On 12/10/23, the FSA announced a 10mg Advised Daily Intake level (ADI) for CBD products.
That figure was derived from three novel foods applications, two focused on CBD isolate, and one on synthetic CBD.
While the FSA identified the sources in all but name, they also stated that the ADI would apply to all CBD extracts and related products, including those that are wholeplant in nature.
That move caused uproar in the industry, which was amplified recently when it became apparent that the FSA are using that 10mg ADI as a barrier to progression in novel foods.
That in itself raises serious concerns of the process, see, the FSA insisted that applicants submitted their own toxicology data, but now it seems that they are waiting at the post with alternative data they would 'ask' companies to 'adopt' before crossing the line towards product authorisation.
This article aims to highlight the absurdity of that approach and asks whether it is truly appropriate to apply the 10mg ADI across all types of CBD extracts.
Below is a list of studies attached to novel foods applications in the EU and UK that are publicly available.
All studies are indexed at the end of this article
Table:
NOAEL (No Observable Adverse Effect Level): The highest dose of a substance at which no adverse effects are observed in the subjects during the study. This dose represents the threshold below which the substance is considered safe.
ADI (Acceptable Daily Intake): The maximum amount of a substance (derived from the NOAEL and applying uncertainty/safety factors) that can be ingested daily over a lifetime without an appreciable health risk to the general population, including sensitive subgroups.
UL (Upper Limit): The highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population.
IDD (Intended Daily Dose): The recommended dose for daily consumption based on the product's intended use. This dose may be set by the manufacturer and may not necessarily align with the ADI or NOAEL from toxicology studies.
H (Human): Indicates that the study was conducted on human subjects.
A (Animal): Indicates that the study was conducted on animal subjects (typically rats or mice in toxicology studies).
Oil Vehicle: If no dilution is specified, an oil vehicle (such as MCT oil or olive oil) is used purely to help administer the extract to test animals and does not alter the concentration of the active compounds. If dilution is specified, the vehicle lowers the concentration of the extract, which needs to be considered when interpreting the NOAEL or ADI.
Other Known Studies
A joint paper from the Committee on Toxicology (CoT) and Advisory Committee on Novel Foods and Processes (ACNFP) that defined the 10mg ADI was published on the day the FSA announced it, which allowed for it to be determined that the ADI was derived from toxicology data from Pureis, Cannaray and EIHA.
ACNFP/CoT Joint Paper announcing the 10mg ADI - Published 12/10/23
However, as you can see, the figures from the three reports were based on an Uncertainty Factor (UF) of 300, whereas the 10 studies above either applied a 100UF when stating an ADI, or provided a NOAEL value that can be used to derive an ADI.
If all the labs were brought in line with a 100UF, Pureis, Cannaray, and EIHA’s suggested ADIs would instead be 16.8mg, 36mg, and 51mg respectively.
Further to this, The Hemp Hound Agency can confirm that the application for the ACI mirrors data from Epidiolex, and provided an ADI of 70mg.
This was a claim made during a webinar in February 2023, attended by the head of novel foods, who did not contest it.
Given that, one would have expected a 100UF to be applied by the ACI, especially as no ADI announcements had been made by that point other than the 70mg figure given on 13/02/20 that was derived from data and literature for Epidiolex.
So what is an 'Uncertainty Factor', and how does it work?
Imagine you are testing how much of a certain substance is safe for animals, like rats.
Even if the substance does not harm the rats, scientists cannot be sure the same amount will be safe for humans. After all, people are different from rats.
This raises questions over the need for animal testing, especially in cases like CBD, where human data already exists. But to be extra safe, they lower the amount a person can safely take by using a safety buffer, which is called an Uncertainty Factor.
If a rat can safely eat 10 chocolate bars without getting ill, that gives a NOAEL value of 10.
Scientists might decide that a human can safely eat only 1 chocolate bar, essentially saying, "We’re being 10 times more careful, just in case humans react differently than rats to chocolate."
But not all humans are the same. Some people might be more sensitive than others, so they add another layer of caution.
