Because of the need for more definitive assessment of antitumor consistent with largely intravascular distribution, although some activity, the trial is now amended to evaluate defined tumor types for a evidence of extravascular distribution is demonstrated by the range more quantitative measure. Two expansion phase cohorts are being of individual distributional volumes. The Vss in our study was evaluated, one for ovarian carcinoma and another for solid tumors determined using noncompartmental analysis. While reasonable for inclusive of triple-negative breast cancer, head and neck squamous cell estimating a pharmacokinetic parameter, this type of model assumes carcinoma, cholangiocarcinoma, pancreatic carcinoma, and colorectal all drug elimination occurs from the central compartment, which for cancer, at the recommended phase II administered activity of antibodies that bind and internalize within cells in tissues sites, tends to 25 mCi/m2 (not exceeding 60 mCi/dose) every 3 months, with also overestimate the Vss. potential for more frequent administration to be explored. All types Further analysis of the antibody pharmacokinetics showed the of ovarian carcinomas are eligible. These expansion cohorts will allow t measured 1/2 for the FF-21101 protein was 2.9 days, which is some- better characterization of safety, as well as provide more robust what less than some approved therapeutic chimeric monoclonal IgG indications for a definitive trial of efficacy. Analysis of P-cadherin t antibodies, but is similar to the 1/2 for agents such as trastuzumab expression continues to be performed for all patients in this expansion (2.7 days) and cetuximab (4.1 days). The half-life of IgG is 23 days, and phase of development. its slow degradation is dependent on protection by FcRn binding to the In summary, data from the dose escalation of 90Y-FF-21101 are Fc portion of the antibody. FF-21101 may be cleared faster than other promising, but preliminary. As noted, patients received a variety of therapeutic antibodies due to reduced affinity for the FcRn receptor. doses, a different number of doses, and had varying results. At this Other factors which may contribute to the increased elimination of early stage, it is not possible to discern a definitive pattern of response FF-21101 compared with other IgG-derived therapeutics include the based on P-cadherin expression score, tumor type, dose, or other

