LEVOTHYROX Crisis - Act II - Atoute.org

I would like to start this new post with a quote from Claude Bernard, father of scientific medicine. From the publication (1865) of his famous Introduction to experimental medicine, he specified its limits with impressive clairvoyance.

This post follows the first one which is still readable here but which did not take into account the detailed data of the bioequivalence studies of the new LEVOTHYROX®

So, faced with the ANSM's inability to reassure patients and properly inform doctors, Agnès Buzyn took charge of the file on September 15 with a commendable efficiency and pragmatic intelligence.

It followed the requests of patient associations and many doctors by asking the ANSM to publish the source data of the studies supposed to prove that the new formula of LEVOTHYROX® was biologically equivalent to the old one. This is a first in France and an excellent decision.

Agnès Buzyn also announced on France-Inter the temporary availability of stocks of drugs identical to the old formula for users in difficulty and the marketing of new generics of LEVOTHYROX, allowing to widen the choice of patients. . Finally, by citing THYROXINE® drops as an alternative, it puts an end to the rumor of a risk of stock shortages for this form initially intended for children.

What does this raw data show?

The 3000 published pages contain all the data collected from the 204 volunteers who tested the old and the new drug. Only the subject numbers were erased, no doubt to avoid reusing the data for other analyzes [1]. But let's not be choosy, it's a first in France and a huge improvement compared to the previous situation. This crisis will at least have had the advantage of shaking up the French health authorities, which are far behind their European counterparts in terms of transparency and access to data, in the right direction.

If the scientific explanations that follow put you off, you can go directly to the summary.

In my first post, I deplored the absence of data on the dispersion around the mean: if the two drugs are bioequivalent on average, what about each individual? Was there a significant percentage of subjects for whom the absorption/diffusion of the active principle differed to the point of being able to cause physical disorders?

By comparing the simple dissolution in water of the old and the new LEVOTHYROX, one understands the importance of the excipients, which can be at the origin of differences in absorption from one subject to another, even if in average, the absorption is identical:

Let's start by acknowledging the quality of the work carried out by the pharmacologists of the Merck Serono Laboratory: it is a solid study, carried out with experimental rigor that does not suffer any criticism. From an ethical point of view, insofar as this is work requested by an Agency, it would have been preferable for it to have been carried out by a team other than that of the laboratory marketing LEVOTHYROX, but there is no reason to doubt the sincerity of the data.

The visualization of the dosage curves shows that for the great majority of the subjects tested, the absorption and blood diffusion of levothyroxine are remarkably superimposable for the old and the new formula. Indeed, if we observe a significant variation in the level of levothyroxine from one subject to another for the two formulas of LEVOTHYROX, for the same subject the blood levels of levothyroxine (T4) are almost identical when we pass from one form to another. It is therefore the favorable hypothesis that I described in my first post that is the rule.

It is enough to be convinced of this to look at the 204 graphs which display the blood levels of each individual for the two formulas of the drug, spread over 3 pdf of 100 pages each [1]: 1 - 2 - 3

Here are three fairly representative examples of the majority of these graphs

The correspondence for the absorption of the two different formulations of LEVOTHYROX in the same subject is impressive.

Beware of a small particularity: levothyroxine is already present in the organism of the volunteers, since it is a natural hormone. The initial rate is not zero. This base rate for each individual is not the same for the two successive tests spaced out over time. It is therefore necessary to take into account the difference between the horizontal dotted lines (initial rate for each formula) to interpret the difference between the two curves. In the graph below, one could erroneously conclude that there is a difference, whereas the starting blood levels show a shift identical to that of the curves. For this subject, the equivalence of blood levels for the old and the new formula of LEVOTHYROX is perfect.

Clearly, in the overwhelming majority of cases, patients should have no problem switching from the old to the new formula. This is also what doctors saw in their patients before the recent media frenzy.

But we must also keep in mind that the conditions of this study are intended to avoid deviations from the average... It concerns 204 subjects under 50 years old, ethnically homogeneous (white), free of thyroid problems, presenting a normal general blood test (including good kidney and liver function), not taking other medications or food supplements, not drinking certain fruit juices, not smoking, not usually following a special diet, etc. The detail is in the protocol from page 30/1628. It's the law of the genre, and there's nothing abnormal or atypical about it. However, this "ideal" population is not at all representative of all LEVOTHYROX users.

As an old researcher said: "It's like describing the life of insects by having studied two dead ants"

This is not to say that the results are unreliable, but that if a few discordant results are found in this context, there will probably be more in "real life" [2].

