Informatics


References 1 https://web.archive.org/ web/20100709180030/http:// www.cancer.ucla.edu/Index.asp x?page=36. 2 https://www.forbes.com/ sites/davidshaywitz/2017/07/26/ the-startling-history-behind- mercks-new-cancer- blockbuster/#23f9e2fe948d. 3 https://labiotech.eu/ interviews/interview-keytruda- cancer-inventors/?_sm_au_= iVVJQMNvk44sTJ8FN4s4kKH FLKVG2. 4 Bzdok, D, Altman, N and Krzywinski, M. Statistics versus machine learning. Nature Methods volume15, pages233- 234 (2018). 5 Stolzenbach, J. The strive to incorporate machine learning into clinical development. Clinical Trials Arena. https://www.clinicaltrialsarena.c om/digital-disruption/the- strive-to-incorporate-machine- learning-into-clinical- development-6218372-2/. Published June 25, 2018. Accessed May 23, 2019. 6 Helfand, C. If pharma looks slow to adopt AI, it’s got good reason, expert says. Fierce Pharma. https://www.fierce pharma.com/marketing/if- pharma-looks-slow-to-adopt- ai-there-s-good-reason-expert. Published May 1, 2019. Accessed May 23, 2019. 7 Clinical Research News. Are we ready for AI in clinical trials? https://www.clinical informaticsnews.com/2018/12/ 14/are-we-ready-for-ai-in- clinical-trials.aspx. 8 https://www.clinical leader.com/doc/machine- learning-in-clinical-trials-what- will-the-future-hold-and-what- s-holding-us-back-0001. 9 https://www.science daily.com/releases/2017/11/171 115091819.htm. 10 https://prahs.com/blog/ 2018/03/23/artificial- intelligence-in-clinical-trials/. 11 https://searchenterprise ai.techtarget.com/feature/Desig ning-and-building-artificial- intelligence-infrastructure. 12 Estimation of clinical trial success rates and related parameters, Biostatistics 2018, DOI: 10.1093/biostatistics/ kxx069.


12


intended end-users and convincing end-users in the value of changing traditional data collection tech- niques10. It is vital to clearly and consistently com- municate and deliver an aligned direction and strat- egy, high impact training and clear incentives and benefits. This will help assure that key stakeholders are on board, ready to ask the right questions and committed to making the most of ML’s benefits.


Appreciating the limits of ML Treating any new technology as ‘the key’ cog with- in the whole research wheel can cause unforeseen issues. ML applications are no different. Scientific testing relies on a lot of inference where mathemat- ical models are used to test a hypothesis and chal- lenge how the system behaves, and lack of an explicit model can make it difficult to directly relate ML solutions to existing biological knowl- edge4. Although we should leverage our scientific and analytical expertise to move into areas of AI and ML with confidence – we should also be mind- ful of its true impact and not consider it a panacea for all analytics needs. Treating it as a stand-alone forecasting tool and ignoring all other information from previously established best practices may become counter-productive in the long run. Instead, we should utilise ML applications as an additional and highly-promising aid that will help us better leverage and make sense of the bigger data without compromising on scientific value.


The way forward Scientific and technological revolutions are well under way and their intersection is poised to give us an unprecedented opportunity to extract more and deeper biological insights from clinical trial data to drive new conclusions, approaches and cures. We need to harness these revolutions in the most


optimal way – leveraging their strong potential but also addressing, head on, the challenges they bring. The latest estimate for the percentage of drug


development programmes that make it from Phase I testing to approval is 13.8% overall and, at 3.4%, oncology drugs have the lowest success rate12. Therefore, we need not only to reduce the time


to market and cost of new therapies, but we also need to rescue the failing clinical pipeline by embracing innovation. There has been a huge shift in how clinical trials


are beginning to leverage a patient’s molecular pro- file to bring therapeutics into the market that are highly successful for targeted patient populations. Furthermore, the industry is increasingly aware that they must better predict clinical trial success and avoid potential risks. The abilities to meet


these demands hinge on how to optimally leverage the historical data sitting in corporate vaults as well as public databases. As an industry, we need to be able to address the


various critical points in the data life cycle, from data collection and access, to data analysis with stakeholder buy-in to truly harness the value of our scientific data investments. Risks to patient safety and privacy inevitably


make Pharma more circumspect when it comes to the application of new technologies. But the indus- try also recognises the huge potential of these tech- nologies such as ML applications. This promise is realised, as we know, to the point that the FDA is developing a new regulatory framework intended to support the use of ML in clinical trials, drug development and regulatory approval8. Furthermore, this is not a one-way street where


technologies are informing the data, but the vast amount of disparate and complex data in Pharma is also pushing these technologies to become better and more productive in the long run. All of this helps ensure that scientific and techno-


logical revolutions are not just running in parallel but can truly engage in a symbiotic evolution. DDW


David Wang is General Manager of Informatics at PerkinElmer. He brings expertise on the transfor- mative value of informatics in delivering smarter decisions and scientific breakthroughs, especially in life sciences. David has held leadership positions at Medtronic/Covidien, J&J and McKinsey. He holds an MBA/Bachelors from the University of Chicago and is completing a Masters in Bioinformatics at Harvard.


Masha Hoffey is Director of Clinical & Translational Solutions at PerkinElmer. She leads the development and productisation of solutions that help translational researchers and clinical study teams derive actionable insights from their data. Masha brings more than 10 years of experi- ence in data analytics, regulatory affairs and prod- uct portfolio development.


Dr Simone Sharma is PerkinElmer’s Strategic Lead in Translational Analytics and focuses on driving product direction for data access, integration and advanced analytics of translational research data. She brings a deep expertise in and holds a PhD from University College London in Molecular Genomics.


Drug Discovery World Fall 2019


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68