Assuring Quality of Dietary Supplements for Cancer Patients: An Integrative Formulary Systems Approach

Abstract

Context

The quality of dietary supplements is highly variable and, therefore, may pose unique risks to cancer patients, who increasingly use these products. Although they are highly regulated, the US Food and Drug Administration (FDA) has reported extensive noncompliance with current Good Manufacturing Practices (cGMPs), which further heightens concerns.

Objective

The study intended to investigate the dietary supplements quality practices of current and prospective suppliers of supplements.

Design

Thirteen manufacturers, marketing 19 dietary supplement brands, were selected for inclusion, and 9 participated.

Setting

This study took place at and was supported by the Cancer Treatment Centers of America (Boca Raton, FL, USA).

Outcome Measures

To ensure patients’ safety, the research team established a dietary supplement formulary committee at the Cancer Treatment Centers of America. A proprietary survey tool was used to measure clinically critical quality markers and compliance with FDA regulations. Information was obtained from suppliers via nondisclosure agreements. Manufacturing documents were audited and compared with responses to the survey. The FDA’s audit reports were obtained by request under the Freedom of Information Act. Several site audits were conducted, and third-party analytical testing was performed as needed.

Results

Although all companies claimed full compliance with cGMPs as of the survey’s date, (1) 3 had received warning letters from the FDA for GMP violations, (2) 2 had recalled a product within the preceding 5 y, (3) 4 had reported products that failed independent testing for potency and purity, (4) 1 did not have product specifications, (5) 1 was found by the FDA to have inadequate testing, (6) 1 was found to have a lack of sufficient controls throughout the supply chain to guard against microbial contamination, and (7) 2 had confirmed melamine contamination or lack of melamine testing for protein powders.

Conclusions

These findings confirm the concern of variable dietary supplement quality and describe a rational process others can use to assess products’ quality and ensure patients’ safety. Although the current study focused on practitioners’ branded products used in an oncology setting, the results are relevant to the use of all dietary supplements in both oncological and nononcological settings.

A majority of Americans, including cancer patients, take dietary supplements, often without their physician’s knowledge.^1-6 Although certain dietary supplements can be very useful as treatment adjuvants to reduce side effects and improve quality of life,^7-9 the quality of dietary supplements can vary widely as a result of the US Food and Drug Administration’s (FDA’s) current Good Manufacturing Regulations (cGMPs), which allow each company to define its own specifications for ingredients and finished products.^10-12

Further exacerbating this problem, the FDA reported that 70% of the cGMP audits of dietary supplement companies in 2013 were noncompliant.¹³ A number of cGMP deficiencies has been found by the FDA in 7 fiscal years as reported in FD483s, written notices of cGMP deficiencies from the FDA to companies: 635 in 2010, 1331 in 2011, 1977 in 2012, 2213 in 2013, 1555 in 2014, 1291 in 2015, and 1608 in 2016. The most common cGMP deficiencies reported during that same period were failure to have written specifications for ingredients and finished products and failure to verify ingredients’ identities (Table 1).

Table 1.

The Most Common cGMP Deficiencies Reported by the FDA, by Fiscal Year

21 CFR No.	Top 3 Deficiencies Reported on FD483s	2010	2011	2012	2013	2014	2015	2016
§111.70(e)	Specifications incomplete	7	31	57	67	58	52	69
§111.75(a)(1)(i)	Failure to verify identity	11	27	67	75	56	46	50
§111.103	Lack of written procedures for quality control operations	13	36	41	59	40	39	48

Note: The table indicates the total number of observations related to cGMPs by fiscal year, found during the FDA’s inspections of dietary supplement companies under 21 CFR §111, and includes reporting of adverse events. The data were extracted and compiled from the FDA’s databases; the accuracy of the data included in the current study is the research team’s responsibility. Copyrighted data is used with permission from Marian Boardley Consulting, LLC.

Abbreviations: cGMP, current Good Manufacturing Practices; FDA, US Food and Drug Administration; CFR, Code of Federal Regulations; FD483s, written notice of deficiencies found during inspection.