And precisely, there are discordant results. Very few, but there are. Since the numbers of the volunteers have been erased, I will designate them by the pages of the report as they appear in the header of each page after the mention EMR200125-001. For example, the topic described on page 956:

Or on page 958

And also, to a lesser extent, topics/pages 927, 926, 934, 935, 944, 947, 971.

We are not clones, and the diversity of humans logically results in the existence of variations from the norm, of subjects who present a difference in absorption/diffusion between the two formulas. The publication of the data shows that these differences fortunately only affect a small percentage of subjects, and that their amplitude is low, contrary to what the omerta around these data which remained inaccessible until now could have led to fear.

Nevertheless, a small difference in some subjects, for a drug with such a narrow therapeutic range, is perfectly compatible with the following consequences (which roughly correspond to the current situation): No problem for 90 to 95% of users. A moderate imbalance for 5 to 10% of users, i.e. all the same 150,000 to 300,000 people A significant imbalance for 1% of users, i.e. 30,000 patients who were deeply destabilized, who only understood the cause a posteriori afterwards weeks of suffering, and who show legitimate anger in social networks and the media.

(Edit on 28/9/2017: these figures have been confirmed by a survey of 2900 doctors)

Rather reassuring data

This reassuring vision is nevertheless based on several prerequisites: That nearly identical blood levels of levothyroxine are well representative of an identical effect on the organism. To free ourselves from this always delicate postulate, it would have been necessary to test the new LEVOTHYROX in patients and over a longer period, and to monitor both their TSH and their feelings. Blood levels measured for three days are sufficient to assess bioequivalence, whereas due to the half-life of levothyroxine of 6 days, blood balance will only be achieved after one month. That the markers used to compare the products are relevant for levothyroxine: area under the curve and maximum concentration. As much as these markers seem suitable for an analgesic or an antibiotic, as much for levothyroxine, we could just as well have been interested in the average of the last two measurements (48 and 72nd hour) which gives an idea of ​​the residual hormone level.

These prerequisites make the term “bioequivalence” qualifying this type of study rather peremptory. A more relevant term might be hematoequivalent or bioapproximate. Only a long-term (and very expensive) study in many patients would confirm true bioequivalence.

Despite these reservations, it appears on reading the detailed data from the main bioequivalence study that the new LEVOTHYROX is perfectly suited to replacing the old formula.

Same problem in New Zealand

The only error finally consisted in not warning users (by an alert on the box) that it was about a new formula likely to modify their hormonal balance, not to communicate on a problem which was however foreseeable ( especially since it had already occurred in other countries in an identical context). This mission was delegated by the ANSM to doctors and pharmacists, without them being sufficiently informed of the risks of notable side effects of the substitution in a small percentage of patients.

The disorders described by users of the new formula of LEVOTHYROX can be due to several causes: Toxicity of the new excipients: highly improbable hypothesis that no scientific evidence currently supports. The ANSM published on 15/9 new controls which it had carried out at the end of August on batches of the new formula. The quantity of mannitol contained in the tablet is insufficient to cause disorders, idem for the citric acid. A more or less intense destabilization of the thyroid hormone balance, worsening over time due to the need for a month for blood concentrations to stabilize, in 1 to 10% of users. It is becoming clear that this is indeed the main explanation for the current crisis. sanitary. This media storm reinforced the negative perception of patients who were not very destabilized and also triggered symptoms in those who were doing well before discovering in the media that their medication could pose a problem: this is the famous nocebo effect. Finally, there is the attribution to the new LEVOTHYROX of responsibility for accidents that occurred just after the change of formula, whether the new drug was really the cause or not [3]. It is not insulting patients to speak of the nocebo effect. All the doctors have been victims of it themselves: during my studies, it was enough for a nurse to discover lice on a hospitalized tramp for everyone to immediately perceive genuine itching... The other frequent option being to feel the symptoms of the diseases we were studying! On the other hand, to insinuate that all the difficulties are due to the nocebo effect is both false and dismissive.

A huge mess

The most terrible thing about this story is that we have hidden reassuring data. ! Competent journalists or information decoders like me could have participated much earlier in calming people's minds if transparency had prevailed.

As for the fact of warning the patients thanks to a mention on the box, the immediate nocebo effect that this mention would have caused should not be underestimated, but the bias of transparency would have made it possible to reassure the patients much more quickly. , to limit the proliferation of conspiratorial hypotheses, and to avoid having to manage the crisis in panic. Trust is inseparable from transparency. This was also the recommendation of American endocrinologists in 2004, faced with the same problem with the various levothyroxine formulas on the market: inform patients of any changes.

In practice

In practice, users who have not experienced any problem with the new formula should do nothing.

