Preparation and characterization of test sample and placebo
The microbial strain used in this work was cultured after inoculation with C. lacerata, originally owned by FugenCellTech, Co. Ltd., Korea, in potato dextrose agar (Difco. Co., Maryland, USA) medium at 25°C for 9 days. As a pre-culture process, a liquid medium of the CLM was mixed with 4 g/L of starch, 20 g/L of glucose, and 600 mL of purified water, following the previous literature [15] for 10 days at pH 5, 23℃, and rotation speed of 300 rpm. After the pre-culture was completed, the mycelium culture medium was transferred to a liquid medium prepared by mixing 12.5 g/L of sucrose, 2.5 g/L of skim soybean meal, 2.5 g/L of starch, 0.125 g/L of antifoam agent, and 400 L of purified water. The culture was further incubated continuously for 9 days by injecting air at a rotational speed of 100 rpm. The submerged culture medium of CLM was freeze-dried and pulverized, and then used according to the capacity of each experimental group based on the dry weight. The raw material was manufactured without using any toxic or unauthorized chemicals and solvents at a highly controlled GMP-certified plant in FugenCellTech Co. Ltd. (acquired permission for functional nutraceutical from the Korean Ministry of Food and Drug Safety on Dec. 2019). CLM tablets were prepared using a freeze-dried culture medium of CLM. As the excipient, it is composed of crystalline cellulose, hydroxypropyl methylcellulose, silicon dioxide, and magnesium stearate. The formulation was stored at room temperature, which is not only easy to supply, but also secured its biological safety through GLP-toxicity tests such as single-dose toxicity tests (rodents), repeated-dose toxicity tests (rodents), and genotoxicity tests (return mutations, micronucleus tests, and chromosomal abnormalities tests). The placebo tablet was composed of lactose, crystalline cellulose, hydroxypropyl methylcellulose, magnesium stearate, and caramel pigments. The formulation was stored at room temperature.
Clinical trial design and process
This study is a single-center, randomized, double-blind, placebo-controlled clinical trial. All processes were conducted at a single center (Kyung Hee University Hospital, Seoul, South Korea) from Feb. 15. 2015 (1st patients screened) to May. 5. 2015 (last patients completed). A total of 113 volunteers who have a glucose intolerance and mild T2D were participated in the screening procedure, and 72 subjects were met the criteria for selection and exclusion and enrolled in the present study.
To calculate the effective number of subjects, the following assumptions were employed.
- The statistical hypothesis test of the evaluation variable is a one-sided test.
- Level of significance is 5%
- Type 2 error (β) is set to 0.2 and the power of the test is maintained at 80%.
- The ratio of the number of test samples between the test group and the control group, e.g., (Jesus of the test group) = (Jesus of the control group), 1:1.
- After ingestion of test food, functional evaluation variables of the test group and the control group were compared.
To estimate the number of subjects, the criteria for evaluating the blood glucose lowering function was used as in the human application test [20], as below:
The hypothesis is as follows.
µt = µc (After the test, the measured value of the endpoint of the test group is the same as that of the control group)
H1: µt <µc (After the test, the measured value of the endpoint of the test group is less than that of the control group)
Assuming the above (1)-(5), the number of test samples required for the clinical trial is as follows (One-sided test)
Formula for calculating the number of subjects is expressed by the difference in the resulting variable as below,
(σ: standard deviation of the post-treatment change value in the previous trial, E: judged to be clinically significant, α =0.05 (Zα=1.645), β = 0.2 (Zβ=0.840), σ = 5.4).
When the number of subjects was calculated as follows by the above equation, the minimum number of subjects is about 23 per group.
The effective number of samples is 23 x (120/100) x (130/100) = 35.88, when calculated considering the 20% dropout rate and 30% compliance rate in the obtained sample number. Thus, the number of subjects to be enrolled in each group is 36, and the total number of subjects is 72.
The 72 selected subjects were randomly assigned to the test group or control group at a ratio of 1:1 based on the following steps: (1) Patients selected as study subjects were assigned to each group using a randomization method based on probability. (2) Random allocation table using a function that generates a random number of the SPSS program. (3) One of the two groups was assigned from the lowest number in the order of date of visit (Day 0). The random number (RN) was assigned according to the randomization table in the order of participation in the group to be assigned to the test group or the control group. (4) The management pharmacist applies the food for human application to be used for the subject according to the randomization plan. Food for the human application test was provided according to the code given to the subject. (5) Randomization of blocks using random code program of statistical program was designed to have the same number of subjects.
All physicians and outcome assessors were blinded to randomization allocation. During the experiment, neither the investigator nor the subjects were aware of each subject was assigned to, and they did not remove their blindfolds until a medical emergency occurred to protect the privacy of the assigned group. To ensure this, research activities, including screening, enrollment, informed consent, baseline data collection, randomization, and medication administration, were conducted solely by research personnel.
After randomly assigning the subjects to the CLM group (n=36) or placebo group (n=36) at a ratio of 1:1, the subjects of each group were administered 2 tablets of CLM (550 mg/tablet) or 2 tablets of placebo (550 mg/tablet) before meal 3 times a day for 6-12 weeks. All subjects were visited at the initial screening, baseline (visit 2), 6 weeks (visit 3), 12 weeks (visit 4) and follow up period.
Clinical trial subjects
Subjects corresponding to all the following criteria were recruited. After the investigator fully explained the purpose and method of this test, possible risks, and rewards to the subjects who wished to participate, a person who agreed to participate in this test in writing was selected as the final subject. The detail test schedule was described in Table S3. No changes after the trial began of the designation of outcomes as primary or secondary were found.
