# # # #
Recently the results of two large clinical trials in Parkinson’s were announced. Both indicated that the therapies involved had not demonstrated any impact on the progression of Parkinson’s. This disappointing news resulted in the usual headlines (“Epic failure” & “Clinical trial tanks”) from news outlets whose editors have obviously never lost anyone they cared about.
In addition, there has also been a useful chorus of “I told you so” and “We’re going the wrong way” coming from the back seat of the car, despite the fact that we haven’t seen any actual trial data yet, or the fact that they can’t propose any viable alternative approaches.
What is missing in all of this noise, however, is a better approach to failure. Not only an open and honest postmortem of what worked and didn’t work in the studies, but also better, more respectful ways of communicating results.
In today’s post, we will discuss our approach to failure.
# # # #
PART 1. POSTMORTEM
The book I gift the most is Sherwin Nuland’s “How we die”.
As a rule, I am selective as to who I gift this to, never for Christmas or birthdays, and I always remind the receiver of the gift that “you should not judge a book by its cover”.
This book is so precious.
A poetic set of reflections from a medical doctor who has sent his entire career watching ‘life’s final chapter’. There is science, wisdom and beauty on every single page. Nuland has such a wonderful way with words, and I find myself constantly going back to this book and finding something new.
Sherwin Nuland (1930 – 2014). Source: Theparisreview
My favourite part of the entire book is chapter 11.
Throughout the first half of the book, Nuland pushes the argument for returning some dignity to our last days of life. Rather than prolonging suffering in a futile effort to extend life a few short months, he implores the reader to let nature simply take its course.
But all of this changes in chapter 11, where he describes the moment his brother Harvey called him on the phone and told him he had been diagnosed with terminal cancer.
In an instant everything changed. The context had shifted, and instead of “let nature simply take its course”, Nuland recalls how his thinking immediately became “we have to do whatever it takes to keep my brother alive”. (And I’m not ruining the book by sharing this spoiler – there is so much more in this book. It should be required reading for first year medical students).
Matthew Syed. Source: Amazon
It investigates how we approach failure, and the first chapter describes everything that is wrong with how we currently conduct clinical trials.
What does it say?
In the first chapter of his book, Syed makes comparisons between the way the aviation industry and the medical profession approach failure, pointing out the processes that follow situations when a disasters occur.
In the aviation industry, when any event occurs there is a major investigative process that starts with the recovery of the “black boxes”. The aviation industry uses this system of investigation to learn from every single incident. It makes the information available to all and this helps with re-thinking everything from cockpit ergonomics and design to air traffic controller procedures. The airline companies are keen to be seen to be involved in this process of investigation, and “failure”, while unfortunate, is not shameful or stigmatising, but rather embraced and viewed as a tool to improve things.
And this process has resulted in the number of deaths per flight decreasing year-on-year as the number of flights has increased (Source).
Note the downward trend on the orange line (deaths) despite the upward trend on number of flights (grey). Source
Now compare that situation with how the medical profession handles failure, which Syed does in the first few pages of his book describing a horrific real life case study. When medical incidents occur (and unfortunately they do), there has previously been little in the way of investigation or feedback to improve practice – as Syed outlines in his book. In addition, it is very difficult for members of the public to access any of the information. And this is particularly evident in the setting of clinical trials, where even when drugs are successful in the clinical trial process there is very little actual reporting of results beyond the major media headlines and a research journal report (that needs translation for most lay individuals).
And the failure to report clinical trial results has previously put the general public at great risk. For example, in 2012 the pharmaceutical company GlaxoSmithKline paid $3 billion after it failed to disclose trial data demonstrating that its antidepressant Paxil was not only no more effective than placebo during their trials, but that it was also linked to increased suicide attempts among teenagers (Source).
Admittedly there are now efforts and policies being put in place to correct this (such as a major change in the rules regarding the reporting of clinical trial results – Click here to read more about this), but the medical profession is still woefully behind the aviation industry in embracing and investigating failure.
‘But medicine is more complicated’ some will say.
Boeing 747-400. Source: FAIB
This is Boeing 747-400.
It has over 6 million individual parts. It also contains 171 miles (274 km) of wiring and 5 miles (8 km) of tubing (Source).
That sounds pretty complicated to me. And while there are contingencies built in (thanks to a process of failure investigation), it is helpful when each of the 6 million parts does exactly what is it supposed to do, based on years of feedback, refinement and review. Such a process has resulted in better success with regards to passenger safety. And it has also led to improvements to the system of investigation, for example as Syed points out that the “black box” recoding systems are no longer black, they are now pink to help visually locate them better (and they collect a lot more information than they used to as well).