Instead of allowing 1 bar, they might say, "Humans can only have 0.1 of that chocolate bar, to account for people who might be more vulnerable."
However, in this case
The FSA, through the CoT and ACNFP, concluded that even this typical 100-fold (10 x 10) Uncertainty Factor wasn’t enough.
"An additional Uncertainty Factor of 3 was used to allow for possible increased sensitivity on chronic exposure and for other data gaps the studies couldn’t cover adequately" (ACNFP/CoT Joint Paper)
In a nutshell, the FSA’s approach means that at best, you will be allowed about 0.033 of that chocolate bar, and that's an extremely cautious and arguably excessive limit.
Disclaimer
The Hemp Hound Agency finds it very disturbing that the FSA insisted on animal studies, especially given that numerous previous studies already showed that CBD is well tolerated by humans.
The CoT and ACNFP cite concerns over chronic exposure, but let’s consider the following:
CBD is widely accepted as a food and a medicine
There is no evidence that CBD has the potential for abuse
Chronic exposure is far more likely to result from the consumption of medicinal CBD products (due to prescribed dosing) rather than from CBD supplements.
And as for the so-called data gaps:
The FSA has already accepted the consumption of CBD through hemp oil before 1997, as evidenced in a draft version of the Article 4 (See below) that initiated novel foods for CBD products
That includes wholeplant hemp oils, where CBD is the most abundant active compound
There is evidence of cannabis consumption as far back as 6,000BC, with recorded recipes using hemp flower from the 14th century
CBD was first isolated in 1940 from wild Minnesotan hemp oil, which produced 33% CBD by weight
So, what are these supposed missing data gaps?
CBD Isolate:
Products made from CBD isolate were available in the UK from before 2016
The FSA classified CBD isolate as a novel food in 2017, but that’s still nearly a decade of consumer use.
Synthetic CBD:
There seems to have been synthetic CBD R&D in 2019
Whilst the FSA deem synthetic CBD as the same as plant derived CBD, that clearly isn't the case, purely on the basis of synthetic CBD not coming from the plant
If there are missing data gaps, they likely pertain to CBD isolate and synthetic CBD. Yet, three toxicology reports, two for isolate and one for synthetic, have been used to define an ADI that applies to all products, including those that are wholeplant in nature.
Given the proven historical consumption of wholeplant extracts (which the FSA itself acknowledges), it’s fair to argue that no such data gaps exist for these products.
This reasons statement in now a validated document, which the FSA supplied The Hemp Hound Agency with in May 2023.
It was confirmed as legitimate through a Freedom of Information (FOI) request response from the FSA in May of this year.
So, based on that document, and previous studies that were publicly available before the novel foods announcement, the FSA is effectively endorsing the needless slaughter of thousands of animals to determine if their livers are affected by being force-fed CBD, even though their livers might not react to CBD in the same way it would react with a human liver.
This is despite the fact that they have already recognized the consumption of CBD in hemp oil before 1997.
So back to that 0.033 bar of chocolate
Imagine a hypothetical novel foods process for chocolate bars, with the FSA fully aware that both humans and rats have consumed more than a few bars of chocolate in a single sitting beforehand.
That’s essentially the situation with wholeplant CBD products, especially those that could be described as redefined hemp oils, which contain CBD, and are derived from traditional processes.
So, let's look at the study data for what it is
11 studies combined give an average ADI of 55.45mg when applying a 100UF, or 18.48mg with a 300UF.
This data does not include the ADIs from Studies 2, 6, and 8, for reasons we'll touch on later.
Compare that against the 10mg ADI we currently have, the 70mg ADI from the FSA between 13/02/20 to 12/10/23, the 200mg ADI from the Medicines and Healthcare products Regulatory Agency (MHRA) between 2016 to 13/02/20, and the 2,400mg ADI suggested by the World Health Organization (WHO) in 2019.
There is also a Low Observable Effect Level (LOAEL) from Study 1, which is crucial for balancing the information. However, it’s worth noting that there are no LOELs (Low Observable Effect Levels) or LOAELs (Low Adverse Effect Levels) attached to the other 13 reports.
