E7386 in patients with advanced solid tumors: results from the dose-escalation part and an expansion part of a phase I study.

Introduction
The multifunctional protein, β-catenin, co-regulates transcription of genes targeted by the Wnt signaling pathway, a tightly regulated signal transduction system required for tissue homeostasis in tumorigenesis and metastasis. Activation of the canonical (β-catenin–dependent) Wnt signaling pathway stabilizes β-catenin by releasing its complex with factors including adenomatous polyposis coli (APC). This causes translocation of β-catenin into the nucleus, where it forms a complex with transcriptional co-factors such as CREB binding protein and activates the transcription of target genes.1, 2, 3 Gain-of-function mutations of β-catenin (CTNNB1) and loss-of-function mutations in APC result in aberrant activation of the canonical Wnt signaling pathway.4 These genetic alterations are often associated with tumor growth in various types of cancers, such as hepatocellular carcinoma (HCC), colorectal cancer (CRC), anaplastic thyroid cancer, endometrial cancer, and desmoid tumors,5, 6, 7, 8, 9 making inhibition or modulation of the canonical Wnt signaling pathway an attractive therapeutic strategy.
E7386 is an orally active protein–protein interaction inhibitor reported to block the CREB binding protein/β-catenin interaction. Notably, E7386 suppressed polyp formation in the intestine of ApcMin/+ mice and inhibited tumor growth in a human gastric cancer cell line (ECC10) xenograft model, and in a breast cancer mouse model with overactivation of the canonical Wnt pathway.10 Similarly to the known clinical effects of Wnt targeting agents,11,12 effects of E7386 administration on bone formation were observed at doses of ≥60 mg/kg in rats (Data on file). E7386 has also been reported to up-regulate the activating transcription factor 4 pathway and activate a cellular integrated stress response in preclinical models of HCC.13 E7386 may therefore impact other pathways in addition to the canonical Wnt pathway, potentially contributing to its antitumor activity.13,14
Here, we present findings from analyses of safety, recommended dose, antitumor activity, and pharmacokinetics (PK) of E7386 in the dose-escalation part and expansion part 1 (hereinafter referred to as expansion part) of the phase I Study 103 in patients with advanced, unresectable, or recurrent solid tumors, including patients with CRC and other gastrointestinal tumors.

Patients and methods

Trial design and ethics
This study (ClinicalTrials.gov identifier: NCT03833700) was a multicenter, open-label, phase I study conducted in Japan to evaluate the safety and tolerability of E7386 monotherapy in patients with solid tumors. The design consisted of an initial dose-escalation part, followed by expansion parts (expansion part 1 is reported here; expansion part 2 will be reported in the future; Figure 1). In the dose-escalation part, dose levels were assigned using a traditional 3 + 3 design. The initial dose level selected was 10 mg twice daily (b.i.d.) continuously in 28-day cycles, which is one level lower than the lowest dose with dose-limiting toxicity (DLT) observed in Study 101 (ClinicalTrials.gov identifier: NCT03264664; i.e. the first-in-human phase I, dose-finding study on E7386 monotherapy conducted in the United States and the UK). In the dose-escalation part, each dose cohort (approximately three patients) was to include at least one patient with CRC. E7386 doses of 10, 15, and 20 mg b.i.d. were predefined for the dose-escalation part, per protocol (Figure 1; Supplementary Appendix, available at https://doi.org/10.1016/j.esmoop.2025.105893). For dosages >20 mg b.i.d., the dose-escalation range was 100% if a treatment-related grade 1 or no toxicity was observed. If a treatment-related grade 2 toxicity occurred in one patient in the E7386 20 mg b.i.d. cohort, the dose-escalation range was reduced to 40%-60%. If a treatment-related grade 2 toxicity occurred in two or more patients, and/or a treatment-related grade 3 or higher toxicity occurred in one or more patient(s) in this cohort, the dose-escalation range was reduced to 25%-39%. Select doses from the dose-escalation part (100 mg b.i.d. and 120 mg b.i.d.) were used for the expansion part to further assess the safety, tolerability, recommended dose, and preliminary antitumor activity of E7386. Any prophylactic treatments to prevent adverse events (AEs) related to E7386 during the DLT evaluation period or until an AE occurred were prohibited in the dose-escalation and the expansion part.
The protocol, informed consent form, and appropriate related documents were reviewed and approved by the applicable local institutional review boards at the National Cancer Center Hospital (approved on 13 February 2019), National Cancer Center Hospital East (approved on 18 November 2019), Kyushu University (approved on 4 October 2021), and Shizuoka Cancer Center (approved on 21 July 2021). This study was conducted in accordance with the standard operating procedures of the sponsor, which are designed to ensure adherence to guidelines as required by the Principles of the World Medical Association Declaration of Helsinki and Good Clinical Practice. All patients provided written informed consent.

