We are pleased to announce our Special Presentations.
by Qian Zhao, Harry Chen and Sumesh Kalappurakal, Statistical Programmers and Analysts at Janssen Pharmaceutica
Traditionally the process for programming ADaM datasets is cumbersome and relies heavily on manual work. Per regulatory requirements clinical programing algorithms should be clearly defined in the analysis specification documents in natural language(human-readable). Programmers spend most of the time developing or updating SAS® code according to specification documents. By adopting Machine Learning and leveraging the power of NLP we could analyze human-readable text from the specification documents, train the machine to convert defined algorithms to metadata and map them to the core pieces of SAS® code.
This paper is part 2 of Metadata-based Auto-Programming Process1, and it shares an approach to automatically generate SAS® code to create ADaM datasets from source SDTM datasets via metadata and NLP methodology. The strategy would be to extract key information from defined algorithms written in human language and existing code, populate metadata and then utilize the metadata to generate code.
by Jiameng Yuan, Statistical Programmer at HighThink Med
One Stop Solution for Bioequivalence Analysis Sample size in BE：what if Highly Variable Drug Average BE (HVABE) or Narrow Therapeutic Index Drug Average BE (NTIABE)? Is WinNonlin the only choice? Is data standardization still bother us in BE? Please look forward to our One Stop Solution for Bioequivalence Analysis.
by Wayne Zhong, Statistical Programmer at Accretion Softworks
Integration and analysis of data across all studies in a submission is a vital part of applications for regulatory approval in the pharma industry. The existing ADaM classes (ADSL, BDS, and OCCDS) already support some simple cases of integration analysis. However, there has been a need for an integration standard that supports the more complex cases. To address this need, the ADaM Integration sub-team is developing the upcoming ADaM Integration standards document. This paper introduces the new IADSL, IBDS, and IOCCDS classes found in this document. IADSL allows for multiple records per subject. IBDS and IOCCDS work effectively with the new IADSL class. This paper also discusses the analysis needs that necessitated the creation of the new classes, and provides examples in the form of usage scenarios, data, and metadata. With them, no future integration will prove too complex. PharmaSUG 2018 Best Paper Winner.
by Wei Dong, Customer Advisory Team Manager at SAS
You might have heard about Artificial Intelligence everywhere these days, but it’s a field of science that has been evolving for sixty years. During this section, I will have a brief summary on AI and its applications in life sciences industry. For SAS, AI naturally a part of advanced analytics. I will introduce how SAS help customers gain tangible value on couple of area such as Predictive Diagnostics, Biomedical Imaging, Health Monitor from their analytics using AI. Lessons learned and best practice advices will also be present in the last part.
by Ruquan You, Biostatistics Group Head at Novartis
Secukinumab is a fully human monoclonal antibody that selectively neutralizes IL-17A, a cornerstone cytokine involved in the development of psoriatic disease. Traditionally, clinical endpoints for psoriasis (Pso) and psoriatic arthritis (PsA) are composite scores meant to summarize in single numbers complex sets of patients’ sign, symptoms, and measures of physical and mental well-being. Here we discuss how to visualize and better assess the outcomes in two key clinical manifestations of psoriatic disease, namely Pso and PsA, their time course, and their treatment response to secukinumab, using innovative visualization mythologies that seek retain as much of the original data captured as possible, yet nonetheless presenting it in an intuitive and informative way.