Opening up drug discovery

Collaboration is vital in modern drug discovery – but how to be transparent yet retain competitiveness? And more importantly, how can the academic-industrial relationship be maintained with integrity? Professor Jackie Hunter tells us about several initiatives set to improve public perception.

It is my natural tendency to believe that transparency in general is a good thing so long as it doesn’t hinder competitiveness or expose people to risk of harm. When I joined the pharmaceutical industry in the 1980s, there was a real sensitivity in sharing any information since the competition between companies was fierce and information was seen as a competitive commodity. Publications were not encouraged and data around pipelines and portfolios a closely guarded secret. ‘Negative’ clinical data was not shared either and this secrecy contributed to the negative view of the pharmaceutical industry that still pervades today.

This is unfortunate because one of the reasons I joined the industry was because I thought discovering and developing new medicines was a laudable and worthwhile endeavour, especially as I myself used medicines for my asthma on a daily basis. So where has the situation improved across the R&D process in the past two decades and where are improvements still necessary? There have been a number of pressures on the industry aside from lack of trust. First the escalating costs of developing new medicines – today it is estimated that this is in the order of $1-$1.5B per drug. This incorporates the cost of failure and ultimately the cost of failure has risen so greatly that it is no longer sustainable. In addition, there have been increasing regulatory demands and uncertainties around reimbursement and pricing. These pressures have led to the adoption of a more open innovation mind set within the industry that has seen much more pre-competitive collaboration and data sharing especially in the early stages of drug discovery and development.

The IMI is a good example of how transparency and collaboration can be of benefit for industry, academia and the wider healthcare landscape

Many of these collaborations have involved academic partners. The Innovative Medicines Initiative (IMI) is the largest example of a public-private partnership between the pharmaceutical industry and academic and other partners, co-funded by the European Commission and EFPIA (the European Federation of Pharmaceutical Industry Associations). This has really helped to drive better understanding of the challenges the industry faces as well as data sharing between companies, academia and other groups such as regulators and patient advocacy groups. With the funding of IMI2 under Horizon 2020, it is anticipated that this collaboration will move further down the drug discovery and development process. The IMI is a good example of how transparency and collaboration can be of benefit for industry, academia and the wider healthcare landscape. In other areas progress has also been made. Many companies now publish their clinical trial results on specific websites. In the early days it was actually quite hard to find these websites from the companies’ main websites but in most cases this is much more transparent now. Clearly companies do not publish all their data especially where drugs are still in development as revealing the results could give away important competitive information. Where patient-level data is made available, it is also important that effective safeguards are in place to protect patient privacy and confidentiality.

[caption id="attachment_55301" align="alignnone" width="620"] In the past, clinical trials were often not published but recently there is a greater transparency around the results arising from trials.[/caption]

On the other hand, not all companies make this data readily available and even for those that do, the amount of data released is not always consistent from one study to another. Few companies post their results alongside the clinical description of the trial on the www.clinicaltrials.gov website. Using our in-house analytics at Stratified Medical, we estimated that approximately 13% of the ~150,000 trials posted in the last 10 years had results posted.  A more comprehensive and consistent approach by the whole industry would be welcome. To support transparency, the All Trials initiative has called for all clinical trials to be registered and results to be made publicly available in a timely fashion. One of the main reasons for this is to avoid unnecessary duplication – for example, many companies have taken vanilloid receptor 1 (TRPV1) antagonists, a family of compounds with potential pain reduction properties, into the clinic since 2000 – none of which appear to have worked since they have not progressed. Of the 23 Phase II studies on the clinicaltrials.gov website only 3 had results – and these were posted some years after study completion by one company. Some of the study results were available on company websites. Had the early study results been more widely available, then it is likely that some of the later studies would not have been started.

Publication of ‘negative data’ in scientific journals is more problematic since most journals are not interested in ‘negative’ data or request additional experiments or analysis which the company personnel are usually unable to do as they would have been moved on to other projects. The British Journal of Pharmacology and the British Journal of Clinical Pharmacology do have a mechanism whereby ‘negative’ studies can be published in a short communication with a light touch review. The only stipulation is that sufficient data should be included to show that the failure in the study was due to a lack of importance of the mechanism in the disease rather than a failure of the molecule to affect that mechanism in vivo.

