Time to rethink GP's advice on weightloss, and ticagrelor data doubts
33:38 Share

Let's talk photos, not just storing them, showcasing them. You've got images that matter. Whether you're a photographer, a business updating your followers, or just someone who wants to share life's moments the right way. So why hand them over to Big Tech's one-size-fits-all cloud? Big Tech companies are the fast food of photo sharing. Quick, easy, but not exactly gourmet. And what about your data integrity? Jalbum.net is the photo sharing solution that puts you in control.

Want to host images on your own server? You can. Want a layout that actually reflects your brand or style? JAlbum's customizability is unmatched. And if you're a business sharing regular photo updates with your audience, this tool was built with you in mind. But don't just take our word for it.

Over 230 million web pages have been created with JAlbum, and it's got stellar reviews on Trustpilot to prove it. So head to jalbum.net to download your free software and try it out. When you're ready to upgrade, use the code podcast for 20% off. Your photos, your layout, your rules, jalbum.net. Welcome to the Medicine Science Podcast from the BMJ. I'm Cameron Abassi, Editor-in-Chief.

This week, we'll hear how the FDA decision to approve the antiplatelet drug Ticagrelor was not only controversial but based on bad data and why routine discussions about weight loss in primary care should be rethought.

Firstly, Ticagrelor. Last December, the BMJ published an investigation into the 2009 Plato trial, exposing serious problems with that study's data analysis and reporting. That investigation also showed that even the FDA's own reviewers didn't support approval of the drug.

Our follow-up investigation shows that those data problems extend to other key supporting evidence in AstraZeneca's initial application to regulators. Peter Doshi, who's one of the editors in the BMJ's investigations unit, joins me now to discuss this. Peter, welcome. Tell us, this is your investigation, tell us about the investigation and what the main messages are.

Well, I think there's probably a lot of takeaways, but I suppose the way I would sum it up is that the investigations that I've led here into these three clinical trials that supported the initial approval, they lead really to the conclusion or they support the conclusion that the data that FDA received for these trials is not trustworthy. There's just too many data integrity problems.

All of this is detailed in the investigations articles that we've published freely available on bmj.com. You said it very briefly. That's got a big statement you made. Yeah, I mean...

The articles that I've written do contain many strong allegations, and we did not make those lightly. You know, they're the result of many months of investigating, checking and double checking our work. So what then led you to draw that conclusion, Peter? What were the elements of those trials that raised those concerns?

So Ticagrelor is this antiplatelet drug that's manufactured by AstraZeneca and it came onto the market in 2011. And the approval was based on this trial called PLATO, which is AstraZeneca's huge 18,000 patient landmark trial that took place across 43 countries and was reported to show this benefit compared to clopidogrel against cardiovascular mortality.

And before I even got started with my own investigation, there were these very troubling signs from the start. First, the FDA approved this drug against the advice of its lead scientific reviewer, who had called AstraZeneca's application one of the worst he had ever seen.

And one of the major issues that got discussed at the time of approval was that in the United States, ticagrelor patients actually fared worse than those given clopidogrel, which is again that standard of care drug. And so my investigation took this issue a bit further, and I discovered that there were four countries where study monitoring was done exclusively by a contract research organization, so a third party.

And in the rest of the countries, monitoring was conducted by AstraZeneca. But where the monitoring was conducted exclusively by a third party company, such as the United States, Ticagrelor fared worse in all four of those countries. OK, so these two issues you raised created some concern about

the trials in your mind? Oh, absolutely. I was able to leverage a recent freedom of information request to the FDA where we obtained the patient-level data that AstraZeneca submitted to FDA. And those data allowed me to look at the work of this independent central endpoint adjudication committee

And there were some quite unexpected patterns and imbalances in those data that point to the possibility of unblinding in the trial. You and I obviously discussed this on several occasions.

It's not an issue that other people weren't talking about, is it, Peter? These concerns have been raised in the past. And one of the problems is that they haven't fully been addressed. That's one of the big issues. And a doctor named Victor Serebrani has been really working hard over the last decade

to show that this trial is unreliable, that the approval is unreliable. And what I did with my investigations, one very core element of it was looking into his allegations and verifying for myself whether or not I agree with them, and I did.

