Safety and Antitumor Activity of Farletuzumab Ecteribulin in Patients With NSCLC: Phase I Expansion Results.

Introduction
Lung cancer is the second most common and most deadly cancer among men and women, with 2,206,771 new cases and 1,796,144 lung cancer–related deaths worldwide in 2020.1 The 5-year survival rate for lung cancer is only 25%.2 Despite advancements in lung cancer therapies, targeted treatment options remain a critical unmet need.3
Farletuzumab ecteribulin (FZEC; formerly MORAb-202) is an antibody–drug conjugate consisting of the humanized anti-folate receptor-alpha (FRα) monoclonal antibody, farletuzumab, and the potent microtubule dynamics inhibitor, eribulin.4 The drug components are conjugated by a cathepsin-B–cleavable linker with a drug-to-antibody ratio of 4:1.4 Notably, FRα is overexpressed in several solid tumors, including NSCLC, with 74% of lung adenocarcinomas being positive for FRα expression.5
Preclinically, FZEC has demonstrated cytotoxicity by directly binding to FRα-positive cancer cells and by inducing a cytotoxic bystander effect on adjacent tumor cells.6 Moreover, FZEC exerted noncytotoxic bystander effects in the tumor microenvironment.6, 7, 8 In the dose-escalation part of the first-in-human Study 101, FZEC demonstrated preliminary antitumor activity in 10 (one complete response [CR] and nine partial responses [PRs]) of 22 Japanese patients with pretreated solid tumors, including two of four patients with NSCLC who had a partial response (NCT03386942).9 FZEC doses ranged from 0.3 mg/kg to 1.2 mg/kg and were delivered intravenously every 3 weeks. The maximum tolerated dose was not reached, and treatment was well tolerated at all doses.
Based on these results, the dose-expansion part of Study 101 reported here explored FZEC safety and antitumor activity in patients with NSCLC treated with FZEC 0.9 mg/kg.

Materials and Methods

Study Design and Patients
The expansion part of Study 101 enrolled patients with platinum-resistant ovarian cancer10 or NSCLC; the latter are reported here. Patients were to receive doses of either 0.9 mg/kg or 1.2 mg/kg administered intravenously every 3 weeks.
However, the FZEC 1.2 mg/kg dose was never commenced for patients with NSCLC, considering preliminary data in other cohorts of the study. Initial treatment of FZEC was to be administered in more than or equal to 60 minutes, and subsequent doses were to be administered in more than or equal to 30 minutes if no infusion reactions were observed. All patients were to receive acetaminophen before the first FZEC infusion. Patients received treatment until criteria for discontinuation were met, including development of grade 3 or 4 pneumonitis or interstitial lung disease (ILD; reported as a group-term ILD/pneumonitis throughout), treatment interruption of more than 12 weeks because of ILD/pneumonitis, progressive disease, patient choice, or the investigator judged that discontinuation was appropriate.
Eligible patients were aged above or equal to 20 years with tumors measurable by Response Evaluation Criteria in Solid Tumors version 1.1 and Eastern Cooperative Oncology Group performance status score of 0 or 1. Patients were histologically or cytologically diagnosed with NSCLC and enrolled into cohorts of patients with NSCLC adenocarcinoma or NSCLC nonadenocarcinoma. Patients with adenocarcinoma were presumed to have high FRα expression and did not require additional testing before enrollment, whereas patients with nonadenocarcinoma were required to have confirmed FRα expression (i.e., ≥5% of neoplastic cells must have demonstrated staining at any intensity), as confirmed by central immunohistochemistry (IHC) assay.
Eligible patients were not indicated for EGFR-, BRAF V600E mutation-, ALK-, or ROS-targeted therapies. They must have had adequate function of major organs within 2 weeks before the first administration of the study drug. Definition of adequate function included the following: hemoglobin more than or equal to 9.0 g/dL, neutrophil count more than or equal to 1.5 × 103/μL, platelet count more than or equal to 10 × 104/μL, total bilirubin less than or equal to 1.5 × upper limit of normal (ULN) in the facility, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) less than or equal to 3.0 × ULN (in the case of liver metastases ≤5 × ULN) in the study facility, serum creatinine less than or equal to 1.5 × ULN in the facility, and albumin more than or equal to 3 g/dL.
Patients with active ILD/pneumonitis, a history of ILD/pneumonitis, who had received radiotherapy to the lung field within 12 months before the first dose of study intervention, or who had current clinically relevant lung disease (e.g., chronic obstructive pulmonary disease) were not eligible. Patients who received previous treatment with eribulin or other folate-receptor–targeting agents were also not eligible.
This study was conducted in accordance with standard operating procedures of the sponsor, which were designed to ensure adherence to the Principles of the World Medical Association Declaration of Helsinki, Japan's Good Clinical Practice, and the following local review boards: The Institutional Review Board of National Cancer Center (reference ID: T4468), the Cancer Institute Hospital of JFCR Institutional Review Board (reference ID: 2020-0019), and the Shikoku Cancer Center Institutional Review Board (reference ID: 2019-13). Written informed consent from all participants was obtained.

