For this clinical study, we recruited 98 volunteers with knee joint pain, all of whom provided written informed consent to participate. Through screening tests, we finally selected 60 healthy Japanese subjects (aged 45–68 years) with subjective symptoms of knee joint pain that did not require hospital treatment, which was determined by a physician. The following volunteers were excluded: (1) individuals receiving treatment/medication for OA; (2) injured subjects who faced problems with daily life activities; (3) subjects receiving treatment and medication for rheumatoid arthritis (RA) or who were suspected to have RA; (4) subjects suspected to have secondary arthrosis; (5) subjects who regularly used anticoagulants, antiplatelet agents, or non-steroidal anti-inflammatory drugs (NSAIDs); (6) those with serious cerebrovascular, cardiac, hepatic, renal, or gastrointestinal disease; (7) those with a history of major surgery related to the digestive system; (8) those with unusually abnormal hematological data or serious anemia; (9) those who regularly used medicines, functional foods, and/or supplements that would improve knee joint function; (10) heavy smokers, alcohol addicts, or subjects with irregular lifestyles; (11) those with a severe allergic reaction to medicine or foods; (12) those who are pregnant or expected to be pregnant, or lactating during the study; and (13) those with other medical reasons as determined by the principal investigator.

The 60 remaining eligible subjects were randomly assigned to the active test food or placebo food groups. A third-party data center assigned these subjects to the aforementioned groups, stratified by sex, age, and the level of the sum score of questions 2-5 on the Japanese Knee Osteoarthritis Measure (JKOM) [10Akai M, Doi T, Fujino K, Iwaya T, Kurosawa H, Nasu T. An outcome measure for Japanese people with knee osteoarthritis. J Rheumatol 2005; 32(8): 1524-32.
[PMID: 16078330] ]. The allocations were computer generated using stratified block randomization. Doctors, nurses, clinical research coordinators, and statistical analyzers had no knowledge of the assignment information during this trial period. This information was only disclosed after the laboratory and analytical data were fixed and the method of statistical analysis was finalized.

This clinical study was conducted as a randomized, double-blinded, placebo-controlled parallel group comparison study at the Hokkaido Information University, Health Information Science Research Center (Ebetsu city, Hokkaido, Japan) in 2016 according to the general clinical trial design for assessing the efficacy of functional food [11Nishimura M, Sugawara M, Kudo M, Nishihira J. A randomized, double-blind, placebo-controlled study to examine the effects of high-isoflavone soybeans “Yukipirika” in climacteric women. Funct Food Health Dis 2017; 7(8): 637-60.]. The examination schedule for this study is shown in Table 1. We performed JKOM and Visual Analog Scale (VAS) questionnaires at weeks 0 (baseline), 4, and 8. A locomotive syndrome risk analysis was performed at weeks 0 and 8 after initiating the test food-intake period. A medical interview and physical, hematological, and biological examinations were performed at weeks 0, 4, and 8. We asked all subjects to consume two capsules of a test food containing either 100 mg of chondroitin sulfate oligosaccharides or placebo. During this study, subjects were asked not to change their daily activities, including food consumption and exercise habits and medication use, and to record a life diary during the study period. The primary outcome was the JKOM score, and the secondary outcome was the VAS questionnaire score and locomotive syndrome risk analysis.

Table 1
Examination schedule.

Chondroitin sulfate oligosaccharides were manufactured as follows: raw high-molecular-weight chondroitin sulfate was obtained by liquefying ray cartilage using protease, removing the peptide, and powdering the extract via spray drying. The powdered high-molecular-weight chondroitin sulfate was dissolved in deionized water and reacted continuously at 213ºC. Thereafter, the reaction solution was cooled, concentrated, and purified using filter pores to remove impurities. This purified solution was then powdered by spray drying to obtain a chondroitin sulfate oligosaccharide powder, which was then placed in capsules (active test food). The placebo food was manufactured according to the same procedure, except that fried wheat was used instead of chondroitin sulfate oligosaccharides. The results of the nutrient composition analyses of the active test food and placebo food used in this study are provided in Table 2. The nutrient composition analyses were performed by the Japan Food Research Laboratories (Chitose, Japan). Chondroitin sulfate oligosaccharides were analyzed using High-Performance Liquid Chromatography (HPLC) methods by Marukyou Suisan Co. Ltd. The active test food and placebo food were manufactured under strict quality control protocols by Marukyou Suisan Co. Ltd. and were identical in appearance.