This is concerning, but in some cases, the data relied on is more of a report (such as Studies 6, 7, and 8), rather than the full study, which may have contained LOEL or LOAEL values.
That being said, all but two of the studies and reports are attached to novel food applications in the UK and EU.
Out of the two:
One is for Epidiolex (Study 3), which gave the 70mg ADI between 13/02/20 to 12/10/23
The other is for an academic institute (Study 5) where a commercial tie-in has not been determined, but it's highly likely due to the focus and construct of the report to be for a novel food application.
Regarding the WHO 2,400mg statement. That seems to be hard to trace, however, that value is accepted by many in the industry who witnessed the WHO statement which came in February 2019, weeks after the EU had announced novel foods.
That figure came with recommendations by the WHO to de-schedule CBD products made up to 0.2% THC.
That recommendation and others were discussed at Vienna in December 2020, with the UK being one of a number of countries who voted against it.
The fact that the 2,400mg figure is hard to trace is concerning, but it was acknowledged by the FSA
In February 2023, The Hemp Hound Agency attended the webinar held by the ACI where they declared in front of the head of novel foods that their toxicology data mirrored that of GW's, and had determined a 70mg ADI.
That was after their toxicology expert had stated that the FSA had refused them the right to use Epidiolex data and literature in the creation of their application, even though an article from 2022 showed that the head of novel foods was advising the ACI and other companies on the requirements of using Epidiolex data in their novel foods applications.
However, at the point of the toxicology experts announcement, The Hemp Hound Agency asked why data was coming back mirroring that of Epidiolex, and not that which was relied on by the WHO to determine 2,400mg a day as safe.
The head of novel foods responded, stating that it was because the WHO data was for pharmaceutical products, and not foods.
That statement is concerning on two levels:
The WHO thought their data was enough to justify the safety of CBD, so why did the FSA dismiss it?
The FSA embraced Epidiolex data in January 2020, which certainly isn't for a food product, and is administered at doses that very few would be ingesting if they were taking CBD supplements as a food-focused product
But then, that's hardly the only case of double standards by the FSA!
The FSA insisted that companies prove consumer sales for their products on 13/02/20 for them to qualify for CBD novel foods, only for Emily Miles, the former CEO of the FSA to state in 2024 that some products on the public list might not have been developed and marketed on 13/02/20
They also selectively enforce the Misuse of Drugs Regulations (MoDR) 1mg rule, and have done so before and after validating Charlotte's Web's products that contain 195mg of THC per container
And most importantly, they take suggestions from the CoT and ACNFP whilst choosing to ignore the undeclared interests of academics that are representing Universities with ties to cannabinoid R&D, and GW Pharmaceuticals
So why did the FSA dismiss WHO data, or more to the point, why did they only use data for Epidiolex to determine the requirements for CBD novel foods and provide the industry with a 70mg ADI between 2020 to 2023?
Back to those historic ADIs from the FSA, MHRA, WHO and the LOEL from Study 1
You might have noticed that the LOEL is only 300mg less than the WHOs 2,400mg ADI, so let me explain that by referring back to the hypothetical 'Novel foods for chocolate bars'.
We all know that chocolate has an effect, but when does that occur, and to what extent?
You could eat one bar, and the anandamide in there, which conveniently for this article is a cannabinoid, might not be felt, but two or three bars (snack size) might lead to a feeling of happiness and a full stomach.
There would be an observable effect there, but not adverse.
For reference, Anandamide is also known as 'the bliss molecule', and is often compared to THC.
You might be able to eat another two and be fine (NOAEL), but past there might come a sugar rush, nausea, and other low observable adverse effects.
So essentially, a NOAEL value doesn't represent a tipping point, and equally, not all observed effects are adverse in nature.
That effectively shows that an ADI, and NOAEL, can be higher than a LOEL for a food.
NOAEL vs. LOEL
Study 1 is a godsend, because it allows us to determine a safety margin.