OF10 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

variables in this small number of patients. Much remains to be non-financial support from Chugai Biopharmaceuticals (honoraria - speaking explored, including, as noted above, the current schedule of treatment engagement), Mayo Clinic, and Rutgers Cancer Institute of New Jersey, and other – from Beth Israel Deaconess Medical Center (travel related) outside the submitted every 3 4 months, which is largely empirical. In the dose-escalation fi – work.D.S.HongreportsgrantsandotherfromP zer (research support to MD study, patients who received multiple (2 5) doses every 4 months Anderson) outside the submitted work; research/grant funding from AbbVie, tolerated that schedule well, with cumulative activities that ranged Adaptimmune, Aldi-Norte, Amgen, AstraZeneca, Bayer, BMS, Daiichi-Sankyo, from 40 to 100 mCi. More frequent dosing may be beneficial and Eisai, Fate Therapeutics, Genentech, Genmab, Ignyta, Infinity, Kite, Kyowa, Lilly, appears to be promising for other radiopharmaceuticals such as those LOXO, Merck, MedImmune, Mirati, miRNA, Molecular Templates, Mologen, fi being evaluated for treatment of prostate cancer (30). Investigation of NCI-CTEP, Novartis, P zer, Seattle Genetics, Takeda, and Turning Point Ther- higher dose levels may also be of value. A cautious limitation for dose apeutics; travel and accommodation expense support from Bayer, LOXO, miRNA, 2 fi Genmab,AACR,ASCO,andSITC;consultingoradvisoryrolesatAlphaInsights, escalation up to a maximum activity of 25 mCi/m was prede ned in Acuta, Amgen, Axiom, Adaptimmune, Baxter, Bayer, Genentech, GLG, Group H, this study based on estimates to remain below standard organ radiation Guidepoint, Infinity, Janssen, Merrimack, Medscape, Numab, Pfizer, Prime dose limits typically used for external beam radiation therapy. How- Oncology, Seattle Genetics, Takeda, Trieza Therapeutics, and WebMD; and other ever, DLTs that have been observed at lower doses with other agents of ownership interests in MolecularMatch (advisor), OncoResponse (founder), and fi this type, such as myelosuppression (24–28), were not observed with Presagia Inc (advisor). D.B. Le reports grants from Fuji lm (to the institution) fi 90Y-FF-21101 even at highest activity level. This improved safety during the conduct of the study. T. Suzuki reports other from Fuji lm Pharma- fl ceuticals U.S.A., Inc. (employed) during the conduct of the study, as well as other margin may re ect a more limited distribution of the target antigen from Fujifilm Pharmaceuticals U.S.A., Inc. (employed) outside the submitted in normal tissue, supporting the prospect for further escalation beyond work. R.A. Subach was a consultant to Fujifilm Pharmaceuticals U.S.A., Inc. until the initial recommended phase II activity level administered every 5/28/2020 and was hired by Fujifilm Pharmaceuticals U.S.A., Inc. on 5/29/2020 3 months. In addition, based on observations thus far that demonstrate and is currently employed full time. F. Nomura reports grants from Fujifilm fi 90Y-FF-21101 is nonmyeloablative, there is potential for combination during the conduct of the study and a patent for JP5380553 licensed to Fuji lm, and is engaged in a job in Perseus Proteomics Inc (Perseus Proteomics was a with other therapies including immunotherapy, targeted therapy, or fi fi fi subsidiary company under parent company Fuji lm until 2018); Fuji lm sup- even chemotherapy. The favorable safety pro le shown in this initial ported preclinical and clinical research with other subsidiary brother companies: clinical dose-escalation study, along with some early signs of antitumor Fujifilm RI Pharma (currently named Fujifilm Toyama Chemical), etc. T. Mat- activity, warrant further evaluation of this RIT approach. suura reports grants from Fujifilm during the conduct of the study, a patent for JP5380553 licensed to Fujifilm, and engagement in a job in Perseus Proteomics Inc. (Perseus Proteomics was a subsidiary company under parent company Disclosure of Potential Conflicts of Interest Fujifilm until 2018); Fujifilm supported preclinical and clinical research with V. Subbiah reports grants from Fujifilm (clinical trials research support) other subsidiary brother companies: Fujifilm RI Pharma (currently named during the conduct of the study; grants from Fujifilm outside the submitted Fujifilm Toyama Chemical), etc. M. Kajita reports a grant from Fujifilm Corpo- work; research funding/grant support for clinical trials from Roche/Genentech, ration during the conduct of the study and outside the submitted work. E. Novartis,Bayer,GlaxoSmithKline,Nanocarrier, Vegenics, Northwest Biothera- Nakamura reports a grant from Fujifilm Corporation during the conduct of the peutics, Berghealth, Incyte, Pharmamar, D3, Pfizer, Multivir, Amgen, Abbvie, study and outside the submitted work. Y. Funase reports a grant from Fujifilm Alfa-sigma, Agensys, Boston Biomedical, Idera Pharma, Inhibrx, Exelixis, Blue- Corporation during the conduct of the