And so part of what we did was we did freedom of information requests, looking at the data that FDA actually received. We're going to hear from him very shortly, but you described really in fact what you found in your initial investigation about the PLATO trial. What have you found this time around in the follow-up investigation?

My follow-up investigation that we just published looks at two platelet studies. These are phase two studies in patients where they're taking a lot of blood draws at different time points to measure platelet aggregation levels after ticagrelor versus clopidogrel. But we found many problems with these trials. For example, the primary endpoint in both of the studies was not accurately reported in the journal publications.

There was also a key but unpublished data adjustment that occurred in one of the trials, which had this effect of obscuring the fact that the data set actually contained many completely implausible data points. And this has implications for the findings, understandably. Yes, absolutely. And so one of the aspects of the investigation looks at the

You know, what I've tried to do in both the original investigation and this one on the platelet studies is obtain original records and compare them to what FDA received. And, you know, investigators were not cooperative at all in terms of being able to provide any data for the platelet studies. However, we did have access to some of the recordings from the actual machines that were used at the principal investigator trial site.

And we were able to see that not all of those recordings made it into the FDA's database. Okay. And of course, I mean, one response to this might be, well, there were these irregularities, but the drug still works or it's better than the comparator. But I think what the argument here is,

is that, in fact, those irregularities do raise questions around whether or not there is a benefit from taking this particular drug. Let's hear Victor, who you just mentioned. You've interviewed him, Peter, for this current issue of the BMJ, and he can explain where this started. Let's listen to that.

I'm a vascular pharmacologist, I would say. In a previous life, a pediatric nephrologist and my PhD in nephrotic syndrome, which is a vascular disease very similar to myocardial infarction in terms of use of antiplatelet agents. And when I arrived in the very early 90s in the United States,

The platelet research became stunningly growing, and there were a bunch of drugs developing. So my knowledge of how to do platelet testing became very valuable in my career, go up, up, and up. And I became the big insider within the big pharma involved in development of antiplatelet agents.

So if you can go back, just take us back in real time. Can you tell us a little bit about how you first started to become concerned about Techagular's evidence base? When I was a big, juicy, you know, farmer, big farmer boy, I was lecturing to them and they paid me very heavily. And I was in internal discussion in the sponsors, American headquarters in Wilmington, Delaware,

And the question was very simple. Victor, what should we do to prove that this drug will be absolutely stunningly great? And I look at the profile and I look at phase two data, which were not impressive at all, the dispersed program. And I told them, if there will be no mortality benefit, this drug will not fly. Period. End of story. And boom, guess what? They got the mortality benefit.

which excites me immediately because I was, wow, that's absolutely great, before we start digging into the data. Why should doctors who are treating patients with antiplatelet drugs, why should they read the BMJ investigations? The benefit of antiplatelet drugs is a very, very small, and it happened very, very early after the stent implantation in the coronary artery. This first week,

is when the events happen and when the antiplatelet drug should protect against repeated myocardial infarction, stroke, or death. Since such events are very, very seldom without stent implantation,

Long-term use of antiplatelet agents is barely change anything. Therefore, any more trials rather than plata per se will not show any deterioration or any benefit. This is exactly what happened.

in the ticagrelor development. So they start a variety of trials in stroke, in stable coronary disease, in diabetes, to repeat the death benefit observed in PLATA. They never receive anything even close to what was reported in PLATA. If there is no mortality benefit,

There is absolutely no reason to use that drug. Thanks. You were listening to Victor Cerabrowni being interviewed by Peter Doshi. Now, also listening to that interview, we have Rita Redberg, a cardiologist and professor of medicine at the University of California, San Francisco, and former editor of JAMA Internal Medicine. Rita, thanks for joining us.