Economic adulteration, adding materials to increase a product’s weight and/or substituting products designed to trick tests for authenticity, together with chemical and biological contaminants, have been widely reported^14-17 and can have adverse clinical consequences, particularly in cancer patients. Dietary supplements adulterated with undeclared prescription drugs have confounded clinical studies, caused harm, and prompted product recalls.¹⁸

Any investigation into quality must first begin with a definition. The FDA defines quality as

the dietary supplement consistently meets the established specifications for identity, purity, strength, and composition and limits on contaminants and has been manufactured, packaged, labeled, and held under conditions to prevent adulteration under sections 402(a)(1), (a)(2), (a)(3), and (a) (4) of the act (CFR Title 21).”¹⁹

Because each company defines its own specifications, the quality of the dietary supplement depends on the suitability of those specifications.

An initial cGMP audit by the FDA can have 1 of 3 possible outcomes (Table 2): (1) no deficiencies found, (2) deficiencies communicated verbally to the company, or (3) preparation of an FD483. In all cases, an Establishment Inspection Report (EIR) that describes the visit is written. EIRs and FD483s are generally available via requests under the Freedom of Information Act (FOIA). These requests can be made online or via direct mail to the FDA.

Table 2.

The FDA’s Typical Enforcement Progression

FDA’s Action	Outcome	Document	FDA’s Inspection Code
Initial audit	The FDA found no deficiencies.	EIR	NAI
Initial audit	The FDA found deficiencies that it mentioned verbally but did not document.	EIR	NAI
Initial audit	The FDA documented observations; the company must notify the FDA of corrective actions taken.	EIR and FD483	VAI
Repeat audit	If the corrective actions satisfy the FDA, no additional enforcement action is taken.	EIR; closeout letter	NAI
Repeat audit	If the response to the FD483 was unsatisfactory, a warning letter may be issued.	Warning letter	OAI
Rejection of response to warning letter	The results are variable. Increasing enforcement efforts may include seizure, injunction, and civil and criminal prosecution, depending on the situation, using resources of the FDA, FTC, and Department of Justice.	Variable

Note: A warning letter is an official notice of a significant violation that directs a company to take corrective action within a set time frame, often 15 days.

Abbreviations: FDA, Food and Drug Administration; EIR, establishment inspection report; NAI, no action indicated; FD483, written notice of deficiencies found during inspection; VAI, voluntary action indicated; OAI, official action indicated; FTC, Federal Trade Commission.

When an FD483 is issued, the supplement company has an obligation to take voluntary action to correct any deficiency and to keep the FDA informed of its plans and progress. In the event an FD483 is issued after an initial audit, the FDA will review the written corrective-action plan and documents submitted by the company. If the corrective-action plan reported to the FDA is insufficient, the FDA may issue a warning letter, which is an official notice of a significant violation that directs the company to take corrective action within a specified time frame, typically 15 days. Issued warning letters are found on the FDA’s Web site; the listing is refreshed weekly.²⁰

If a company’s response to a warning letter is insufficient, or if the violation is considered a significant risk to public health, the FDA has many tools available to protect the public, including a product’s seizure, an injunction, and civil and criminal prosecutions using resources of the FDA, the Federal Trade Commission, and the Department of Justice.

It is important to note that a warning letter does not always mean that a clinically significant quality problem exists. For example, in many cases, warning letters are issued after the FDA finds illegal health claims on a company’s Web site. Phrases such as “reduces LDL cholesterol” are implied health claims, rendering the product an unapproved new drug, which is not a clinical matter. The most clinically significant violations include those found in Table 1. Without proper specifications, testing, and compliant quality-control operations, it is impossible to have any confidence in product quality. Warning letters that name those deficiencies are a serious reason for concern.

The current study intended to investigate the DS quality practices of current and prospective suppliers of supplements to the Cancer Treatment Centers of America^® (CTCA).

Methods

This study took place at and was supported by the CTCA^® (Boca Raton, FL, USA).