Inclusion criteria
(1) Subjects with 20-75 years of age, excluding illiterate people, and non-nursing women with no possibility of pregnancy.
(2) Subjects with 100-140 mg/dL of FBG, who do not take diabetic drugs. Subjects with less than 7.0% of HbA1c.
(3) Subjects with less than 110 mg/dL of FBG but 6.5%-7.0% of HbA1c.
(4) After fully understanding the detailed explanation of this clinical trial, the subjects voluntarily decide to participate. Those who have agreed in writing to comply with these precautions.
Exclusion criteria
(1) Subjects who experienced adverse reactions such as allergies when taking medicines, health functional foods, etc.
(2) Subjects with hypersensitivity to mushrooms or a history of the same reaction.
(3) Gastrointestinal diseases that may affect the absorption of the test product for human application (e. g., Crohn's disease) or a person with a history of gastrointestinal surgery (except for simple appendectomy or hernia surgery).
(4) A person who shows the following results in a diagnostic medical examination.
- AST, ALT> 2 times the upper limit of the normal range.
- Other significant diagnostic medical examination findings.
(5) Those who have the following clinically significant diseases; Diabetes patients taking hypoglycemic drugs or insulin, patients with uncontrolled hypertension (over 140/90 mmHg), patients with blood LDL-cholesterol over 160 mg/dL, patients with thyroid dysfunction, depression, schizophrenia, alcoholism, drug addiction, heart failure, angina pectoris, cardiovascular disease, or acute and chronic liver disease (chronic hepatitis B, chronic hepatitis C, various cirrhosis, liver cancer, etc.).
(6) Subjects who have taken anti-obesity, antidepressants, contraceptives, oral steroids, or female hormones. Subjects who have taken merchant hormones, or who have taken drugs that affect the absorption, metabolism, and excretion of the test food, or drugs that may affect blood sugar reduction.
(7) Pregnant women and lactating women.
(8) Those who participated in other clinical trial within 1 month before the first intake date.
(9) Subjects who cannot follow the requirements of clinical trial by investigator
(10) Subjects who proved to be inappropriate by other doctors.
Evaluation of efficacy and safety
For evaluation of anti-diabetic activity of CLM, laboratory test for FBG, HbA1c, insulin, and C-peptide were performed at each visit. Oral glucose tolerance test (OGTT) (2-hr postprandial glucose test) was performed at the baseline (visits 2) and after 12 weeks (visit 4).
HOMA-IR index is the most used method for estimating insulin resistance and was calculated as follows. The product of basal glucose (mmol/L) and fasting insulin (lU/mL) divided by 22.5. HOMA-IR by C-peptide was analyzed by replacing insulin with C-peptide in HOMA-IR formula. For safety evaluation, vital sign (systolic and diastolic blood pressure, pulse rate), electrocardiogram, laboratory test (complete blood cell count, chemistry laboratory test, urinalysis) and adverse events were thoroughly checked for subjects.
Statistical analysis
All data were statistically processed using SPSS Statistics (ver. 21.0) for Windows. Data processing for efficacy was based on an ITT analysis of subjects taking the test product for which at least one primary endpoint was measured. In addition, PP analysis, which analyzes data obtained from subjects who completed the study according to the human application test plan, was used as an auxiliary data. To evaluate the efficacy of the dropout subjects, the analysis was performed by applying the LOCF (Last Observation Carried Forward) method, which took the final record evaluated after administration of the test product or the control product. In all tests, a statistical treatment result of p<0.05 was considered significant. Less than three decimal places were not displayed.
(1) Primary efficacy evaluation method
The change in FBG at the end point (6 weeks, 12 weeks (LOCF)) after administration of the test group and the control group compared to the baseline value (visit 2) was analyzed by repeated ANOVA measurement. Student's t-test was used to analyze the amount of change after 12 weeks compared to the baseline. In addition, the results of ANCOVA analysis after correcting the baseline value and total calorie intake that may affect the efficacy indicators are also presented.
(2) Secondary efficacy evaluation method
The change in the parameters (postprandial blood glucose, glycated hemoglobin (HbAlc), insulin, C-peptide, lipids) at the end of dosing compared to the baseline value was analyzed using student's t-test, and the change amount for 12 weeks in each group was analyzed using the paired t-test. In addition, the results of ANCOVA analysis after correcting the baseline value and total calorie intake that may affect the efficacy indicators are also presented.
Furthermore, a stratified analysis was performed based on fasting hyperglycemic subjects with a FBG >110 mg/dL on the basis that the median FBG at the baseline was 110 mg/dL. The median HOMA-IR at baseline was 1.66, which was clinically consistent with the value determined by the Japanese Diabetes Association for normal insulin resistance, thus sub-group analysis was performed based on HOMA-IR>1.66 as well as HOMA-IR<1.66. Descriptive statistics on the changes in glucose, insulin, C-peptide, HOMA-IR, and HOMA IR by C-peptide after ingestion of samples at 6-12 weeks compared to the baseline were achieved for each group, and the degree of change before and after the sample intake within the group was measured by paired t-test. Intergroup comparisons of changes at each time point were also presented to evaluate statistically significant difference by performing two-sample t-test or Wilcoxon rank-sum test depending on whether normality was satisfied. Analysis of HbA1c was performed by Chi-square test.
(3) Safety endpoint and analysis method
For safety analysis, chi-square test was performed by identifying the number of subjects with adverse reactions by group based on laboratory test items. For 12 weeks, we analyzed whether there were significant differences between the two groups in adverse reactions, laboratory test results, and vital signs between the test group and the control group.