What do you think of this Ticagrelor story, particularly in relation to the debate around mortality benefit? Sure. First, it's a pleasure to be here and congratulations to the BMJ on this investigation. I think it's really important work. As Peter had detailed in his article, Ticagrelor was a blockbuster drug generating millions of dollars in revenue.

for the company AstraZeneca and widely used in the treatment of patients with acute coronary syndrome, very common and part of the dual antiplatelet therapy for patients getting cardiac stents and other kinds of stents. I mean, as always, it's really important to know for any drug that the safety and effectiveness can be assured. And that is what people expect from an FDA approval. And so the concerns raised

And detailed in Peter's two articles and investigations about the safety and effectiveness and the assurance that the FDA gave without the data actually supporting it is a great concern and concern.

Then magnified by the publications, which omitted key parts, omitted key data, it appears from this investigation, omitted key conflicts of interest. The drug was recommended in cardiology professional society guidelines and many patients received this drug. So there was great concerns raised by these pair of articles.

Yeah. Peter, you've, I mean, clearly you investigated this, you've, you know, with Victor, looked into the detail of this whole story. And you tried to have this investigated by responsible authorities. Where are you at with that?

Yeah. So, you know, as I was saying, the pieces that I've published make very strong allegations and we did not make those allegations lightly. And so still, even though you've checked, you've double checked all your work, you want to make sure that we can.

really stand that up. And so you have to, you know, go to the investigators because there's always this doubt that somehow maybe there's, you know, there's some other way to interpret what I'm seeing. So it's really key to speak with those accused parties, not just allow them the quote unquote right of reply, but, you know, have an open conversation with them and hear their side of the story. And I think one of the most concerning aspects of

of my work over the past year on this has been how many people are basically turning their back on the situation. And I was thinking about all the various parties and there's actually so many people that I kind of divided into a few buckets here. The first bucket are the principals, right? The study investigators, the primary endpoint adjudication committee, the sponsor, the lead investigators of all three of these studies.

these three trials, they have to date never responded to us. They either do not respond or through a public relations office, they formally decline an interview. So that's one layer. The principals don't want to speak.

Then we have the oversight bodies, right? And I think amongst these parties, the most stunning for me really was the chair of the Institutional Review Board, something that's called the Independent Ethics Committees.

At the hospital where the platelet studies occurred, that chair refused to speak with me even after I wrote saying that we have concerns about the data integrity and trials that occurred in this hospital. Declined to be interviewed, did not answer any questions that we sent. That to me is just quite unbelievable.

And then the third bucket I thought about are the journals. You know, what role do the journals have here? And the journals publish these trials. And these are the leading journals in their area. You know, it's the New England Journal of Medicine for the Plato study and its circulation for onset, offset and response studies. And we have verifiable errors in all of these articles. You know, for example, in my latest investigation, the primary endpoint in the articles reported as statistically significant

But according to the company's internal report, it was a non-significant result.

And the journals, in both of these cases, these two journals have done nothing to correct the record. There's no expression of concern. The New England Journal did say that they were in touch with the authors at one point, but then basically said, we're not going to do anything because they said it's an, quote, active topic of discussion in the literature, and that was the best place for further inquiry.

And so there hasn't been an expression of concern, no correction on the article, which is what I think we would do if this were a trial that we had published. Okay. So it's been quite difficult to get anybody to engage with you, let alone take some responsibility for this. Rita, as an experienced clinician and editor, what do you think needs to happen from here? Well,

Certainly, I think it's important to have an accurate record. As Peter noted, this particular drug seems to be about to go off patent and become generic.

But I think the original FDA approval needs to be accurate and based on accurate data. What's in the literature needs to be accurate, and our guidelines need to be based on accurate data. And then of more concern is how do we know that this hasn't happened for other drugs and there just hasn't been an investigation? It really undermines sort of the information.

about integrity and assurance of the entire process because so much of it is conducted without public scrutiny and without investigations. And we rely on the integrity. And I hope that that is a reasonable assumption, but it does raise questions that need to be answered for us to make improvements in the system and assure this doesn't happen again. And then again, the post-market surveillance part I think is important because

transparency of the data is important so that the data that is obtained in the trial needs to be the same data that goes to the FDA and is available to the public so we could have improvements in all of that so that these kind of things cannot happen so easily again. Yeah, I think that's right. And also I think from a journalist's perspective,

Certainly, as Peter intimated, if there were these concerns raised about papers that we'd published now, I guess someone will give me an example where we haven't done this. But I think it's the duty of a journal and the journal's editorial team to take those seriously and go back to your original response, Rita, was correct the record. That's fundamentally what's important here. I agree. I think a journal's primary mission is to create accurate scientific record.