Procedures

Definition of Quality

Recognizing the structural limitations of CFR Title 21 Part 111, which allows each company to define its own quality specifications, the current research team defined the quality of a dietary supplement with respect to 3 characteristics—authenticity, potency, and purity—as follows: (1) ingredients must always be proven to be authentic, using scientifically valid methods; (2) all label claims for potency must be supported by scientifically valid data from each product/lot release into commerce through the product’s expiration date; and (3) purity refers to a consistently proven, maximum freedom from chemical and biological contaminants, such as pesticides, solvent residues, aflatoxins, and other reasonably anticipated contaminants (RACs).

An RAC is a harmful substance associated with dietary ingredients for economic reasons (eg, dilution of protein powders with melamine to fool analytical tests or contamination found in the supply based on origin, processing, and storage, or the like). It is only through strict control over these 3 variables that a predictable clinical outcome can reasonably be expected. A careful review of specifications and testing is fundamental to evidence-based quality and a novel aspect of the current investigation.

Data review and inspections were performed by a qualified independent technical expert who was certified in cGMP training for dietary supplements and qualified by >30 years of experience in managing manufacturing, packaging, and distribution operations for pharmaceuticals and dietary supplements.

Creation of and Performance of Survey

A multidisciplinary dietary supplement formula committee (DSFC)—consisting of pharmacists, naturopathic doctors, dietitians, and a subject matter expert—created a survey tool consisting of 75 questions and data requests. The proprietary survey outlined a number of the critical quality practices that the research team wanted to evaluate (Table 3).

Table 3.

Sections of the Survey of CQPs of Supplement Manufacturers

Section	Section Name	Purpose	Example of Requested Documents
I	Company information	Provide contact information, scope of activities performed at address, number of employees, brands, etc.	Certificate of product-liability insurance
II	Products sold	Indicate categories of products sold (eg, botanicals, OTC drugs, homeopathic remedies, or Rx drugs).	A product catalog
III	Dosage forms sold	Indicate dosage forms and their packaging (eg, unit-dose oral solids, liquids, or topicals).	A product catalog; verification of Web site
IV	Company registrations	Indicate if the company is registered with the FDA, TGA, and USDA.	Proof of all current registrations
V	Company and product certifications	Indicate certifications for company and products; indicate if any product failed testing by consumerlab.com.	Copies of all certifications; for certified cGMP compliance programs, copies of most recent audit report(s)
VI	FDA audit history	Indicate if company has been audited by the FDA.	FD483s (if any) and your company’s responses to them
VII	Product recalls	Indicate if a recall history exists; confirm that the company can track a lot of 1 specific ingredient into all products or lots for which that ingredient lot was used.	Information on all recalls within the 5 y prior to completion of this survey
VIII	Quality practices-cGMPs	Identify scope of quality practices, including quality unit composition, specifications, etc.	Current table of contents for your SOPs
IX	Use of subcontractors	Identify areas where subcontractors are used.	SOPs for qualifying and validating subcontractors
X	Dietary-ingredient testing programs	Discuss your understanding of the complete scope of your dietary-ingredient testing practices.	All relevant SOPs
XI	Finished-goods testing programs	Discuss your understanding of complete scope of your finished-goods testing programs, including stability programs.	All relevant SOPs
XII	Packaging and labeling operations	Discuss your understanding of all critical aspects of packaging and labeling, including storage conditions, master formula control, use of UPC numbers, etc.	SOP for metal detection in packaging operations
XIII	Request for product documentation	Provide documentation for up to 3 representative products.	Finished product specifications, analytical results for the current lot in commerce, and specifications and analytical results for each lot of dietary ingredient used to manufacture the finished product or lot under review

Abbreviations: CQPs, critical quality practices; OTC, over the counter; Rx, prescription; FDA, US Food and Drug Administration; TGA, Therapeutic Goods Administration (Australia); USDA, US Department of Agriculture; cGMP, current Good Manufacturing Practices; FD483s, written notice of deficiencies found during inspection; SOPs, standard operating procedures; UPC, universal product code.

