Coriolus versicolor ( Yunzhi ) Use as Therapy in Advanced Hepatocellular Carcinoma Patients with Poor Liver Function or Who Are Unfit for Standard Therapy

Background

Hepatocellular carcinoma (HCC) is endemic in Asia and increasing in the United States. ¹ Advanced HCC is often not amenable to resection or local regional treatments and outcomes are dismal. ² Despite a low clinical response and modest survival benefit, sorafenib has become the standard of care for Child–Pugh A advanced HCC. ³ In reality, majority of advanced, inoperable HCC have Child–Pugh B or C liver cirrhosis and end up on supportive care alone. Many turn to consuming alternative medicine. Yet, literature surrounding its use in cancer treatment is limited.

Coriolus versicolor (CV), a fungus of the Basidiomycetes family, has historically been used for medicinal purposes. ⁴ In China it is called Yunzhi and in Japan, kawaratake. Polysaccharide peptide (PSP) isolated from CV has shown potential in cancer treatment with immune-modulating properties such as reducing the rate of cell proliferation, induction of apoptosis, and suppression of DNA and RNA synthesis. ^5,6 Of interest, PSP induced apoptosis of human promyelocytic leukemia HL-60 cells but not normal human T lymphocytes. ⁶ This selectivity offers great potential in cancer treatment.

The authors report the use of CV in individuals with inoperable HCC and Child–Pugh Class C liver cirrhosis; and HCC patients with Child–Pugh A or B liver cirrhosis who have failed or were unfit for standard therapy. HCC was diagnosed either histologically or by imaging evidence of a space-occupying lesion in the liver together with a serum alpha fetoprotein (AFP) concentration >200 μg/L (subjects with chronic hepatitis B or C infection) or >400 μg/L (without chronic hepatitis B or C infection).

Methods and Results

Fifteen subjects were randomized 2:1 to CV or placebo (Table 1). The CV arm had a higher proportion with prior liver resection, chemotherapy, and radiotherapy. CV was given as a standard continuous daily dose of 2.4 g. Treatment on either arm was continued till disease progression or occurrence of unacceptable toxicities. Median treatment duration was 1.5 cycles (5.9 weeks) and 3 cycles (12.1 weeks) for placebo and CV, respectively. Seventy percent terminated treatment due to progressive disease or death. Both arms tolerated the treatment equally with no dose reductions or delays.

The median time to progression (TTP) was compared. Secondary endpoints include response rates, toxicity, quality of life (QOL) using the FACT HEP and EORTC QLQ-C30 questionnaires, progression-free survival (PFS), and overall survival (OS). TTP, PFS, and OS were analyzed by Kaplan–Meier and the hazard ratio (HR) estimated by Cox regression. QOL was assessed over several time points, however, due to the small numbers and high discontinuation rate, all postbaseline time points were summarized into a single score for each subject. A linear regression model was used to obtain the estimated difference in worst scores between both groups. All analyses were performed on an intention-to-treat basis.

No significant difference in TTP was observed between treatment groups. Median TTP was 2.5 (1.4–5.3) and 4.2 (0.4–4.2) months in the CV and placebo group, respectively, HR 0.70 (0.16–3.05), p = 0.63. Overall response rate was 11.1% (0.3–48.2) and 16.7% (0.4–64.1) in the CV and placebo group, respectively. There was a trend toward longer PFS and OS in the CV arm. Median PFS was 2.5 (1.4–5.3) and 1.1 (0.4–4.2) months, HR 0.42 (0.13–1.34), p = 0.144; and median OS was 6.5 (3.3–24.1) and 2.2 (0.8–23.3) months, HR 0.35 (0.10–1.25), p = 0.105 in the CV and placebo arm, respectively.

CV subjects reported better physical, emotional, cognitive, and social functioning compared to the placebo group during treatment on EORTC scores (Table 2). The lowest emotional and social functioning scores reported during treatment were on average 7.3 (−16.1, +30.6) points and 10.4 (−5.7, +26.5) points higher in the CV group compared to placebo. CV subjects reported on average 40.5 (−3.2, +84.3) points higher in terms of cognitive functioning on treatment compared to placebo. Lower average symptom scores for nausea, vomiting, pain, insomnia, constipation, and diarrhea were reported by CV patients on treatment compared to placebo. CV subjects experienced significantly less pain and appetite loss compared to placebo subjects during treatment. Using the FACT HEP questionnaires, the lowest physical and emotional well-being scores reported by CV subjects were on average higher than those reported by placebo subjects on treatment.

Correlative studies looking at serum concentrations of potential HCC markers were performed. CV subjects had reduced interleukin (IL)-17F and monocyte chemoattractant protein-1 (MCP-1) levels, and increased prolactin and TNF-related apoptosis-inducing ligand (TRAIL) R1 levels at the end of treatment compared to placebo subjects. CV subjects had a lower mean level of IL-17F compared to the placebo group after 4 weeks of treatment, ratio 0.61 (0.01, 33.60). This reduction in ratio was more pronounced at treatment completion, ratio 0.16 (0.004, 7.67). Prolactin levels were consistently higher in the CV group compared to placebo at all time points. MCP-1 levels were marginally lower in the CV group compared to placebo at baseline and this difference widened on treatment completion, −9.81 (−34.45, +14.84). TRAIL R1 ratio increased on treatment completion. Subjects with stable disease appeared to have lowered serum IL-17F, prolactin, and TRAIL R1 concentration ratios at end of treatment compared to subjects with disease progression. In contrast, AFP levels did not appear to change significantly with CV use.

Discussion

The conduct of studies on traditional medicines is often challenging with problems of drug standardization and various confounders. Herbal medicines are often used in various combinations, which makes it difficult to determine the true efficacy of each individual compound. This study investigated CV efficacy using a standardized herbal preparation with only one compound as the treatment drug.

The CV arm contained a larger percentage of cases with poorer liver function and prior treatment. Despite this, a trend toward improvement in PFS and OS was seen. These observations are interesting and warrant further larger studies. Further treatment postprogression may have affected these OS results. However, the authors do not have details of postprogression treatment although the numbers who received further therapy are likely low given the limited treatment options in this poor risk population. Data regarding the optimal dosing of CV are limited. This dose was based on recommendations by alternative medicine practitioners from whom the treatment drug was supplied.

These results suggest that advanced HCC patients taking CV had better social and emotional functioning scores and less appetite loss and pain symptoms compared to placebo subjects during treatment. These QOL results are limited by small patient numbers and the authors have tried to mitigate the effect of imbalances by adjusting for baseline in the analysis.

These data suggest that CV may inhibit the increase in IL-17F, maintain prolactin levels, and elevate TRAIL R1 in HCC patients. This is consistent with the previously reported role of these immunological markers. IL-17F increase has been reported in tumor growth and invasion. ⁷ Prolactin is significantly upregulated in HCC, affecting cell proliferation through PRLR/JAK2 signaling. ⁸ TRAIL is known to selectively induce apoptosis in transformed cell lines. ⁹ Previous reports suggest a decrease in TRAIL receptor 2 after treatment, with possible enhancement of tumor cell apoptosis. ¹⁰ These data suggest a similar interaction in HCC patients following CV treatment. Use of CV appears to stabilize IL-17F, prolactin, and TRAIL R1 levels with a trend toward improvement in PFS and OS. The authors are limited by the small numbers and the end of treatment time point differed for each subject making it difficult to draw conclusions about the role of these markers as surrogates for disease response.

In conclusion, patients with advanced HCC taking CV generally had similar efficacy with better QOL on treatment, compared to placebo subjects. The utility of this supplement in patients whose primary treatment goal is palliation should be further explored.