Ginkgo Extract Is Safe in Patients with Hepatocellular Carcinoma and Does Not Affect Tumor Response, but It May Increase Survival

Hepatocellular carcinoma (HCC) is a type of liver cancer and is the third greatest cause of cancer-related death worldwide. When diagnosed in the advanced stage, it is associated with a poor prognosis and death after a median of 6 months if untreated. A common therapy is a sorafenib, which is an oral multikinase inhibitor that increases median survival times by 2-3 months. HCC involves many signaling pathways, so researchers are looking at combining sorafenib with other therapies to increase the benefits. Many in vitro studies have shown that ginkgo (Ginkgo biloba, Ginkgoaceae) leaf extract (GBE) has activities against a variety of cancer cells, and it is reported to inhibit the development of HCC in a rat model. Hence, the purpose of this prospective, open-label, non-randomized study was to evaluate the efficacy and safety of a combination of GBE and sorafenib in treating patients with advanced HCC.

Patients (n = 27 in Phase 1 and n = 32 in Phase 2, aged ≥ 20 years) diagnosed with HCC by histological examination or typical diagnostic images participated in this study conducted at Tianjin First Center Hospital; Tianjin, China. Study dates were not given. Included patients had no indication for surgical resection or any other local therapy (e.g., ablation, chemoembolization, radiation therapy); had had no systemic chemotherapy; had measurable disease based on Response Evaluation Criteria in Solid Tumors (RECIST) of ≥ 1 untreated target lesion and ECOG (Eastern Cooperative Oncology Group) performance status 0-1; had neutrophil count ≥ 1500/μL, hemoglobin ≥ 9.0 g/dL, platelets ≥ 50,000/μL, total bilirubin < 3.0 mg/dL, aspartate aminotransferase (AST)/alanine aminotransferase (ALT) < 5x the upper limit of normal (ULN), albumin ≥ 2.8 g/dL, serum creatinine ≤ ULN, prothrombin time ≥ 40%, and Child-Pugh score A; were able to take food and drugs orally; and had life expectancy of ≥ 12 weeks. Patients were excluded if they had received previous therapy for HCC, an albumin preparation or blood transfusion, or any major surgery ≤ 30 days before study entry; had any surgery ≤ 15 days before study entry; had portal vein tumor thrombus in the primary trunk; had uncontrollable hypertension; had clinically significant pleural effusion, ascites, or pericardial fluid; had symptomatic hepatic encephalopathy, bone metastasis, brain lesions, or any central nervous system tumor; had active infection other than hepatitis B virus and hepatitis C virus (HCV) infection; had serious gastrointestinal bleeding ≤ 30 days before study entry; had gastroesophageal variances requiring treatment; were pregnant, lactating, or did not agree to use contraception during study treatment; or had a second primary cancer, except for in situ carcinoma or a cancer treated > 5 years ago without recurrence.

The first phase was to determine the maximum tolerated dose of GBE. A dosing scheme common to cancer trials was employed. Dosing was increased until a dose-limiting toxicity (DLT) occurred. DLTs included non-hematologic grade 3 or grade 4 toxicities that required treatment interruption for ≥ 7 days or that took ≥ 7 days to resolve to ≤ grade 1 despite treatment (excluding a list of common chemotherapy side effects, such as alopecia, hypertension, hypercholesterolemia, hand-foot skin reaction, and vomiting); hematologic toxicities of grade 4 or of grade 3 with specified complications or needing dose interruptions for > 7 days; increase in bilirubin or creatinine to ≥ 2x ULN; and hepatitis B reactivation or HCV flare. All patients received 400 mg sorafenib 2x/day. The first dose of GBE (obtained from Xi’an Honson Biotechnology Co.; Xi’an, China) was 60 mg/day (Cohort 1, n = 6). The second dose was 120 mg/day (Cohort 2, n = 8). The third dose was 240 mg/day (Cohort 3, n = 10). The fourth dose was 360 mg/day (Cohort 4, n = 3). The enrollment schedule was planned to minimize exposure of more patients than necessary to a dose that proved to be toxic. Patients remained on treatment until disease progression or toxicity. In the second phase, which enrolled 32 patients, the efficacy, and safety of the optimal GBE dose were evaluated. All patients were followed for 1 year for survival. Physical examination, vital signs, height and weight, ECOG performance status, and laboratory parameters were assessed. Best overall response rate (efficacy) was evaluated by RECIST with tumors assessed by standard computed tomography/magnetic resonance imaging of the abdomen at baseline and at 8-week intervals until the end of treatment. Blood was collected for pharmacokinetic assessment.

In Phase 1, no patient in Cohort 1 or 2 had a DLT. In Cohort 3, 1 patient had a DLT (grade 4 AST/ALT elevation). In Cohort 4, 2 of 3 patients had a DLT (grade 3 hand-foot skin reaction [Note: The paper elsewhere suggested that such reactions were not enumerated among DLTs.] and grade 3 gastrointestinal bleeding). Due to the high rate of DLTs in Cohort 4, it was decided that the dose used in Cohort 3 (240 mg/day GBE plus 400 mg 2x/day sorafenib) would be the maximum tolerated dose and the dose used in Phase 2.

In Phase 2, 2 patients (6%) stopped treatment because of adverse effects (AEs). Treatment-related AEs included hyperbilirubinemia (94% of patients), AST elevation (94%), thrombocytopenia (84%), anemia (72%), ALT elevation (66%), hand-foot skin reaction (59%), and fatigue (53%). The most common grade 3/4 AEs were AST elevation (28%), thrombocytopenia (19%), neutropenia (19%), hyperbilirubinemia (13%), and ALT elevation (13%). This toxicity profile was comparable to the expected toxicity of sorafenib monotherapy, except that the frequency of elevated liver enzymes was higher than that expected from either sorafenib or GBE monotherapy.

Regarding efficacy, 3 patients (9.4%) had a partial response, 21 patients (65.6%) had stable disease, and 8 patients (25%) had progressive disease. Mean overall survival was 11.6 months. Time to disease progression was a median of 2.5 months, which is similar to sorafenib monotherapy, and the tumor response rate and disease control rate did not significantly differ from those reported in a previous study of sorafenib monotherapy. However, median survival time was 11.6 months, which represented a “slightly” improved overall survival compared with the prior cohort treated with sorafenib alone. In terms of pharmacokinetics, the increase in the maximum concentration of unspecified compounds with increasing dose was usually proportional to the dose, whereas the increase in minimum concentration was more than dose-proportional.

The authors conclude that since treatment-related toxicity was similar to that observed with sorafenib monotherapy, the combination was safe and tolerable. Since the addition of GBE to sorafenib did not appear to improve tumor response, yet might have modestly improved median survival by some other mechanism, the authors suggest that further research is warranted. Limitations of the study include the small sample size and those only Chinese patients were included. The results could differ in other populations. The authors state no conflict of interest.


Cai Z, Wang C, Liu P, Shen P, Han Y, Liu N. Ginkgo biloba extract in combination with sorafenib is clinically safe and tolerable in advanced hepatocellular carcinoma patients. Phytomedicine. November 15, 2016;23(12):1295-1300.