Workshop

Workshop Training Series

Analytical method validation is critical for drug development, diagnostic test, medical device, and general biomedical research. The evolving guidelines such as those developed by CLSI (Clinical Laboratory Standards Institute) and USP (US Pharmacopeia) have become widely used. Yet to many users these guidelines are still relatively new, and more importantly, there are always some challenges in implementing the guidelines in reality and interpreting the results. 

Stat4ward provides workshop training teaching a variety of topics. In particular, some of the recurrent topics include:

  • Assay validation according to CLSI guidelines 
  • Flow Cytometry assay study design and data analysis
  • Immunogenicity assay validation and cut point determinations
  • A life-cycle approach to process validation (for CMC and LDT operation)
  • Clinical biomarker development- Assay validation and algorithm development
  • Statistics in Non-clinical studies

These workshops take place in conferences such as NextGen Dx Summit, Immuno-Oncology Summit, PharmaED, The Molecular Medicine Tri Conference, and a few workshops organized by Stat4ward.

An Example

Biotherapeutic proteins induce undesired immune responses that can affect drug efficacy and safety. For this reason, immunogenicity assessment is an integral part of drug development and is mandated by the regulatory authorities. One of the statistical tasks is to determine the cut point (for screening, confirmation, titer, or neutralizing assay).The final formulae for the cut points all take the form of CP=Mean +/- k*SD, where k is the critical value of the distribution so that the probability coverage satisfies Pr(X>CP) = α. 

It appears to be so simple since almost everyone knows how to calculate Mean and SD, and the k value can be easily looked up from distribution tables. Then why people spent so much effort discussing such an easy formula?

In addition to validation of all the performance characteristics of an analytical assay (specificity, sensitivity, accuracy, linearity, robustness, etc) which can benefit from statistical design and analysis, most of the cut point calculation methods require the verification of certain statistical assumptions (such as normality, variance homogeneity, and absent of outliers) and calculation of total error (e.g. using mixed models). The concept of confidence/tolerance interval suggested by FDA adds another layer of complexity to this task.

The workshop uses real life data and simulated data to provide training on 

    • Workflow for determining cut points
    • Assay design (e.g. plate layout, sample size)
    • Statistical principals and procedures for outlier detection, data transformation, normality testing, equal variance testing, mixed model for variance components analysis, cut point determination
    • Cut point verification using known samples

Have a Project on mind?

We can help you bring your ideas to life. Let’s talk about what we can build and raise together.