The surveyed vendors manufacture finished dosage forms for packaging and, in some cases, purchase finished dosage forms (eg, soft gels, from other manufacturers for packaging and labeling). In other cases, the supplier subcontracts manufacturing processes, such as blending, tableting, and coating, to third-party subcontractors, which introduces additional supply-chain variables that might impact safety.

Thirteen current and prospective vendors, representing 19 brands of dietary supplements, were surveyed and reviewed. Vendors selected for inclusion were either current suppliers or prospective suppliers recommended by Naturopathic Medicine. They were prioritized by DSFC review. The invitation to participate, along with each survey and a nondisclosure agreement (NDA) were distributed by our technical expert. Executed NDAs were managed by Cancer Treatment Centers of America’s vice president of integrative medicine.

Laboratory Analysis

Independent third-party analytical and microbiology laboratories were used on an as-needed basis to test for potency and freedom from chemical or biological contaminants. Analytical laboratories used were Flora Research Laboratories (Grant Pass, OR, USA) and Tampa Bay Analytical Research (Clearwater, FL, USA). KML Labs, Inc (Bonners Ferry, ID, USA) was selected for microbiological testing.

Requests for Information

Between 2010 and 2013, for 3 of each manufacturing company’s representative supplements, the research team requested copies of the specifications and analytical results for all dietary ingredients and each finished product; these data were audited against survey responses to test consistency. Information received from participating vendors was obtained under nondisclosure agreements.

Since an enforcement action by the FDA is a marker of noncompliance that may have clinical significance, the DSFC integrated the FDA’s oversight and enforcement actions into its process. The research team used the FOIA to obtain all EIRs and FD483s issued to each company during the 5 years before the current study. The research team compared company-provided reports against those obtained directly from the FDA. Recommendations of the independent technical expert were provided to the DSFC for final determination.

Onsite Quality Audit

Of the 9 vendors who participated, 4 were subjected to an onsite quality audit because (1) the vendor’s products at the time of the survey represented >50% of the Cancer Treatment Centers of America’s purchases of supplements, (2) the prospective vendor had moved into new facilities and had received an FDA warning letter, (3) the prospective vendor was a small company and used subcontractors, and (4) the survey’s results for a current vendor showed major discrepancies between the responses and the manufacturing records.

Limited Investigations

An additional 4 vendors were selected for an investigation limited in scope to a single product because (1) the product was a new supplement, the use of which had been requested by a clinician at one of the Cancer Treatment Centers of America’s hospitals; (2) the product was an existing supplement, the use of which had been requested by clinician at one of the Cancer Treatment Centers of America’s hospitals; (3) reports had been published indicating the supplement’s subpotency and contamination; and (4) the supplement was an existing protein product with potential safety risks related to melamine.

Use of Results

At the end of the study, the surveyed vendors were classified as (1) preferred, a first-choice source; (2) approved for products not available from the preferred vendor; or (3) not approved (ie, rejected). Vendors selected for the survey were practitioners’ brands prioritized by their reputation and clinicians’ demands for their products.

Outcome Measures

Results were compared against primary quality benchmarks developed internally by the DSFC: (1) specifications must exist for all dietary ingredients, components, and finished products; (2) all lots of dietary ingredients must be tested for identity and, where known, for economic adulterants, using scientific methods fit for the purpose; (3) all lots of dietary ingredients for which a potency claim is made on the label must be tested for potency, using scientific methods fit for purpose; (4) all lots of herbal dietary ingredients must be tested for solvent residues; for herbicide, fungicide, and pesticide residues if not certified as organic; for aflatoxins; and for microbiological contaminants, including yeasts and molds; (5) nonherbal ingredients subjected to solvent processing must be tested for solvent residues; (6) all lots of ingredients at risk for contamination with heavy metals and other RACs must be tested for them; (7) oils must be tested for rancidity markers; (8) fish oils must be tested to be in compliance with standards of the Global Organization for EPA and DHA; (9) protein powders must be tested for presence of melamine; (10) all finished dietary supplements must be tested for potency and microbiological contamination; (11) all dietary supplements having expiration dates must have stability data to support them; (12) suppliers must share the results of internal quality audits; (13) metal detection must be used in final packaging; (14) suppliers must demonstrate compliance with cGMPs and internal standard operating procedures (SOPs); and (15) product audit results must be consistent with quality practices and SOPs reported by the company.