study and outside the submitted work. print Medicines, Loxo Oncology, Medimmune, Altum, Dragonfly Therapeutics, G. Ravizzini reports grants from Fujifilm during the conduct of the study and Takeda, National Comprehensive Cancer Network, NCI-CTEP, UT MD Ander- outside the submitted work. No potential conflicts of interest were disclosed by son Cancer Center, Turning Point Therapeutics, and Boston Pharmaceuticals; the other authors. travel support from Novartis, Pharmamar, ASCO, ESMO, Helsinn, and Incyte; consultancy/advisory board relationships with Helsinn, Loxo Oncology/Eli Lilly, R-Pharma US, Incyte, QED Pharma, Medimmune, Novartis, and Signant Health; Authors’ Contributions and other from Medscape. W.D. Erwin reports grants from Fujifilm Pharma- V. Subbiah: Conceptualization, resources, formal analysis, supervision, funding ceuticals U.S.A., Inc. during the conduct of the study, as well as grants from acquisition, validation, investigation, methodology, writing-original draft, project Oncosil Medical, Ltd., Ipsen Pharma SAS, and Alfasigma S.p.A. outside the administration, writing-review and editing. W. Erwin: Conceptualization, resources, submitted work. Q. Mawlawi reports grants from Fuji Pharmaceuticals during data curation, software, formal analysis, methodology, writing-review and editing. the conduct of the study. A. McCoy reports other from Fujifilm (study sponsor) O. Mawlawi: Resources, investigation, visualization. A. McCoy: Data curation, during the conduct of the study. D. Wages reports personal fees from Salarius validation. D. Wages: Formal analysis, investigation, writing-original draft, Pharmaceuticals, Acerta Pharmaceuticals, and Novella outside the submitted writing-review and editing. C. Wheeler: Formal analysis, investigation. work. C. Gonzalez-Lepera reports grants from Fujifilm (grant to produce the C. Gonzalez-Lepera: Validation, investigation. H. Liu: Investigation, drug in-house) during the conduct of the study.H.A.Macapinlacreportsgrants methodology. H. Macapinlac: Supervision, methodology, project administration. from Fujifilm during the conduct of the study. F. Meric-Bernstam reports personal F. Meric-Bernstam: Resources, funding acquisition, investigation, methodology, fees from Aduro BioTech Inc. (consulting), Alkermes (consulting), DebioPharm project administration. D. Hong: Resources, investigation, writing-original draft. (consulting), F. Hoffman-La Roche Ltd. (consulting), Jackson Laboratory (con- S. Pant: Resources, investigation. D. Le: Resources, supervision, project sulting), Kolon Life Science (consulting), OrigiMed (consulting), PACT Pharma administration. E. Santos: Data curation, software, formal analysis. J.F. Gonzalez: (consulting), Parexel International (consulting), Pfizer Inc. (consulting), Samsung Resources, data curation, formal analysis. J. Roszik: Software, formal analysis, Bioepis (consulting), Seattle Genetics Inc.(consulting,advisorycommittee),Tyra validation. T. Suzuki: Formal analysis, investigation, resources, writing-original Biosciences, Xencor, Zymeworks, Immunomedics, Inflection Biosciences, Mer- draft. R. Subach: Formal analysis, project administration. T. Madden: Data sana Therapeutics, Spectrum Pharmaceuticals (advisory committee), and Zentalis curation, formal analysis, supervision, investigation, methodology, validation, (advisory committee), grants and personal fees from Effector Therapeutics writing-original draft. M. Johansen: Data curation, formal analysis, investigation, (consulting, sponsored research), Genentech Inc. (consulting, sponsored writing-review and editing. F. Nomura: Conceptualization, resources, data curation, research), IBM Watson (consulting), Puma Biotechnology Inc. (sponsored validation, visualization. H. Satoh: Conceptualization, data curation, formal analysis, research, advisory board), and Silverback Therapeutics (advisory committee), funding acquisition, validation, methodology. T. Matsuura: Conceptualization, data grants from Aileron Therapeutics (sponsored research), AstraZeneca (sponsored curation, software, formal analysis, project administration. M. Kajita: Data curation, research), Bayer Healthcare Pharmaceutical (sponsored research), Calithera formal analysis, funding acquisition. E. Nakamura: Resources, formal analysis, Biosciences (sponsored research), Curis Inc. (sponsored research), CytomX funding acquisition, visualization. Y. Funase: Resources, formal analysis. Therapeutics Inc. (sponsored research), Daiichi Sankyo Co. Ltd. (sponsored S. Matsushima: Resources, data curation, formal analysis. G. Ravizzini: research), Debiopharm International (sponsored research), Guardant Health Inc. Conceptualization, resources, data curation, formal analysis, supervision, (sponsored research), Millennium Pharmaceuticals (sponsored research), Novar- validation, investigation, visualization, methodology, writing-original draft, project tis (sponsored research), and Taiho Pharmaceutical Co. (sponsored research), administration, writing-review and editing.

