Speed to market in drug development is vital to both the bottom line and to improving patient's lives
How Natural Language Processing (NLP) helped Eli Lilly accelerate the design of their trial
The clinical trials stage of drug development is a costly business. Speed to market is vital both to the bottom line and to improving patients’ lives. Technologies that expedite and focus competitor analysis are key enablers, as Eli Lilly and Company’s Principal Research Scientist Eric Su was only too aware.
Charged with finding a scalable solution to extracting summary statistics from oncology and diabetes
trials in clinical trial databases (ClinicalTrials.gov and TrialTrove), Eric was looking for a solution that was agile and scalable, and that could extract facts and synthesize knowledge. Finding such a solution would enable Eli Lilly to better understand the competitive landscape, where to aim future clinical trials and how to best enable its meta-analysis. Ultimately, it would enable Eli Lilly to save time and money, and, potentially, contribute to better patient outcomes.
Eric realized that text mining was the way forward and that a real-time, natural language processing (NLP)-based solution would save time and money, and also reduce laborious manual effort for the clinical trial teams. He chose such a solution: the Linguamatics NLP platform.
The Linguamatics NLP platform is an agile, scalable, real- time NLP-based text mining solution. It is currently used by 18 of the world’s top 20 pharmaceutical and biotech companies during many stages of the drug development pipeline. Various government bodies, including the FDA, and healthcare providers such as Kaiser Permanente, also rely on the NLP platform.
Linguamatics NLP provides data that would take tens or hundreds of times longer with tedious manual work. It enables downstream calculations to provide insight. Some work would not have been done or done comprehensively without [the platform].
— Eric Su, Principal Research Scientist, Eli Lilly and Company
Supporting business and healthcare decisions
Solutions based on keyword searching simply retrieve sets of documents that researchers have to read through. Linguamatics NLP-based text mining solution, however, can identify the key facts, interpret the meaning, and extract and present facts in a structured form for researchers to review, analyze and summarize.
Using the NLP platform’s text mining capabilities would accelerate the Eli Lilly research team’s ability to synthesize knowledge across clinical trial databases, and generate actual insights that would support business and healthcare decisions.
Improving drug trials’ efficiency and effectiveness, increasing speed to market
The Linguamatics NLP platform helps clinical researchers to pull out precise information, numeric and otherwise. For example, inclusion and exclusion biomarker metrics can be standardized, meaning a precise, semantically normalized, structured data output, rather than unstructured text that is hard to research.
Improving the efficiency and effectiveness of drug trial planning in this way reduces the cost of the whole trial, and has the potential to minimize the time before the drug reaches the market.
Speeding up and future-proofing research, reducing errors and cost
Eric describes accessing key clinical endpoints with and without NLP:
“NLP enables us to extract efficacy endpoints in the form of summary statistics. Before [the NLP platform], some of my colleagues in clinical statistics were going into databases and clinical journals and copying and pasting relevant data into Excel. This is intensive, repetitive and very boring for PhD level statisticians. It is also prone to error. Pasting data into the wrong cell as the spreadsheet grows is typical.
“The alternative solution, pre-[NLP], was to outsource the extraction of summary statistics data. This is expensive, you don’t know the quality standard, and, within a few months, the report from the vendor is out of date.”
Enabling improved clinical trial design and informed answers for payers
The NLP platform’s interactive information extraction tools further improved Eric’s results, by increasing recall and precision, integrating key information from different clinical trial sources and ensuring future data could be easily captured through periodic running of the queries.
“A public law that has been in place since 2007 requires companies and academic institutes to post results of all their clinical trials of approved drugs in ClinicalTrials. gov, making this a really amazing treasure trove. Linguamatics NLP enables us to go into this database to extract data from hundreds of thousands of clinical trials and transform data into knowledge for drug development,” concludes Eric.