How similar is similar enough? A sufficient similarity case study with Ginkgo biloba extract

Highlights

• Botanical dietary supplements with similar labels can vary widely in content.

• Approaches to determine sufficient similarity of complex mixtures are presented.

• Chemical and biological activity data are integrated for similarity evaluation.

• Ginkgo biloba extract is used as a case study to illustrate methods.

• While a majority of Ginkgo samples were similar, some were notably different.

Abstract

Botanical dietary supplements are complex mixtures that can be highly variable in composition and quality, making safety evaluation difficult. A key challenge is determining how diverse products in the marketplace relate to chemically and toxicologically characterized reference samples (i.e., how similar must a product be in order to be well-represented by the tested reference sample?). Ginkgo biloba extract (GBE) was used as a case study to develop and evaluate approaches for determining sufficient similarity. Multiple GBE extracts were evaluated for chemical and biological-response similarity. Chemical similarity was assessed using untargeted and targeted chemistry approaches. Biological similarity was evaluated using in vitro liver models and short-term rodent studies. Statistical and data visualization methods were then used to make decisions about the similarity of products to the reference sample. A majority of the 26 GBE samples tested (62%) were consistently determined to be sufficiently similar to the reference sample, while 27% were different from the reference GBE, and 12% were either similar or different depending on the method used. This case study demonstrated that approaches to evaluate sufficient similarity allow for critical evaluation of complex mixtures so that safety data from the tested reference can be applied to untested materials.

Introduction

Botanical dietary supplements are used by approximately 18% of adults (Clarke et al., 2015), or over 40 million people, in the United States. Therefore, the safety of these products is an important public health concern. For each type of botanical dietary supplement, there are numerous products available in the marketplace. These products are complex mixtures that can vary significantly depending on the ingredients (e.g., plant part, growing conditions), formulation (e.g., presence of different excipients like stabilizing agents or fillers), processing procedures, and storage conditions. Adulteration, either economically-motivated (i.e., addition of less expensive plant material) or with pharmaceutical compounds, adds to the variability in botanical dietary supplements. A companion paper (Shipkowski et al., 2018), provides a detailed discussion of the many challenges associated with assessing the safety of botanical dietary supplements.

Studies to evaluate the toxicity (or efficacy) of a botanical dietary supplement are typically limited to a single sample (e.g., unfinished extract from a certain supplier or finished product from a specific manufacturer) based on the inherent assumption that the selected sample is representative of all products with the same name on the label. However, there has been little effort to systematically evaluate this assumption. Two major challenges in determining whether toxicological data from a tested sample can be used to evaluate the safety of an untested sample are 1) the development of methods for evaluating sufficient similarity (i.e., what are the best methods to compare complex mixtures?), and 2) the establishment of categorical or quantitative similarity criteria (i.e., what level of similarity is sufficient?).

The term sufficient similarity, also referred to as phytoequivalence in the parlance of botanical dietary supplements, applies to complex mixtures and indicates that two mixtures are similar enough that the toxicity or efficacy data from one of the mixtures (reference mixture) can be used as a surrogate for the mixture(s)-of-interest. A reference mixture can be selected based on a pre-determined quality standard or a high level of characterization (i.e., the known to which the unknown is being compared). To our knowledge, there has not been a concerted effort to evaluate sufficient similarity of botanicals using both chemical and biological-response measures. However, several organizations have made significant progress in establishing the authenticity of botanicals and identifying potential adulterants based on chemical analysis of samples, including the USDA (particularly the Food Composition and Methods Development Laboratory) (Harnly et al., 2012, 2013, 2016), the Association of Official Analytical Chemists (AOAC) International Guidelines for Validation of Botanical Identification Methods (2012), The United States Pharmacopeial Convention (2015), and the American Botanical Council (Gafner et al., 2015). These efforts have been instrumental in providing manufacturers with the tools needed to evaluate the quality of their products.

The approaches developed to evaluate authenticity and detect adulteration of botanical dietary supplements have relied on both untargeted and targeted chemistry. Untargeted chemistry is defined here as any method that is used to detect the presence and relative abundance of constituents without prior knowledge of their identity. Untargeted chemistry approaches are intended to characterize as much of the sample composition as possible. These methods make no attempt to identify or quantify individual constituents within the mixture. However, they are not truly “untargeted” in that the methods used for sample preparation and the specific detection device used will influence the range of chemical structures included in the chromatographic profile. For example, the extraction process used to prepare samples for chemical analysis by chromatography will target components of a polarity range determined by the solvent. A true untargeted approach would employ multiple solvents with a range of polarities. In contrast to untargeted chemistry, targeted chemistry uses methods that have been qualified for each constituent-of-interest and requires authentic standards to quantify select marker constituents. Targeted chemistry can serve as a complement to untargeted chemistry by confirming the identity of specific analytes noted in chromatographic profiles generated using untargeted chemistry. Targeted chemistry is often used to quantify the purported active constituent(s) or marker constituents (if the active constituent is unknown) of a botanical product to confirm that the appropriate concentration is present in a standardized sample. For example, organizations such as the U.S. Pharmacopeia (http://www.usp.org/verification-services) and NSF International (http://www.nsf.org/services/by-industry/dietary-supplements) use targeted chemistry techniques to verify that the contents of a dietary supplement match label claims in their certification programs.

While there is not a history of sufficient similarity work involving botanical dietary supplements, there have been a limited number of examples evaluating similarity of complex environmental chemical mixtures such as water disinfection byproducts, petroleum substances, and pesticide mixtures (Rice et al., 2009; Marshall et al., 2013; Murray et al., 2013). These approaches have typically relied on quantification and comparison of select constituents in the mixture (Marshall et al., 2013) or structurally-defined classes of constituents (Murray et al., 2013), inferring that the toxicologically-active constituents are captured by the chemical analysis. However, it is important to note that there is usually a very large unidentified fraction in botanical products, and the active constituents are often unknown. Therefore, targeted analysis of select chemical constituents could be inadequate for determining similarity. In addition to examples that rely on chemical analysis alone, a few studies with environmental mixtures have also included in vitro biological data in the evaluation of similarity (Schenck et al., 2009; Grimm et al., 2016). In this manuscript, we explore sufficient similarity in terms of both chemical and biological-response similarity as applied to a complex botanical mixture: Ginkgo biloba extract (GBE).