AACRJournals.org Clin Cancer Res; 2020 OF11

Subbiah et al.

Acknowledgments interpretation of the data; the writing of the manuscript; or the decision to submit The study was funded by FUJIFILM Pharmaceuticals U.S.A., Inc. The authors the manuscript for publication. acknowledge the patients and their families for participating in this radiophar- maceutical ﬁrst-in-human clinical trial. V. Subbiah is supported by NIH/NCI The costs of publication of this article were defrayed in part by the grant 1R01CA242845-01A1. This work was supported in part by The Cancer payment of page charges. This article must therefore be hereby marked Prevention and Research Institute of Texas (RP1100584), the Sheikh Khalifa Bin advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate Zayed Al Nahyan Institute for Personalized Cancer Therapy, 1U01 CA180964, this fact. NCATS grant UL1 TR000371 (Center for Clinical and Translational Sciences), and The MD Anderson Cancer Center support grant (P30 CA016672). The Received January 4, 2020; revised April 26, 2020; accepted August 14, 2020; funders had no role in the design of the study; the collection, analysis, and published ﬁrst August 18, 2020.

References 1. Shapiro L, Weis WI. Structure and biochemistry of cadherins and catenins. 20. Stabin MG, Sparks RB, Crowe E. OLINDA/EXM: the second-generation per- Cold Spring Harb Perspect Biol 2009;1:a003053. sonal computer software for internal dose assessment in nuclear medicine. J Nucl 2. Paredes J, Figueiredo J, Albergaria A, Oliveira P, Carvalho J, Ribeiro AS, et al. Med 2005;46:1023–7. Epithelial E- and P-cadherins: role and clinical significance in cancer. 21. Emami B, Lyman J, Brown A, Coia L, Goitein M, Munzenrider JE, et al. Tolerance Biochim Biophys Acta 2012;1826:297–311. of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 1991;21: 3. Usui A, Ko SY, Barengo N, Naora H. P-cadherin promotes ovarian cancer 109–22. dissemination through tumor cell aggregation and tumor-peritoneum interac- 22. Weichselbaum R, Chen G, Hallahan D. Biological and physical basis of radiation tions. Mol Cancer Res 2014;12:504–13. oncology. In: Holland J, Frei EI, Bast RJ, Kufe D, Morton D, Weichselbaum R, 4. Imai K, Hirata S, Irie A, Senju S, Ikuta Y, Yokomine K, et al. Identification of a editors. Cancer medicine. 4th edition. Baltimore, MD: Williams & Wilkins; 1997. novel tumor-associated antigen, cadherin 3/P-cadherin, as a possible target for p. 697–726. immunotherapy of pancreatic, gastric, and colorectal cancers. Clin Cancer Res 23. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, 2008;14:6487–95. et al. European Organization for Research and Treatment of Cancer, National 5. Park J, Park E, Han SW, Im SA, Kim TY, Kim WH, et al. Down-regulation of Cancer Institute of the United States, National Cancer Institute of Canada. J Natl P-cadherin with PF-03732010 inhibits cell migration and tumor growth in Cancer Inst 2000;92:205–16. gastric cancer. Invest New Drugs 2012;30:1404–12. 24. Forero A, Meredith RF, Khazaeli MB, Shen S, Grizzle WE, Carey D, et al. Phase I 6. Kumara H, Bellini GA, Caballero OL, Herath SAC, Su T, Ahmed A, et al. study of 90Y-CC49 monoclonal antibody therapy in patients with advanced non- P-Cadherin (CDH3) is overexpressed in colorectal tumors and has potential as a small cell lung cancer: effect of chelating agents and paclitaxel co-administration. serum marker for colorectal cancer monitoring. Oncoscience 2017;4:139–47. Cancer Biother Radiopharm 2005;20:467–78. 7. Mandeville JA, Silva Neto B, Vanni AJ, Smith GL, Rieger-Christ KM, Zeheb R, 25. Gulec SA, Cohen SJ, Pennington KL, Zuckier LS, Hauke RJ, Horne H, et al. et al. P-cadherin as a prognostic indicator and a modulator of migratory Treatment of advanced pancreatic carcinoma with 90Y-clivatuzumab tetraxetan: behaviour in bladder carcinoma cells. BJU Int 2008;102:1707–14. a phase I single-dose escalation trial. Clin Cancer Res 2011;17:4091–100. 8. Sakamoto K, Imai K, Higashi T, Taki K, Nakagawa S, Okabe H, et al. Significance 26. Milowsky MI, Nanus DM, Kostakoglu L, Vallabhajosula S, Goldsmith SJ, Bander of P-cadherin overexpression and possible mechanism of its regulation in NH. Phase I trial of yttrium-90-labeled anti-prostate-specific membrane antigen intrahepatic cholangiocarcinoma and pancreatic cancer. Cancer Sci 2015;106: monoclonal antibody J591 for androgen-independent prostate cancer. J Clin 1153–62. Oncol 2004;22:2522–31. 