Patients
In the dose-escalation part, eligible patients were aged ≥20 years and had a histological and/or cytological diagnosis of advanced, unresectable, or recurrent solid tumor, for which no alternative standard or effective therapy exists. Eligible patients in the expansion part had histological and/or cytological diagnosis of advanced, unresectable, or recurrent CRC after two or more prior systemic anticancer therapies. Patients with other gastrointestinal tumors such as small bowel carcinoma (SBC) and gastrointestinal neuroendocrine tumors who received at least one prior systemic anticancer therapy were also included. Additional inclusion criteria, irrespective of patients’ tumor types, included an Eastern Cooperative Oncology Group performance status of 0 or 1, adequate renal/liver/bone marrow functions, and adequate serum mineral concentrations. Patients with bone disease or conditions (e.g. diagnosed osteoporosis with a T-score of < −2.5 by dual energy X-ray absorptiometry (DXA) scan and bone metastasis not treated with a bisphosphonate or denosumab) were not eligible. Patients continued to receive study drug until disease progression, development of unacceptable toxicity, patient request, withdrawal of consent, or study termination by the sponsor.

Study objectives and assessments
The primary objective of the study was to assess the safety and tolerability of E7386 in patients with solid tumors. Secondary objectives were to evaluate the PK profile and preliminary antitumor activity of E7386.
Safety assessments consisted of monitoring and recording AEs of all severity grades including DLTs, and serious AEs graded according to the Common Terminology Criteria for Adverse Events of the National Cancer Institute version 5.0. DLTs were defined as AEs related to E7386 occurring during cycle 1 (28 days; Supplementary Appendix, available at https://doi.org/10.1016/j.esmoop.2025.105893). Patients who prematurely withdrew from the study for reasons other than a DLT, or who failed to complete ≥75% of the planned total dose without DLT in cycle 1, were replaced.
Bone assessments including DXA scan and serum C-terminal telopeptide of type I collagen β isomer (β-CTX) levels a marker of increased bone turnover, were included. In the dose-escalation part, bone density was to be evaluated by DXA scan at screening, on cycle 2 day 1 (C2D1) and every 12 weeks (±7 days) from C1D1, and fasting β-CTX in serum was to be evaluated at screening, C1D15, C2D1, C3D1, and then every 8 weeks (C5D1, C7D1, etc.); evaluation was also to occur at the end of treatment visit. In the expansion part, bone density was to be evaluated by DXA scan at screening, on C2D1 (±7 days), every 12 weeks (±7 days) from C1D1, end of treatment visit and if clinically indicated, and fasting β-CTX in serum was to be evaluated at screening, C2D1, C3D1, and then every 8 weeks (C5D1, C7D1, etc.); evaluation was also to occur at the end of treatment visit and as clinically needed. Blood plasma samples for circulating tumor DNA (ctDNA) analyses were collected predose at baseline; C2D1; D1 of subsequent cycles and at the end-of-treatment visit; and conducted by the GuardantOMNI™ assay (Guardant Health, CA). OncoKB™15,16 annotation was used to filter variants and oncogenic or likely oncogenic mutations were included. Blood samples for PK analyses in the dose-escalation part were collected predose and 0.25, 0.5, 1, 2, 4, 6, 8, and 12 h after the first dose of C1D1 and C1D8, as well as predose on D1 of each cycle of C2-6. In the expansion part, samples were collected predose and 0.5, 1, and 2 h after the first dose on C1D1 and C1D8, as well as before the first dose of C2D1 onward. At the specified visits, all hematology, blood chemistry, and urinalysis (including pregnancy test, as applicable) samples were to be obtained before study drug administration, with results reviewed before study drug administration at the beginning of each treatment cycle (except C1D8).
Tumor assessments were carried out with computed tomography/magnetic resonance imaging every 8 weeks, at the end-of-treatment visit, and if clinically indicated by investigators, based on images acquired by the same imaging modality and under the same operating conditions as the screening scan. The efficacy endpoints were: the proportions of patients with an objective response [complete response (CR) or partial response (PR)], disease control [CR, PR, or stable disease (SD)], and clinical benefit (CR, PR, or durable SD lasting ≥23 weeks), assessed by investigators using Response Evaluation Criteria in Solid Tumors version 1.1.

