Can open science help patients and save pharma? | Opensource.com

Can open science help patients and save pharma?

Posted 19 Jun 2014 by 

Rating: 
(9 votes)
Image by : 

opensource.com

submit to reddit

Open science research and development hybrid development model can protect pharma company profits while reducing costs of medicines for consumers


View the complete collection of Open Science Week articles

The current model for pharmaceutical development is time-consuming, expensive, and inefficient: developing a new pharmaceutical therapy costs on average more than $1 billion and takes 12-15 years to go from from lab concept to approved drug on the pharmacy shelf. Furthermore, the majority of that $1 billion cost goes towards the recovery of research and development (R&D) costs for drugs that fail to get approval—the profits from each approved drug must cover the costs of all the drugs that failed. And, contrary to what you might expect, research has gotten less efficient over the last 60 years, despite innovations in clinical research: the number of drugs approved annually has remained relatively static, while the financial resources required for R&D have soared at a rate well beyond inflation. And the high cost of R&D contributes to the high cost or prescriptions: projections estimate that by 2016, global spending on pharmaceutical development will exceed $1.2 trillion annually,3 placing a burden on patients and overall global health resources.4 Meanwhile, pharma companies are under increasing pressure to reduce the price of drugs, thanks to the combination of generic drug competition and the increasing reluctance of insurance companies to reimburse for expensive new therapies unless they are superior to less expensive alternatives. The downward pressure being placed on the cost of drugs to consumers plus the upward spiral in development costs means that in order to stay competitive, pharma companies must find cost savings.

One potential source of cost reduction is to improve efficiency in pharma R&D while protecting patient safety and research quality. But how to create those savings?

Approach one: Fully open source model

One model proposed to improve R&D efficiency and quality is to make the process completely transparent and collaborative so that researchers—even those from competing pharmaceutical companies can freely share information on their research designs, processes, and outcomes. Akin to the process of open source software development, researchers would have full access to all data on a potential molecule or compound, including patented information such as chemical structure and manufacturing techniques. For champions of transparency, this sounds like a wonderful idea. So, why not adopt open source pharma R&D immediately?

First, for-profit pharma is unlikely to give away their patents and trade secrets, at least for more common (and profitable) diseases, thus this open source model has only been successful under a limited range of circumstances: For rarer diseases where there tends to be less commercial interest (e.g. neglected tropical diseases (NTD), “orphan” diseases). When researchers pursuing open source drug discovery into full clinical development are volunteers or perhaps supported by grants or government funding. A great example of open source R&D for malaria is the recent Opensource.com article by Alice Williamson. True open source therefore is likely relevant to only a small number of compounds targeted for rare diseases and will only benefit a relatively small number of people from a global population health perspective—albeit among the most underserved. Second, in 2012, the top 10 pharma companies alone reinvested ~$70 billion of their profits into R&D. If pharma cannot protect its profits, the most likely result would be a more than $70 billion dollar reduction in research funding an enormous sum not easily recovered from foundations or government sources.

Approach two: Open science, a hybrid development model

A hybrid approach to greater transparency and collaboration shows promise for pharma and, more importantly, patients. Called by some "open science" R&D, the hybrid approach proposes that the "source"—the molecule and the manufacturing processes remain protected. The pharma developer would still own the drug and only they would know how to make it; however, trade secrets and "know-how information that pharma cannot patent but they attempt to keep secret would be shared.

In this scenario, developers could freely share: study protocols and  data analysis techniques

  • results
  • communications with regulatory agencies (such as FDA, EMA, etc.)
  • interactions with payers such as insurance companies or national health plans that typically pay for therapies

Much of this information is currently shared, but in an incredibly inefficient, "under-the-covers" fashion via what is arguably industrial espionage, but in reality is better characterized as researchers loosely sharing information even though they have signed confidentiality agreements not to. So, open science simply proposes a more organized and efficient exchange of this information to drive a more efficient R&D process overall.

Why would increased transparency and collaboration help reduce drug costs to patients? Recall that each approved drug costs, on average, over $1 billion to develop, but a great deal of that investment goes to recover costs for drugs that fail to get approval. The process to "kill" a less promising candidate drug can often take more time than it should because research teams are committed to their projects and want them to succeed —creating pressure to  extend clinical trials beyond the stage that’s warranted based on the data alone. With more eyes looking at the data critically, it’s more likely that poorer candidates for further development would be weeded out earlier in the process, saving time and money (more eyes means fewer bugs).

