The Cytel blog keeps you up to speed with the latest developments in biostatistics and clinical biometrics.
The Head of the DIA’s Adaptive Design Working Group Asks Us to Consider 5 ‘Soft-Skills’ All Effective Statisticians Should Cultivate
The advent of adaptive designs has meant that statisticians have a new role to play in the drug development process. Not only are they responsible for tackling the precise statistical issues that can arise during the course of a study, but their knowledge and vision can result in innovations that substantially alter study design, strategic decision-making, probability of trial success and level of revenue.
Given this new role played by statisticians, what skills do statisticians now need to cultivate to be as effective as possible in the boardroom and on their project teams?
As we head into the new year, the life-sciences industry can look forward to some long awaited advances in pharmaceutical drug development. Here are our predictions for top trends to look out for in 2015.
1: The Arrival of Adaptive Financing:
Adaptive financing was an idea that was cultivated and nurtured by members and affiliates of Cytel. It is the idea that a financial strategy can be built around outcomes that are observed at certain decision points during clinical development (for example, at interim analyses.) Constructing designs with adaptive financing in mind helps provide potential investors and study sponsors with a monetary ballpark of how much innovation will cost, and how much they have the potential to save. This helps overcome one of the biggest bottlenecks in implementing safer and more efficient adaptive trials, namely the perception by many in the industry that these trials are risky.
Richard Branson once wrote: “I have always valued capability over expertise. While you may need to hire specialists for some positions, take a close look at people who have thrived in different industries and jobs – they are usually more versatile, have transferable skills, and can potentially tackle problems creatively.”
He goes on to write: “Obviously a healthy mix of experience and novel thinking is the ideal, but on balance I would anticipate more fresh and objective solutions to flow from the smart and curious inexpert outsider than the ‘been there done that’ experts.”
Although the versatility to which Branson alludes is instrumental for successful statistical consulting, it is also obvious that an inexpert outsider would not be able to waltz into the drug development industry and make successful contributions. The expertise that is required is simply too much for a non-specialist. This raises an important question:
How should we, as an industry, walk this fine line between specialized expertise and versatile capabilities?
Consider, for example, the list of seven questions provided below.
During a recent DIA webinar on reinventing the clinical trial, Laurie Halloran (President of the Halloran Consulting Group) and Irving Dark (Senior Vice President at Cytel) weighed in on a wide array of recent technological breakthroughs that life sciences companies can leverage to simplify clinical development. Their discussion ranged from simple techniques to cut down the 80,000 sheets of paper used for the average clinical trial (“Take it to the Cloud!”) to remote centralized monitoring techniques that make use of cutting edge statistical innovations.
A key stage of exploratory drug development is implementing a proof-of-concept study to demonstrate the safety of a drug. Given the importance of accurate dose-finding for Phase 3 success, methodological improvements to proof-of-concept studies in Phase 2 can translate into greater likelihood of getting a drug to market.
A few weeks ago, we posted a synopsis of an event held at ISCB Vienna in which statisticians from Cytel, SAS and Stata spoke about processes for developing software for statisticians. The button below will guide you to a copy of the presentation by Cytel's Vice President of Consulting, Yannis Jemiai.
In the presentation, Yannis talks about the decision-making process behind software development, and how it aligns with a general philosophy to empower, simplify and educate. He explains how software developers choose from a variety of statistical algorithms in putting together a classic software package. "Just because we can," he cautions, "doesn't mean that we should."
As more clinical trials make use of adaptive designs, investors have come to realize that high quality trial designs can result in significant improvements to a trial’s financial risk profile. Regardless of a trial’s eventual success or failure, a well-constructed design provides a drug with the highest possible probability of success while mitigating financial risk.
Statisticians on Software Development Part I: Statisticians from Cytel, SAS and Stata talk Software Development
During an invited speakers session at the lnternational Society for Clinical Biostatistics, Cytel VP Yannis Jemiai was joined by R.N. Rodriguez from the SAS Institute and IR White (formerly of Stata), to discuss innovations in software for clinical trials.
Beyond Wild Horses: Developing Innovation at Cytel