Statistical methods
The DLT analysis set included all patients who received study drug as planned in C1 and who experienced a DLT regardless of study drug adherence in the dose-escalation part. All tolerability analyses were to be carried out on the DLT analysis set. The safety analysis set included all patients who received at least one dose of the study drug. This was the analysis set for all safety evaluations except DLTs. The PK analysis set included all patients who received at least one dose of study drug and had at least one evaluable plasma concentration. The efficacy analysis set included all patients who received at least one dose of the study drug.
Plasma PK parameters of E7386 were derived from plasma concentrations by noncompartmental analysis using actual times based on plasma concentration data in the dose-escalation part. PK parameters calculated of E7386 included maximum observed plasma concentration (Cmax), time to reach maximum (peak) plasma concentration (tmax), area under the plasma concentration-time curve (AUC), and accumulation ratios.

Results

Patients
From March 2019 to June 2022, a total of 55 patients (dose-escalation part: 36 patients; expansion part: 19 patients) were enrolled and received at least one dose of E7386. The cut-off date for data analysis, which occurred 24 weeks after C1D1 of the last enrolled patient in Study 103, was 19 July 2024; one patient with a desmoid tumor treated with 80 mg b.i.d. from the dose-escalation part had treatment ongoing with E7386. All patients were Japanese; the majority were male (≥60% in both parts) and had an Eastern Cooperative Oncology Group performance status of 0 (dose-escalation part: 86.1%; expansion part: 73.7%; Table 1). The median age was 61.5 years (range: 36-81 years) in the dose-escalation part and 70.0 years (range: 37-81 years) in the expansion part (Table 1). In the dose-escalation part, the most frequent types of tumors included were CRC (20 patients; 55.6%) and biliary tract cancer, endometrial cancer, and SBC (two patients each; 5.6% each; Table 1). Per study design, the most frequent primary tumor type in the expansion part was CRC (16 patients; 84.2%).
Over half of those enrolled were heavily pretreated and had received three or more prior systemic anticancer medications (dose-escalation part: 86.1%; expansion part: 73.7%). In the dose-escalation part, three to four patients each received E7386 at doses ranging from 10 to 80 mg b.i.d.; six and eight patients received E7386 at 100 and 120 mg b.i.d., respectively. Two patients received E7386 160 mg b.i.d. The median duration of treatment was 1.81 months (range: 0.5-46.1 months) in the dose-escalation part. In the expansion part, 9 patients received E7386 100 mg b.i.d. and 10 received 120 mg b.i.d.; the median duration of treatment was 1.71 months (range: 0.2-7.3 months).
Based on baseline ctDNA mutation status, in the dose-escalation part, 19 patients (52.8%) had mutations in APC and 1 patient (2.8%) had CTNNB1 mutation(s); 22 patients (61.1%) had mutations in KRAS and 24 patients (66.7%) had mutations in TP53 (Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2025.105893). Correspondingly, in the expansion part, 12 patients (63.2%) had mutations in APC and 2 patients (10.5%) had mutations in CTNNB1; 9 patients (47.4%) had mutations in KRAS and 15 patients (78.9%) had mutations in TP53 (Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2025.105893).

Dose-limiting toxicities
Grade 3 decreased appetite (caused by uncontrollable nausea and/or vomiting despite treatment with antiemetics) occurred in two patients in C1 of the E7386 160 mg b.i.d. cohort; both events met the criteria for a DLT. Conversely, no DLTs were observed in the E7386 10-120 mg b.i.d. cohorts. Per protocol, the expansion part commenced at E7386 100 mg b.i.d. (after the tolerability of this dosage was confirmed in the dose-escalation part) in parallel with the dose-escalation part at E7386 120 mg b.i.d. Since the dosage of 120 mg b.i.d. was also considered tolerable per data from the dose-escalation part, this dose was additionally selected for evaluation in the expansion part to further investigate the safety and tolerability of E7386.