Objectives
The primary objective of Study 101 was evaluation of the tolerability and safety profile of FZEC in patients with solid tumors, and it has been reported previously.9 Here, we assess safety and report on secondary objectives, including evaluation of objective response rate (ORR), duration of response, progression-free survival (PFS), and overall survival (OS) in patients with NSCLC.

Assessments
Tumor response was assessed by the investigators per Response Evaluation Criteria in Solid Tumors version 1.1. Tumors were assessed at screening and every 6 weeks until week 36 and then every 8 weeks until treatment discontinuation. Patients were followed for survival every 12 weeks post-discontinuation up to 18 months after the last enrolled patient was first treated with FZEC. Safety was assessed by monitoring and recording all adverse events (AEs) and serious AEs (SAEs), in addition to regular laboratory monitoring and physical examinations. The severity of AEs was graded according to the Common Terminology Criteria for Adverse Events version 4.03. Assessment for ILD/pneumonitis was performed by investigators and an external ILD expert committee (two radiologists and one pulmonologist) based on computed tomography (CT) scans and clinical information, retrospectively. FRα status of tumor samples was assessed by IHC (FRα IHC assay kit [mAb 26B3.F2]).

Pharmacokinetic Analysis
Blood samples were collected on day 1 before dosing, at the end of infusion during cycles 1 to 6, and on day 8 at a single time point if the visit occurred. Additional samples were obtained within 7 days after treatment discontinuation and at the final observation, 30 days after the last dose.
FZEC area under the curve (AUC) of serum concentration–time at steady state was predicted using a previously reported population pharmacokinetic (PK) model.11

Statistical Analyses
Best overall response (BOR), ORR, disease control rate, and clinical benefit rate (CBR) were summarized with corresponding Clopper and Pearson’s exact two-sided 95% confidence intervals (CIs). ORR was defined as a BOR of CR or PR, disease control rate was defined as a BOR of CR, PR, or stable disease (SD), and CBR was defined as a BOR of CR, PR, or durable SD lasting more than or equal to 23 weeks. Responses of CR and PR were confirmed at subsequent tumor assessment after more than or equal to 4 weeks. Duration of response, PFS, and OS were summarized and plotted over time by the Kaplan–Meier method. To calculate CIs of their median, Greenwood’s formula12 and Brookmeyer and Crowley method13 were used. In the expansion part, an enrollment of 15 patients with adenocarcinoma for each dose group and up to 15 patients with nonadenocarcinoma (until completion of enrollment for adenocarcinoma) for each dose group was planned to allow for further evaluation of safety and preliminary antitumor activity. Assuming 15 patients per dose group, a certain clinical event rate (e.g., AEs of special interest) and an ORR of 20% in the lower dose group and 30% in the higher dose group, the probability of observing an equal or higher adverse event rate and ORR with the higher dose compared with that of the lower dose is 80%. Ultimately, only patients with adenocarcinoma were enrolled to receive FZEC 0.9 mg/kg every 3 weeks.