An open science R&D model, while not completely open source, allows the sort of data-sharing that currently only occurs with rare diseases, thereby improving overall R&D efficiency. It could also protect the margins for pharma companies (which realistically must happen in order to gain pharma support). More importantly, if pharma passes on savings to patients, public health would benefit from reduced pharmaceutical costs.

Is open science R&D feasible? Interviews with decision makers

If senior leaders are not open to open science, it doesn’t matter how much an open science model could improve the R&D process. Therefore, to explore whether open science could be an acceptable alternative to current pharmaceutical R&D practices that keep "competing" scientists in the dark, I interviewed senior leaders from academia, industry, and regulatory agencies, including C-suite level executives from top-5 pharma and contract research organizations (CROs). These interviews were confidential to encourage candor. Prior to starting the interviews, I assumed that academics and regulators generally would support the concept, and industry leaders would not.

When asked about the efficiency and costs of the current R&D process, most decision makers recognized there was substantial room for improvement:

The clinical side of it keeps getting longer. Well, not so much longer, but costlier and with poorer success rates. That is a big concern. (a senior academician)

R&D is very slow" and the" costs are ungodly. (a Vice President at a large pharma company)

It’s terrible because it is so costly and [pharma has] such poor success rates – the predictability of their models is so bad. (a senior regulator of the FDA)

It is worth noting that both pharma executives (88%) and academics/regulators (83%) opined that open science could have a positive impact on speeding up R&D and reducing costs; however, some concern was expressed around information overload or "analysis paralysis":

If you put five companies together, instead of getting one wise entity you simply have five entities coming together and still muddling through. (a CEO of a small pharma company)

Either [open science] could be refreshingly revelatory and encourage people to be hyper vigilant about the quality of the work that they do or it could have exactly the opposite effect and all work would essentially grind to a halt because [pharma] would be afraid of exposing a vulnerability. (a CMO of a large CRO)

As originally assumed, regulators and academics were very positive in terms of open science and efficiency:

So I think in process innovation, [open science] can be very valuable… I think it could be significant… I think there is a lot that could be done to speed up the process and also to make it more targeted… if you could decrease cost by 20%, that is a couple hundred million dollars." (a senior academician)

[Open science] is definitely beneficial. There are currently a number of areas where investments by companies are duplicative, even if they are each aiming for somewhat different molecules. (a senior regulator)

But, surprisingly, there was more support (87%) than concern among the pharma leaders.

In response to the concern of information overload: you always have a choice about what pieces of information you want to spend a lot of time analyzing and pursuing. I would rather be given the choice of looking at as much information I chose to look at, rather than being in a position where I was not allowed to look at some information that might be helpful. (a CEO of a large pharma company)

And in response to the challenge that perhaps open science only makes sense for rarer diseases, this same CEO shot back:

It is illogical to me to say we believe that the [open science] model is right for orphan or niche diseases [but not] right for bigger diseases. We are seeing with these orphan diseases data that improves the outcome in terms of approval times, time to market, and patient benefit. [Therefore], I find it illogical to say that the benefits [of open science] should not be extended to broader populations. (a CEO of a large pharma company)

This was bolstered by two of the CEOs interviewed:

I think that, if I were a dictator of the world, I would probably give a try or at least analyze the [modified open science] model that we just talked about. (a CEO of a small pharma company)

I think there is openness to it now that five years ago frankly would not have been there. (a CEO of another small pharma company)

So, while the results showed that the senior leaders were concerned that for-profit pharmaceutical companies would not voluntarily embrace open science or perhaps be overwhelmed by additional data, the results also revealed that:

  1. Open science should be more efficient, and therefore better, in terms of R&D costs,
  2. although not widely known, many open science-type activities are already in place ( e.g. TransCelerate, DNDi, CEO LSC, iSAEC, OMOP, etc . ),
  3. even more transparency is probably inevitable (think WikiLeaks), and
  4. senior leaders, including Pharma Execs, are open to exploring opport unities for broad transparency and collaboration such as those envisioned in open science.

These study results support that transparency and collaboration such as that envisioned by open science would be positive for: 1) R&D efficiency and costs, 2) science, 3) patients as individuals, and 4) population health as a whole. This finding perhaps is not remarkable. However, open science could also be positive for the pharma industry itself in terms of the bottom line. This has the potential to have a revolutionary impact on the way that drugs are researched, reported on, and approved, with the possibility of both maintaining pharma profitability while reducing the costs of medicines to everyone, everywhere.

It isn’t a question IF change is happening, it is what will pharma do in light of it. We are committed to open innovation in clinical research. We look forward to enabling and collaborating with others to accelerate clinical research and greater meet patient needs.” —Thomas Krohn, head of Eli Lilly’s Open Innovation Unit


1 Morgan S, Grootendorst P, Lexchin J, Cunningham C, Greyson D. The cost of drug development: A systematic review. Health Policy. 2011 Apr;100(1):4-17. Epub 2011 Jan 21.

