Liquid Biopsies Move In on Tissue Biopsies for Cancer Testing

In a typical surgical biopsy, a tissue sample is removed for analysis, by a procedure that can be costly, risky, and painful. Liquid biopsy assays offer a much-heralded alternative. They are usually less invasive, quicker, and more easily repeatable because they can analyze blood, urine, saliva, or cerebrospinal fluid.

Potentially revolutionary, liquid biopsy assays promise to dramatically improve disease diagnosis, surveillance, and survival prediction. Currently, they are used most often in cancer testing, but as technology advances, scientists believe that liquid biopsies could be used to detect a wide range of diseases, including diabetes, cystic fibrosis, and periodontitis.

The global market for liquid biopsy diagnostic and monitoring tests had estimated revenues of approximately $394 million in 2016.

The global market for liquid biopsy diagnostic and monitoring tests had estimated revenues of approximately $394 million in 2016, according to a recent Kalorama Information report. Kalorama expects a $1.3 billion market for these technologies by 2021, and possibly a $3.3 billion market in 2027, if liquid biopsies live up to their hype.

Liquid biopsies target an assortment of biofluid constituents, including cell-free circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), exosomes, proteins, and RNA.

Circulating tumor DNA

ctDNA is fragmented DNA released from cancer cells into the plasma fraction of blood during apoptosis/necrosis. ctDNA fragments are extremely rare, making their isolation and identification very difficult. Because of the small proportion of ctDNA present in the total cell-free DNA (cfDNA) samples obtained, the use of highly sensitive techniques, such as PCR and next-generation sequencing (NGS), are recommended. The use of ctDNA in the clinical management of lung cancer patients as well as for monitoring disease status in advanced melanoma is well documented. 

In May, Cancer Genetics, Inc. (CGI) introduced Liquid::Lung-cfDNA™ to detect lung tumor-derived cfDNA. This assay reportedly enables analysis of frequently mutated single-nucleotide variants and short indels (insertions and deletions) in 11 genes with significant clinical relevance to non-small-cell lung cancer. The test is based on Thermo Fisher’s Oncomine™ Lung cfDNA assay, which is used to detect lung tumor-derived DNA (ctDNA) in cfDNA.

“Liquid biopsy technology using either NGS or digital PCR is showing significant advantages over traditional tissue biopsy samples,” explained CGI CEO and president Panna Sharma. “We think that once a baseline diagnosis or therapy plan is established, liquid biopsy can provide a compelling triple advantage by lowering the cost and burden associated with deriving the patient sample; providing a more thorough, consistent, and longitudinal biomarker profile of the patient; and improving the sensitivity and ability for earlier detection of biomarker changes.”

Also focusing on ctDNA, a team from University of California, San Diego sought to find biomarkers that could predict responses to immune checkpoint inhibitors, to which only about 20% of patients respond. The team hoped to identify an easily obtainable and easily interpretable biomarker that could predict response to checkpoint inhibitor therapy.

Using blood samples from 69 patients with different types of cancer treated with immune checkpoint inhibitor therapy, the investigators analyzed the ctDNA by NGS. Specifically, they counted the number of variants of unknown significance (or VUS, which are alterations/mutations in the DNA whose association with disease risk is unknown). They found that 29% of patients had more than three VUS alterations and 71% had three or fewer VUS in their ctDNA.

After being treated with an immune checkpoint inhibitor, patients with more than three VUS in their ctDNA had significantly higher response rates (45%) compared with those who had three or fewer VUS (15%). The median survival was not reached among patients who had higher numbers of alterations of the VUS type, while those who had lower numbers of alterations of the VUS type had a median survival of only about 11 months.

“Our study demonstrated that high ctDNA alteration number was associated with response to checkpoint inhibitor therapy,” Razelle Kurzrock, M.D., director of the Center for Personalized Cancer Therapy, said. “Though the study was small, a significant difference could also be identified for a longer survival without progression of cancer and even a longer overall survival in the high versus low VUS alteration group,” added Yulian Khagi, M.D., a hematology-oncology fellow at UCSD. “It is likely that, with larger studies, this difference will become more significant.” 

