Phase 2 study of lenvatinib monotherapy as second-line treatment in unresectable biliary tract cancer: primary analysis results

Background Biliary tract cancer (BTC) has a poor prognosis and lacks a standardized second-line therapy. Vascular endothelial growth factor (VEGF), fibroblast growth factor receptor (FGFR) 4, and platelet-derived growth factor receptor (PDGFR) are highly expressed in BTC. Therefore, lenvatinib (a known inhibitor of VEGF receptors 1–3, FGFRs 1–4, and PDGFR-α) was evaluated for second-line treatment of BTC. Methods In this single-arm, multicenter, open-label, phase 2 study, patients with BTC received lenvatinib 24 mg orally once daily in 28-day cycles. The primary endpoint was objective response rate (ORR). Secondary endpoints included overall survival (OS), progression-free survival (PFS), PFS rate at 12 weeks, disease control rate, clinical benefit rate, safety and pharmacokinetic profiles. Results Twenty-six Japanese patients were enrolled and treated; 3 had a confirmed partial response per investigator assessment and per independent imaging review (IIR); ORR was 11.5% (90% confidence interval [CI]: 3.2–27.2). Median PFS was 3.19 months (95% CI: 2.79–7.23) per investigator assessment and 1.64 months (95% CI: 1.41–3.19) per IIR. Median OS was 7.35 months (95% CI: 4.50–11.27). Grade ≥ 3 treatment-emergent adverse events (TEAEs) occurred in 21 patients (80.8%) and included hypertension (n = 10 [38.5%]), proteinuria (n = 3 [11.5%]), palmar-plantar erythrodysesthesia (n = 3 [11.5%]), decreased appetite (n = 3 [11.5%]), and anemia (n = 3 [11.5%]). Two deaths occurred due to TEAEs between treatment initiation and 30 days after last dose, but neither were considered treatment related. Conclusions Lenvatinib demonstrated antitumor activity in BTC, with a tolerable safety profile, and should be further evaluated as potential second-line therapy for this difficult to treat population. Trial registration ClinicalTrials.gov NCT02579616. Date of registration: October 19, 2015.


Background
Biliary tract cancer (BTC) is the second-mostcommon hepatobiliary cancer worldwide [1,2] and includes gallbladder cancer, intrahepatic cholangiocarcinoma, and extrahepatic cholangiocarcinoma [3]. Ampulla of Vater cancer is sometimes characterized as a biliary tract cancer [1,3], although the National Comprehensive Cancer Network does not include it under hepatobiliary cancers [4]. BTC incidence increased 25% worldwide between 2007 and 2017 (approximately 174,000 deaths) according to a Global Burden of Disease study [5]. A recent US study reported increasing incidence rates of gallbladder cancer in younger patients (< 45 years of age; 1.8% increase per year) and African Americans [3]. Patients with BTC have a poor prognosis and a shortened life expectancy (typically ≤1 year following diagnosis) [1,6].
Currently, radical surgery is the only potentially curative therapy, but this is not an option for many patients who present with advanced disease [7]. The standard first-line therapy for BTC is gemcitabine and cisplatin (GC) [4]. Gemcitabine plus S-1 (GS) [8] and GC plus S-1 have demonstrated potential as first-line therapies [9]. Unfortunately, for patients who progress on or after first-line therapies there are no approved subsequent treatment options [4]. Thus, there is an unmet need for an effective second-line therapy for patients with BTC.
The overexpression of VEGF, FGFR, and PDGFR in BTC suggests that lenvatinib could play a role in the treatment of BTC. This phase 2 study evaluated the safety and efficacy of lenvatinib as second-line therapy for patients with BTC [19].