There's a big gap between the FSAs 10mg ADI, and 2,100mg as a LOEL, which is derived from 30mg/kg bw/day being applied to a 70kg adult.
But, we need to be mindful of the FSAs 300UF, which would bring the LOEL for humans to 700mg.
So this means that we can establish that somewhere around 350mg is the tipping point, where the likelihood of a low adverse effect should be considered.
And that's convenient
According to online data, Epidiolex dosing is suggested at anywhere between 5mg to 20mg/kg bw/day. That's between 350mg to 1,400mg for a 70kg adult.
The MHRA Yellow Card list for adverse reactions to CBD somewhat verifies this to be the case. The list itself was opened since 2006 and is redacted. The CBD industry itself didn't start to emerge until 2013, with it becoming established by 2016.
Some time after that, CBD supplemental products were added to the list, but that means anything between 2006 to 2016 is a catalogued reaction to a medicine or vaccine, and the only company who had interests with CBD throughout that time is the company whose 16 lab reports were used to give the hemp and CBD industry the 70mg ADI between 13/02/20 and 12/10/23.
Past that point, though, the list becomes unreliable due to it being redacted, the inclusion of CBD supplemental products, and the further confusion due to high THC medicinal products, some with little to no CBD in at all, also being included.
But let me ask you this:
How many people do you know that are ingesting 350mg of CBD as a food per day?
It could be argued that there would be very few, because at that level, it's most likely that it's being taken in the hope that it could help manage a medical concern.
The question then becomes economic in focus: Is it cheaper to buy supplemental products daily, or seek a route to prescription?
The MHRA, between 2016 to the start of 2020, maintained a 200mg ADI for consumer products. That value would no doubt have come with considerations for the reactions recorded on the Yellow Card list for CBD.
They wouldn't have allowed that number if there was the potential of an adverse reaction, which suggests that the safety profile of CBD is misrepresented by the FSAs 10mg ADI that was derived from a 300UF.
A further indicator of that misrepresentation
One of the biggest concerns quoted for CBD products is liver toxicity. This issue would be recorded under Hepatobiliary disorders on the MHRAs Yellow Card list. However, between 2006 to 2016, there are no reported reactions under that category whatsoever.
From 2016 to the current day, there have only been 5 recorded reactions in that category out of a total of 609 reactions, or 1302 reactions over the lifetime of the list.
So where do the concerns for liver toxicity come from?
The answer to that question is from Epidiolex data, which again is designed to be taken at higher levels than a CBD supplement would be ingested.
So again, is that 300UF that the FSA seem intent on applying appropriate?
let's dissect the data
First and foremost, the 100UF and 300UF figures are included. This is due to the FSA's use of the 300UF, which will be argued against further in this article.
But straight away, the data paints an interesting and somewhat counter-intuitive picture:
Wholeplant extracts have been shown in multiple studies to be significantly safer than CBD isolate, yet there's only 2.68mg separating the two at 61.8mg and 59.12mg respectively, which is an interestingly narrow gap
CBD isolate seems to be safer than refined full-and broad-spectrum extracts, with a gap of 8.65mg from 59.12mg and 50.47mg respectively. However, Study 1, which represents the latter, comes with LOEL data which in itself suggests that the tested extract would be tolerated at significantly higher doses.
There's something further to consider regarding wholeplant full- and broad-spectrum extracts, and that revolves around the extraction type.
This highlights a potential concern for the safety profile of CO2 extracts vs. those that are derived from traditional methods.
Disclaimer
Study 7 is vague, in that it doesn't truly define what extract type it is. It is assumed to be CO2 derived on the basis of most extracts being as such.
If it is an ethanol derived extract, it would lower the ADI slightly for full and broad-spectrum ethanol extracts to 92.52mg, but the ADI for CO2 extracts would plummet to 31.07mg.
So what about Studies 2, 6 & 8
Something does not feel right about those reports. There's repetitive data for what are vastly different extracts, and the suggested ADIs for studies 6 and 8 are the same, and are higher than the recorded NOAEL value would allow.
