Prof Coles says “They should be written by authors with complete access to the data from the trial, with appropriate expertise and rigorous disposition, who are free from industry influence”.
This is an extremely admirable aim but without seeking to be too critical, perhaps there might be a case of being blind to your own faults here? As we have seen (and subsequently Team G have recently reported from the original trial data) with the phase II and III alemtuzumab trials, surely we would/should have been told about the binding and neutralizing antibody responses, which probably dictated the treatment protocol, to expose the irony that the World’s first humanised antibody (CAMPATH-1H) designed to minimise immunogenicity, is perhaps one of the most immunogenic humanised antibodies:-(.
So perhaps, this paper is a result of personal experience about pharma meddling in clinical trial write-ups. I can see a few familar names on the alemtuzumab trial papers that could feature in the top ten.
However, companies have to be very careful about what is written. There are rules what they can say. This is one of the reasons why trial reports are often so bland and are often written by professional writers aware of these rules.
Such papers have demographics (this really should be supplementary data), efficacy and side-effects.
Anyway it is said
“The integrity of this process is potentially undermined by the fact that authors are chosen by the sponsor, most usually a pharmaceutical company, and that authorship of such papers leads to high citation rates and promotes career development”
Yes but it could be said that if get your papers in certain Journals, work in certain Institutions, have certain mates in influential places then your career can be promoted, there are more ways to do this and being involved in clinical trials is not the only way to do this.
1. “Publications describing phase 3 clinical trials are key to pharmaceutical companies’ interactions with the prescribing community and regulatory bodies. Great rewards follow successful licensing”.
Yep, for example Cambridge University will received millions to create a Chair (Professorship) after the phase III success:-).
2. “There is a commercial incentive for companies to ensure their trials are published as prominently as possible and describe their products to best effect.” “One consequence of these pressures is the phenomenon of “guest authors”: celebrated “key opinion leaders” who do not contribute to trial design or execution, or manuscript drafting, but whose name lends gravitas to the study”.
Oh my God! Have I made Prof G a guest author?
That's marketing I guess and it's where trial reports can stray into over-emphasising the positives whilst under-emphasising any negatives and in some cases squirrelling away negative data that requires Sherlock Holmes style detective skills to uncover with judicious use of freedom of information legislation.
For the record, Prof G was instrumental in the licensing of process for alemtuzumab and Barts was a major recruting site, but anyway. Back to the story.
3. “The International Committee of Medical Journal Editors has attempted to avoid this by strict definitions of authorship; nonetheless, a recent study showed that 13% of industry studies, with collaboration, consisted of guest authors as either first or last authors”.
However, journals get big impact factors if they publish articles that get cited a lot, allowing them to charge companies loads for the reprints and loads for advertising within the journal and high amounts for publication costs.
4 “The current ICMJE criteria for authorship requires substantial contributions to trial design, manuscript drafting, approval of the final manuscript and agreement to be accountable. A significant proportion of authors fail to meet these criteria”.
Authorship has always been an thorny issue for papers and is true whether or not they are phase III trial papers, I would put money that there are thousands of authors that get a relative free-lunch every single day, where they are on publications but didn't do too much except be boss of the lab. However, keeping the lab running is important.
However, If that isn’t bad enough
5 “Interaction between the funder’s commercial goals and the professional rewards of authorship may lead to another phenomenon: a small group of academics, trusted by sponsors, who are repeatedly gifted trial leadership positions. Although such individuals may fulfil the ICMJE authorship criteria, their long-term dependence on further trial commissions might not foster critical independent analysis of trial data”.
So having set out his stall, Prof C now goes for the jugular.
So in this study, the authorship of phase 3 clinical trials of drug treatment of multiple sclerosis (1993-2016) was analysed, to identify if such a group of favoured authors exist. 27 phase 3 trials have been conducted on the clinically isolated syndrome, relapsing-remitting, primary and secondary progressive forms of multiple sclerosis.
“Authors” were defined as those listed as authors on PubMed, or as members of the Writing Committee in the manuscript. These 26 trials have a median of 12 authors (range 5-36), leading to a total listing of 366 authors. 197 individuals are authors of these trials; 145 of these appear as authors on only one paper and 27 people have authored two trials (of which the author is one). Just 25 people (12% of authors) are authors on three or more trials and make up 127 of the 366 (35%) author listings”.
“A select group of 10 people (5% of the total authors) are named as authors on 83 occasions (23%)”.
“At least one member of this group is an author on 22 of the 26 trials (85%)”.
“One prominent expert is an author of 12 of the 26 (46%) trials, taking a leading position (first or last author) on 6 of the 26 (23%) of the trial manuscripts”.
“All of these ten are clinical neurologists; none have particularly rare skillsets”. Ouch!
“It is not possible to tell, from the trial reports, whether these prominent authors were high recruiters to the clinical trials concerned”.
“It seems that a small group of academics, chosen by the sponsoring pharmaceutical companies, dominate authorship of phase 3 clinical trials in multiple sclerosis.”
Surprise Surprise...well not really if you have read them as we see the same names over and over again.
Just like we can see who are on grant panels and then who get the grants.
Likewise, there are names that appear time and time again on academic projects. This may be because they are the best scientists, but could be they are good science politicians.
