Non Clinical Statistic
Statistics for (Bio)Analytical methods
The objective of an analytical method is to be able to determine as accurately as possible each of the future unknown quantity that the laboratory will have to quantify.
The objective of validation is to give to the laboratory as well as to the regulatory bodies guarantees that every single result that will be performed in routine will be close enough from the unknown « true value » of the sample.
Pay attention to the result, not the method : a bias smaller than 15% combined with a precision smaller than 15% do not guarantee that your results will be within [-15%,15%] of the true value. The quality of the results produced by analytical methods is the focus since the results are the very reason (intent of use) of an analytical method.
Arlenda proposes to validate your analytical method on the basis of the error concept and tolerance intervals approach. The same principles are used to assess the robustness of your analytical methods or to confirm that your analytical method can be transferred.
Statistics for (Bio)Analytical methods
Softwares are already available for validation of physico-chemical methods, immunoassays, transfers of methods. Please, visit the softwares page to know more about them.
Arlenda can provide you advices to design your validation/robustness/transfer/stability studies, interpret the results or improve your analytical method validation/transfer process.
Quality by Design for processes and methods
In the report "Pharmaceutical cGMPs for the 21st Century: A Risk-Based Approach", FDA states that companies should begin "the implementation of robust manufacturing processes that reliably produce pharmaceuticals of high quality and that accommodate process change to support continuous process improvement". Quality by Design principle is the full understanding of your process and of all the characteristics that influence that process, rather than just testing the resulting product at the end, hoping that your process has performed as expected.
Thanks to its combined knowledge in pharmacy, chemistry, design of experiments and Bayesian statistics, Arlenda may contribute to the practical implementation of the Quality by Design concept in your organization, by developing robust optimized processes and methods.
Design Space for process and methods
ICH Q8 proposes to use the Design Space (DS) risk-based methodology to provide assurance of quality: "Understand and gain knowledge about a process/method to find a parametric region of reliable robustness for future performance of this process/method ".
Arlenda rephrases as: "Understand and gain knowledge about a process/method to find the set of conditions such that the predictive probability of being in the specifications is greater than a specified minimal quality level or risk". The set of conditions are your operating conditions, the specifications are the acceptance limits and the quality level is the maximal risk of being outside, that you tolerate.
Are you aware of your risk? Arlenda can compute it by estimating the predictive probability that your results will fall within your specifications in the future.
Bayesian methodology is perfectly suited to determine Design Space in that perspective, since it allows to identify the predictive distributions of responses. Once these distributions are available, predicting future performance given pas experiments becomes straightforward.
Biomarker validation and use
Biomarkers may have different purpose: it can serve for early detection screening, for diagnosis or prognosis, risk prediction, treatment selection,..
Whatever the objective, Arlenda can help you in the design, analysis or report of your biomarker study.