Study design
This study (Study 215; NCT02579616) was a singlearm, multicenter, open-label, phase 2 study in Japanese patients with unresectable BTC. Patients received lenvatinib 24 mg orally once daily in 28-day cycles. Treatment continued until development of an unacceptable toxicity, disease progression, withdrawal of consent, or documentation of significant violations of the prespecified inclusion/exclusion criteria.
The primary endpoint was objective response rate (ORR). Secondary endpoints included overall survival (OS), progression-free survival (PFS), PFS rate at 12 weeks, disease control rate (DCR), clinical benefit rate (CBR; the proportion of patients with complete response + partial response + durable stable disease [≥ 23 weeks]), and safety and pharmacokinetic profiles. Tumor assessments were performed every 6 weeks until week 24, and then every 8 weeks thereafter, utilizing Response Evaluation Criteria In Solid Tumors (RECIST), version 1.1, by investigator assessment for the primary analysis. Independent imaging review (IIR) was utilized to support the post hoc analysis. Complete or partial responses required confirmation ≥28 days after the initial response.
The safety profile was assessed by monitoring and recording all adverse events (AEs), including all Common Terminology Criteria for Adverse Events, version 4.03, grades and serious AEs; periodic laboratory evaluations for hematology, blood chemistry, and urine values; periodic measurement of vital signs; electrocardiograms; and physical examinations. Toxicity was managed by supportive medications, treatment interruption, dose reduction (to 20 mg, 14 mg, or 10 mg; re-escalation was not allowed), and/or treatment discontinuation in accordance with protocol-prespecified dose-modification guidelines. Briefly, hypertension was managed by initiating antihypertensives if blood pressure was ≥ 140 mmHg (systolic) or ≥ 90 mmHg (diastolic), and then by dose interruption and reduction if blood pressure was ≥ 160 mmHg (systolic) or ≥ 90 mmHg (diastolic), despite optimal management with antihypertensive medications. Lenvatinib was discontinued upon occurrence of any grade ≥ 4 treatmentrelated AEs.
Plasma samples were collected from all patients on cycle 1, day 1 (C1D1; postdose), C1D8 (predose), C1D15 (pre/postdose), and C2D1 (predose) to assess the pharmacokinetic profile. Validated liquid chromatography with tandem mass spectrometry was utilized to determine lenvatinib plasma concentrations. Plasma concentrations were compared to the levels observed in patients from Study 303 (a phase 3 study of patients with DTC treated with lenvatinib 24 mg once daily) [20].
The primary analysis was performed as planned at the data cut-off (November 22, 2016) when all patients had finished their week 32 tumor assessment or had discontinued treatment. All patients provided written informed consent. The study protocol, informed consent form and any related documents were submitted to an Institutional Review Board for approval. This study was conducted in accordance with the World Medical Association Declaration of Helsinki, Good Clinical Practices, and local ethical/ legal requirements.

Eligibility
Patients enrolled must have experienced disease progression or treatment failure following 1 prior gemcitabine-based chemotherapy regimen (in combination with cisplatin or other platinum agent/fluoropyrimidine agent). Pathologically or cytologically confirmed unresectable adenocarcinoma of BTC, measurable disease per RECIST version 1.1, and an Eastern Cooperative Oncology Group performance status (ECOG PS) score of 0 or 1 were required. Additionally, eligible patients were required to be ≥ 20 years old with adequately controlled blood pressure (≤ 150/90 mmHg), adequate blood coagulation, and major organ function.
Key exclusion criteria comprised: any anticancer treatment within 21 days prior to the first dose of study drug, bleeding/thrombotic disorders, meningeal carcinomatosis, unstable brain/subdural metastases, or New York Heart Association Class ≥ 2 heart failure.

Statistical analysis
The sample size was determined based on the width of confidence interval (CI) using the 1-sample binomial distribution. Approximately, 25 patients were to be enrolled and if the true ORR was 15%, 3 (90% CI: 3.4-28.2) to 4 (90% CI: 5.7-33.0) responses were expected. Efficacy and safety, and pharmacokinetic assessments were performed on all patients who received at least 1 dose of study drug. The ORR, DCR, CBR, and corresponding exact 2-sided 90% CIs, were calculated using the Clopper-Pearson method. These endpoints were evaluated by investigator assessment and IIR. The Kaplan-Meier method was utilized to summarize OS, PFS, and PFS rate at 12 weeks. The Greenwood formula and log-log transformation were used to calculate the 95% CIs.

Patients
This study enrolled 26 Japanese patients, and all patients received at least 1 dose of lenvatinib. Primary tumor locations included gallbladder (n = 10), extrahepatic bile duct (n = 8), intrahepatic bile duct (n = 6), and the ampulla of Vater (n = 2). Most patients were male (57.7%), had an ECOG PS score of 0 (73.1%), and had metastases of the lymph nodes (61.5%) or liver (57.7%) ( Table 1). Additionally, the following baseline characteristics have previously demonstrated correlations with OS and/or PFS, and are shown in Table 1 according to the cutoff values used in previous research: white blood cell count, hemoglobin, alkaline phosphatase, albumin, and lesion size [7,21].

Pharmacokinetic profile
Lenvatinib trough plasma concentrations in Japanese patients with BTC were comparable to the levels seen in a previous study of Japanese patients with DTC (Study 303) [20]. Plasma concentrations and body-weightadjusted plasma concentrations from both studies at C1D15 are shown in Fig. 4.