Peter, finally, who else might be interested in looking into this in the U.S.?

I think people who have been affected by this drug, whether directly from receiving it, those families, or whether indirectly by being a taxpayer where this country has spent billions of dollars on this drug, they ought to be concerned. And the way you go further, in my opinion, is you have to break through the stonewalling. It cannot be the end of the story that the investigators responsible for this simply refuse to speak.

The journals refused to act. That cannot be the end of this. And as you mentioned, Cameron, in your editorial, the courts have a powerful lever here and near exclusive power, really, to address the situation. You absolutely need some kind of institution with subpoena authority to

to make people discuss, to have them show us their records. That is what I see as absolutely needed. I see no alternative and patients deserve no less. All right. Listen, seems to me this is to be continued. Peter and Rita, thank you so much for joining us. My pleasure. Thanks, Cam. And those BMJ investigations into Thai Kagra law are online now and free to access at bmj.com.

Now, a new analysis paper in the BMJ this week is bound to catch the attention of doctors and patients. A group of international authors argue that weight loss advice given in primary care might actually be doing more harm than good. It's ineffective and potentially reinforces damaging stigma.

To discuss this fascinating issue, I'm joined by Juan Franco, who is the editor of one of our journals, Evidence-Based Medicine. He's also a practicing GP in Germany, as well as being one of our editors on the BMJ. Juan, how do you fit all of that in? It sounds pretty improbable that you could do all of those things. You don't have to answer that now. And Emma Graham from the Center for General Practice at the University of Copenhagen. So welcome to both of you.

Can I start with you, Emma, and just tell us what your paper is, what message is it trying to get across?

Thank you. Our main focus is that there's a pervasive focus on weight right now, especially in the light of the new GLP-1 receptor agonists and new weight loss drugs. And Juan and I originally had the idea that overweight might be a usual suspect of overdiagnosis because of the lowering of thresholds for BMI and overweight disease mongering. Now that

Overweight is considered a chronic disease in ICD-11 and also just the medicalization of large bodies as in need of medical interventions. And we reviewed guidelines and found that most of these guidelines offer lifestyle advice as their mainstay and that the evidence does not really back this as an effective solution to losing weight.

Okay, let's start with one of the statements you made just there. I think you were implying that being overweight isn't necessarily a bad thing. Is that what you're saying? That's correct. Let me correct my statement. Being defined as being overweight. Because I think that's the crux of your argument here, isn't it? Yes, that was how our interest in this topic sparked.

that maybe we are over-medicalizing people with high weight or large bodies and

And that kind of got us into thinking, how do we handle high weight in practice? And we found that the evidence show that you can be healthy and still have a large body and the other way around. Okay, Juan, this all relates to BMI. Could you explain what that is and why you think it's overused in this context?

So BMI is a relationship between height and weight, and it's one of the proxy measures that clinicians usually use to categorize a person as normal weight, overweight, or obese in different categories. We know that BMI as a measure is not perfect because you can have a bodybuilder that has a high BMI and have very low adiposity and not be obese. But it's

in general, it is a proxy measure of adiposity. And adiposity is what defines obesity. That's why our paper says Beyond BMI, because BMI is sort of the usual go-to measure that clinicians use to have conversations with patients about treatment goals. In this case, for example, losing weight with lifestyle changes, which is usually very frustrating for the patient and the clinician as well.

