NEW YORK – Backed by a recently closed $46 million financing round, cancer diagnostics firm PrognomiQ is in the middle of large-scale multiomic discovery studies supporting its test development program.
The company declined to say what cancers or clinical questions it is targeting, but Philip Ma, PrognomiQ's president, founder, and CEO, said it is currently looking to tailor its tests to specific cancers, as opposed to developing a pan-cancer assay, and that it is targeting a "a variety of different applications within those cancers."
In an interview last summer, Tomer Berkovitz, a general partner at Israeli investment fund aMoon, which led San Mateo, California-based PrognomiQ's initial $55 million funding round in 2020 and participated in the recent round, said the company is pursuing as its lead product a test for distinguishing between cancerous and benign lung lesions detected on CT scans, though Ma did not confirm this.
PrognomiQ is a spinoff of proteomics tools firm Seer, which maintains a roughly 19 percent stake in the company. As such, Seer's ProteoGraph platform is central to PrognomiQ's biomarker discovery efforts. The ProteoGraph platform is based on the observation that nanoparticles, when incubated in a biological sample, collect proteins that form a "corona." These nanoparticles can serve as an enrichment tool, allowing researchers to pull proteins out of a sample to identify and quantify, using technologies like mass spec.
From a protein biomarker discovery perspective, the Seer system offers the potential for deeper analysis of plasma samples with higher throughput than has been possible with traditional mass spectrometry workflows. This allows researchers to perform discovery experiments in large cohorts (Ma said PrognomiQ is working with hundreds to thousands of samples) while still getting deep proteomic coverage. This combination of high throughput and deep coverage is also important for the types of multiomic methods being used by PrognomiQ as it allows the company to generate proteomic data that is more comparable in depth and scale to genomics and transcriptomics data.
In a 2020 paper published in Nature Communications, Seer employees and affiliated researchers used the platform to analyze 141 plasma samples from non-small cell lung cancer patients and healthy controls, identifying an average of 1,664 proteins per subject and more than 2,000 proteins across all subjects combined.
This month, the company published a study in the Proceedings of the National Academy of Sciences detailing the performance of its platform. They found that in undepleted plasma the system could quantify 1,706 proteins using a 2.5 hour data-independent acquisition (DIA) mass spec workflow and 2,014 proteins using a 10-hour data-dependent acquisition (DDA) workflow. That compared to 684 proteins measured in a conventional 30-minute DIA mass spec run and 1,761 proteins in a 36-hour DDA mass spec run.
Next-generation sequencing has allowed "large-scale, unbiased approaches to looking at different [health] states," Ma said. "The natural extension of that is to take it to proteins, which is what Seer's technology enables. And then the extension beyond that is to take all of that unbiased, large-scale information together to provide a truly complementary angle, a window into biology where you can see what information is provided by mutations, what is provided by transcripts, what is provided by metabolites, [and] what is provided by protein."
Ma said that in addition to proteomics, PrognomiQ is generating nucleic acid-level data including transcriptomic and methylation information as well as large-scale metabolomic and lipidomic data.
"We are interrogating different biological samples in a large-scale discovery program and finding those multi-analyte markers," he said. "And that, we hope, is going to lead to higher sensitivity and better tests for various cancers."
Affinity-based platforms from SomaLogic and Olink have also drawn interest from researchers as tools for multiomic biomarker discovery, with both companies offering higher throughput and broader coverage than Seer has demonstrated in mass spec-based experiments with ProteoGraph. Olink's Explore platform measures up to 3,000 proteins per sample while SomaLogic's SomaScan measures roughly 7,000 proteins.
Ma indicated, though, that PrognomiQ has no plans at the moment to use either of these platforms for proteomics work.
"You always want to be aware of other technologies and see where they are, but for the moment we are very confident in what Seer's platform provides."
According to Seer's full-year 2021 financial results, the company generated $2.2 million in revenues from product sales to PrognomiQ last year.
Ma said the decision to spin off PrognomiQ from Seer followed several rationales, including the fact that the proteomics tools business and diagnostics business have different funding requirements, different time horizons, and different regulatory concerns.
"It gives us the ability to just focus on applying the technology along with other omics technologies and our data science to come up with the best [diagnostic] products," he said. He cited the example of Illumina's initial spinning off of Grail, though he noted that Illumina's subsequent decision to bring Grail back into the fold had not changed his or PrognomiQ's thinking about the matter.
Ma said PrognomiQ aims to have tests on the market "in the next couple of years." He did not say whether it plans to bring them to market as laboratory-developed tests or take them through the US Food and Drug Administration. He declined to say whether the company will build a CLIA facility as part of its efforts.
Ma said PrognomiQ has no plans to offer its biomarker discovery platforms to outside companies for fee-for-service work but said it would consider strategic partnerships in cases, where, for instance, an outside party could help it access clinical samples or data or technologies that would help its own diagnostics development projects.