An investigation into possible reasons behind inconsistent clinical outcomes of glucosamine in osteoarthritis patients

Osteoarthritis (OA) is an irreversible joint disease, characterised by the gradual depletion of articular cartilage. The main symptoms of OA include pain and stiffness of the affected joint, impaired joint function and ultimately, physical disability. Glucosamine products are widely used for the management of OA symptoms. Unlike various pharmacological agents, glucosamine products are relatively safe, cost-effective and are proposed as a symptomrelieving and structure modifying agent in OA. Apparently, the natural occurrence of glucosamine in the articular cartilage could be one of the reasons for its suggested role in OA. Many non-clinical studies have reported that glucosamine could alter the biochemical/pathophysiological changes in OA. However, clinical studies have presented inconsistent clinical outcomes; therefore, the efficacy of glucosamine as a structure- or symptom-modifying agent in OA is still unclear. There could be a number of possible reasons for the reported inconsistent clinical outcomes, including the study designs and/or study duration, use of different salts, dosage forms, dose regimens and/or different brands of glucosamine. Apart from this, one of the possibilities is the use of either pharmaceutical or non-pharmaceutical grade glucosamine products, since glucosamine is marketed as non-pharmaceutical grade, over-the-counter (OTC) products in the majority of non-European countries. Hence, the thesis hypothesised that the difference between the labelled and the actual amount of glucosamine present in the marketed products and/or poor dissolution profile of the product could be a possible reason for inconsistent clinical outcomes. On the other hand, glucosamine has low oral bioavailability and erratic first-pass metabolism; therefore, the thesis hypothesised that another possible reason for the discrepancies reported in the clinical results could be due to the differences in the pharmacokinetics, including bioavailability and elimination, of glucosamine in individuals. Therefore, the overall aim of this thesis was to investigate possible reasons behind inconsistent clinical outcomes of glucosamine, including assessing the quality of OTC glucosamine products, determining the association of different salts, doses, dosage forms, brands, and dosage regimens of glucosamine with the plasma levels of glucosamine in OA patients taking the therapy, and analysing pharmacokinetic differences between pharmaceutical and nonpharmaceutical grade glucosamine in healthy individuals. The specific aims of this thesis were to: i) develop a novel analytical method for quantification of underivatised glucosamine in dietary supplements (Chapter 2); ii) analyse the content and dissolution profile of glucosamine products commercially available in Australia and India (Chapter 3); iii) develop a novel and sensitive method for quantifying underivatised glucosamine in human plasma (Chapter 4); iv) investigate variations in plasma levels of glucosamine in OA patients with chronic dosing in a real-world scenario (Chapter 5); and v) investigate the inter-brand and inter-individual variations in the pharmacokinetics of glucosamine in healthy volunteers (Chapter 6). To fulfil the first objective, a new analytical method using high-performance liquid chromatography-coupled (HPLC) with a Corona charged aerosol detector (CAD) was developed and validated to quantify glucosamine in commercially available OTC products (Chapter 2). The developed method was found to be linear (r\\(^2\\) > 0.99, 10-200 µg/mL glucosamine), and the relative standard deviations (%RSD) for intra- and inter-day accuracy, precision and reproducibility were all less than 4%. Unlike previous HPLC methods, this method does not require pre-derivatisation and can separate glucosamine from both hydrochloride and sulphate salts, and from other common ingredients such as chondroitin sulphate and excipients present in glucosamine products. The second objective was to assess the quality of commercially available glucosamine products (Chapter 3). This objective was completed by applying the previously developed HPLC-CAD method to perform the content and dissolution analysis of Australian (n=25 brands) and Indian (n=21 brands) glucosamine products according to the criteria specified in the United States Pharmacopoeia (USP). Overall, the results showed that the majority of the brands satisfied the content (84% Australian and 75% Indian brands) and dissolution (82% Australian and 81% Indian brands) standards specified by the USP. Also, the extent of deviation shown by the brands that failed to fulfil either one or both USP-specified criteria was not enough to conclude that glucosamine product quality alone would have a substantial impact on the broad variability