Mm-hmm. So you're saying you measure with somebody's BMI or you assess somebody's BMI and you see they're either overweight or obese and you immediately say right this is the treatment pathway or this is the pathway you now need to be on but you're arguing that that should not necessarily be the case. I mean I can understand it from a sort of on a borderline sort of case but let's say someone comes to you with a BMI of 35, 40, you know surely

I accept there's the patient preference side of it, but surely you have a responsibility as that person's doctor to advise them about their weight. Isn't that the case, Emma? I must make the disclaimer that I'm not a doctor.

But I think what we're trying to argue here is that new evidence points to that weight might not be as an important indicator of health that we have previously assumed. And therefore, patient-centered care, evidence-based care should focus on broader outcomes or broader ranges of health outcomes than simply weight.

And we should also acknowledge the kind of cutbacks of the mainstay of providing lifestyle interventions or lifestyle advice for these patients. Okay, I'm going to come back to you and ask you about the lifestyle interventions. You mentioned about the evidence base for those. But let me put that same question to Juan, who assures me is a doctor. So, Juan, if you're...

Don't deny it now. If you see a patient with a very high BMI, surely it's your duty of care to that individual to advise them about their weight. Yeah, and I completely understand that. And just another disclaimer is that in our analysis, we're not trying to

blame the doctors in any point. We're doing what we can, you know, uh, on the, on the contrary, we're trying to, to, to think together as doctors, uh, from our own frustration of having that conversation with patients where they come with a BMI of 37 and we feel like they're in urge to say, yeah, you need to lose weight. Yes, you need to lose weight. And,

And if we can stop for a moment and think about the story of that person and where do they come from, what does the weight mean for them? Because it is very likely that if you have someone with a BMI of 38 or 37, they spend all their life with doctors telling them to lose weight.

And if you don't acknowledge that, you don't pause that and try to think about the story, you need to start asking, what does weight mean to you? And people may just tell you the story. And you may get patients that say, yes, I do want to lose weight and blah, blah, blah. And you may end up with the decision of losing weight through biotic surgery. But you may also have patients that say, look, I've tried everything my whole life. And for me, the priority now is just...

is that I have a problem in my skin and I need a cream for that. And for me, I don't want to talk about weight right now. And that's also okay. And so I'm not trying to neglect that side of us that we identify. And we put it in a paper that in extreme high,

high weight, there is a risk management situation that was not the main focus of our paper. But we acknowledge that physicians feel that way and pausing and thinking, what is the story of that patient that comes to us is really, really important for a patient-centered care.

Yeah, and I guess you're suggesting that in the situation where somebody who is overweight, which is not that they have a particularly high BMI, but it's not within the designated normal range, it's high, but not in the normal range, that they might be living a perfectly active, fulfilling life and receiving weight loss advice is something exactly they don't need and don't want.

Yeah, and just to add another thing, and we also mentioned in our paper that a patient may come with the desire to lose weight, but it's very important to have a dialogue about what is the meaning of that for the patient. Yeah, how it impacts them. Yeah, in the sense that a lot of patients...

do go to the doctor because they feel an extreme, extreme external pressure to lose weight. And it's not a priority for them. But when they go to school, when they go to work, when they go to the eye doctor, the eye doctor says, but you should lose weight. So they go to the GP and say, oh, the eye doctor told me to lose weight. But it's not a priority for them. They're being told to lose weight. And if we as doctors say, okay,

Okay, the eye doctor said you should need to lose weight. Is that important to you? No, it's not important to me. Okay, let's have a discussion about weight that is not conditioned by what the eye doctor had told you. And again, we're not blaming the poor eye doctor, but it's just saying that people receive a lot of inputs and trying to understand the motivation

and contextualize it and not just assume that their body shape is wrong and they just need to fit in. Okay. Emma, can I come back to the point you made right at the beginning almost now, which is about you looked at the lifestyle interventions and you worked out that the

evidence-based underpinning those isn't particularly sound. Could you explain what you mean by that? Yes. So as we saw that what doctors have as their primary recommendation for people with high weight with lifestyle interventions, we looked into the evidence behind this and found that systematic reviews of RCTs found that on average you can lose 1.8 kilos over the course of two years of

being on lifestyle interventions meaning that isn't very much no I don't think that is very clinically relevant especially if you weigh let's say 100 kilos and you lose maybe 2 kilos that does not make a difference to you

potentially, but it might make a difference to you in terms of internalized stigmatization, might impact you negatively in a psychological way. And we also found evidence that there's actually evidence for physical harms as well, that being on lifestyle interventions, meaning increased physical activity and restricted calorie intake,

might lead to disordered eating, self-harm, even suicidal ideation. So there's both the psychological side of blaming yourself for not being able to lose weight, and then there's the physical harms resulting from that kind of internalized weight stigma.

