Perspective on Improving the Relevance, Rigor, and Reproducibility of Botanical Clinical Trials: Lessons Learned From Turmeric Trials

Abstract

Plant-derived compounds, without doubt, can have significant medicinal effects since many notable drugs in use today, such as morphine or taxol, were first isolated from botanical sources. When an isolated and purified phytochemical is developed as a pharmaceutical, the uniformity and appropriate use of the product are well defined. Less clear are the benefits and best use of plant-based dietary supplements or other formulations since these products, unlike traditional drugs, are chemically complex and variable in composition, even if derived from a single plant source. This perspective will summarize key points-including the premise of ethnobotanical and preclinical evidence, pharmacokinetics, metabolism, and safety-inherent and unique to the study of botanical dietary supplements to be considered when planning or evaluating botanical clinical trials. Market forces and regulatory frameworks also affect clinical trial design since in the United States, for example, botanical dietary supplements cannot be marketed for disease treatment and submission of information on safety or efficacy is not required. Specific challenges are thus readily apparent both for consumers comparing available products for purchase, as well as for commercially sponsored vs. independent researchers planning clinical trials to evaluate medicinal effects of botanicals. Turmeric dietary supplements, a top selling botanical in the United States and focus of over 400 clinical trials to date, will be used throughout to illustrate both the promise and pitfalls associated with the clinical evaluation of botanicals.

Keywords: botanical; clinical trial; curcumin; curcuminoids; dietary supplement; turmeric.

Figure 1

In a pre-clinical arthritis model, consistent with curcuminoid blockade of intraarticular NF-κB activation (not shown), curcuminoid (CURC) treatment significantly altered (A) gene expression in arthritic joints, including suppression of over 40 NF-κB regulated genes, and (B) inhibited bone-resorbing osteoclast formation in arthritic joints, which is also NF-κB mediated. ns, not significant or ^***p < 0.001 vs. control (13).

Figure 2

Differential anti-arthritic effects of components extracted from turmeric rhizome in a pre-clinical arthritis model. The anti-arthritic effects of turmeric extracts of different chemical composition were assessed using an ip dosing strategy given reports of altered oral curcuminoid bioavailability when combined with essential oils. Each of turmeric's secondary metabolites [curcuminoids (orange triangles) and essential oils (blue squares)] had significant anti-arthritic effects when administered separately. Interestingly, differential effects were noted for a chemically complex curcuminoid extract (brown triangles) devoid of essential oils but containing polar compounds, such as polysaccharides. Anti-arthritic curcuminoid efficacy was confirmed with oral dosing [50% inhibition; human equivalent dose (HED) of 1 g/d], while protection from oral essential oils was much reduced (20%; HED of 5 g/d) (13).

Figure 3

Metabolic conjugation of curcumin and deconjugation for quantitative analysis. Curcumin is consumed in its free (aglycone) form and undergoes rapid phase II conjugation following intestinal absorption. The glucuronide conjugate accounts for about half of the circulating conjugates while sulfate and other, more complex conjugates account for the rest. Free curcumin is low or undetectable in plasma samples. For quantification plasma samples are often deconjugated using β-glucuronidase but this achieves only incomplete hydrolysis of sulfate and complex conjugates; more complete hydrolysis of all conjugates can be achieved using sulfatase. Direct analysis of conjugates is preferred but hampered by the large number of conjugates, lack of standards, and technical challenges. Reduction of the aliphatic double bonds, a significant route of metabolism in vivo, and other metabolic events are not illustrated.

Figure 4

Pharmacokinetics of oral curcuminoids in mice. (A) Circulating curcumin-glucuronide (CG) levels predominate and are sustained for up to 24 h in mice following a single oral curcumin dose (HED 2.5 g), with sustained, albeit lower, levels also documented in bone, where ß-glucuronidase-dependent GC hydrolysis to form aglycone curcumin occurs, resulting in higher aglycone concentrations than those perfusing bone. (B) Following curcumin ingestion in mature ovariectomized (OVX) mice modeling menopausal bone loss (vs. controls), the capacity of bone to deconjugate curcumin glucuronide distributing to this site persists and is higher in OVX mice. Interestingly, curcumin concentrations in mouse bones are also highest in trabecular bone compartments where menopausal bone loss is most pronounced. Figures are reproduced with permission from John Wiley and Sons (62).

Figure 5

Metabolism of curcumin and the effect on biological activity. Curcumin is reduced and conjugated in vivo as shown by the detection of the corresponding metabolites in plasma samples. Deconjugation of circulating and inactive curcumin-glucuronide by β-glucuronidase may contribute to a tissue- or disease-specific effect. Oxidation of curcumin is prominent in buffer and in cultured cells but awaits to be proven in vivo. Adduction of curcumin to protein has been described, and at least in part depends on oxidation of curcumin to a quinone methide or other electrophilic oxidation product, i.e., spiroepoxide, that target redox-sensitive cysteine residues and soluble thiols like glutathione. The quinone methide radical and spiroepoxide are unstable intermediates in the oxidation of curcumin to the stable end-product, bicyclopentadione. Curcumin also exerts biological effects through mechanisms not involving metabolic transformation.