Results

Patients
A total of 15 patients (all with adenocarcinoma) were enrolled into the NSCLC expansion part of Study 101 between July and November 2020; each patient received at least one dose of FZEC 0.9 mg/kg. The median age of patients was 68.0 years (range, 45–78), 53.3% had an Eastern Cooperative Oncology Group performance status of 0, and 46.7% had FRα expression in more than or equal to 75% of neoplastic cells (Table 1). Most patients (n = 10, 66.7%) identified as former smokers, whereas the rest (n = 5, 33.3%) had never smoked. All patients (100%) had received at least two previous anticancer regimens (range, 2–7).
All patients had discontinued FZEC by the end of the study period. Of these patients, 14 (93.3%) discontinued because of radiological disease progression and one (6.7%) discontinued because of AEs (ILD/pneumonitis, discussed subsequently).

Safety
The median number of cycles was 2.0 (range, 2–13), and the median duration of treatment was 1.41 months (range, 1.4–9.9). The mean relative dose intensity weighted by duration of treatment was 95.4%.
Any-grade treatment-emergent AEs (TEAEs) were reported in 93.3% of patients (n = 14), and 13.3% of patients (n = 2) experienced a TEAE of grade more than or equal to 3 (grade 3 pulmonary embolism and grade 4 hyperuricemia [both n = 1]) (Table 2). No patients had a fatal AE. The most common TEAEs experienced by more than or equal to 20% of all patients were ALT increased (n = 6), AST increased (n = 5), ILD/pneumonitis (n = 5), and pyrexia (n = 3) (Table 3). Neutropenia and peripheral sensory neuropathy each occurred in 6.7% of patients (n = 1).
Two patients (13.3%) had treatment-emergent SAEs, both of which were ILD/pneumonitis (one was grade 1, the other was grade 2). One patient (6.7%) had a TEAE (grade 2 ILD/pneumonitis) that led to discontinuation of FZEC. TEAEs led to FZEC dose interruption in four patients (26.7%; ILD/pneumonitis [n = 2], epistaxis [n = 1], and pneumonia [n = 1]).
Among the five patients (33.3%) with ILD/pneumonitis (grade 1, n = 4; grade 2, n = 1), the median time to ILD/pneumonitis from first dose of FZEC was 36 days (range, 23–176). Three patients had recovered from ILD/pneumonitis and two were recovering (including a change from grade 2 to 1) by the end of follow-up. Three patients received corticosteroids (all were oral prednisolone) for management of ILD/pneumonitis and continued to receive them by the end of follow-up. The median initial corticosteroid dose was 60.0 mg/d (range, 30–60 mg/d) and 0.92 mg/kg/d (range, 0.4–1.0 mg/kg/d). The median duration of total use, including a tapering period, was 124.0 days (range, 64–145 d).
The independent ILD expert committee identified eight patients (53.3%) with ILD events. All patients who were identified by investigators were also identified by the independent ILD expert committee; conversely, three of these patients were judged to not have ILD/pneumonitis by investigators.

PK Analysis
Based on the observed PK values, the median of the individual model-predicted FZEC AUC at steady state was 3590 μg·h/mL (range, 1680–4370 μg·h/mL).

