Welcome to Part 2 of Novo Nordisk’s masterful marketing push to turn their weight loss drugs into cardiovascular life savers. You can read Part 1 here.
|
|
We’ll start with an overview of the clinical trial at the centre of Novo’s PR stunt, the ‘Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity Trial.‘ The acronym is SECOPOOT, but they chose to call it SELECT.
A SELECTive Sample
SELECT was a 5 year, multi-country study including 17600 subjects. The largest group was from the USA (3652), followed by Russia (1491) and then South Africa (797). About 30% of the countries involved were predominantly non-white, including Japan, India, and Thailand. 327 Aussies were in the mix. No information is provided to describe how people were recruited, but hundreds of small groups (often < 10) were drawn from places such as medical centres, hospitals, research units, universities, and individual physicians.
Participants had to be older than 45, not diabetic, have a BMI over 27, and have cardiovascular disease – either a previous heart attack, stroke or symptomatic peripheral arterial disease. About ⅔ had insulin resistance. Unlike the weight loss trials for Wegovy, which recruited mainly women (74%), SELECT subjects were mostly men (72%), and overwhelmingly white (84%) – quite a feat given that one third of the countries involved were non-white.
Novo ended up with a highly SELECT group of older (average age was 61) white men with insulin resistance and cardiovascular problems – a carefully stacked deck with which to provide Novo speedy, favourable results. Such pre-selection of subjects is simply not OK, especially considering the phenomenal amount of money and limitless access to patients this company has. Modern science has an ethical responsibility to ensure that medications intended for huge markets should be tested on a diverse range of humans – not simply the ones who’ll give the pharmaceutical company a quick return on investment.
SELECT was earmarked to run for 5 years. Recruitment took place between October 2018 and March 2021 – before it was approved for use as a weight loss drug by the FDA or in Australia. Everyone in the experiment was being treated ‘off-label’, but this isn’t mentioned in the paper.
Half the subjects received weekly placebo injections and half received Wegovy. Everyone then sat back, waiting for the heart attacks and strokes to happen. And Novo were betting that Wegovy would reduce the risk.
Shareholder Science
Let’s be frank: Novo didn’t run SELECT because they care about health. They did it to show the regulators tangible health benefits beyond temporary weight loss, bolstering their case for subsidisation of their expensive drugs. Hoping for a strong result, Novo’s shareholders expected SELECT to stop before the 5 year mark. But in a 2022 meeting, Novo revealed that after ‘interim analysis,’ their independent data monitoring committee (IDMC) recommended that SELECT should continue for the full 5 years*.
This decision was a clue that Wegovy wasn’t having a massive impact on cardiovascular risk reduction and that Novo needed to buy more time. This was not good news for the money people, and several investors quizzed Novo’s board, pressing them to reveal exactly what they knew about the SELECT data. The execs denied specific knowledge, claiming that only the IDMC had seen the raw data. This seems like a bit of a stretch given that Novo own the database and employ the statisticians!
SELECT Outcomes
As we saw in Part 1, the main outcome was a 1.5% absolute risk reduction in Major Adverse Cardiac Events (MACE) between the placebo and Wegovy groups. Novo’s strategic decision to put out a press release claiming 20% risk reduction without simultaneously releasing the data was diabolical. Because three months later, when the research was published, that ‘key talking point’ of a 20% risk reduction had been repeated so many times that it was both embedded and yesterday’s news. By then, no-one was interested to hear about the real, much less impressive, data.
Apart from small MACE reductions, no other outcomes in SELECT were scientifically significant. Although touted as an electrifying breakthrough, the SELECT trial was little more than an overhyped advertorial.
|
|
Where’s The Weight Loss?
The SELECT trial rested on the assumption that weight loss is the major driver behind cardiovascular risk reduction. In a 2020 paper outlining its rationale and trial design, the authors specify that:
‘sustained, effective weight loss may have independent cardiovascular benefit.’ – (emphasis added)
Given the apparent centrality of weight loss in their hypothesis, it’s weird that in the published outcome paper, weight loss isn’t mentioned in the abstract, the accompanying ‘video explainer‘, or the research summary infographic.
In fact, weight loss warranted just 3 sentences squished into the results section. One lone statistic – average weight loss % at 2 years – was provided: a very odd choice given that the SELECT trial ran for 4 ¾ years.
At 2 years the Wegovy group lost an average of 9.39% and the placebo group lost 0.88%. Buried on page 36 of the Supplementary index is this:
 |
In typical Novo Nordisk vague-booking, there’s no corresponding list of actual numbers, so we’re left to rely on visuals. But we can see that weight loss bottomed out at about the 1 year mark, and then bumped around somewhere in the range of 8-9%ish. In the Wegovy weight loss trials, average weight loss at week 68 was 14.9%. SELECT weight loss was much less impressive.
By the four year mark (week 221), weight data was available for a mere 157 people. Novo claimed a 97% completion rate in the Wegovy group (which started with 8800), so the lack of weighed-in people at 4 years means most people in SELECT were recruited later (ie, closer to 2021 than 2018). And because the trial was stopped in mid 2023, 4 year data for the majority of participants weren’t available, and now that the study has ended, this will never be collected. Novo’s decision to stop the trial as soon as they’d reached a shareholder-friendly statistical end point, rather than running it for as long as possible in order to collect data on long-term weight loss and its apparent link to heart health, is scientifically and ethically questionable.
A bit more information appeared in a different paper on long-term weight loss in the SELECT trial. Once again Novo’s statisticians have gone out of their way to render simple weight loss information almost indecipherable, but the take home message is this:
The vast majority of the Wegovy group in the SELECT trial only lost a small amount of weight. 2 years in, one third (32.2%) lost less than 5% of their starting weight, and more than half (55%) lost less than 10%.