Results

Nine of the 13 possible vendors (69%) participated in the survey (Table 4). Of the 9 participating vendors, 5 (56%) were approved for use: 1 was selected as the preferred vendor, and 4 were approved for use for supplements not available from the preferred vendor.

Table 4.

Results of Vendors’ Survey

Type of Vendor	No. of Companies	No. of Brands	No. of Field Audits	No. of FOIA Reports
Selected as preferred vendor	1	2	1	1
Approved as vendors	4	9	2	4
Not approved as vendors	4	4	1	3
Totals	9	15	4	8

Abbreviation: FOIA, Freedom of Information Act.

Although all the companies claimed full compliance with cGMPs as of the survey’s date, (1) 3 had received warning letters from the FDA for GMP violations, (2) 2 had recalled a product within the preceding 5 years, (3) 4 had reported products that failed independent testing for potency and purity, (4) 1 did not have product specifications, (5) 1 was found by the FDA to have inadequate testing, (6) 1 was found to have a lack of sufficient controls throughout the supply chain to guard against microbial contamination, and (7) 2 had confirmed melamine contamination or lack of melamine testing for protein powders. The companies’ claims did not always stand up to an audit. Only one company met the research team’s internal standards for authenticity, potency, and purity of dietary ingredients and finished products.

Of the supplements from the 4 vendors that were selected for an investigation that was limited in scope, (1) the first, a new product, was rejected because third-party testing proved illegal contamination; (2) the second, an existing product, was discontinued because it was deemed a safety risk for patients; (3) the third had negative reports of quality, and the vendor was barred because its products were deemed a safety risk for patients; and (4) the vendor of the fourth was barred because it conducted no melamine testing and its lack of cGMP compliance was deemed a safety risk.

Discussion

Unlike previously published efforts to survey quality practices for dietary supplements,²¹ the current survey included a combination of a survey, a documentation review, freedom-of-information requests, research into the international biomedical literature, and audits of facilities. This combination provided the team with deep insights into the critical quality and compliance practices of the Cancer Treatment Centers of America’s vendors, empowering its clinical teams to make more fully informed decisions.

Although FDA regulations require that every company define specifications for its dietary ingredients, the current research team confirmed that some companies fail to do this required work (Table 1). Lack of adequate specifications, together with failure to perform identity testing of ingredients, are the 2 most commonly found cGMP failures reported by the FDA.¹³ One company that the current study subjected to a limited review did not have specifications nor did they have SOPs required by the FDA’s regulations; as a result, they were barred from the Cancer Treatment Centers of America’s hospitals.

The concept of RACs is clinically relevant. This concept requires the manufacturer to consider what sorts of contaminants can adulterate a product and set limits for those contaminants considered potentially dangerous, and it requires tests for compliance with all specifications.²²

In a review of a medicinal mushroom product in the current study, for example, research into its manufacturing process, suggested that both cyclosporine and melamine could be considered to be RACs. Thorough analysis of the supply chain revealed that this product was produced in a plant that manufactured cyclosporine and also had a history of FDA compliance issues. This Chinese product was cultured in a growth medium to which milk powder was added. The current research team had the finished product sent to qualified independent laboratories for analysis.

Although the product was negative for cyclosporine, it was found to contain illegal and potentially dangerous levels of melamine. The analytical laboratory we selected, using a FDA analytical method to test for melamine in dry-powder samples, reported that an average of 3 replicates (samples) contained 76 parts per million (PPM) of melamine, roughly 30 times higher than the FDA’s limit of 2.5 PPM. The manufacturer also had this product tested at a different, major analytical laboratory; their finding on a single sample of <2 PPM was due to the fact that the laboratory used an FDA analytical method developed for use in liquids, not powders. This finding shows that the analytical methods used are critically important.