For this reason, the ADIs from those studies have not been included in any data to this point.
But despite the raised uncertainties, they highlight a wider concern with CO2 extracts. If the NOAEL derived suggested ADIs were included with all the other values for CO2 products, the overall ADI for them would be 23.5mg, and that's with a 100UF applied. If a 300UF was applied, that would come to 7.83mg a day for a 70kg adult.
That is rather low for a food that has been consumed for millennia
But then, CO2 extracts have not been consumed for very long. The process itself was developed in the 1940s and the earliest known use of it with hemp comes from a beer that was launched in 1996.
Regardless of the state of the data, studies 2, 6 and 8 further raise concerns about CO2 extracts, and with or without LOEL or LOAEL values that would indicate whether a higher ADI could be tolerated, a question needs to be asked on why the values for those extracts are so low in comparison to ethanol extracts.
Round up
There are two points that need to be addressed:
Is a 300UF appropriate for all CBD extracts
And by extension, is the 10mg ADI appropriate for all CBD extracts
The answer to both has to be a resounding 'no'.
Regarding the 300UF - it may well be appropriate for synthetic CBD, but is it for CBD isolate, which seems to have a significantly better safety profile?
It may also be appropriate for CO2 extracts, but it doesn't seem to be the case for ethanol extracts.
And as for the 10mg ADI - if the 300UF is appropriate, why has synthetic CBD been allowed an ADI that is nearly double the level suggested from toxicology data?
On the flip-side, how are CBD isolates, refined full- and broad-spectrum extracts, and wholeplant full-and broad-spectrum extracts being diminished despite the safety data suggesting an ADI of between 4 to 6 times higher is more appropriate than the current 10mg?
Further more, if we were to consider ethanol vs. CO2 extracts, is it appropriate for the former to be lumbered with an ADI that suggests a 10th of that determined in toxicology data?
Conclusion
The blanket application of the 10mg ADI diminishes the safety of certain extracts. That could also be argued as the case for the application of the 300UF, especially on extracts that are derived from traditional processes and have a recognised history of consumption.
The absolute polite way to describe this situation is that the FSA is misguided, but there's an issue with that politeness:
It doesn't factor in that they are advised by the CoT and ACNFP, which are panels populated by academics who represent Universities with interests in CBD and cannabinoid-based medicinal products and associated R&D.
That being said, it's up to the FSA if they accept any recommendations from the CoT and ACNFP, which is the same with the Home Office and whether they accept any recommendations from the ACMD.
Allow me to bring you back to the FSA's legitimised draft version of the Article 4 that launched novel foods in January 2019.
The FSA accepts the consumption of CBD through hemp oil, which has been widely used before 1997.
It doesn't matter that it was derived from cold pressed hemp, because hemp is a recognised whole food, and a recent judicial review on monk fruit decoctions led to the FSA being reminded that if an extract from a recognised non-novel wholefood is derived from traditional processes and isn't significantly altered, that extract in itself is also recognised as a non-novel food.
So why is there no recognition of the historic consumption of CBD through hemp oils in that 10mg ADI?
Final thoughts
Does anyone remember the FSA discouraging hemp and CBD companies from submitting novel applications for ethanol extracts?
The Head of novel foods himself was in that number, but as we can see from the draft version of the Article 4 above, ethanol extraction was defined as a non-novel process in a document that focused on the novel status of CBD.
Suspicious?
let me ask you this: Could it be that the FSA knew that CO2 extracts would have a lower safety threshold?
It's possible, but then it's also possible that they tried to push applicants towards CO2 extracts knowing that it's not a traditionally accepted process, and that there would be no way for a company to argue against novel foods if CO2 extracts were used as the base for their products.
The lack of recognition of historic CBD consumption
It's crazy to think that at one point, full- and broad-spectrum CBD products were deemed as non-novel.
Interestingly, it seems to be something that the FSA don't want to admit to, but then there are publicly available documents to show them and EU member states trying to backtrack from that distinction.
