The draft Personal Care Products Safety Act proposes that FDA review cosmetic ingredients against a standard that has proven challenging for the dietary supplement industry and could set unrealistically high consumer expectations, experts suggest.

The bill would require FDA to evaluate the safety of five cosmetic ingredients or non-functional constituents annually, considering whether evidence is adequate "to support a reasonable certainty among competent scientists that the ingredient is not harmful under the recommended or suggested conditions of use, or that a non-functional constituent is not harmful under the recommended or suggested tolerance levels."

The de facto application of a reasonable-certainty-of-no-harm standard in FDA's review of New Dietary Ingredient Notifications in the supplement sector has led to high failure rates – around 75%, according to George Burdock, president of regulatory and toxicology consulting firm the Burdock Group.

Reportedly, the rejection rate in 2014 reached 88%.

The Dietary Supplement Health and Education Act of 1994 (DSHEA) established that dietary ingredients in supplements must "reasonably be expected to be safe," which industry expected to pose a lower evidence burden than the reasonable-certainty-of-no-harm standard FDA uses in assessing Generally Recognized As Safe notifications for food ingredients.

However, FDA has suggested that the two standards represent a "distinction without a difference," and many stakeholders agree that NDIN's are being weighed against the latter, contributing to the relatively low percentage of notifications that receive "no objection" determinations from the agency.

Burdock noted in an April 6 email that FDA expects GRAS filings to contain "the same quantity and quality of data" mandated for food-additive preapproval, which otherwise is required under federal food and drug law. Companies can convene their own experts to self-affirm that a substance is GRAS and notify FDA of the finding. In some cases, FDA has found that such notices do not provide a basis for GRAS determination.

In essence, Burdock said, "FDA has set itself up as the final arbiter of the requirements of a GRAS and presumably the judge as to whether the candidate ingredient has met the safety standard. The lesson here is, 'Don't ask the question if you are not ready to hear the answer.'"

"There are a lot of things in this bill that I just shake my head at. I'm not sure that this level of complexity or oversight is really necessary." – EAS Consulting Group's John Bailey

The dietary-supplement and food industries' experience raises questions as to how FDA would apply a reasonable-certainty-of-no-harm metric in cosmetic ingredient review were the PCPSA or related legislation to pass, and the level of scientific evidence industry would be expected to provide in support of an ingredient's safety.

In an April 13 interview, John Bailey, senior advisor for colors and cosmetics at EAS Consulting Group and former director of FDA's cosmetics program, expressed uncertainty regarding the need for and feasibility of FDA ingredient review against the standard.

He suggested that the standard's implementation could deemphasize benefits in the risk/benefit calculations typically employed in cosmetic ingredient safety assessment and regulate ingredient use more in line with the precautionary principle supported by many public advocacy groups – a view that the Environmental Working Group seems to be taking on the bill as well Also see "Feinstein, EWG Gunning For Spring Senate Hearing On Cosmetics Bill" - HBW Insight, 1 March, 2016..

Bailey noted that the proposed standard differs from conventional safety-assessment approaches for cosmetics in that it would require industry to provide data proving a negative – no harm.

In the area of foods, he said, meeting the standard does not mean ensuring absolute certainty of safety. Rather, it signals that reasonable proof exists that a substance will not be harmful.

"Well, what's reasonable?" he asked. "And who decides what's reasonable, and what happens if somebody thinks it's unreasonable, if somebody disagrees with what you've decided is reasonable and is in a position to hold you accountable?"

In contrast to the reasonable-certainty-of-no-harm approach, subjectivity is minimized in conventional risk analysis, Bailey said.

From his perspective, the proposed standard's application to FDA ingredient review could only work if it were accompanied by guidance development enabling experts to weigh in on what "reasonable" means in the context of ingredient safety substantiation.

"It really comes down to how it's applied," he said. "In the absence of a clear, concise interpretation it could be something of a moving target and one that I think is just going to slow things down and make it more difficult to actually accomplish anything. I just don't see it as something that, number one, is necessary, and number two, is really going to contribute to overall consumer safety."

Burdock believes a standard of "reasonable certainty of no risk" could potentially achieve a more appropriate balance, "but the word 'certainty' might be misleading and tend to give the alarmist culture a foothold."