Antitumor Activity
Among all patients, there was one responder (Fig. 1A), for an ORR of 6.7% (95% CI: 0.2–31.9; duration: 2.3 mo) (Table 4). This responder had 100% FRα expression and had received more than or equal to four lines of prior therapy; tumor imaging is found in Figure 1B. Three patients had SD for more than or equal to 23 weeks, for a CBR of 26.7% (95% CI: 7.8–55.1). Median PFS was 2.7 months (95% CI: 1.2–4.9) (Fig. 2A), and median OS was not estimable (NE) (95% CI: 9.4–NE) (Fig. 2B).
Among the 11 FRα-positive patients, the ORR was 9.1% (95% CI: 0.2–41.3), the median PFS was 2.7 months (95% CI: 1.2–5.1), and median OS was NE (95% CI: 9.4 mo–NE) (Supplementary Table 1). Among the four FRα-negative patients, the ORR was 0% (95% CI: 0–60.2), the median PFS was 2.1 months (95% CI: 1.1–8.5), and the median OS was NE (95% CI: 8.2 mo–NE).

Discussion
This phase I study assessed the safety and antitumor activity of FZEC in patients with heavily pretreated NSCLC. Two patients (13.3%) experienced a grade more than or equal to 3 TEAE, two experienced a SAE, and no TEAEs leading to death were observed. The most common TEAEs were ALT increased, AST increased, ILD/pneumonitis, and pyrexia. The observed AEs were generally manageable—only one patient (6.7%) discontinued study treatment because of ILD/pneumonitis. ILD events occurred in 33.3% of patients (grade 1, n = 4; grade 2, n = 1). These ILD events were carefully monitored and managed by investigators according to severity, including consultation with a pulmonologist, regular follow-up visits, and treatment with corticosteroids if needed. These management strategies may have resulted in ILD events being predominantly low grade. The underlying mechanism of FZEC-induced ILD is not clearly understood. Previous studies of patients treated with either eribulin or farletuzumab monotherapy reported low incidences of ILD/pneumonitis (≤5%),14,15 suggesting a different mechanism of action for FZEC-induced ILD specific to the antibody–drug conjugate.
Although neutropenia and peripheral neuropathy are common AEs in patients treated with eribulin (the payload of FZEC),16 incidences of neutropenia and peripheral sensory neuropathy (both 6.7%) were low in this study. This may be because systemic exposure of free eribulin in patients treated with FZEC has been found to be considerably lower than with the approved dose of eribulin.9,17
The median AUC at steady state was 3590 μg·h/mL, with a range of 1680 to 4370 μg·h/mL. This value was comparable to the AUC reported for the 0.9 mg/kg cohort in the dose-escalation part of Study 101.9 Furthermore, no differences in PK levels were observed between NSCLC and other tumor types.
Although patient numbers were limited, modest antitumor activity was observed in one of 15 patients (6.7%) in this expansion part. Although the preliminary antitumor activity in the expansion part was not as pronounced as in the dose-escalation part, it is noteworthy that the tumor FRα expression rate in the patient who responded in the expansion part was 100%. Tumor responses were observed in two patients (with FRα expression levels of 90% and 100%) among four patients with advanced NSCLC in the previously reported dose-escalation part of this study.9 Thus, the pooled dose-escalation plus expansion part response rate was 15.8% (3/19), with each responder having more than or equal to 90% FRα expression. However, in the expansion part, a total of six patients exhibited FRα expression of 90% to 100%. In addition to the patient who achieved a BOR of PR, three of these patients had SD and two had PD. Therefore, when limited to cases with FRα expression of 90% to 100%, the ORR in the expansion part was 16.7% (one of six patients). Considering that the response in this dose-expansion part was found in one of four patients (25%) who received more than or equal to four lines of prior therapy, response rates overall and by prior treatment status were comparable to later-line chemotherapy for NSCLC18,19 and eribulin monotherapy in phase 3 study for heavily pretreated NSCLC.20 The relationship between FRα level and antitumor activity could not be adequately determined because of the small sample size. Therefore, further investigation of the correlation between the efficacy of FZEC and FRα expression level is needed, especially given the known bystander effect of FZEC.6
Although planned, FZEC 1.2 mg/kg was not administered to patients with NSCLC in this study per safety findings from the platinum-resistant ovarian cancer cohorts. Moreover, a population PK model of FZEC has demonstrated that dosing based on bodyweight alone results in a dose exposure that increased with increasing bodyweight, potentially leading to a greater risk of ILD in patients with higher bodyweights.11 Body surface area–based dosing is predicted to lower this exposure-dependent risk of ILD in patients with higher bodyweights and, thus, should be considered for FZEC in the future.
This study includes a small sample size of patients with NSCLC, which limits the generalizability of the results. Further studies with larger patient populations and a control arm are needed to support findings. Despite these limitations, the expansion part of Study 101 in patients with NSCLC demonstrated some modest antitumor activity, and the observed AEs were generally manageable. These findings are consistent with the results of the dose-escalation part9 and warrant further investigation to optimize dosing and to determine the potential of FZEC as a new cancer drug for NSCLC. In particular, the patient who had a response had a high level of FRα expression, which was consistent with the results of the dose-escalation part. Evaluation of the FZEC treatment regimen is ongoing in another clinical trial (NCT04300556). Further investigation is warranted to optimize the balance between antitumor activity and ILD risk and to assess the potential of FZEC as a new treatment for NSCLC in the appropriate population based on FRα expression.