But regardless, the FSA has been shown to accept that cold press hemp oil which was consumed before 1997 contained CBD, so why wasn't that historic consumption data considered when determining the 10mg ADI?
It's shocking that they wouldn't consider that, but considering the novel foods process as it stands, it's hardly surprising because as 12/10/23 shows when the 10mg ADI was announced, the FSA quietly redefine non-novel hemp products in line with an EU amendment to their novel food catalogue made earlier in that year.
Are the FSA and EU member states trying to diminish the authenticity of wholeplant hemp extracts and products? I believe they are, which is why there is no representation of such extracts and products inside of the 10mg ADI, even though it applies to all CBD extracts and associated products.
Animal testing
Why do animals need to die if scientists can't be certain if the tested substance will react in the same way to humans?
I can't be the only one who is flummoxed by that logic.
There is safety data already through human consumption. In fact, out of all the dynamics of extracts identified in this article, the only type that could be argued where testing is needed is synthetic CBD.
Cautionary science
As discussed, if a substance is determined to have a NOAEL of 10, that doesn't mean that 10.1 becomes the LOAEL value.
So why are most studies not providing a LOEL or LOAEL value?
There are four studies that seem to have a curiously low ADI. Two are reports for studies, and two are complete studies.
The two reports are from Charlotte's Web, and show no LOEL or LOAEL, and one of the complete study's are from them, and also with no LOEL or LOAEL.
So why not include that data? It's important after all because it gives a better picture of the safety profiles of the tested extracts.
Is there a reason for that data not being included? Possibly, just as there might be a reason why the determined NOAELs are low.
Maybe there is a wider issue with CO2 extracts, but then, there are studies with those types of extracts with a higher NOAEL and subsequent ADI.
Either way, there are more questions surrounding those low ADIs than answers supplied in the study data.
Could some company's have low-balled their findings? It's a valid question!
That being said, the NOAEL suggested ADIs from Charlotte's Webs extracts do not match that which they have stated in their reports, which they seemingly argue for on the basis of further studies that were conducted by others.
Forgive me, but that’s like saying I’m an amazing bagpiper because I’ve listened to some amazing pipers, or claiming I’m good at pool because I’ve been watching snooker on TV.
For those who are curious, if studies 2, 6 and 8 were added to the overall ADI determined with the other studies, it would go down from 55.45mg to 47.52mg.
Even with the uncertainties surrounding their reports, that number is still significantly higher than the 10mg ADI as it stands.
Data gaps
It's highly possible that the mention of data gaps refers to the state of EIHAs application. It's for 10% CBD isolate in hempseed oil, yet there are products on that application that are far from that description.
There are also full- and broad-spectrum products on that application, but for some reason the FSA has allowed for it to be progressed to the point of authorisation, without indicating any outcome for listed products that exceed the specified dynamics.
However, it's more likely that the data gaps are from all three applications that have been processed.
What it can't be is a reference to the state of all novel food applications, because they haven't all been thoroughly reviewed. And what that means by extension is that the FSA is forcing the hemp and CBD industry to accept what they have already deemed to be incomplete data, even though separate applications might include the data the FSA claims is missing.
That's not just misrepresenting the novel food process, that's misrepresenting all of the companies who are going through that process.
The 'other' data gaps
The ACNFP have recently discussed the potential of THC being defined as a contaminant, which at best can only be described as short sighted when considering wholeplant products, and that THC and all other naturally occurring controlled cannabinoids have been ingested as long as the plant has been provenly consumed.
The really sad thing in this is that you would expect scientists to understand the history of the compounds they are discussing, as well as for that which they're derived from.
But that being said, they and the CoT are effectively a panel of conflicted academics who the FSA suggest don't have to declare the interests of their employers.
The University of Reading is the absolute case and point for that comment. They publicly announced in November 2019 that they had helped with the development of Epidiolex from 2007, but in January 2020 when the CoT discussed GW Pharmaceuticals literature and toxicology data for Epidiolex, faculty members from Reading had nothing to declare, and neither have they from beyond that point, even when highly criticised for not doing so by The Hemp Hound Agency.