He also noted that while the PCPSA only explicitly calls for the reasonable-certainty-of-no-risk standard to be used by FDA in its ingredient evaluations, the bill conceivably could be read as extending the standard to manufacturers' own safety-assurance work.

Among its provisions, the proposed legislation would mandate "written determinations" from companies that the cosmetics they market are safe under the conditions of use recommended in product labeling.

"Companies have to submit basically an affidavit that the safety has been substantiated," Bailey explained. In doing so, "some idea of reasonable certainty of no harm may be invoked, and in the absence of what that means, there's probably some risk in how that's interpreted from companies' perspective, FDA's perspective and probably from the consumer groups' perspectives as well."

Burdock also expressed concerns about how the proposed reasonable-certainty-of-no-harm standard would be understood by consumers, suggesting that it "would falsely raise the expectations of the public – i.e., the public believing absolutely no harm," a categorical impossibility.

A revised standard of "reasonable certainty of no risk" could potentially achieve the appropriate balance, he suggested, "but the word 'certainty' might be misleading and tend to give the alarmist culture a foothold."

Though unwieldy, the optimal phrasing might be "reasonable expectation of acceptable risk to the average consumer," he said. "I believe manufacturers would feel they could meet the standard, and consumers would understand the message."

Why Reinvent The Wheel?

The PCPSA designates the first five cosmetic ingredients that would be subject to FDA review if the legislation were to be enacted:

Diazolidinyl urea;
Lead acetate;
Methylene glycol/methanediol/formaldehyde;
Propyl paraben;
Quaternium-15.

The bar for a safe finding, under the reasonable-certainty-of-no-harm model, ostensibly would be set higher than that currently observed by the Cosmetic Ingredient Review Expert Panel in its evaluations.

CIR procedures define "safe" to mean "no evidence in the available information that demonstrates or suggests reasonable grounds to suspect a hazard to the public under the conditions of use that are now current or that might reasonably be expected in the future – e.g., a low incidence of minor adverse reactions."

Absence of the term "certainty" in that guideline and its deference to reasonable expectations aligns the CIR experts' approach with the risk-based model advocated by cosmetics manufacturers and perhaps more closely with the framework that supplement stakeholders thought they were getting under the DSHEA.

The proposed PCPSA does specify that FDA "may not consider an ingredient or non-functional constituent harmful solely because it can cause minor adverse health reactions, such as minor transient allergic reactions or minor transient skin irritations, in some users."

However, the legislation also would require FDA to consider the probable cumulative and aggregate effect of exposure to ingredients or non-functional constituents, or to any chemically or pharmacologically related substances, from use of cosmetics or other products with similar routes of exposure.

"In appropriate cases, the [FDA] may consider available information on the total exposure to an ingredient or non-functional constituent from all sources," the bill states.

The CIR Expert Panel normally does not take cumulative and/or aggregate exposure into account in its safety assessments.

Compiling data of that nature is "extremely difficult," Bailey pointed out. "When you're talking about a substance that may be used in foods, drugs, household products, agriculture and every possible thing that presents a possible exposure, that's virtually impossible to measure."

He added that conservative steps in conventional risk assessment already take such unknowns into account and factor them into final safety determinations. While it's a concern frequently raised by NGOs, "I think that to some degree this whole notion of cumulative exposure is a bit of a red herring," Bailey said.

"How can you put an (alleged) endocrine disruptor in cosmetics and still identify the products as having met the 'reasonable certainty of no harm' standard'?" – Burdock

Generally, the consultant, who left the Personal Care Products Council in 2011 after serving nine years as the trade group's chief scientist, sees existing institutions in the US as largely sufficient for ensuring cosmetic ingredient safety.

He noted that FDA already has the power to take action against cosmetic ingredients it deems harmful, and characterized the CIR program as one in which he has "a great deal of confidence."

He continued: "CIR is the only systematic ingredient review program in the world. Everyone else does reviews for cause – there's a reason to do it, and they will address it accordingly."

CIR, on the other hand, does not wait necessarily for safety questions to arise, but evaluates ingredients on a quarterly basis, taking into account frequency of use, new information that becomes available and other factors. The data it generates then can be highly useful if an issue does emerge, Bailey noted.

While the program was founded and is funded by industry, its independent experts are recognized globally as leaders in their fields, he said. The panel "really finds the problems if they exist and highlights them and makes the industry and authorities aware of those risks. So I'm not sure it's really necessary to reinvent the wheel here."