CRediT Authorship Contribution Statement
Tatsuya Yoshida: Conceptualization, Data curation, Investigation, Resources, Writing – original draft, Writing – review & editing.
Toshiyuki Kozuki: Data curation, Investigation, Resources, Writing – review & editing.
Hiroki Ikezawa: Data curation, Formal analysis, Investigation, Resources, Writing – review & editing.
Wataru Yusa: Data curation, Investigation, Resources, Writing – review & editing.
Takuma Miura: Data curation, Investigation, Resources, Writing – review & editing.
Yohei Otake: Data curation, Investigation, Resources, Writing – review & editing.
Makoto Nishio: Data curation, Investigation, Resources, Writing – review & editing.

Disclosure
Yoshida reports receiving grants or contracts from Novartis, AbbVie, Amgen, Daiichi Sankyo, AstraZeneca, Merck Sharp & Dohme, Chugai, Astellas, Medpace, Boehringer Ingelheim, Bristol Myers Squibb, Ono, and Merck; receiving honoraria from Novartis, Daiichi Sankyo, AstraZeneca, Merck Sharp & Dohme, Chugai, Bristol Myers Squibb, Ono, Takeda, Pfizer, Lilly, and Merck; and having participated in advisory boards for Novartis, Merck Sharp & Dohme, Amgen, Chugai, Pfizer, and Boehringer Ingelheim. Kozuki reports receiving honoraria from AstraZeneca, Chugai Pharmaceutical, Eli Lilly Japan, Ono Pharmaceutical, Nippon Kayaku, Bristol Myers Squibb, Merck Sharp & Dohme, Pfizer, Nippon Boehringer Ingelheim, Daiichi Sankyo, Merck Biopharma, Takeda Pharmaceutical, Taiho Pharmaceutical, Bayer, Novartis, Amgen, AbbVie, and Sawai Pharmaceutical; receiving research funding from AstraZeneca, Chugai Pharmaceutical, Bristol Myers Squibb, Amgen, Ono Pharmaceutical, Daiichi Sankyo, Taiho Pharmaceutical, AbbVie, Sanofi, Merck Sharp & Dohme, Eisai, Eli Lilly Japan, Merck Biopharma, Kyowa-Hakko Kirin, and Labcorp Development Japan. Ikezawa, Yusa, Miura, and Otake report being employees of Eisai. Nishio reports receiving honoraria from Ono Pharmaceutical, Chugai Pharmaceutical, Taiho Pharmaceutical, Bristol Myers Squibb, Daiichi Sankyo, Lilly, AstraZeneca, Merck Sharp & Dohme, AbbVie, Takeda, Pfizer, Boehringer Ingelheim, Novartis, Nippon Kayaku, Merck, and Janssen.