LOB-vs
Download Lectures on Biostatistics (1971).
Corrected and searchable version of Google books edition

Download review of Lectures on Biostatistics (THES, 1973).

Latest Tweets
Categories
Archives

Clinical trials

Jump to follow-up

Synexus is "The world’s largest multi-national company entirely focused on the recruitment and running of clinical trials company that runs clinical trials and screening programmes".

logo

I should say at the outset that I’m deeply impressed by our local GP practice. I can’t imagine a better GP than mine; he has the ideal mix of knowledge and empathy. I do, however, worry about the fragmentisation of the NHS and its creeping privatisation.

I came across Synexus because my wife had a letter (on our GP practice letterhead) inviting her to go for osteoporosis screening, and possibly to "take part in a study". Download the letter.

Notice that the form gives no idea of what the "study" might be. Notice also, more seriously, the small print on the second page of the form. Here it is in normal size print.

"If you contact Synexus and/or return the attached tear-off slip Synexus may, with your consent, use the data you provide for the purposes of informing you of the study, of medical products and processes that might be of interest to you. Your information will be held by, and access to it limited to, Synexus Ltd and/or companies within the Synexus group of companies and/or third parties acting on their behalf"

You are invited, in near-illegible small print, to allow all your medical data to be handed over to Synexus [see comment, below], and an unspecified number of other companies and third parties. It also gives the company permission to "use the data you provide for the purposes of informing you. . . of medical products and processes that might be of interest to you". This appears to mean that in the future you’ll be pestered with mailings that bypass your GP and advertise (private?) screening etc. For the purposes of screening there should be no need to hand over any data whatsoever (and the practice manager ensures me that they don’t).

My wife asked my advice about whether she should sign up for "the study" if invited to do so, so I asked the GP practice what the trial was about. Rather to my surprise, they didn’t know. Neither did Hertfordshire NHS. So I asked the National Osteoporosis Society, and they didn’t know either. After several emails and a phone call, I eventually got the details from Synexus.

I have two concerns about this. One is the argument that’s been raging about the value of indiscrimate screening, The case against it has been put perfectly in Margaret McCartney’s recent book, The Patient Paradox. There’s a good case that too much money is spent on people who are well, and not enough on those who are ill. Of course prevention is better than cure. The problem is that in many cases the screening tests aren’t accurate enough, so many people get diagnosed and treated when they are not actually ill.

On top of that, there is now a serious worry about screening tests promoted by private companies, for profit. Lifeline has been criticised, for good reasons. The men’s health charity, Movember, promotes PSA screening for prostate cancer, one of the most unreliable tests in existence. There is now a web site that collates evidence about private health screening. Many of the tests are available on the NHS, and the NHS advice about them is being re-written so that it gives information about risks as well as benefits.

The NHS advice on screening for osteoporosis is still ambiguous. The evidence for benefit of screening at age 60 is not clear.

The main question, though, is this. If my wife were offered an opportunity to "take part in a study", should she say yes, or no? My first inclination was to say yes. Clinical trials are the only way to find out whether treatments work or not. If people don’t volunteer for trials, we’ll never know. But before saying yes, one would want to know that the trial was organised properly, so that it could answer a relevant question. That’s why I was surprised when I found it so hard to discover the details. Nobody seemed to know even where the trial was registered. It’s no use searching trial registers for "Synexus": you need to know who is paying for it.

Eventually Dr John Robinson of Synexus turned out to be very helpful. The protocol number is 20070337 with a EudraCT number 2011-001456-11. The trial is registered at ClinicalTrails.gov and it has ethical approval. It’s a trial of a new osteoporosis treatment made by Amgen, AMG 785. It’s a monoclonal antibody against sclerostin, a protein that inhibits bone formation. It sounds like a good idea, but we won’t know how well it works until it’s been tested. The allocation of patients to AMG 785 or placebo is randomised and double blind. The patient Information sheet for participants looks pretty good to me.

Nevertheless, I have some reservations about the trial. First, its organisation is odd. “After taking AMG 785 or placebo for one year, all study participants will be taking denosumab for the following year”. Denusomab is another product of the same company, Amgen. It has already been approved by NICE.  When I asked Dr Robinson why this arrangement had been chosen, this is what he said.

