International study results show need for universal set of quality control materials for liquid biopsies

A public-private partnership including scientists from the Frederick National Laboratory for Cancer Research determined quality control materials for liquid biopsies performed differently depending on the assay platform utilized and showed uneven results. 

Liquid biopsies are emerging as valuable tool in diagnosis of cancer and can identify mutations that can guide treatment decisions. The seven-year project to develop and validate liquid biopsy quality control materials was published in the Journal of Clinical Oncology Precision Oncology. 

The partnership, which formed under the Foundation for the National Institutes of Health’s Biomarker Consortium, was co-led by the national laboratory’s Molecular Characterization Laboratory and included pharmaceutical companies, the National Cancer Institute, and U.S. Food and Drug Administration (FDA). 

The need for quality control materials  

Liquid biopsies detect circulating tumor DNA (ctDNA), fragments of tumor DNA shed into the patient’s bloodstream.  

However, validating a liquid biopsy assay to ensure it provides accurate and reliable results is challenging, as is comparing different assays. Such studies require high volumes of clinical samples, and even patients with advanced cancer have a low volume of ctDNA detected in their blood. 

“It’s very difficult to source samples from clinical studies,” said Thomas Forbes, scientific project manager at Molecular Characterization Laboratory, an author on the paper. “Any full-scale validation with clinical samples is almost prohibitively expensive and difficult.” 

Therefore, having a set of well-characterized quality control materials, which are standardized materials meant to act like clinical samples, would be an asset to the liquid biopsy community, enabling laboratories to more easily and economically validate their tests and compare their results.  

Three phases 

In the first phase of the study, the Biomarker Consortium team worked with three commercial manufacturers to develop quality control materials and to test them in four core reference laboratories. The results were published in 2021. These materials, which include 14 common variants, are now commercially available from at least one of the manufacturers.  

Next, Forbes led a functional characterization study to determine how these materials performed in different assays compared to the examined clinical samples. They tested three different assays: digital droplet polymerase chain reaction (PCR), hybrid capture next-generation sequencing (NGS), and target-amplicon sequencing.  

Finally, they performed a clinical pilot in which they tested the quality control materials and clinical samples in 11 laboratories across four continents.  

Unexpected results 

Scientists were surprised to find that while the quality control materials approximated the clinical ctDNA samples, their performance depended on the assay platform utilized. In particular, the materials tested with the hybrid-capture NGS assay behaved differently across clinical samples. Results also varied across laboratories.  

“Laboratories that are developing tests for liquid biopsies really need to be aware of these results,” said Chris Karlovich, director of the Molecular Characterization Laboratory and senior author on the paper. “If they’re developing an assay that’s based on digital PCR, for example, they would want to look at our paper and look at how these quality control materials performed when tested with digital PCR assays. In the paper, the data are described, so if you’re developing your own digital PCR assay, you know what to expect.” 

Karlovich noted that there are currently no universal quality control materials, but a universal set would be a valuable tool to compare assays, such as when a company submits a new assay for FDA clearance. 

He said these materials have the potential to become part of such a set.  

“These are, to my knowledge, the most well-vetted set of quality control materials that are out there for liquid biopsies,” Karlovich said.  

“I’m hoping that this work that we’ve done has shown that while the quality control materials might not work exactly like clinical samples, they could be used as a sort of surrogate, with some caveats,” Forbes said.  

Public–partnership 

Forbes noted that without the generosity of pharmaceutical company partners and the National Cancer Institute’s Research Donor Program, it would have been impossible to secure sufficient clinical samples to run this study—demonstrating both the need for quality control materials and the value of collaboration in biomedical research.  

This project was part of a highly collaborative effort. The Foundation for the National Institutes of Health and Frederick National Laboratory for Cancer Research signed a cooperative research and development agreement in 2019 to support the project, and Biomarker Consortium members included partners from academia, industry, and the public-sector.