of clinical outcomes observed in patients with OA. Hence as the second hypothesis proposed, the inter-individual differences in the pharmacokinetics of glucosamine could, in part, be another possible reason behind inconsistent clinical outcomes. This hypothesis was examined by completing the last three objectives of this thesis. To fulfil the third objective of the thesis, a new, simple and sensitive hydrophilic interaction liquid chromatography (HILIC) coupled with a CAD method was developed and validated to analyse underivatised glucosamine in human plasma (Chapter 4). The developed method was found to be linear (r\\(^2\\) = 0.999, 12.5-800 ng/mL), and the %RSD for intra- and inter-day accuracy, precision and repeatability were all less than 6%. The method was sensitive (lower limit of quantitation of 12.5 ng/mL) and can be applied for determining the glucosamine plasma concentrations in pharmacokinetic studies. The fourth objective was fulfilled by determining the exogenous steady-state minimum plasma concentrations (Actual Css min) of glucosamine in OA patients who were taking the supplement and examining the variability in exogenous plasma levels between patients (Chapter 5). A total of ninety-one patients with OA (aged 68.2 ± 7.6 years, 70% females), who had been taking oral glucosamine (750 or 1500 mg/day) for at least one week were recruited. Blood samples of the participants were collected 24 hours after the ingestion of the previous dose of glucosamine and analysed using the previously-developed HILIC-CAD method (Chapter 4) to determine the Actual Css min of glucosamine at steady-state. A substantial variation (106-fold, 45% coefficient of variation) was observed between the Actual Css min levels (3-320 ng/mL) in participants. No significant association of Actual Css min was observed with patient-related (age, gender) or dose-related (daily dose, mg/day and standardised daily dose, mg/Kg/day) variables. The observed high variability could be due to inter-patient differences in the absorption and elimination of glucosamine. Therefore, the results support the hypothesis indicating variability in the pharmacokinetics of glucosamine in patients could be a cause for inconsistent clinical outcomes. Since, this study was performed in a real-world setting, with patients taking different brands of OTC glucosamine products and the majority of the recruited patients were elderly, were suffering from other health conditions and taking prescription medications, there is a possibility that glucosamine plasma levels are affected due to these factors. However, no significant differences in glucosamine plasma levels were observed between a) patients taking commonly-used brands, and b) patients who were not taking, and who were taking, commonly-used prescription medications. Hence, to further support the current finding, it was imperative to examine the inter-brand (between pharmaceutical and non-pharmaceutical grade glucosamine) and inter-individual variability in the pharmacokinetics of glucosamine in healthy individuals. The final objective was fulfilled by recruiting fourteen healthy participants (aged 35.5 ± 8.75 years, 64.2% were males), who ingested 1500 mg/day of each of Blackmores brand (nonpharmaceutical grade) and DONA brand (pharmaceutical grade) glucosamine individually for six days to attain steady-state plasma levels (Chapter 6). The blood samples were collected before and after the ingestion of the last dose for up to 12 hours for each brand and analysed using the previously-developed HILIC-CAD method (Chapter 4) to determine pre- and postdose glucosamine plasma levels. The pharmacokinetic parameters at steady-state including the minimum (C\\(_{ss}\\) \\(_{min}\\)) and maximum (C\\(_{ss}\\) \\(_{max}\\)) plasma concentration of glucosamine, time to reach C\\(_{ss}\\) \\(_{max}\\) post-dosing (T\\(_{ss}\\) \\(_{max}\\)), and area under the plasma concentration versus time curve (AUC\\(_{ss}\\) \\(_{0-12}\\)) were calculated and statistically compared. No significant differences were observed in the pharmacokinetic parameters between the two brands; however, for both brands, the coefficient of variation for C\\(_{ss}\\) \\(_{min}\\), C\\(_{ss}\\) \\(_{max}\\), T\\(_{ss}\\) \\(_{max}\\) and AUC\\(_{ss}\\) \\(_{0-12}\\) exceeded 20%, indicating considerable differences in the parameters between participants. No significant association of the pharmacokinetic parameters was observed with various dosing- and patient related variables. Therefore, the results further substantiate our hypothesis that interindividual differences in the absorption and elimination of glucosamine could be a cause of variable clinical outcomes in OA. Based on the current findings, it is proposed that future studies should investigate the relationship between glucosamine plasma levels and its clinical outcomes in a large cohort of patients with OA.