9. Vieira AF, Paredes J. P-cadherin and the journey to cancer metastasis. 27. Vallabhajosula S, Goldsmith SJ, Kostakoglu L, Milowsky MI, Nanus DM, Bander Mol Cancer 2015;14:178. NH. Radioimmunotherapy of prostate cancer using 90Y- and 177Lu-labeled 10. Zhang CC, Yan Z, Zhang Q, Kuszpit K, Zasadny K, Qiu M, et al. PF-03732010: a J591 monoclonal antibodies: effect of multiple treatments on myelotoxicity. fully human monoclonal antibody against P-cadherin with antitumor and Clin Cancer Res 2005;11:7195s–200s. antimetastatic activity. Clin Cancer Res 2010;16:5177–88. 28. Wong JYC, Chu DZ, Yamauchi DM, Williams LE, Liu A, Wilczynski S, et al. A 11. Williams R. Discontinued drugs in 2011: oncology drugs. Expert Opin Investig phase I radioimmunotherapy trial evaluating 90yttrium-labeled anti- Drugs 2013;22:9–34. carcinoembryonic antigen (CEA) chimeric T84.66 in patients with metastatic 12. Strosberg J, Krenning E. 177Lu-dotatate for midgut neuroendocrine tumors. CEA-producing malignancies. Clin Cancer Res 2000;6:3855–63. N Engl J Med 2017;376:1391–2. 29. Picozzi VJ, Ramanathan RK, Lowery MA, Ocean AJ, Mitchel EP, O'Neil BH, et al. 13. Strosberg J, El-Haddad G, Wolin E, Hendifar A, Yao J, Chasen B, et al. Phase 3 (90)Y-clivatuzumab tetraxetan with or without low-dose gemcitabine: A phase trial of (177)Lu-dotatate for midgut neuroendocrine tumors. N Engl J Med 2017; Ib study in patients with metastatic pancreatic cancer after two or more prior 376:125–35. therapies. Eur J Cancer 2015;51:1857–64. 14. Hofman MS, Violet J, Hicks RJ, Ferdinandus J, Thang SP, Akhurst T, et al. 30. Dolgin E. Radioactive drugs emerge from the shadows to storm the market. [(177)Lu]-PSMA-617 radionuclide treatment in patients with metastatic cas- Nat Biotechnol 2018;36:1125–7. tration-resistant prostate cancer (LuPSMA trial): a single-centre, single-arm, 31. Kim C, Subramaniam D, Liu S, Giaccone G. P2.12–03 phase I/II trial phase 2 study. Lancet Oncol 2018;19:825–33. of 177Lu-DOTA0-Tyr3-octreotate (lutathera) and nivolumab for patients 15. Emmanouilides C, Witzig TE, Wiseman GA, Gordon LI, Wang H, Schilder R, with extensive-stage small cell lung cancer (ES-SCLC). J Thorac Oncol 2018; et al. Safety and efficacy of yttrium-90 ibritumomab tiuxetan in older patients 13:S791. with non-Hodgkin's lymphoma. Cancer Biother Radiopharm 2007;22:684–91. 32. Ocean AJ, Pennington KL, Guarino MJ, Sheikh A, Bekaii-Saab T, Serafini AN, 16. Lehnert M, Ludwig H, Zojer N. Update on the rational use of Y-ibritumomab et al. Fractionated radioimmunotherapy with (90) Y-clivatuzumab tetraxetan tiuxetan in the treatment of follicular lymphoma. Onco Targets Ther 2009;2: and low-dose gemcitabine is active in advanced pancreatic cancer: a phase 1 trial. 199–208. Cancer 2012;118:5497–506. 17. Hino A, Naruse M, Funase Y, Nagano A, Kasahara H, Matsuura T, et al. 33. Bauer R, Valletta D, Bauer K, Thasler WE, Hartmann A, Muller M, et al. Preclinical evaluation of anti-cadherin 3 human-mouse chimeric monoclonal Downregulation of P-cadherin expression in hepatocellular carcinoma induces antibody FF-21101 radioimmunotherapy. J Nucl Med 2013;54:1367. tumorigenicity. Int J Clin Exp Pathol 2014;7:6125–32. 18. Satoh H, Nomura F, Aikawa Y, Kotaka R, Watanabe M, Saito Y, et al. Anti- 34. Bialucha CU, Collins SD, Li X, Saxena P, Zhang X, Durr C, et al. Discovery and Cadherin 3 antibody labeled with Y-90 effectively inhibits tumor growth of non- optimization of HKT288, a cadherin-6-targeting ADC for the treatment of small cell lung cancer xenograft in nude mice by radioimmunotherapy. J Nucl ovarian and renal cancers. Cancer Discov 2017;7:1030–45. Med 2010;21:1151. 35. Yoshioka H, Yamamoto S, Hanaoka H, Iida Y, Paudyal P, Higuchi T, et al. In vivo 19. Sgouros G. Bone marrow dosimetry for radioimmunotherapy: theoretical con- therapeutic effect of CDH3/P-cadherin-targeting radioimmunotherapy. siderations. J Nucl Med 1993;34:689–94. Cancer Immunol Immunother 2012;61:1211–20.