Let me put this to you: What if there are no data gaps on the basis of Novel food Regulations, and actually, 'someone' is over-applying pharmaceutical data and asking a food industry where their equivalent data is?
Uncertainty Factors
Most new foods are judged by a 100UF, so what makes CBD so different that a 300UF needs to be applied?
The EU has recently been releasing reports that call for caution with CBD products as well as controlled cannabinoids, which have been seemingly mirrored by the FSA through the CoT and ACNFP, but again there's no consideration for the types of products that they defined as non-novel in 2017.
The head of novel foods, Paul Tossall, stated to The Hemp Hound Agency in 2023 that as soon as CBD products appeared on the market, they were instantly novel.
The Hemp Hound Agency only agrees to this up to a point, and that would have been as soon as it was realised that wholeplant hemp oils and extracts are redefined when being sold as CBD oils and extracts, and a redefined food is not a novel food.
So on that basis, it could be adequately argued that such products made up to 10% and with up to 0.2% THC should not be judged with any uncertainty factor, especially as they were consumed regularly, and all around the world, several thousands of years before the FSA and Novel Food Regulations even existed.
The worth of a single study
The FSA used one study for synthetic CBD to form their opinion that led to the 10mg ADI.
Simply put, that means that the data from Study 10 has as much weight, and it reflects a truly non-novel extract in comparison to others that could and should be defined as novel.
Subsequently, it could be argued that the study shouldn't be subject to a 300UF, or a 100UF when viewed in comparison to similar extracts that were utilised prior to 1997.
Favouritism
The 10mg ADI diminishes all types of extracts bar one, synthetic CBD, which the FSA suggest is identical to plant derived CBD.
This is specific to one company, Pureis, who have been seen to have benefitted in other ways through the novel foods process, specifically by being allowed into a process they didn't qualify for due to having no products available to the consumer on or before 13/02/20.
The FSA claim that is OK because Pureis had products 'on the market' on 13/02/20, even though the company itself wasn't launched until July of that year. What's more, their products weren't available to the consumer until after they had been validated by the FSA in March 2021.
They hold one of a number of applications the FSA has validated and now authorised that proves the FSA is creating a cartel market.
Charlotte's Web are also in that number, as are another two companies whose studies are referred to in this article.
On that very basis, companies should be questioning the 10mg ADI, and whether it's appropriate for their products. More so because regardless of what the FSA say, synthetic CBD is not the same as that which is derived from the plant.
Semantics
Novel foods for CBD products is being ran on the basis of there being no recognised sales of CBD products before 1997, even though the FSA determined wholeplant CBD products as non-novel in 2017 due to the sale of hemp oil which contains CBD prior to 1997.
I think I've said enough there...
But at the end of the day
The data presented in this article shows that the 10mg ADI is over cautious, and when someone in Government, or a Government department is being overly cautious, they're not doing it on the basis of consumer safety, they're doing it with commercial interests in mind.
Call to action
It’s becoming increasingly clear that the intent behind the novel foods process is unravelling.
The Hemp Hound Agency has consistently argued that, contrary to what the FSA claims, novel foods regulation is not about public safety. Instead, it’s about protecting and promoting the commercial interests of a select few, rather than fostering the growth and success of the industry as a whole.
In reviewing the data, it’s worth noting that some of the companies contributing this data have been implicated in complaints against the FSA for allegedly creating a cartel market. However, their data still provides valuable insights that highlight the inconsistencies in the FSA’s approach.
The 10mg ADI simply does not hold up, even for synthetic CBD, especially if a 300UF is likely excessive. Furthermore, the FSA’s exclusion of real-world data from wholeplant extracts in determining the ADI demonstrates a deliberate narrowing of the scope by treating CBD as a uniform substance across formats, when that’s clearly not the case.
The Hemp Hound Agency has warned of these issues since 2022, when it first became evident that the FSA was working with a select few, rather than addressing the needs of the wider industry.