Upon the current research team’s notification, the company withdrew its product from the US market, and the current research team’s work protected Cancer Treatment Centers of America’s patients from receiving dangerously contaminated product. The team believes that the root cause was inadequate specifications for the milk powder, which was likely contaminated with melamine. The mushroom also appeared to have concentrated melamine from its environment, as it often does with other substances found in its growing environment.

Similarly, botanical extracts, by their very definition, use solvents during processing. The issue of residual toxic solvents has received increasing attention from a collaborative effort involving the University of Mississippi’s National Center from Natural Products Research, the American Botanical Council, and the American Herbal Pharmacopeia.²³ Excessive solvent residues can be hazardous, putting all patients at risk. Endocrine-disrupting chemicals can be found in dietary ingredients, and research suggests that they may have harmful acute, chronic, and/or transgenerational effects, even in low doses.²⁴

Curcumin extract, for example, for which the US Pharmacopeia’s allowable limit in drugs is 5 PPM, was reported in a 2010 publication to contain dangerous amounts of the Class 1 solvent 1,2-dichloroethane at levels exceeding 3270 PPM²⁵; this amount translates into a clearly toxic dose of 98 milligrams per 3 grams.

Desk audits and on-site quality audits revealed material discrepancies between the quality practices claimed on the current study’s survey and the supplement companies’ actual practices. One company, for example, claimed that it performed finished-goods potency analysis on all products, possessed complete stability studies, and performed residual toxic-chemical testing on all lots of incoming botanicals. None of those claims were true. It was only through a careful audit against the Cancer Treatment Centers of America’s primary quality benchmarks that the current research successfully uncovered these clinically relevant misleading or untruthful claims.

Limitations of the Study

Limitations included a small sample size, a small scope largely limited to health care practitioner branded products, the inability to share specific results due to nondisclosure agreements, limited resources, and an incomplete longitudinal follow-up.

Summary

Providing safe and helpful dietary supplements to cancer patients requires careful formulary oversight. An integrative team having technical expertise can support formulary efforts and provide recommendations for safe dietary supplements for clinicians and their patients. Because it is impossible to compare quality practices objectively based on label or advertising claims, the Cancer Treatment Centers of America’s 15 primary quality benchmarks can be used as critical proxies for quality. Suppliers who do not have or refuse to share specifications and/or analytical results are completely contrary to the principles of evidence-based quality outlined herein and should be avoided.

Conclusions

The current study’s findings confirmed that quality practices for dietary supplement vary widely due to regulatory limitations and willful noncompliance with cGMPs. Using a novel, evidence-based quality approach uncovered quality issues that would not have otherwise been found and which could have put patients at risk had they not been uncovered. This unique integrative approach, which included a sharp focus on specifications and analytical results, succeeded in raising awareness of a growing problem and empowered clinicians to make informed decisions. This approach prevented cancer patients from receiving illegally contaminated dietary supplements. The current study also alerted the Cancer Treatment Centers of America’s vendors to safety and compliance issues and empowered them to improve their quality practices. Although the current study focused on practitioners’ branded products used in an oncological setting, the results are relevant to the use of all dietary supplements in oncological and nononcological settings.

Biographies

Carolyn A. Lammersfeld, MS, RD, is vice president of integrative medicine

Michael D. Levin, BA, is the founder of Health Business Strategies, LLC, in Clackamas, Oregon

Paul Reilly, ND, is a naturopathic physician at the Cancer Treatment Centers of America, at the time of study, and a naturopathic physician at Salish Integrative Oncology Care Center in Fife, Washington.

Joseph W. Coyne, RPh, was vice president of pharmacy at the time the study was conducted at the Cancer Treatment Centers of America and a consultant at Clinical IQ, LLC, in Mundelein, Illinois.

Timothy C. Birdsall, ND, is senior vice president of information services and chief medical information officer

Maurie Markman, MD, is president of medicine and science a the Cancer Treatment Centers of America in Boca Raton, Florida.