2 Munos B. Lessons from 60 years of pharmaceutical innovation. Nature Reviews in Drug Discovery, 2009, 8, 959-968.

3 The global use of medicines: Outlook through 2016. IMS Institute for Healthcare Informatics. July 2012.

4 Snow W. FDA Strangling Consumer Health. From http://www.supplementquality.com/news/skyrocketing_drug_costs.html , accessed 5 November 2011.

5 Munos B. Can open-source R&D reinvigorate drug research? Nature Reviews Drug Discovery AOP, published online 18 August 2006; doi:10.1038/nrd2131.

6 Ortı´ L, Carbajo RJ, Pieper U, Eswar N, Maurer SM, et al. (2009) A Kernel for Open Source Drug Discovery in Tropical Diseases. PLoS Negl Trop Dis 3(4): e418. doi:10.1371/journal.pntd.0000418.

7 Guth RA. Glaxo tries a Linux approach. Wall Street Journal. 26 May 2010.

8  Biopharma's Top R&D Spenders - 2012. FierceBiotech 18Mar13. http://www.fiercebiotech.com/special-reports/biopharmas-top-rd-spenders-2012 accessed 10 Jun13.

View the complete collection of stories for Open Science Week.

submit to reddit

4 Comments

lintqueen
Open Minded

Neat research! Do you think that if pharmaceutical companies are more transparent about their work it will increase consumer trust? Right now it feels to me like a very adversarial relationship, and it doesn't seem like it has to be...

Vote up!
4
Vote down!
0
Timothy King
Open Enthusiast

And the news keep getting worse for pharma, whose reputation puts them right down at the bottom with the US Congress and Big Oil (see this post from just today -- http://www.fiercepharma.com/story/can-pharma-salvage-its-tarnished-rep-s...). As for my research, several of the execs I've interviewed spoke to this point -- here is a quote from a CEO: “[open science] can help pharma from a public relations perspective... to show that pharma cares about patients, not just profits.”

Vote up!
1
Vote down!
0
Matthew Todd

Thanks for the interesting article, Timothy. I work with the Open Source Malaria team you mention in the article. You say that such an approach has the capacity to benefit only a small number of people, but I'd say the exact opposite. Diseases such as malaria and TB affect millions of people, and would never be classified as "rare". The extension of the open principle to other disease areas is something we'll be talking about at a forthcoming meeting: http://intermolecular.wordpress.com/2014/05/13/open-source-pharma-and-pr...

I wanted to ask about the specifics of what you refer to as "open science". To my mind that term implies a sharing of data - actual results. That is the transformative impact of openness - for example one company revealing the shortcomings of a certain molecule series, saving others the expense of repeating the work. Without the specific data at that granularity, what's your vision for where the efficiency savings would come from? Are you talking about protocols and best practice?

Vote up!
0
Vote down!
0
Timothy King
Open Enthusiast

Matthew -- thanks for the great malaria article and for your questions too. I agree that malaria and TB are not orphans -- they're better classified as NTD, as you know. I hope the comment about open-source being "only" geared for orphan/NTD didn't seem pejorative. I am a proponent of openness and collaboration and am proposing ways to get all Pharma on board. I am working towards an "and" approach here -- more openness for both orphan/NTD therapies AND, well, every therapy.

To the specifics of your question about what gets shared -- I recommend that Pharma consider sharing everything but the "source" -- the patents and CMC (basically so they still "own" the therapy and only they know how to make it). That is why I refer to this as open science rather than open source. Everything else would be shared, in particular the data on an ongoing basis (or once it is unblinded for blinded studies). So, just as you say, researchers would have a better understanding of potential drug class effects, and a better understanding of what the regulators expect.

Finally, on a related point, I wonder whether the greatest power of open science is not in making it (much) faster or cheaper to get therapies approved, but instead to make it easier and quicker to cull out therapies that are unlikely to have a good efficacy/safety profile.

Vote up!
0
Vote down!
0

Comment now

tim king: painting by kimberly wheaton

Dr. King, who has been in the field of global medical research since 1990, is passionate about finding ways to promote open innovation, unite for-profit and non-for-profit health research, and ultimately break down health disparities and encourage health for all. He has built and led operational groups of up to 225 people and ~$300M research budgets in a variety of therapeutic areas including hematology/oncology, infectious diseases, cardiopulmonary, and internal medicine.

Holiday gift guide promo