Safety
Treatment-emergent adverse events (TEAEs) occurred in 35 (97.2%) of 36 patients in the dose-escalation part and in 18 (94.7%) of 19 patients in the expansion part (Table 2). The most frequent TEAEs were nausea (dose-escalation part: 83.3%; expansion part: 84.2%) and vomiting (dose-escalation part: 61.1%; expansion part: 73.7%). Nausea and vomiting were mostly of grade 1 or 2 severity (i.e. one patient had grade ≥3 nausea in the expansion part) and were generally well controlled with serotonin (5-HT3)-receptor antagonists, except in two patients in the 160 mg b.i.d. cohort (both patients had DLTs of intolerable decreased appetite caused by uncontrollable nausea and/or vomiting). The most frequent grade ≥3 severity TEAE in the dose-escalation part was anemia, which occurred in four patients; in the expansion part, two patients each had grade ≥3 anemia or hyponatremia (Table 2). Overall, 15 patients in the dose-escalation part and 10 patients in the expansion part had grade ≥3 TEAEs (Table 2). No bone fractures were reported in either part of the study, although 7 (12.7%) of 55 patients showed asymptomatic, post-baseline elevations in β-CTX >1000 pg/ml and 1 (1.8%) of 55 patients had a T-score < −2.5 as per DXA scan during the study (spine: −2.4 at screening to −2.6 on C2D1). No patients died due to TEAEs. TEAEs led to dose interruption in 13 patients (36.1%) in the dose-escalation part and in 5 patients (26.3%) in the expansion part. TEAEs led to dose reduction in one patient (2.8%; decreased appetite) in the dose-escalation part and four patients [21.1%; nausea (n = 2), diarrhea, vomiting, alanine aminotransferase increased, and decreased appetite (n = 1 each)] in the expansion part. TEAEs led to study-drug discontinuation in two patients [5.6%; pneumonia and hemorrhagic cerebral infection (n = 1 each)] in the dose-escalation part, but these events were not considered to be related to treatment. No patient in the expansion part discontinued the study drug due to TEAEs.
A summary of treatment-related AEs is included in Supplementary Table S2, available at https://doi.org/10.1016/j.esmoop.2025.105893.

PK analysis
E7386 was administered b.i.d. during the study period. E7386 exposure generally increased with dose over the dosing range of 10-160 mg b.i.d. following single and repeat dose administration, although large inter-subject variability was observed (Table 3). The maximum plasma concentration of E7386 was reached ∼0.5-2.0 h after oral administration on C1D1; a similar trend was observed on C1D8. The geometric mean accumulation ratios of E7386 for each dose were 1.12-3.81 for AUC and 0.65-3.97 for Cmax.

Antitumor activity
In the dose-escalation part, two patients achieved PR (one patient with SBC in the E7386 15 mg b.i.d. cohort and one patient with a desmoid tumor in the E7386 80 mg b.i.d. cohort; Figure 2A). Both patients with PR had APC mutation(s); the responder with SBC had additional mutations in KRAS and TP53, all detected by ctDNA analysis at baseline. The patient with SBC had a PR on C3D1 (maximum tumor reduction on C7D1) and discontinued study drug on C8D15 due to progressive disease; the treatment period was ∼7.5 months in this responder (Supplementary Figure S1A, available at https://doi.org/10.1016/j.esmoop.2025.105893). As of the data cut-off date, the responder with a desmoid tumor continued treatment with E7386, and the treatment period lasted >3.5 years. Additionally, one patient with solid pseudopapillary neoplasm of the pancreas, in whom no known mutations were detected by ctDNA analysis, had durable SD with a treatment duration of >2.5 years in the E7386 30 mg b.i.d. cohort (Supplementary Figure S1A, available at https://doi.org/10.1016/j.esmoop.2025.105893). This patient experienced a reduction in tumor size (Figure 2A) although they did not achieve PR.
No PRs were observed in the expansion part (Figure 2B and C). Among patients with CRC in the expansion part, one patient with APC/KRAS/TP53 mutations receiving E7386 100 mg b.i.d. had durable SD with a treatment duration of ∼7 months (Supplementary Figure S1B, available at https://doi.org/10.1016/j.esmoop.2025.105893) but no tumor size reduction (Figure 2B). Among patients with other tumors in the expansion part, one patient with HCC and no known ctDNA mutations receiving E7386 120 mg b.i.d. showed durable SD [with tumor size reduction (Figure 2C)] with the longest treatment duration of ∼7 months (Supplementary Figure S1C, available at https://doi.org/10.1016/j.esmoop.2025.105893).