The best solution for patients in difficulty consists in seeking with their doctor a new balance by taking the new LEVOTHYROX, even if it means relying as much on the feelings as on the TSH, the meaning of which is not absolute, contrary to what is still believed. many of my colleagues. The use of the old formula, which can only be temporary, should be reserved for those who absolutely cannot manage to rebalance themselves with the new LEVOTHYROX. This cannot be a lasting solution. Sometimes the negative anchor associated with a product is such that it triggers real symptoms, much like people who have eaten a spoiled oyster and cannot eat it for years without getting sick. In this case, it will be possible for them to turn to new drugs containing levothyroxine which will be marketed in a few months.

The lessons of the LEVOTHYROX affair

The first lesson concerns transparency: failing to be ethical, hiding information that could cause concern is no longer possible in the age of the Internet and the rapid circulation of information. The best way to defuse health crises is to release any information that can be released, including if it bothers certain manufacturers. It is all the intelligence of Agnès Buzyn to have understood this. I am afraid that it is too late to apply this principle to the vaccine crisis which will explode with the new obligation, but I think that a page is being turned and that France will evolve towards a democracy health which finally puts into practice the principles it claims.

It is not necessarily a question of publishing everything, but in the present case, of giving fundamental information which was the dispersion of individual differences around the average of 0.7%. One page would have sufficed (note that this information is absent other than graphically among the 3000 pages put online). I still remember a meeting at the AFSSAPS (ex ANSM) in 2011 where it was a question of communication around generics. I asked the manager why the Agency did not put the bioequivalence studies online in order to reassure the public and doctors. His surreal answer was "That would take up too much space"... But so, we are progressing.

Finally, this case demonstrates that the term "bioequivalent" must be understood for what it means. This is not a total equivalence of effect, nor an equivalence of effect for all patients. We discover that a bioequivalence study is only intended to verify that the new drug is well dosed "on average" and that there are no major interactions with the excipients. It's useful, but it doesn't mean that it won't cause problems in some atypical patients.

For the majority of generic drugs (tested in the same way), the small differences observed in some do not always translate into symptoms. But there will always be a few patients who deviate from the norm and experience actual symptoms during a substitution. The psychology and the understandable feeling of a drug "at a discount" compared to generics does not explain everything.

Summary for Dummies in Pharmacology

Thanks to Agnès Buzyn, we have the complete and detailed results of the bioequivalence studies of the new LEVOTHYROX. The unjustifiable secrecy on these capital data is finally lifted.

These results show that the new LEVOTHYROX reproduces almost identically to the old formula the blood levothyroxine (T4) levels in the vast majority of patients, who should therefore not suffer any discomfort when switching to the new formula. .

These results are compatible with the possibility of a moderate hormonal imbalance in 5% of users of the new LEVOTHYROX, or even significant in 1% of them. Due to the number of French people using this drug, this may therefore represent 150,000 and 30,000 users respectively, which seems compatible with the situation currently observed.

The media "noise" around this drug has amplified the negative reactions in some patients, but this effect alone cannot explain the current situation.

There is no evidence to suggest that new LEVOTHYROX is a bad drug or that its excipients could be toxic in any way at the minute doses at which they are present in the tablets.

Users who do not feel anything abnormal, and who are the vast majority, need not worry.

Those who feel abnormal disorders should take stock with their doctor. Independently of this bioequivalence study, it should be kept in mind that TSH is useful for monitoring hormonal treatment, but that it is not infallible and that one should not hesitate to test a small increase or decrease in dosage in case of symptoms suggestive of hypo or hyperthyroidism. The best solution remains to seek to rebalance the treatment with the new LEVOTHYROX and there is no rational reason for this not to work.

For those who really do not want to try to rebalance themselves with the new LEVOTHYROX, it is possible to use L-THYROXINE SERB in drops while waiting for the imminent but transitory marketing of a drug from the Merck Serono laboratory identical to the old LEVOTHYROX , under the name EUTHYROX or EUTIROX [4]. They can then refer to other drugs containing levothyroxine which should soon be marketed in France.

Addition of 2/10/2017: The ANSM is posting an update on the availability of the different formulas of levothyroxine. See the patient information sheet and the one for health professionals at the bottom of the page, which are rather well done.

Acknowledgements: I would like to thank Régis Bouquié, pharmacologist, Olivier Gaget, public health intern and my wife Claire Dutrey-Dupagne (multi-skilled!), for their scientific proofreading. Christian Funck-Brentano (pharmacologist) was unable to proofread this version before its publication, but he had given me major help in understanding bioequivalence studies when writing the first article.

Link of interest: I have no link of interest with the pharmaceutical industry. I am a consultant for the editions of VIDAL. My wife, a cardiologist, worked in development in the pharmaceutical industry for 25 years, before leaving in 2011 to devote herself to geriatrics in nursing homes.