Is this Prof G again?. As one of a small group of neurologists I must admit I see him on telecoms and we know he goes to plenty of advisory board meeting to advise pharma...so what's his letter in the list?
“There are several positive interpretations of this observation. These select ten authors are likely to be the appropriate experts in multiple sclerosis trial design, execution and analysis”. They may also run large centres capable of high recruitment to such trials.
"However, the disproportionate dominance of these few lead authors raises some reasonable concerns".
"However, leaders of writing committees are in a powerful position to sway choices over the reporting of secondary and tertiary outcomes, and more especially, the writing of subsidiary papers with the results of ad hoc analyses, follow-up extension studies and reports of post-marketing experience".
“In these choices, conflicts may arise between rigorous reporting of scientific results and the commercial interests of the sponsor. There is a risk, or at least the perception of a risk, that clinicians who are more sympathetic to the needs of the sponsor may be repeatedly invited to lead authorship”.
A few horror stories are given that could be a chapter in Ben Goldacre's Bad Pharma, "where appeasing a pharmaceutical company’s position on trial data may conflict with an investigator’s wider responsibilities".
“Another influence of pharmaceutical companies on trial authors may come indirectly from prominent investigators’ commitment to professional associations”.
Some of the select ten multiple sclerosis authors have close connections to organisations like ECTRIMS (European Committee on Treatment and Research in Multiple Sclerosis) and ACTRIMS (Americas Committee for Treatment and Research in Multiple Sclerosis) which draws significant income from pharmaceutical companies.
“For instance, ECTRIMS receives 37% of its income directly from industry......It is, after all....a trade show rather a forum for novel ideas. Isn't it?....." with significant indirect income from sponsoring of attendees at conferences.
Company sponsorship has been reduced due to the introduction of new rules.
Is this why ECTRIMS is a day shorter this year?
"It would be naïve to imagine that this does not influence the relationship between author and sponsor”.
However, there is a realisation
“Without pharmaceutical companies, drugs would not be developed and patients would suffer. They are essential to medicine. But the commercial instinct to market products may lead to choices about academic publications that need to be challenged”
“For this, informed and independent investigators are critical",
"yet this is potentially undermined by the situation in which a small group of people are repeatedly asked to lead writing committees by sponsoring companies”.
“There are no easy solutions. Both regulatory authorities and journal editors have the power over sponsors to modify authorship behaviour”.
“If trial publications did not list individual authors, potentially harmful behaviours, designed to promote repeated invitations to lead authorship, would be minimised; but this would also eliminate transparency.”
If you applied this to all publications, you wouldn't have the Research Assessment Exercise/Research Excellence framework in the UK. This costs tens of millions to say Oxford & Cambridge are ace:-(
I wonder also if certain names were removed from papers...would we still have the dogmas that can be a hindrance to effective treatment?
These papers would get rejected and the Lemming effect, where scientist all follow the same idea and sometimes say the wrong thing, would not occur.
“Journals could consider undertaking their own analyses and scrutiny of clinical trial data”
This is an interesting one, however it means that you would be in the hands of the journal and their biases. It takes months to go through data. We got our hands on the CARE-MS trial data and focused our reporting on potential safety issues...not mentioned in the original trial reports.
“Structures could be created for investigators to report concerns about biases in trial reporting, free from reprisal".
Surely this could be done now anyway, if you are an author and you are not happy with content, you can address issues during the writing process. If you are not satisfied you can remove yourself from the paper, you can write to editors once submitted and you can write reviews to set the record straight.
However, I doubt anything would be free of consequences if unwelcome reports are made. Many people provide paid service to the companies and if the companies don't like the service they get they will go elsewhere. It;s simple economics. Likewise, if they like the advice they comeback for more. I guess the ten speak words of wisdom.
"Above all, awareness needs to be raised about this particular way in which trial reporting may be influenced”.
So the question is who on the list?..........
Maybe, I am being too presumptive about suggesting that ProfG could be a Trial Junkie.
Maybe he'll smack my wrists for the suggestion:-( , but science is the freedom to say things and then to expect people to challenge them.
What do you think?
Are you shocked?
Authorship is a tool to promote recruitment, where the first x number of recruiters get to be on the paper and the rest go into the Trial Investigators list. This makes people recruit quickly if they want to be on the paper.
Is this the beginning of the end of professional trialist?
I doubt it.
Does Prof C get a knighthood or has he really put the cat among the pigeons with his colleagues?
What else will get revealed?
Is this issue just an MS issue?.......I doubt it.
More on the same subject
Vera-Badill FM et al. Honorary and ghost authorship in reports of randomised clinical trials in oncology. Eur J Oncol 2016.66:1-8
There is a rejoinder to Prof Coles's article in the same journal (which will be the subject of a post here) that attempts to rebut the concerns made plain in his piece which perhaps shows that this article has really ruffled a few feathers. Well done to Prof Coles for that and for raising the subject it's certainly got people talking and it's nice to see people putting their heads above the parapet, even if they risk having their heads shot off.
What do you think?
- Honorary and ghost authorship are prevalent.
- Medical writers are employed commonly to assist with reporting of commercially sponsored clinical trials.
- We identified honorary authorship in 33% of the trials analysed, compared with 9–19% reported previously.
- Assistance of medical writers was only identified in reports of commercially sponsored trials.
- A statistician should always be included as an author to ensure accountability on the statistical analysis.