NEW YORK – Developers said last week that they have achieved a new milestone in their collaborative pilot effort — spearheaded by the nonprofit Medical Device Innovation Consortium — to develop well-characterized reference samples using gene-edited cell lines that can be used by diagnostic developers to validate or harmonize next-generation sequencing tests.
Following a meeting at the MDIC's annual public forum, Maryellen de Mars, MDIC's program director of clinical diagnostics, said in an interview that the consortium's Somatic Reference Samples Initiative recently confirmed that it has successfully engineered the 10 target reference lines and is ready to start its next steps to confirm its results and establish use cases.
The SRS Initiative, formed initially in 2019, involves a variety of stakeholders, including the US Food and Drug Administration, the National Institutes of Health, diagnostic manufacturers, and payors. It officially launched in 2022, initially in partnership with Horizon Discovery, part of Revvity, then called PerkinElmer, aiming to engineer a cell line using CRISPR to feature a set of different cancer-associated genetic variants.
"The idea with the pilot project is that we have identified 10 variants that are highly associated with cancer, but they represent different types of mutations like multi-nucleotide variants, SNPs, fusion proteins, rearrangements, and indels," de Mars said.
"We're trying to get a good representation of different types of mutations, different technical difficulties, different doses, [to have] a repertory to represent as much as we could," she said.
According to de Mars, a challenge for tissue-based sequencing diagnostics is obtaining appropriate validation samples. "Usually, what people are doing is relying on extracting from a discarded sample [or other] one-offs. … People are relying on samples that they can get their hands on," she said. "What we're providing [is] a resource that has been highly characterized … and then there's going to be datasets that go with that, that provide data analysis and integration that makes these tools very valuable."
"We have 10 cell lines that each have a single variant, and then we're going to be combining them in mixtures so that you have different genomes with variants at different allele frequencies," de Mars added.
Revvity began the project's CRISPR work in 2022, but there were some challenges. "The good news with the cell line we are using is that it was very highly characterized. The bad news was that it doesn't have the best behavior in culture, but I'm happy to report that we have now, as confirmed by Sanger sequencing, all 10 variants engineered," said de Mars.
The group is now in the process of sending samples out for whole-genome sequencing to reconfirm with a broader view of the 10 variant genomes. The team should have those results within the next three weeks or so.
In the meantime, bioinformaticians from MDIC and the National Institute of Standards and Technology are working on pipelines for sequencing analysis, de Mars added. "We're making great progress on that so that when we get the data back, we will be ready to do the analysis and look at the actual sequence of these variants in these cell lines to see if we got exactly what we intended."
"And if we didn't get what we intended, we'll be able to see if the changes are still acceptable and still fit for purpose," she said.
If everything goes to plan technically, the group will then pivot to defining use cases.
The consortium's central aim is to supply tools for test validation, but the group believes that the reference lines could also support aspects of regulatory review and reimbursement decisions.
"All along the product lifecycle these could be tools to look at performance of these platforms," de Mars said.
The SRS Initiative hopes to find a sponsor or laboratory interested in validating a specific NGS diagnostic to pilot the samples. It is also working on recruiting for an inter-lab validation study, not as a test of the laboratories themselves, but rather of the samples.
"We have to think about this a little bit because right now we just have very small quantities that are being sequenced from the initial batches, but the idea is that they would then go to batch production," de Mars said. "We wanted to do this early sequencing because we're so excited and want to see what we have, but how much we can do and what we should do with just this initial batch is still to be determined."
Regarding regulatory applications, de Mars said that the consortium is hoping to explore whether it can be part of pre-submission meetings where it could introduce these as a potential tool for a sponsor for validation, but there are also other internal programs at the FDA where she and her colleagues believe their reference samples could be useful, such as the Medical Device Development Tools program.
De Mars said that the team has discussed potential next steps to use its cell-line references to develop synthetic samples for liquid biopsy test validation, but it is holding off on that for the time being.
"We thought it would be more beneficial to get this technology a little bit more developed … and then we would start taking a look at how we could adapt these and make these useful for or with liquid biopsies," she said.