Discussion
E7386 was evaluated as monotherapy at doses ranging from 10 to 160 mg b.i.d. in 55 patients with advanced solid tumors in the dose-escalation part and expansion part of Study 103. Over half of the patients enrolled in both parts were heavily pretreated and had received more than three prior systemic anticancer medications. E7386 treatment was generally well tolerated. The most frequent TEAEs were nausea and vomiting in both the dose-escalation part and expansion part. These TEAEs were mostly of grade 1 or 2 severity and well managed with 5-HT3 receptor antagonists in patients receiving up to 120 mg b.i.d.; the recommended dose of E7386 as monotherapy was determined to be 120 mg b.i.d. in this study. These results are similar to those from a study on a different drug (DKN-01, a DKK-1 antibody17) targeting the Wnt signaling pathway, which revealed TEAEs of nausea (incidence: 48.3%) and vomiting (incidence: 24.1%)18; treatment with a TBL1 inhibitor, tegavivint, was also related to nausea (incidence: 33%) in patients with desmoid tumors.19
In a 7-day repeated-dose toxicity study in rats, effects of E7386 administration on bone formation were observed at doses of ≥60 mg/kg (Data on file). Moreover, bone fragility has been noted in previous clinical trials of other agents targeting the Wnt pathway,11,12 indicating that bone assessments may be important when evaluating E7386. Of note, no bone fracture was observed within either part of Study 103.
PK exposure, as measured by Cmax and AUC, increased with doses over the dosing range of 10-160 mg b.i.d. following single and repeat dose administrations, although large inter-subject variability was observed.
Several agents related to the Wnt signaling pathway20 are under development including DKN-01,17 MTG201 (a REIC/DKK-3 adenovirus),21 RXC004 (a porcupine inhibitor),22 tegavivint,23 and CWP23229124, but thus far, there have been no globally approved oncology-related drugs targeting the Wnt signaling pathway. Generally, antitumor activity of various Wnt signaling inhibitors and modulators as monotherapy has been limited. The objective response rate (ORR) with DKN-01 was 7.7% (2 of 26) in endometrial cancer18 and 4.2% (1 of 22) in non-small-cell lung cancer.25 Moreover, ORR was 0% (0 of 13) with MTG201 in pleural mesothelioma26 and 0% (0 of 25) with RXC004 in advanced solid tumors.22 Results from our study follow a similar pattern: preliminary antitumor activity of E7386 was observed in two patients with PRs in the dose-escalation part [one with SBC (APC/KRAS/TP53 mutations) and one with a desmoid tumor (APC mutation)]. It is plausible that these patients both had a high dependency on the Wnt signaling pathway, especially as the patient with a desmoid tumor had familial adenomatous polyposis, a disorder for which E7386 showed anti-polyposis activity in a preclinical mouse model.10
Mutations in APC are typically early initiation events in tumorigenesis of CRC, leading to the development of adenomas; multiple genetic alterations in genes such as KRAS and TP53 are further accumulated with progression.27 While mutations in APC might still play a role in progression, the dependency on these mutations might be lower than at the initial stages of tumorigenesis. Targeting the Wnt signaling pathway therefore may not induce antitumor activity in heavily pretreated patients with CRC. In our study, E7386 did not indicate antitumor activity in patients with CRC, irrespective of APC mutation status. Despite these findings, early clinical data suggest a more promising role for E7386 in combination with other anticancer agents. A study evaluating the combination of E7386 and lenvatinib is under way (NCT04008797) and has shown promising antitumor activity of the combination in patients with advanced endometrial cancer28; the dose-optimization part in patients with endometrial cancer is currently in progress.29

Conclusions
This phase I study indicates that E7386 has a manageable safety profile, with the majority of TEAEs being low-grade; symptoms were managed with interventions, and study drugs were interrupted and/or reduced as needed. In this heavily pretreated population, E7386 showed a dose-dependent PK profile and PRs in two patients: one with SBC (with mutations in APC/KRAS/TP53) and another with a desmoid tumor [with mutation(s) in APC]. Durable SD was observed in one patient with solid pseudopapillary neoplasm of the pancreas and in one patient each with HCC and CRC. Further exploratory analyses of biomarkers are ongoing, and assessments of patients with advanced, unresectable, or recurrent solid tumors expected to be highly dependent on the Wnt/β-catenin signaling pathway in expansion part 2 will be reported in the future.