If this debacle is ever to be resolved, the industry must find its voice and challenge the FSA’s actions. Now is the time to either speak up, or stand behind those who are fighting for the industry’s interests, like The Hemp Hound Agency.
1 - Tallon, M. J., & Child, R. (2023). Subchronic oral toxicity assessment of a cannabis extract. Regulatory Toxicology and Pharmacology, 144, 105496. https://doi.org/10.1016/j.yrtph.2023.105496
Notes: Industry reports and the study itself suggested an ADI of 161mg for a 70kg adult, yet that doesn’t seem to reflect that the 230.25mg/kg bw/day NOAEL was determined after the extract had been diluted to between 31% to 33% with MCT oil. The study was conducted on behalf of, and funded by cbdMD.
2 - Dziwenka, M., Coppock, R., Alexander, M., Palumbo, E., Ramirez, C., & Lermer, S. (2020). Safety assessment of a hemp extract using genotoxicity and oral repeat-dose toxicity studies in Sprague-Dawley rats. Toxicology Reports, 7, 376-385.
Notes: The NOAEL’s for the raw extract as well as a diluted extract were reported to be identical.
The study was conducted on behalf of, and funded by Charlotte’s Web
3 - European Medicines Agency. (n.d.). Epidyolex, INN-cannabidiol: Summary of product characteristics. Retrieved from https://www.ema.europa.eu/en/medicines/human/EPAR/epidyolex
Notes: This data was used by the CoT in January 2020 to define a 70mg ADI which was announced on 13/02/20.
4 - Henderson, R. G., Welsh, B. T., Rogers, J. M., Borghoff, S. J., Trexler, K. R., Bonn-Miller, M. O., & Lefever, T. W. (2023). Reproductive and developmental toxicity evaluation of cannabidiol. Food and Chemical Toxicology, 113786. https://doi.org/10.1016/j.fct.2023.113786
Notes: The study was conducted on behalf of, and funded by Canopy Growth.
5 - Marx, T. K., Reddeman, R., Clewell, A. E., Endres, J. R., Béres, E., Vértesi, A., Glávits, R., Hirka, G., & Pasics Szakonyiné, I. (2018). An assessment of the genotoxicity and subchronic toxicity of a supercritical fluid extract of the aerial parts of hemp. BioMed Research International, 2018, 8143582. https://doi.org/10.1155/2018/8143582
Notes: The study was sponsored by CV Sciences.
6 - European Commission. (2021). Summary of the application: Extract of Cannabis sativa L. extract with Tetrahydrocannabinol (THC) removed. Novel Foods. Retrieved from https://food.ec.europa.eu/document/download/f891b257-727d-4ddf-b991-bb078f098fa5_en
Notes: The study was conducted on behalf of, and funded by Charlotte’s Web
7 - European Commission. (2022). Summary of the application: Oil extract distillate from Cannabis sativa L. Novel Foods. Retrieved from
Notes: No vehicle or dilution indicated, neither is extraction method. Study details as part of an EU novel foods application for ARCO.
8 - European Commission. (2021). Summary of the application: CO2 extract from Cannabis sativa L. Novel Foods. Retrieved from https://food.ec.europa.eu/document/download/80546dc5-b916-4afa-b0a9-43cc3b561bd3_en
Notes: The study was conducted on behalf of, and funded by Charlotte’s Web
9 - Dziwenka, M., Dolan, L., & Mitchell, J. (2021). Toxicological safety of VOHO Hemp Oil; a supercritical fluid extract from the aerial parts of hemp. PLOS ONE, 16(12), e0261900. https://doi.org/10.1371/journal.pone.0261900
Notes: Study conducted on behalf of Verdant Nature LLC in collaboration with HempFusion.
10 - Dziwenka, M., & Lermer, S. (2023). Toxicological evaluation of HempChoice® full-spectrum hemp oil extract: A 90-day repeated dose oral toxicity study in rats. Heliyon, 9(5), e04120. https://www.cell.com/heliyon/fulltext/S2405-8440(23)04120-8
Notes: Study conducted by Geocann.
Comments