"Previous studies have shown that the maximal benefit on bone density is seen after 12 months and that treatment after this period shows a lower increase, it is for this reason treatment with AMG 785 is for 12 months in this study.

Other studies have also shown benefit in further improving and maintaining the increase in bone density and reducing fracture risk by subsequently treating patients with Alendronate after 12 months of AMG 785. This study is investigating whether similar or better findings occur with denosumab."

This does not make any sense to me. If the object is to compare AMG 785 with denusomab, they should be compared side by side, not sequentially. That brings us straight to the main problem with the trial design. It asks the wrong question. What the doctor needs to know is whether AMG 785 is more effective than existing treatments, not whether it is better than placebo. When I asked Dr Robinson about this, he said

"To quote from the protocol: A placebo-controlled study was chosen because it permits a minimally confounded demonstration of efficacy and safety of AMG 785 in the treatment of PMO. Using an active control such as a bisphosphonate means that more patients have to be enrolled to show benefit from AMG 785. The study already plans to enrol 6000 women. Increasing this number would add to the time required to complete the study. In addition the use of a placebo control is also within regulatory guidelines. "

What this means, in plain English, is that they are expecting a rather small difference between AMG 785 and existing treatments. It would take a very large number of patients to show this difference. If the difference is indeed small, it would be hard to justify the (doubtless eye-watering) cost of AMG 795 (denusomab costs £185.00 per dose). Testing a new drug against placebo, or against a low dose of something not very effective, is one of the stratagems listed in Chapter 4, Bad trials, in Ben Goldacre’s Bad Pharma. It makes the new drug look good, but it asks the wrong question.

The National Osteoporosis Society should be an organisation to which patients could turn to for advice in cases like this. In this case they were not helpful. They didn’t know much about the trial. I hope that this is not related to the fact that they get a lot of funding from Synexus. I noticed too that one of their advisors is the infamous Professor Richard Eastell, who admitted in print to lying in a paper, about a drug for osteoporosis made by Proctor & Gamble. It’s getting quite hard to find a medical charity that isn’t in the pocket of Big Pharma. or quacks (or even occasionally, both).

Conclusion. The trial asks the wrong question. On those grounds alone, I think that my advice would be not to volunteer for the trial.

Follow-up

I should have mentioned an interesting and relevant Cochrane review, New treatments compared to established treatments in randomized trials (2012), The authors’ conclusions are as follows.

“Society can expect that slightly more than half of new experimental treatments will prove to be better than established treatments when tested in RCTs, but few will be substantially better. This is an important finding for patients (as they contemplate participation in RCTs), researchers (as they plan design of the new trials), and funders (as they assess the ’return on investment’).”

15 May 2013. As noted in the comments, Synexus has been censured by the Advertising Standards Authority, because the ASA judged that they did not give sufficient prominence to the fact that there advertising of free screening was actually a way to recruit people into clinical trials.

Jump to follow-up

This is a very important book.

Buy it now (that link is to Waterstone’s Amazon don’t pay tax in the UK, so don’t use them).

When you’ve read it, do something about it. The book has lots of suggestions about what to do.

bg gif

Stolen from badscience.net

 

Peter Medawar, the eminent biologist, in his classic book Advice to a Young Scientist, said this.

 “Exaggerated claims for the efficacy of a medicament are very seldom the consequence of any intention to deceive; they are usually the outcome of a kindly conspiracy in which everybody has the very best intentions. The patient wants to get well, his physician wants to have made him better, and the pharmaceutical company would have liked to have put it into the physician’s power to have made him so. The controlled clinical trial is an attempt to avoid being taken in by this conspiracy of good will.”

There was a lot of truth in that 1979, towards the end of the heyday of small molecule pharmacology.  Since then, one can argue, things have gone downhill.

First, though, think of life without general anaesthetics, local anaesthetics, antibiotics, anticoagulants and many others.  They work well and have done incalculable good.  And they were developed by the drug industry.