Discussion
Here, we report the results of a phase 2 study evaluating lenvatinib as a second-line treatment option in    Recent studies have investigated other second-line therapy options: one phase 3 study (NCT01926236) suggested that the modified FOLFOX (mFOLFOX) chemotherapy regimen consisting of oxaliplatin and 5-fluorouracil should be considered the default second-line treatment for advanced/metastatic BTC [22]. The results of this study, which evaluated active symptom control (ASC) versus ASC plus mFOLFOX, were presented at ASCO 2019. The ASCO presentation reported a median OS of 6.2 months [22], median PFS of 4.0 months, ORR of 5%, and DCR of 33% in the ASC plus mFOLFOX arm. A statistically significant (P = 0.031) and clinically meaningful improvement in OS was observed in patients treated with ASC plus mFOLFOX versus ASC alone. Another chemotherapy regimen, FOLFIRINOX, demonstrated efficacy in a phase 2 trial of patients with BTC who had experienced disease progression following treatment with cisplatin and gemcitabine: median PFS and OS were 6.2 and 10.7 months, respectively [23,24]. Also, Abou-Alfa et al. [25] evaluated ivosidenib versus placebo in patients with advanced cholangiocarcinoma (primarily intrahepatic) and an isocitrate dehydrogenase 1 (IDH1) gene mutation. Of note, this population differed from our study, which enrolled patients without regard to a specific gene mutation. This phase 3 study, which allowed crossover from placebo to ivosidenib, reported the first positive PFS data of molecularly targeted therapy in cholangiocarcinoma [25]. Median PFS was 2.7 months versus 1.4 months in the ivosidenib and placebo arms, respectively (hazard ratio 0.37; 95% CI: 0.25-0.54; P < 0.001). Median OS was longer in the ivosidenib arm versus placebo (10.8 vs 9.7 months) but these results were not significant (P = 0.06) [25].
Previous studies have evaluated the efficacy of other tyrosine kinase inhibitors in the treatment of patients with advanced BTC [26][27][28][29]. A phase 2 study of regorafenib as second-line treatment in 43 patients with metastatic BTC demonstrated favorable results with 11% (n = 5) of patients achieving a partial response and a median PFS of 15.6 weeks (90% CI: 12.9-24.7 weeks) [27], which is approximately 3.9 months. Sunitinib demonstrated marginal efficacy as second-line treatment in a phase 2 study (n = 56) of metastatic BTC with a median time to progression of 1.7 months (95% CI: 1.0-2.4) and an ORR of 8.9% [28]. Lastly, a phase 2 study of sorafenib in patients with unresectable or metastatic gallbladder carcinoma and cholangiocarcinoma demonstrated a median PFS of 3 months (95% CI: 2-4) but was terminated early because it failed to meet the primary objective (ORR of 20%) [29]. However, a pilot study of sorafenib versus best supportive care in patients with advanced intrahepatic cholangiocarcinoma demonstrated sorafenib has antitumor activity with a median PFS of 3.2 months (95% CI: 2.4-4.1) and median OS of 5.7 months (95% CI: 3.7-8.5) [26]. Additionally, several phase 2 trials evaluating tyrosine kinase inhibitors, such as apatinib (NCT03521219), infigratinib (NCT02150967), derazantinib (NCT03230318), erdafitinib (NCT02699606), and pemigatinib (NCT04256980), as second-line treatment options in patients with BTC are currently ongoing [23]. In addition to direct antitumor activity, lenvatinib has also demonstrated immunomodulatory activity. Research has shown that lenvatinib modulates cancer immunity in the immunocompetent tumor microenvironment by reducing the population of tumor-associated macrophages    IIR (B). a One patient was assigned a BOR of "unknown" by IIR and was excluded from this analysis. BOR, best overall response; IIR, independent imaging review; PD, progressive disease; PR, partial response; SD, stable disease (TAMs) and increasing the population of interferon-γand granzyme-B-producing CD8+ T cells [30,31]. Further, lenvatinib demonstrated enhanced antitumor activity preclinically via the interferon-signaling pathway in combination with a programmed cell death-1 (PD-1) inhibitor [31]. Atanasov et al. [32] evaluated the relationship between the prevalence of TAMs and tumor growth in patients with hilar cholangiocarcinoma, a subtype of BTC. This study reported that overall tumor recurrence was significantly higher in patients with high levels of TAMs at the tumor invasive fronts compared with patients with low levels of TAMs (69.2% vs 33.3%; P = 0.015). Patients with high levels of TAMs experienced worse survival outcomes [32]. These preclinical data suggest that lenvatinib in combination with a PD-1 inhibitor may demonstrate further improved outcomes and therefore should be investigated. Further, a phase 2 trial of lenvatinib in

Conclusions
Lenvatinib demonstrated promising antitumor activity in patients with unresectable BTC who had failed gemcitabine-based chemotherapy. Additionally, the safety profile of lenvatinib in patients with BTC is similar to that previously reported in patients with other tumor types [34][35][36], and no new safety signals were identified. Toxicities were manageable with treatment modifications, dose reductions, or discontinuations. Noteworthy limitations of this study include the small sample size (n = 26), and the large diversity in BTC subgroups that were included (gallbladder cancer, n = 10; intrahepatic cholangiocarcinoma, n = 6; extrahepatic cholangiocarcinoma, n = 8; and ampulla of Vater cancer, n = 2). However, the inclusion of several BTC subgroups allows for easier trial enrollment, which in turn enables earlier clinical advances in the treatment of BTC. Further investigation of lenvatinib monotherapy or combination therapy with a PD-1 inhibitor as a potential second-line treatment option for patients with unresectable BTC is warranted, and ideally should be conducted in larger patient populations of each BTC subgroup.