He acknowledged that CIR doesn't have the authority to require companies to comply with its recommendations. "Maybe that's where the emphasis should be," he suggested.

Bailey pointed out that developing the infrastructure for FDA ingredient review as proposed by the PCPSA will be no small undertaking, even with the $20.6 million in funds the bill would provide from industry registration fees.

"These things don't just turn off and on; they have to be built," he said.

He noted that reviewing five ingredients each year could be a tall order for FDA. "It might sound like very little, but when you bring to bear the procedures that FDA has to go through, it really is going to be a slow process."

Bailey has advocated for a formal bridge between FDA and CIR to capitalize on the latter's review activities and bring added authority to its safety conclusions.

"CIR is much more agile and able to adapt to changes or new information," he said.

Congressional appropriators likewise have proposed a system wherein FDA would formally recognize CIR's ingredient safety assessments. However, in its fiscal 2017 budget justification, FDA said such a framework would be "inappropriate" and could raise "significant appearance concerns and resource issues" Also see "FDA Balks At CIR Public-Private Partnership In FY 2017 Budget Request" - HBW Insight, 15 February, 2016..

Preservatives Would Put Standard To The Test

Another question that warrants consideration is how FDA would weigh the importance of effective product preservation against the potential hazards of proven preservatives in the context of a reasonable-certainty-of-no-risk assessment, with three preservatives – diazolidinyl urea, quaternium-15 and propyl paraben – making the list of ingredients positioned for first-year FDA review under the PCPSA.

Currently, those preservatives are recognized by FDA as commonly used options in the cosmetics sector, as opposed to non-traditional preservatives the agency has prioritized for attention in its monitoring and enforcement work due to safety and efficacy concerns Also see "FDA Has Eye On 'Non-Traditional Preservative' Use, Updates Internal Guide" - HBW Insight, 1 March, 2016..

At the same time, the three have been demonized by NGOs due to alleged research linking formaldehyde-releasing preservatives to cancer risks and parabens to hormonal effects.

"These three preservatives have been on the activist groups' hit lists for some time. But let's face it, these are probably, in a cosmetic preparation, the most important ingredient," Bailey said.

"The skill with which they're used protects the consumer to an extremely high degree. It's very rare to have microbial contamination injure the consumer today, and it's because the industry has a tool chest of preservatives that will handle all of the widely varying conditions of use" for cosmetics on the market.

Tightening regulations and increasing consumer pressure already have narrowed industry's preservative options, and from Bailey's point of view, it would be "foolish" to deplete that palette further. But the proposed reasonable-certainty-of-no-harm standard could have that effect.

Bailey, who has advocated for a formal bridge between CIR and FDA, said the proposal that FDA review five ingredients each year "might sound like very little, but when you bring to bear the procedures that FDA has to go through, it really is going to be a slow process."

The issue of endocrine disruption in general, beyond its relevance to propyl paraben, could pose challenges in assessing cosmetic ingredient safety against the proposed standard.

Efforts to reform the US Toxic Substances Control Act in recent years have included proposals for holding new chemicals to such a standard. Experts tracking the legislative developments have suggested that it would be virtually impossible for chemicals exhibiting endocrine-disrupting potential to meet the criterion.

Differing TSCA-reform bills that passed in 2015 in the Senate and House – the chambers currently are working to reconcile the two versions – abandoned the reasonable-certainty-of-no-risk standard in favor of criteria designed to prohibit "unreasonable risk" to consumer health.

Given mounting concerns about endocrine disruption, Burdock put the operative question thus: "How can you put an (alleged) endocrine disruptor in cosmetics and still identify the products as having met the 'reasonable certainty of no harm' standard'?"

Lastly, Burdock observed that the level of data that could be required for meeting the proposed safety threshold across toxicity endpoints may conflict with the PCPSA's objectives to reduce animal testing.

Bailey maintains that the PCPSA generally may be overkill in terms of what is needed in a cosmetic regulatory update.

"There are a lot of things in this bill that I just shake my head at. I'm not sure that this level of complexity or oversight is really necessary," he said.

PCPSA's 'No Harm' Standard Would Set High Bar For FDA Ingredient Review

More from Legislation

More from Policy & Regulation