Reviewed: Ladas EJ, Kroll DJ, Oberlies NH, et al. A randomized, controlled, double-blind, pilot study of milk thistle for the treatment of hepatotoxicity in childhood acute lymphoblastic leukemia (ALL). Cancer. 2010;116(2):506-513.

Milk thistle (Silybum marianum, Asteraceae) is a medicinal plant prized for its hepatoprotective effects. A case report has suggested that it may be beneficial in treating liver toxicity caused by cancer chemotherapy.¹ During the maintenance phase of treatment, chemotherapy in children with acute lymphoblastic leukemia (ALL) is often interrupted due to liver toxicity, which is the main cause of cumulative long-term drug withdrawals that can potentially increase the risk of bone marrow relapse in pediatric ALL patients.

In this randomized, double-blind, multi-center, pilot clinical trial, researchers examined the safety and effectiveness of a proprietary milk thistle extract in reducing hepatotoxicity in pediatric ALL patients receiving maintenance-phase chemotherapy.

The milk thistle seed (technically the fruit) extract product used in this study was Siliphos_®* (marketed by Thorne Research;^† Sandpoint, ID; produced by Indena SpA, Milano, Italy), a proprietary 1:2 mixture of silibinin and soy phosphatidylcholine. Silibinin is a 1:1 mixture of silybin A and silybin B, which are two of the 7 flavonolignans found in milk thistle extract. (The nomenclature of various milk thistle constituents is somewhat confusing; a recent review explains the differences in detail.²)

The researchers conducted in vitro bioassays using CCRF-CEM T-cell ALL cells to investigate potential antagonism of the cytotoxic chemotherapeutic agents used in ALL treatment. The trial was conducted on children, adolescents, and young adults, aged 1-21 years, who were in the maintenance phase of chemotherapy and receiving treatment according to the protocols of the Children’s Cancer Group, the Children’s Oncology Group, or the Dana Farber Cancer Institute ALL Consortium. These protocols utilized the following pharmaceutical drugs: vincristine, prednisone or dexamethasone, 6-mercaptopurine or thioguanine, and methotrexate. The participants were eligible for the study if they had a hepatic toxicity grade of 2 or more according to the Common Toxicity Criteria of the National Cancer Institute on at least 1 of 3 tests: the levels of liver enzymes aspartate amino transferase (AST), amino alanine transferase (ALT), or total bilirubin (TB).

The patients were randomized to receive either Siliphos (n=23) or a placebo (n=26) orally for 28 days beginning the day after administration of intravenous chemotherapy. The researchers measured hepatic toxicity on days 0, 28, and 56. Weekly telephone interviews and returned medication containers were used to monitor compliance, and the researchers defined “adherence” as the completion of at least 80% of the assigned treatment. A standardized questionnaire administered by a research assistant over the telephone was used to monitor adverse effects.

Each Siliphos capsule was designed to contain 240 mg of Siliphos standardized to 80 mg of silibinin, with a target dose of 5.1 mg/kg/day silibinin. The doses were 80 mg/day for 15-20 kg patients, 160 mg/day for 21-40 kg patients, 240 mg/day for 41-60 kg patients, and 320 mg/day for 61-70 kg patients. The placebo capsules were similar in odor and appearance to the Siliphos capsules. The purity, quality, and stability of the Siliphos and placebo capsules were assessed by the North Carolina laboratory of Nick Oberlies, PhD, one of the co-authors of this trial. Dr. Oberlies has been studying the chemistry of milk thistle for over 5 years and has found “a slight but consistent” overfill of the Siliphos capsules, resulting in 281.6 mg capsules with 97 mg silibinin. Each capsule provided 42.4 mg silybin A and 54.6 mg silybin B. The stability of the material was consistent over the 21-month course of study.

The in vitro results revealed that Siliphos does not reduce the cytotoxicity of the chemotherapeutic drugs vincristine or L-asparaginase. The results also showed “a degree of antileukemic synergy between vincristine and silibinin,” based on fixed concentration-ratio experiments and cellsurvival data. Between May 2002 and August 2005, 50 patients were recruited for the study, and 1 patient withdrew due to refusal to take the Siliphos. At baseline and at day 28, there were no significant differences in ALT, AST, or TB between the groups. (While the abstract states that on day 56, AST was significantly lower in the Siliphos group compared to the placebo group [P=0.05], the text states that the P value was 0.04.) The authors did not observe significant differences between the groups in the changes in ALT and AST levels from baseline to day 28. There was a trend toward greater reduction of ALT (P=0.07). The researchers did not observe significant inter-group differences in TB levels. By day 28, they observed a reduction in TB greater than 50% in 5 patients in the Siliphos group and none in the placebo group (P<0.007).

There were no significant differences in the toxicities, rates of infection, chemotherapy doses, reductions of doses, or delays in treatment between the groups. Patients in the Siliphos group reported diarrhea, flatulence, irritability, and stomachache, while the placebo group reported decreased appetite, diarrhea, stomachache, and soft stools. All patient-reported adverse side effects were mild and pre-existing complaints. Compliance was only 68% in the Siliphos group (nonadherence) and 96% in the placebo group.

The authors conclude that Siliphos “may be a safe, effective, supportive-care agent.” The results show that Siliphos reduces AST levels in children receiving chemotherapy for ALL and does not antagonize the actions of vincristine and L-asparaginase. The authors state that the study was not sufficiently powered to detect significant treatment effects at day 28. In addition, the lower level of compliance in the Siliphos group may have obscured treatment effects. Further research is needed to confirm these results and to determine the appropriate dose and duration of treatment. The authors suggest future clinical trials on Siliphos in cancer patients for whom “hepatic toxicity prevents provision of the recommended chemotherapy.”