But remember also that remarkably little is known about medicine.  There are huge areas in which neither causes nor cures are known.  Treatments for chronic pain, back problems, many sorts of cancer and almost all mental problems are a mess.  It just isn’t known what to do.  Nobody is to blame for this.  Serious medical research has been going on for little more than 60 years, and it turns out to be very complicated.  We are doing our best, but are still ignorant about whole huge areas. That leads to a temptation to make things up. Clutching at straws is very evident when it comes to depression, pain and Alzheimer’s disease, among others.

In order to improve matters, one essential is to do fair tests on treatments that we have.  Ben Goldacre’s book is a superb account of how this could be done, and how the process of testing has been subverted for commercial gain and to satisfy the vanities of academics.

Of course there is nothing new in criticisms of Big Pharma.  The huge fines levied on them for false advertising are well known.  The difference is that Goldacre’s book explains clearly what’s gone wrong in great detail, documents it thoroughly, and makes concrete suggestions for improving matters.

Big Pharma has undoubtedly sometimes behaved appallingly in recent years. Someone should be in jail for crimes against patients.  They have behaved in much the same way that bankers have. In any huge globalised industry it is always possible to blame someone in another department for the dishonesty.  But they aren’t the only people to blame.  None of the problems could have arisen with the complicity of academics, universities, and a plethora of regulatory agencies and professional bodies.

The biggest scandal of all is missing data (chapter 1).  Companies, and sometmes academics, have suppressed of trials that don’t favour the drugs that they are trying to sell.  The antidepressant drug, reboxetine, appeared at first to be good. It had been approved by the Medicines and Healthcare products Regulatory Agency (MHRA) and there was at least one good randomized placebo-controlled trial (RCT) showing it worked.  But it didn’t.  The manufacturer didn’t provide a complete list of unpublished trials when asked for them.  After much work it was found in 2010 that, as well as the published, favourable trial, there were six more trials which had not been published and all six showed reboxetine to be no better than placebo .  In comparisons with other antidepressant drugs three small studies (507 patients) showed reboxetine to be as good as its competitors.  These were published. But it came to light that data on 1657 patients had never been published and these showed reboxetine to be worse than its rivals.

When all the data for the SSRI antidepressants were unearthed (Kirsch et al., 2008) it turned out that they were no better than placebo for mild or moderate depression. This selective suppression of negative data has happened time and time again. It harms patients and deceives doctors, but, incredibly, it’s not illegal.

Disgracefully, Kirsch et al. had to use a Freedom of Information Act request to get the data from the FDA.

“The output of a regulator is often simply a crude, brief summary: almost a ‘yes’ or ‘no’ about side effects. This is the opposite of science, which is only reliable because everyone shows their working, explains how they know that something is effective or safe, shares their methods and their results, and allows others to decide if they agree with the way they processed and analysed the data.”

 

“the NICE document discussing whether it’s a good idea to have Lucentis, an extremely expensive drug, costing well over £ 1,000 per treatment, that is injected into the eye for a condition called acute macular degeneration. As you can see, the NICE document on whether this treatment is a good idea is censored. Not only is the data on the effectiveness of the treatment blanked out by thick black rectangles, in case any doctor or patient should see it, but absurdly, even the names of some trials are missing, preventing the reader from even knowing of their existence, or cross referencing information about them.Most disturbing of all, as you can see in the last bullet point, the data on adverse events is also censored.”

nice

 

The book lists all the tricks that are used by both industry and academics. Here are some of them.