Side Effects
About one third of the Wegovy group experienced ‘serious adverse events.’ The rate of discontinuation was twice as high in the Wegovy group compared to placebo – 16.6% of Wegovy users had to stop taking it, mostly because of gastrointestinal disorders (10%) or gallbladder problems (2.8%). Nasty side effects – mostly gastrointestinal – saw more than 1 in 5 Wegovy users unable to tolerate the full 2.4mg dose.
Perhaps reflecting how unpleasant living with chronic gastrointestinal distress can be, in the real world, discontinuation rates of semaglutide are extremely high. According to USA prescription data, at 2 years, 75% of Wegovy users had discontinued it. Once people stop, weight regain is inevitable – around 2/3 of weight lost comes back within a year -even with continued dieting.
In a report questioning Novo’s cardiovascular claims, The Danish Health Authority pointed out that 40% of weight loss in the original Wegovy trial was muscle mass, contrasting this with the 11-25% ‘typically seen’ in chronic dieting. Muscle mass loss is of course bad news for everyone, but particularly in older populations such as the one in SELECT.
If we add in the exorbitant cost of semaglutide, and consider the detrimental long-term impact of weight loss on metabolism, none of which are discussed in the SELECT paper, there’s a raft of downsides, and we must consider whether the cure is worse than the condition.
SELECT Is Scientifically Useless
There’s a laundry list of problems with Novo and the SELECT trial, but there’s one issue so fundamental that the entire study should go straight into the bin.
In a scientific study, confounding occurs when something other than the assumed active ingredient is the real agent of change. In SELECT, the confounding issue is that we already know that semaglutide at lower doses in non-diabetic people reduces cardiovascular risk.
The modest SELECT risk reduction seen at the 2.4 mg dose could have occurred at the diabetes (0.5-1 mg) dose. But because SELECT didn’t control for dose -i.e., there was no third semaglutide group who received a lower dose – it’s impossible to conclude that the improvement was definitely due to the weight loss dose.
|
|
The fact that Novo consulted with an academic steering committee, had the study design approved by countless ethics boards, made it through peer review and then successfully published, in spite of being structured so poorly that its methodology would have failed a high school science project goes to show how incredibly low the standard of ‘obesity research’ is.
SELECT’S Dirty Little Secret
No triumphant press release accompanied a subsequent analysis of the SELECT data which finally examined the relationship between weight loss and cardiovascular risk. The paper appeared as a ‘late breaking abstract’ (no time for the press to catch on?!) in a 5 minute presentation during the morning coffee break at the European Congress on Obesity (ECO) in Venice, Italy, in May 2024.
|
|
This hasty overview is in stark contrast to the 90 minute Novo sponsored presentation at Obesity Week, where participants enjoyed a free dinner and a gamified obesity escape room, while a panel of Novo-sponsored obesity experts ‘educated’ them on SELECT – an astounding event given that the trial results hadn’t even been published. The experts were literally speculating about Novo’s press release.
Also in stark contract to Novo’s usual lightening pace when bringing positive data to publication, this late-breaking study still hasn’t been published, so it’s impossible to analyse the findings in any detail.
But the abstract shows that cardiovascular risk reductions in the Wegovy group were independent of weight loss. Those who lost more than 5% on Wegovy had exactly the same reduction in MACE as those who lost less than 5% – or even gained weight.
It cannot be over emphasised that this is extremely important information. In SELECT, weight loss had nothing to do with cardiovascular risk reduction – it was all because of the drug itself.
Vital scientific information being buried in 5 minute presentations is absolutely outrageous. Believe me I’ll be on the look out for the paper, and once it’s published, will shout it from the rooftops!
$ELECTive Science
It’s clear now why Novo named this trial SELECT. Using SELECT groups of paid-off researchers and committees, they SELECTed a stacked deck of subjects to ensure a quick result, SELECTed terrible statistics to communicate risk reduction, SELECTed shareholder price-friendly sound bytes to release to the media, and SELECTed out all of the unfavourable bits. It’s peak $ELECTive science, conducted in the name of profits.
The weight loss industry has a long history of hiding the facts about the relationship between weight and health. It’s been interesting to see them confess that in spite of diet culture’s widespread belief that weight loss reduces the risk of heart problems, actual data on this is lacking. The SELECT co-authors even admit that:
‘‘treatments that produce effective, durable weight loss and the impact of weight reduction in reducing cardiovascular risk have been elusive.’
And now the SELECT trial, which researchers assumed would prove their weight-centric views, has shown the opposite.
If we look purely at the data, this is a wonderful opportunity for weight-inclusive care. If semaglutide can help, even a little, to reduce risk of a secondary cardiovascular event, then this could apply (possibly at lower doses) to people of all sizes, without any weight loss requirement. It’s truly ironic that SELECT – a trial funded by the weight loss industry – has provided evidence supporting the weight inclusive approach.
Part 3 of Risky Business is coming very soon – for early access, subscribe to my newsletter – by downloading my free E-Book ‘Everything You’ve Been Told About Weight Loss is Bullish*t’, you’ll be added to the newsletter list. Click here to download and join!
*What’s an IDMC? In large clinical trials, IDMC’s are set up to monitor participant safety and trial integrity by monitoring data, assessing adverse events, performing interim analyses, and making recommendations about trial continuation, modification, or termination. The SELECT trial included 7 IDMC members, all of whom have received large and repeated payments from Novo. Dr Christopher Granger, for example, received a total of almost $US100 000 from Novo, with one third of that for ‘travel and lodging”. This seems incredibly luxurious for ‘safety monitoring’ and this is clearly not an independent committee.
If you like to listen as well as read on this topic, don’t miss my All Fired Up podcast episode on this with the fabulous Ragen Chastain. Listen here !