Table 2
Nutrient compositions of the test foods (per day).

To evaluate the effects of chondroitin sulfate oligosaccharides on knee pain, the subjects completed a VAS questionnaire comprising three questions to assess (1) knee pain at rest, (2) knee pain at walking, and (3) knee pain while climbing up and down stairs. Subjects were instructed to mark a cross on a 100-mm line in response to each question based on their current health condition. Here, the left end of the line (0 mm) was defined as the worst condition and the right end (100 mm) as the best condition. The questionnaire results were assessed by evaluating the length from the left-hand start of the line to the cross. An increase in the VAS score indicated an improvement in that symptom.

The locomotive syndrome risk test comprised three parts: (1) a questionnaire regarding locomotive syndrome, (2) the two-steps test, and (3) the stand-up test [12 Otsuka Pharmaceutical Co. L. Locomotive Syndrome Rsik Test [Available from: https://www.otsuka.co.jp/en/health_illness/locomo/, 13Conference JLCP. Conference JLCP. Locomotive syndrome [Available from: https://locomo-joa.jp/en/index.pdf]. The locomotive syndrome questionnaire comprised 25 questions for the assessment of body pain and other factors. A decrease in this questionnaire score indicated an improvement. The two-step test measures walking ability including muscular strength, balance, and flexibility of the lower limbs. Subjects took two long strides, the longest they could without losing their balance. This test was performed twice, and the better result was recorded. The two-step score was calculated by dividing the length of both strides (cm) by the subject's height. The stand-up test assesses leg strength by having the subject stand up on one or both legs. Subjects stand up from each seat, starting at 40 cm, first with one leg. If subjects could stand up, and maintain the position for 3 seconds, next, they were made to stand up at 30 cm with one leg. If subjects could not stand up, they were made to stand up with both legs from the same-height seat which they could not stand up with one leg. The high score was of the following order: 10 cm with one leg, 20 cm with one leg, 30 cm with one leg, 40 cm with one leg, 10 cm with both legs, 20 cm with both legs, 30 cm with both legs, and 40 cm with both legs. We compared the 8-week score to the 0-week score, and evaluated “worse”, “no-change” and “improvement”.

Blood was collected after a 12-h fast. General blood tests were performed, including the measurement of complete blood counts (CBCs; white blood cells, red blood cells, hemoglobin, hematocrit, and platelets), liver function (aspartate aminotransferase, alanine aminotransferase, gamma glutamyl transpeptidase, alkaline phosphatase, and lactate dehydrogenase), and renal function (blood urea nitrogen, creatinine, and uric acid). Blood tests were performed by the Sapporo Clinical Laboratory, Inc. (Sapporo, Japan).

Each subject’s body composition and blood pressure were measured using a Body Composition Analyzer DC-320 (Tanita Corp, Tokyo, Japan) and an Automatic Blood Pressure Monitor HEM-7080IC (Omron Co., Ltd., Kyoto, Japan) respectively.

All subjects provided written informed consent prior to undergoing any of the tests related to this study. The Ethics Committee of Hokkaido Information University approved the study protocol conforming to the Helsinki Declaration and Ethical Guidelines for Medical and Health Research Involving Human Subjects (approval date, Jul 21, 2016; approval number 2016-04). This study was registered with UMIN Clinical Trials Registry (approval number UMIN000023492).

The sample size was statistically determined to obtain a power of 80% and a two-sided significance level of 5%. To demonstrate JKOM 2-5 at week 8, which was postulated to have an intergroup difference of 10 points with a standard deviation (SD) of 12 points, a sample size of 54 (27 in each group) was required. Assuming a 10% loss in the follow-up rate, 60 subjects were selected.