  • Regulatory agencies like the MHRA, the European Medicines Agency (EMA) and the US Food and Drugs Administration (FDA) set a low bar for approval of drugs.
  •  Companies make universities sign gagging agreements which allow unfavourable results to be suppressed, and their existence hidden.
  • Accelerated approval schemes are abused to get quick approval of ineffective drugs and the promised proper tests often don’t materialise
  • Disgracefully, even when all the results have been given to the regulatory agencies (which isn’t always). The MHRA, EMA and FDA don’t make them public. We are expected to take their word.
  • Although all clinical trials are meant to be registered before they start, the EMA register, unbelievably, is not public.  Furthermore there is no check that the results if trials ever get published.  Despite mandates that results must be published within a year of finishing the trial, many aren’t.  Journals promise to check this sort of thing, but they don’t.
  • When the results are published, it is not uncommon for the primary outcome, specified before it started, to have been changed to one that looks like a more favourable result.  Journals are meant to check, but mostly don’t.
  • Companies use scientific conferences, phony journals, make-believe “seed trials” and “continuing medical education” for surreptitious advertising.
  • Companies invent new diseases, plant papers to make you think you’re abnormal, and try to sell you a “cure”.  For example, female sexual dysfunction , restless legs syndrome and social anxiety disorder (i.e. shyness).  This is called disease-mongering, medicalisation or over-diagnosis. It’s bad.
  • Spin is rife. Companies, and authors, want to talk up their results. University PR departments want to exaggerate benefits. Journal editors want sensational papers. Read the results, not the summary. This is universal (but particularly bad in alternative medicine).
  • Companies fund patient groups to lobby for pills even when the pills are known to be ineffective.  The lobby that demanded that Herceptin should be available to all on the breast cancer patients on the NHS was organised by a PR company working for the manufacturer, Roche.  But Herceptin doesn’t work at all in 80% of patients and gives you at best a few extra months of  life in advanced cases.
  • Ghostwriting of papers is serious corruption.  A company writes the paper and senior academics appear as the authors, though they may never have seen the original data.  Even in cases where academics have admitted to lying about whether they have seen the data, they go unpunished by their universities. See for example, the case of Professor Eastell.
  • By encouraging the funding of “continuing medical education” by companies, the great and the good of academic medicine have let us down badly.

This last point is where the book ends, and it’s worth amplification.

“So what have the great and good of British medicine done to help patients, in the face of this endemic corruption, and these systematic flaws? In 2012, a collaborative document was produced by senior figures in medicine from across the board, called ‘Guidance on Collaboration Between Healthcare Professionals and the Pharmaceutical Industry’. This document was jointly approved by the ABPI, the Department of Health, the Royal Colleges of Physicians, Nursing, Psychiatrists, GPs, the Lancet, the British Medical Association, the NHS Confederation, and so on. ”

“It contains no recognition of the serious problems we have seen in this book. In fact, quite the opposite: it makes a series of assertions about them that are factually incorrect.”

“It states that drug reps ‘can be a useful resource for healthcare professionals’. Again, I’m not sure why the Royal Colleges, the BMA, the Department of Health and the NHS Confederation felt the need to reassert this to the doctors of the UK, on behalf of industry, when the evidence shows that drug reps actively distort prescribing practices. But that is the battle you face, trying to get these issues taken seriously by the pinnacle of the medical establishment.”

This is perhaps the most shameful betrayal of all.  The organisations that should protect patients have sold them out.

You may have been sold out by your “elders and betters”, but you can do something. The “What to do” sections of the book should be produced as a set of flash cards, as a reminder that matters can be improved.

It is shameful that this book was not written by a clinical pharmacologist, or a senior doctor, or a Royal College, or a senior academic.  Why has the British Pharmacological Society said nothing?

It is shameful too that this book was not written by one of the quacks who are keen to defend the $60 billion alternative medicine industry (which has cured virtually nothing) and who are strident in their criticism of the 600 billion dollar Pharma industry.  They haven’t done the work that Goldacre has to analyse the real problems.  All they have done is to advocate unfair tests, because that is the only sort their treatments can pass.

It’s weird that medicine, the most caring profession, is more corrupt than any other branch of science.  The reason, needless to say, is money. Well, money and vanity.  The publish or perish mentality of senior academics encourages dishonesty. It is a threat to honest science.

Goldacre’s book shows the consequences: harm to patients and huge wastage of public money.

Read it.

Do something.

 

Follow-up

7 October, 2012, The Observer

Goldacre wrote

"I think it’s really disappointing that nobody, not the Royal Colleges, the Academy of Medical Sciences, the British Pharmacological Society, the British Medical Association, none of these organisations have stood up and said: selective non-publication of unflattering trial data is research misconduct, and if you do it you will be booted out. And I think they really urgently should."

Exactly.