Okay, so I think finally then what does this mean for BMI? Because it's something that doctors are very familiar with, all health professionals are very familiar with the concept of BMI, but equally patients are, or the public are. I don't think it would be a rare occurrence for somebody to look up their own BMI on the internet and then think about how they might, you know, what they might need to do. So

What does it mean for BMI, do you think, as a measure or as an indicator? Yeah, I think the pervasive focus on being a specific weight, being within a specific bracket of a BMI that is an unreliable measure of health, I think that has medicalising effects on how people view themselves or how they perceive themselves. And as weight is in society,

coupled with ideas about intelligence or even a discipline or people with large bodies are often labelled as either lazy. And that's a kind of certain type of epistemic injustice where people are kind of perceived as less knowledgeable. So I think, you know, focusing strictly on BMI as a diagnosis can have harmful medicalising effects.

Okay, Juan, what's your take on that? I fully agree. I fully agree. And it could be a conversation starter if people want to discuss, because patients know about that and if they use their own like a glitter, but to zoom out as Emma mentioned, I think from that single number and considering all the other factors that affect people's life, age, cardiovascular risk factors, their socio-economical circumstances and comorbidities and so on,

And just their preferences at the end of the day is the most important thing. Okay. So if I were to ask both of you, and you don't have to give the same answer to this, what would be the single message you'd want people to take away from your paper? I think that...

Guidelines currently miss a golden opportunity to support clinicians in reducing weight stigma in the management of obesity or people with larger bodies. And they also miss the opportunity to support doctors and practitioners in providing better evidence-based care and more compassionate care.

Thank you. Juan? I can't do better than that. Sorry. I can't do better than that. I could tell by your face. Okay, listen, thank you both very much for joining us today. Appreciate your time. Thank you. Thank you so much. That paper, Beyond BMI, Rethinking Doctors' Advice for Weight Loss, is also online at bmj.com.

That's it for this episode. Everything we've talked about is now on our website, bmj.com. We'll be back in a fortnight with a look at the future of the NHS. Until then, I'm still Cameron Abassi. Thanks for listening.

Hello, I am Sarah Edwards, the Social Media Editor for the Emergency Medicine Journal. And I'm Rick Boddy, the Deputy Editor, and together we host the EMJ podcast. Every month we do a primary survey of the papers in the latest issue of the journal and we cover exciting topics in the field of emergency care. So if you work in a busy emergency department, the pre-hospital environment, or if you're otherwise interested in emergency care, then this is a great way to keep up to date with the latest advances.

Make sure you're part of our community and subscribe wherever you usually listen so that you never miss an episode. The EMJ podcast is published at the start of every month.

Let's talk photos, not just storing them, showcasing them. You've got images that matter. Whether you're a photographer, a business updating your followers, or just someone who wants to share life's moments the right way. So why hand them over to Big Tech's one-size-fits-all cloud? Big Tech companies are the fast food of photo sharing. Quick, easy, but not exactly gourmet. And what about your data integrity? Jalbum.net is the photo sharing solution that puts you in control.

Want to host images on your own server? You can. Want a layout that actually reflects your brand or style? JAlbum's customizability is unmatched. And if you're a business sharing regular photo updates with your audience, this tool was built with you in mind. But don't just take our word for it.

Over 230 million webpages have been created with JAlbum, and it's got stellar reviews on Trustpilot to prove it. So head to JAlbum.net to download your free software and try it out. When you're ready to upgrade, use the code PODCAST for 20% off. Your photos, your layout, your rules. JAlbum.net.