From Cell & Gene Therapy to Biostatistics, a Conversation with Nancy Sajjadi, Independent Quality Consultant

Kaitlyn Barago:
Hello everyone, and welcome to this podcast from Cambridge Health Tech Institute for the Immunogenicity and Bioassay Summit, which is taking place this October 15th to the 18th in Washington D.C. My name is Kaitlyn Barago and I'm the project manager for production initiatives here at CHI. I'm joined by Nancy Sajjadi, an independent quality consultant who has over 30 years of experience in biopharmaceutical product development. Nancy has authored several articles pertaining to bioassays and viral gene therapy, and has served on five advisory panels for the USP. She teaches introductory courses in bioassay design, development, validation and fundamental concepts in biostatistics for non statisticians. Nancy, thank you for joining us today.

Nancy Sajjadi:
Thank you Kaitlyn. Thanks for having me.

Kaitlyn Barago:
Could you please share a little bit about your background and how you came to start your consulting business?

Nancy Sajjadi:
Sure. Well, I'm originally Canadian and I graduated from the University of British Columbia with a Master's in genetic engineering. I moved to San Diego and took a job briefly doing malaria vaccine research for our recombinant vaccine, and then I took a job with a small cell and gene therapy company called Viagene. And I started off there doing building packaging cell lines for MLB based vectors. And then I followed the lead product candidates through development. So I did a lot of assay development and supported preclinical studies in animals, and then moved into developing assays for lot release for the products to go into phase one, phase two trials and also specialized clinical tests to monitor those gene therapy products in patients. And then we were acquired, and at that point I was put in charge of asset development, clinical testing and quality control, and that's basically where I got to in the year 2000.

Kaitlyn Barago:
So you have a history of working with cell and gene therapies then?

Nancy Sajjadi:
Yes, we worked with two different vector systems at least that got far enough along into development, and several different cell therapies that were gene modified.

Kaitlyn Barago:
What do you see as some of the current challenges in developing bioassays for these cell and gene therapies versus other biologic drugs?

Nancy Sajjadi:
Well, from my experience, it's really the complexity of the biology and the complex mechanism of action of the products. Most often what ends up in the vial has many steps before it actually turns into the treatment. So whether it's a vector that has to get in to the body and then find the target cell and then integrate or express the genes to the right level or at the right time, or to be regulated and in response to something, or if it's a stem cell that's gene modified that has to be transplanted and then repopulate an entire lineage and do something specific, or the product has some sort of 3D structure and has multiple functions once it's implanted, that all of that makes the design of the bioassay rather complex and difficult to usually manage in practice. But that's one aspect.

The other aspect would be what I consider regulatory paradigm issues. So if you think about it, the biologics regulation, the six hundreds, really were a response to the vaccine accidents of the 1950s with polio vaccine. And in that case you had these large production lots and materials that you would test for potency. But if you think something like an autologous therapy, cell therapy or gene modified cell therapy, you have patient to patient variation. So you don't really have one product, you have multiple products. You could have extremely short time frames. For example, products now are being generated bedside in the hospital setting.

For other products, there are constraints in sampling them, such as taking a sample will destroy the integrity. So you have to figure out maybe how to set up parallel cultures and what that might look like from a validation standpoint, or the sample size is very limiting, so being able to conduct the type of testing you with a very small sample. And then finally things like products that require animal models that can't easily be scaled for whatever reasons, whether it's the nature of what you're looking at in the animals or the nature of the animals themselves. But I think those are the biggest challenges I've seen.

Kaitlyn Barago
Switching directions a little bit. One of your passions is teaching biostatistics courses for non statisticians, and so why do you think it's important that bench scientists have an understanding of biostatistics?

Nancy Sajjadi
I just want to comment on your first part of your question there, I consider myself as much a motivational speaker for biostatistics as I am as an educator, and the reason is that I just think it's extremely important. I guess the short answer is because the bench scientists need to be able to communicate information about the quality of the data they generate that they report to stakeholders. Quantifying measurement uncertainty definitely is critical in justifying the R&D and QC decisions that ultimately will support products being administered to human beings. So there's a huge responsibility that goes with that, and that's shared between the biologist, the bench scientists and the statisticians and other stakeholders.

Kaitlyn Barago
Can you elaborate on some of the other key areas in bioassay development where scientists need to partner with their statisticians?

Nancy Sajjadi
I think there's so many different places where the statisticians come in. My own recommendation is that scientists should start talking to statisticians as soon as they have gone from concept to lab and they have sort of a basic version of the candidate bioassays up and running, they've done their one factor at a time experiments that they like to do and kind of get a feel for the assay. And then from that point on, the statisticians become key, they are helpful in characterizing the assay, which to me is identifying which factors, in an efficient way like through design of experiment, identify factors that influence the assay performance, help you screen for a factor interactions that occur a lot in biological essays to optimize that essay and figure out how to get the conditions such that it's not sensitive to small changes. And then moving into a qualification stage, which is not only qualifying the assay for its performance, but also figuring out how you're going to qualify critical reagents, whether it's cells or antibodies, or most importantly the reference material.

And then also learning at that point how you're going to qualify your analysts, as you move into clinical studies and you're going to need to have people trained to run the assay. And then you move into what people typically think of as a statistically intensive phase, which is the validation of the assay and monitoring it's performance. You know by that point in time you're really looking at trying to take the clinical data and set specifications for the product and design your validation of your assay to support that.

Kaitlyn Barago
And what are some ways that you can see these collaborations between the bench scientists and the statisticians being strengthened?

Nancy Sajjadi
I would say probably most importantly, it's by expecting that relationship to be growth fostering from the start and really working to make that happen. In my experience, the more each person knows about the entire process of what's going on, the greater the potential for the synergy between them. And to really strengthen any professional collaborations, the real first step is to overcome the language barriers that exist between people who have deep knowledge and usually then lots of jargon associated to their expertise. And for bench scientists and statisticians, in reality, that means that bench scientists really need to learn to understand, think, and speak in statistical terms as well as really learn to accept that statistical approaches are subject to debate among statisticians, and that if we as bench scientists don't really understand the fundamental issues, then we're not going to be able to defend the positions taken, and we all want to come having a united front and really feel confident in what we're doing.

Kaitlyn Barago
Wonderful. Thank you so much for your time and your insights today, Nancy.

Nancy Sajjadi
You're welcome.

Kaitlyn Barago
That was Nancy Sajjadi. She will be instructing a workshop at the upcoming Immunogenicity and Bioassay Summit this October in Washington D.C., on biostatistics for beginners. If you'd like to hear her in person, go to immunogenicitysummit.com/workshop. Thank you for joining us.