One area where I think academia could learn from industry in terms of transparency and process is in the design and reporting of preclinical experimental studies, especially those using animals

Several companies have provided clinical study data via an EFPIA sponsored portal (www.clinicalstudydatarequest.com) and requests to access the data for further research purposes can be made via this portal. The phase of development where companies place their data does vary; some putting early trials on as well as those for marketed compounds whereas others only do this for marketed compounds. It is important that companies do take the time and effort to report their clinical failures for the reasons stated above.  It will also provide further proof that drug discovery and development at the moment is a highly risky business. The failure rate is such that many people in the industry never work on anything that makes it to the market. It is also important to share the data on the failures of molecules in toxicology and carcinogenicity studies pre-clinically since the precompetitive value of sharing this data to enable better predictions means that such studies have value even though the compounds have stopped progressing. There are a number of such initiatives but there is room for even greater sharing of this information.

One area where I think academia could learn from industry in terms of transparency and process is in the design and reporting of preclinical experimental studies, especially those using animals. Recently the ‘reproducibility’ problem has been the subject of much attention and several reports. The Academy of Medical Sciences, BBSRC, MRC and Wellcome held a symposium in 2015 to explore these concerns – and I was pleased to speak at that meeting. Greater openness and transparency around methods and data, including publication of ‘negative’ results, were identified as important steps in improving the reproducibility and reliability of biomedical research. Academia and industry will need to work together, alongside others, to improve research reproducibility. In my experience of preclinical research in large pharmaceutical companies, the design and conduct of studies is extremely rigorous and well documented. I think there are best practises that can be adopted here by academia – many of which are already documented but not always followed.

[caption id="attachment_55302" align="alignnone" width="620"] A variety of websites now exist that have a collection of clinical trials, and it is hoped more will show negative results - as this can reduce duplication of experiments.[/caption]

In the past there has been lack of transparency between the pharmaceutical industry and the academic scientists and clinicians that the industry collaborate with – especially on the commercial side. Unsurprisingly this lack of transparency has raised concerns in the press and society. These concerns include the fact that funding from industry could bias the reporting of results or bias comments and reviews from academics or clinicians of other studies. They could lead to at best a lack of objectivity and at worst withholding of information. The advent of Disclosure UK , an initiative by the Association of the British Pharmaceutical Industry (ABPI), is therefore welcome. This is part of a European-wide initiative designed to increase transparency about the relationships between industry, and healthcare professionals and organisations. It is based on disclosure of 'transfers of value': payments and other benefits in kind made as part of these partnerships. This is an important step but is no substitute for transparency in declaring interests, that could lead to conflicts depending on the circumstances, in other fora – such as on review panels, presentations etc.

Certainly within the Research Councils in the UK, interests and potential conflicts concerning relationships with industry as well as other academic affiliations have to be reported for members of panels or Governing Bodies. The Academy of Medical Sciences is exploring how to best manage interests, including potential conflicts, as part of its workstream on ‘How can we all best use evidence to judge the potential benefits and harms of medicines?’ and recently held a workshop on this matter. Greater transparency around industry-academia relationships was identified as an important first step in addressing interests, with further steps required to ensure they are effectively managed. Improving the health and wellness of the population is one of the major challenges of the 21st century. To tackle this challenge as rapidly and efficiently as possible will need collaboration between a huge range of players including industry, healthcare professionals, regulators and academia. There have been many benefits to such collaboration in the past and perhaps these need to be highlighted more often. However, there is still the perception in some quarters that collaborating with industry is undesirable and creates bias – anything that can be done by industry and academia to improve transparency will be a positive step in dispelling this perception.

Useful Sites www.alltrials.net www.clinicalstudydatarequest.com www.abpi.org.uk

Author

Professor Jackie Hunter FMedSci, is CEO at Stratified Medical.