OF12 Clin Cancer Res; 2020 CLINICAL CANCER RESEARCH

Phase I Study of 90Y-FF-21101 in Solid Tumors

36. Malik P, Phipps C, Edginton A, Blay J. Pharmacokinetic consideration for with distinct histologic types of ovarian carcinomas. Hum Pathol 2006;37: antibody-drug conjugates against cancer. Pharm Res 2017;24:2579–95. 1042–9. 37. Lobo ED, Hansen RJ, Balthasar JP. Antibody pharmacokinetics and pharma- 39. Darai E, Scoazec JY, Walker-Combrouze F, Mlika-Cabanne N, Feldmann G, codynamics. J Pharm Sci 2004;93:2645–68. Madelenat P, et al. Expression of cadherins in benign, borderline, and malignant 38. Sarrio D, Moreno-Bueno G, Sanchez-Estevez C, Banon-Rodriguez I, Hernandez- ovarian epithelial tumors: a clinicopathologic study of 60 cases. Hum Pathol Cortes G, Hardisson D, et al. Expression of cadherins and catenins correlates 1997;28:922–8.

AACRJournals.org Clin Cancer Res; 2020 OF13

Phase I Study of P-cadherin−targeted Radioimmunotherapy with 90Y-FF-21101 Monoclonal Antibody in Solid Tumors

Vivek Subbiah, William Erwin, Osama Mawlawi, et al.

Clin Cancer Res Published OnlineFirst August 18, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-20-0037

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2020/08/18/1078-0432.CCR-20-0037.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2020/09/29/1078-0432.CCR-20-0037. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.