A Phase I Study of the Pan-Notch Inhibitor CB-103 for Patients with Advanced Adenoid Cystic Carcinoma and Other Tumors

Abstract

Purpose: CB-103 selectively inhibits the CSL-NICD (Notch intracellular domain) interaction leading to transcriptional downregulation of oncogenic Notch pathway activation. This dose-escalation/expansion study aimed to determine safety, pharmacokinetics, and preliminary antitumor activity.

Experimental design: Patients ≥18 years of age with selected advanced solid tumors [namely, adenoid cystic carcinoma (ACC)] and hematologic malignancies were eligible. CB-103 was dosed orally in cycles of 28 days at escalating doses until disease progression. Notch-activating mutations were required in a dose confirmatory cohort. Endpoints included dose-limiting toxicities (DLT), safety, tumor response, pharmacokinetics, and pharmacodynamics. Exploratory analyses focused on correlates of Notch and target gene expression.

Results: Seventy-nine patients (64, 12 dose-escalation cohorts; 15, confirmatory cohort) enrolled with 54% receiving two or more lines of prior therapy. ACC was the dominant tumor type (40, 51%). Two DLTs were observed [elevated gamma-glutamyl transferase (GGT), visual change]; recommended phase II dose was declared as 500 mg twice daily (5 days on, 2 days off weekly). Grade 3-4 treatment-related adverse events occurred in 15 patients (19%), including elevated liver function tests (LFTs), anemia, and visual changes. Five (6%) discontinued drug for toxicity; with no drug-related deaths. There were no objective responses, but 37 (49%) had stable disease; including 23 of 40 (58%) patients with ACC. In the ACC cohort, median progression-free survival was 2.5 months [95% confidence interval (CI), 1.5-3.7] and median overall survival was 18.4 months (95% CI, 6.3-not reached).

Conclusions: CB-103 had a manageable safety profile and biological activity but limited clinical antitumor activity as monotherapy in this first-in-human study.

Significance: CB-103 is a novel oral pan-Notch inhibitor that selectively blocks the CSL-NICD interaction leading to transcriptional downregulation of oncogenic Notch pathway activation. This first-in-human dose-escalation and -confirmation study aimed to determine the safety, pharmacokinetics, and preliminary antitumor efficacy of CB-103. We observed a favorable safety profile with good tolerability and biological activity but limited clinical single-agent antitumor activity. Some disease stabilization was observed among an aggressive NOTCH-mutant ACC type-I subgroup where prognosis is poor and therapies are critically needed. Peripheral downregulation of select Notch target gene levels was observed with escalating doses. Future studies exploring CB-103 should enrich for patients with NOTCH-mutant ACC and investigate rational combinatorial approaches in tumors where there is limited success with investigational or approved drugs.

FIGURE 1

Waterfall (A) and swimmer (B) plot among all n = 40 patients with evaluable ACC color coded by best overall response. Time to progression is denoted along with patient survival and Notch signaling alteration status at last data cutoff (March 5, 2022). C, Case example of a patient with metastatic ACC with rapid progression prior to study entry showing −25.7% tumor regression in her dominant liver lesion (26.1–19.4 mm from left to right, top) and stable disease over 8 months on study; and a patient with locoregionally recurrent ACC (NOTCH-activating mutation present) with −17% tumor regression in a right parapharyngeal mass (30 to 25 mm from left to right, bottom) with stable disease over 3 months while on study.

FIGURE 2

PFS (A) and OS (B) among the ACC subgroup (n = 40). Kaplan–Meier method; number of events (%), and median and 95% CIs shown.