How do you validate 3D in vitro culture systems?

[Gautam Mahajan]

Spheroids, 3D tissue constructs, and technologies like Organ-Chips are increasingly recognized as valuable models for myriad purposes. But, labs often use bespoke or modified methods to create a model that specifically fits their needs, such as the addition or subtraction of disease-specific cell lines.

The variability between labs seems inevitable, but it does present an interesting challenge to reproducibility of specific experiments, and more broadly to the acceptance of 3D models as bona-fide, valuable culture systems. Harvard’s Don Ingber said in a recent review that a key challenge facing the field is the need to “identify the critical design criteria and performance parameters that must be met to qualify an Organ-Chip model as functional at a level necessary for adoption by pharmaceutical and biotechnology companies as well as regulatory agencies.”

While Ingber is specifically discussing Organ-Chips, his point is clearly relevant to all 3D systems.

In short, there are many different variations on each model type. So, how do we standardize validation across laboratories? How do we prove that each variation is a valid model? Naturally, this means defining what ‘valid’ means. Presumably, this definition would include a list of biomarkers and tasks that the model would have to demonstrate in order to be considered a valid model of the human tissue/disease of interest.

So, my question for you all: How do you define ‘valid’ for your specific model? What criteria do you use to ensure that your in vitro model is a robust representation of in vivo conditions? And, have you seen any efforts to standardize across your community (i.e. shared validation criteria for establishing a multicellular tumor spheroid model of Ewing Sarcoma, for example).

[Christopher Carman]

This is an important issue. In our work to model inflammatory bowel disease, we are looking for signature cytokine and biomarker and permeability responses compared against those that are clinically validated for inflammatory bowel disease (IBD). In addition, we further validate against the ability of our model to capture effects and MOA of established clinically relevant IBD.

[Stephen Riffle]

@Chris — That makes a lot of sense. Is there any benchmark for determining what minimum criteria the model should meet? For example, if the Gut-Chip shows signature cytokine and permeability responses, but is lacking a cell-surface biomarker, would the model be invalidated or just caveated? Not sure if that example makes complete sense, but I’m more diving in on how much similarity with in vivo conditions is enough, and how much dissimilarity is too much?

My guess is that there’s no hard cut off. But do you think there’s the potential to work towards an IBD checklist/grading scale, so to speak, to allow for the same minimal validation criteria to be met across laboratories?

[Ben Swenor]

@Stephen – My perspective would be that the model would then be caveated for that particular marker. My main reasoning being that it isn’t necessarily a flaw of the model, but a donor variability. It certainly could be a deficit of the model, but that is a large component of why having multiple donors for a given model is so critical. It allows us to account for this variability in response and determine whether it is an outlier or commonality. Additionally, for determining the criteria for a model, that is why organizations such as IQ MPS and others like it are so valuable. You have an independent group that sets a standard to be met that addresses the needs of that particular specialty and researchers on 3D in-vitro systems can then work towards them. With further understanding, the guidelines/requirements can be adjusted and improved upon, but it sets a precedent.

[Author]

Posts