Bioanalytical Community

Hello Bioanalytical community, 

When developing and validating flow cytometry panels, or other assays, for clinical trials for which the matrix is hard to obtain in sufficient quantity for the needs, what approach do you favor? Do you perform initial experiments with a surrogate matrix or a model system and then use the clinical matrix in a proof-of-concept experiment?  Would you use the same surrogate sample to validate the panel? For example, using PBMC to develop a bone-marrow aspirate panel.

Wow, that is a pretty clear example of something you really cannot acquire - would be unethical to source such a painful collection just for assay characterization. In cases like that, I think the surrogate matrix in development and qualification/validation is the best you can do. What you might want to plan is some additional in-study validation experiments such as incurred sample reproducibility, parallelism, etc.
Flow assays can be very difficult to source QCs for anyway, so do the best you can achieve practically and good luck! Reaching out for regulator feedback on an analytical plan might also be wise. There are some new meeting types for the FDA that may fit this better than the legacy meeting types.

Hello Bioanalytical community, 

When developing and validating flow cytometry panels, or other assays, for clinical trials for which the matrix is hard to obtain in sufficient quantity for the needs, what approach do you favor? Do you perform initial experiments with a surrogate matrix or a model system and then use the clinical matrix in a proof-of-concept experiment?  Would you use the same surrogate sample to validate the panel? For example, using PBMC to develop a bone-marrow aspirate panel.

Thanks for the question and great timing- the manuscript below just came out in Cytometry part B. Carefully chosen surrogates are sometimes the only available option- between healthy donor samples, cell lines, stimulated cells, or entirely synthetic hydrogels, there are some new options to allow an assay to be developed and validated to understand how far from "normal" a sample would need to be for detection, however patient variability is always a wildcard.

When in-study validation is needed, the two biggest takeaways for me are to clearly justify and document any in-study work planned ahead of time and to use multi-factorial experimental design to minimize the amount of sample needed in these cases. As Joleen mentioned, the more regulator and quality feedback on the plan ahead of time the better.

Reference- validations in rare matrices

 Implementation of flow cytometry testing on rare matrix samples: Special considerations and best practices when the sample is unique or difficult to obtain - Devitt - Cytometry Part B: Clinical Cytometry - Wiley Online Library

Reference- in-study validation

Best practices for the development, analytical validation and clinical implementation of flow cytometric methods for chimeric antigen receptor T cell analyses - Sarikonda - 2021 - Cytometry Part B: Clinical Cytometry - Wiley Online Library

Wow, that is a pretty clear example of something you really cannot acquire - would be unethical to source such a painful collection just for assay characterization. In cases like that, I think the surrogate matrix in development and qualification/validation is the best you can do. What you might want to plan is some additional in-study validation experiments such as incurred sample reproducibility, parallelism, etc.
Flow assays can be very difficult to source QCs for anyway, so do the best you can achieve practically and good luck! Reaching out for regulator feedback on an analytical plan might also be wise. There are some new meeting types for the FDA that may fit this better than the legacy meeting types.

Hello Bioanalytical community, 

When developing and validating flow cytometry panels, or other assays, for clinical trials for which the matrix is hard to obtain in sufficient quantity for the needs, what approach do you favor? Do you perform initial experiments with a surrogate matrix or a model system and then use the clinical matrix in a proof-of-concept experiment?  Would you use the same surrogate sample to validate the panel? For example, using PBMC to develop a bone-marrow aspirate panel.