A dedicated SAS Programming Group working in a pharmaceutical Modeling & Simulation organization - Current role, experience and prospects
Vincent Buchheit, Marie-Odile Lemarechal, Caterina Ferioli, Aurelie Gautier, Gregory Pinault, Kathy Chen*, Goonaseelan (Colin) Pillai, Jean-Louis Steimer
M&S, Novartis, Basel Switzerland, * East Hanover USA
Objectives: The objective is to raise awareness of the benefits and advantages of having SAS programmers working in Modeling and Simulation (M&S). We would like to share our experience and vision, to solicit feedback and to encourage collaboration with other similar groups.
Methods: Modeling and Simulation as implemented within Novartis consists of 4 subgroups: Biology, Pharmacology, Statistics and the SAS Programming group. Programmers work in close collaboration with the 3 other modeling subgroups.
Results: The current primary task of the SAS Programming group is to liaise with the modelers in order to organize the transfer of data. Due to their knowledge of the different company databases (Pre-clinical, Clinical, Watson Lims, Legacy, Biomarkers) they currently extract data from these databases, to harmonize them and often to pool them into a suitable format for further analysis in NONMEM, Matlab, S-Plus, SAS, Berkeley Madonna, or other relevant software. This activity is done in compliance with health authority regulations and in-house pharmaceutical standard operating practices, in a validated audit trail environment.
The role of the programming scientists within M&S (M&S/Prog) is currently expanding. For some projects, M&S/Prog conducts exploratory statistical analysis, produces graphics using SAS or S-Plus, codes some literature models or conducts some modeling work in collaboration with an experienced modeler.
M&S/Prog also increases the skills of the modelers with respect to data manipulation by providing internal courses (e.g. in SAS) targeted to meet modelers needs of compliance with internal processes. They also train modelers in the use of the company’s data repository systems.
To meet the goal of M&S integration into all aspects of research and development, it is vital that hardware and software match the needs of M&S. A major IT project is currently on-going, with M&S/Prog as a business leader on the project. M&S/Prog also tracks software validation and facilitates software installation.
In 2006, M&S/Prog was involved in nearly 60% of the projects carried out within M&S, with a ratio of 1 programmer for 6 modelers. The primary focus was on late phase drug development projects, due to the higher complexity on the data generation and programming effort required for regulatory submission work. Conclusion: An integrated programming group within M&S increases the efficiency of the overall M&S organisation while ensuring reproducibility, quality and compliance with relevant regulatory guidelines or processes .
Conclusions: An integrated programming group within M&S increases the efficiency of the overall M&S organisation while ensuring reproducibility, quality and compliance with relevant regulatory guidelines or processes.