Whether successful or not, one question that must be asked by those conducting clinical trials (everyone: participants, clinicians, sponsors and funders) is why is there no formal postmortem analysis of each trial. We certainly comb through data conducting post-hoc analyses, looking for new insights in the results, but why is the same effort not made with the clinical trial process itself.
The current system of clinical trials is analogous to building a football stadium for just one game and then tearing it down again, before planning another stadium for a different game – without much of a review process regarding how good the previous stadium was.
Building stadiums for a single game of football. Source: WWM
And a review process shouldn’t be looked upon like a witch hunt, looking for things to blame for any bad outcome. Questions should also focus on what worked well and how can we improve on them going forward?
Parkinson’s research advocate Sue Buff recently proposed the idea of a patient community driven forum for evaluating failed trials on Twitter:
This section of the post was titled “Postmortem” – a word that traditionally refers to a situation “done, occurring, or collected after death”. More recently, however, the word has also come to embody “an analysis or discussion of an event after it is over” (Source). I’m not a big fan of Winston Churchill, but he is correct when he said “Those that fail to learn from history are doomed to repeat it”. We need to start digging deeper into the information that clinical trials are providing us – beyond just the results – otherwise we will be having this conversation over and over again.
And I for one am tiring of it. Matthew Syed’s “Black Box Thinking” was published in 2015, and we’re still waiting for the medical research community to embrace any airline incident investigation-like approach.
PART 2 – BETTER COMMUNICATION
In January, I wrote the annual ‘Wish list’ post (Click here to read that), and in it I bemoaned the quality of communication in modern world.
We have all this amazing technology, and massive Government departments dedicated to the task of disseminating information. And yet all too often there is a complete lack of clear communication from parties in positions of authority. And the communication of a clear, prudent message has regularly been trumped by political and economic agendas.
In addition, there is the age old problem of the “this is how we have always done it” approach.
And this last one in particular relates to Parkinson’s research (and how we approach failure).
Last year I (like many others) was disappointed by the way that the negative results from the highly anticipated PASADENA study were disclosed early in the year: There was a single sentence in the sales results of the Pharmaceutical company Roche‘s 1st Quarter report (Click here to read a SoPD post on this).
A multi-year study, 316 participants (and their families) involved, monthly treatments & regular clinical assessment, and a single sentence statement in a quarterly company report announces the result.
Now, to be fair to Roche, later in the year their announcement of post hoc analysis findings – indicating that their treatment may have have had some potential effects in the PASADENA trial (Click here to read more about that) – was handled in a better fashion (a full press release, etc).
And I don’t want to be seen to be picking on just Roche here, they are not alone in this disclosure situation.
Other more recent examples include the pharmaceutical company Biogen who announced the result of their highly anticipated SPARK study (Click here to read a previous SoPD post about this) as a single sentence buried deep in their annual results (PDF).
And just days after that, the biotech company Sanofi announced in a single sentence of their annual report that “the venglustat Phase 2 trial in Parkinson’s with GBA mutations did not meet the primary endpoint (end-January) & the indication was halted” (Click here to read more).
Again, I am not trying to pick on these particular biotech companies, they are simply the most recent Parkinson’s examples. The disclosure of clinical trial results by the pharmaceutical companies is an industry-wide problem. From oncology to diabetes, single sentence statements in the financial reports are a common theme.
And I can appreciate the industry’s point of view. Such information has a very real impact on the well being of these companies (impacting share prices – and thus pension schemes and the company’s ability to raise funds – as well as wider societal implications, such as job losses, etc). There is also the PR aspect. All of these companies want to be seen as funding winning research.
But I hope we can all agree that an approach that employs a single sentence to share important information is an rather insensitive means of informing the patient community about the outcome of a particular study.
The participants (and their families) who have bravely volunteered to take part in this necessary research and dedicated 1 or 2 years of their lives to the endeavour.
They deserve better.
A culture change is required in the pharmaceutical industry. And it would not be a difficult task for some of these enormous companies. Simply coordinating the release of a clinical trial participant information sheet of the top line results to coincide with the release of release of the all important financial disclosure. One email sends a media press release to relevant news organisations and another sends an information sheet to the clinical trial participants. How hard is that?
Perhaps regional Government go-betweens (eg the FDA or EMA) could be commissioned with taking on and coordinating such a role.
A little effort would go a long way though. The pharma industry business model depends on engagement with the patient community, and they should always be seeking methods of nurturing and encouraging that relationship.
Enough of the rant – let’s get back to the research.
All of the material on this website is licensed under a
Creative Commons Attribution 4.0 International License
You can do whatever you like with it!
The banner for today’s post was sourced from Youinc