Values are presented in the tables as means ± standard deviations. Changes in subject values were analyzed using Mann-Whitney U test for evaluation of JKOM, VAS, and the questionnaire of locomotive syndrome, and Student’s t test for the evaluation of two-step test between active test food and placebo food at each evaluation point without statistical consideration for multiplicity. Frequency values were obtained with chi-square test for the evaluation of the stand-up test. p value <0.05 was considered to be significant. Statistical analyses were performed using SPSS Statistics 20 (IBM, Armonk, NY, USA).

Although we initially enrolled 60 subjects (active test food, n = 30, placebo food, n = 30), two subjects withdrew for personal reasons before the trial began. Additionally, two subjects withdrew during the trial (abnormal liver function value before test food intake, n = 1; injury, n = 1). Consequently, 56 subjects completed the trial (placebo food group, n = 26; active test food group, n = 30). No subject was excluded from efficacy analysis (placebo food group, n = 26; active test food group, n = 30). The subjects who ingested food even once were included in the safety analysis (placebo food group, n = 28; active test food group, n = 30). The study flow diagram is shown in Fig. (1). Table 3 presents the mean age, height, body weight, body mass index, score for JKOM 1, and summed score for JKOM 2-5 for each group. These data did not differ significantly between the groups, which confirmed the appropriateness of subject allocation. Moreover, the active test food and placebo food groups did not differ regarding the intake rate (Table 3).

Fig. (1) Flow diagram of the trial.

Table 3
Characteristics of subjects in the active test food and placebo groups.

Values are shown as means ± standard deviations. Student’s t test was used for analyses of age, height, body weight, and body mass index, and the chi-square test was used to evaluate sex. The Mann-Whitney U test was used to analyze JKOM scores and the intake rate. n, number of subjects.

We evaluated the effects of chondroitin sulfate oligosaccharides on JKOM scores (Table 4), but observed no differences between the active test food and placebo food groups regarding changes in JKOM. However, among subjects with worse VAS scores (using the median baseline total VAS score for stratification), the sum of JKOM scores 2–5 improved significantly in the active test food group relative to the placebo food group at weeks 8 (change from baseline to week 8: placebo, −3.85 ± 9.09 points; active, −11.33 ± 6.58 points, p = 0.016). Moreover, the JKOM score 1 tended to improve with active test food intake relative to the placebo at week 4 (change from baseline to week 4: placebo, −3.69 ± 20.18 mm; active, −18.40 ± 16.86 mm, p = 0.07).

Next, we confirmed the effects of chondroitin sulfate oligosaccharides on knee pain. We found no differences in the changes in VAS scores between the active test food and placebo food groups (Table 4).

Table 4
JKOM and VAS questionnaire scores.

To evaluate the effects of chondroitin sulfate oligosaccharides on locomotive syndrome, the subjects completed the questionnaire of locomotive syndrome, two-steps test, and stand-up test. The results of stand-up test exhibited a differing trend between the active test food group and placebo food group (p = 0.054) (Table 5). Next, to confirm whether the frequencies of the subjects’ score to the stand-up test differed between the two groups, we performed a residual analysis, which revealed that “worse” was significantly reduced in the active test food-intake group, compared to the placebo food group (adjusted residual: ±2.2). Although the active test food group exhibited improvements in the questionnaire of locomotive syndrome and two-steps test outcomes, these differences were not significant (Table 5).

Table 5
Locomotive syndrome test scores.

week 8, change from baseline to week 8. Values are shown as means ± standard deviations for the questionnaire of locomotive syndrome and two-steps test, and as frequencies for the stand-up test. The mean changes in subject values at each evaluation point were compared between the two groups using the Mann-Whitney U test for the questionnaire of locomotive syndrome and Student’s t test. The frequency values from the stand-up test were compared using the chi-square test.

We further evaluated blood pressure, body composition, CBCs, and liver and renal function after the subjects ingested the test foods (Table 6). Only minimal changes within the normal ranges were observed. Moreover, few subjects exhibited adverse effects, and these effects were mild and resolved within a few days. Therefore, the principal investigator determined that no adverse events were related to the ingestion of the test food. Additionally, no abnormal changes or severe adverse events were observed during the physical, hematological, and biological examinations or the medical interview. Accordingly, the 8-week intake of 100 mg of low molecular chondroitin sulfate was considered safe.

Table 6
Blood pressure, body composition, CBC parameters, liver, and renal function.