Circulating tumor cells

CTCs are intact tumor cells that have shed into the vasculature from primary or secondary tumors and circulate in the bloodstream. By definition, the presence of CTCs suggests the patient has metastatic disease. CTCs are low in abundance in most cancer patients, thus a CTB biopsy typically begins with an enrichment step to raise the concentration of CTCs. The main clinical utility of CTC testing to date has been to provide prognostic survival projections in proven cancers.

A research team from Complete Genomics and the University of California, San Francisco recently reported that it detected clinically relevant mutations using just five CTCs. For the study, they evaluated CTCs from two liquid biopsies drawn from a 61-year-old female patient with estrogen receptor-positive (ER+)/human epidermal growth factor receptor 2-negative (HER2–) metastatic breast cancer. First, they isolated 34 highly pure CTCs. Then they performed advanced whole-genome sequencing by splitting the genomic DNA from the CTCs into 3,072 individual compartments, with each compartment containing approximately 5% of the cancer genome. The DNA in each compartment was subsequently labeled with a unique barcode, the compartments were combined, and the genomic DNA and barcodes were sequenced.

“From 34 cells, we accurately detected mutations present in as few as 12% of CTCs, established the tissue of origin, and identified potential personalized combination therapies for this patient's highly heterogeneous disease,” noted Brock Peters, Ph.D., senior director of research at Complete Genomics.

According to Peters, this research was made possible by the use of the company’s long fragment read (LFR) technology. “That our sequencing method could detect the most important somatic mutations from just five CTCs in a noninvasive liquid biopsy is important, demonstrating cost-effectiveness and utility in clinical settings,” he added.

Proteomic-based technology

Proteins have better signal-to-noise expression levels in comparison to other biological molecules, such as DNA or RNA, that may be found in the circulation, in tissue, or in cells, according to Al Luderer, Ph.D., CEO of IndiDx. “Proteins undergo numerous steps in their biosynthetic evolution before they are made available to a cell and either remain within the cell, are transported, or shed into circulation. This unique processing means that a protein exists for a given functional reason, is generally long lived, and there is little to no detectable synthetic byproducts that potentially mask the detection of a given protein or that is biologically relevant.”

IndiDx leverages advances in proteomics to better diagnose complex diseases. Its experimental work emphasizes a systems biological approach to disease coupled with advanced instrumentation to measure multiple proteins simultaneously. IndiDx was co-founded by Leroy Hood, M.D., Ph.D., whose vision for the future of medicine is embodied in P4 medicine, which is predictive, preventive, personalized, and participatory.

The company’s breakthrough product is Xpresys® Lung 2, which measures blood proteins and identifies lung nodules with a high probability of being benign. “When lung cancer presents with symptoms, it is already Stage III/IV disease,” Luderer observed. “If lung nodules can be found when they’re small, e.g., dime-sized, and without symptoms, then there’s a much greater chance of removal for cure. But small lung nodules present a diagnostic challenge. The patients that will benefit from our test are seen after a computed tomography (CT) radiographic find of a lung nodule. This patient is then typically referred to a pulmonologist.”

The company has pipeline activities that include central nervous system and other oncology indications outside of lung cancer.

RNA biomarkers

Researchers at The University of Texas at Austin have taken a different tack and are looking at an ancient enzyme in bacteria that detects a range of RNA biomarkers with high accuracy.

An ancient bacterial enzyme (grey) crawls along a tangled strand of RNA (orange), creating a complimentary strand of DNA (blue). Lambowitz and his team think this enzyme—called GsI-IIC RT and part of a group of enzymes known as TGIRTs—have novel properties that make it easier to detect RNA biomarkers from cancer and other disorders. Postdoctoral researcher Jennifer Stamos revealed for the first time the molecular structure of this enzyme. Image courtesy of Jennifer Stamos at the University of Texas at Austin.

“DNA biomarkers are static. They provide information about mutations that cause a disease, but they don't provide information about the effect of these mutations on cellular processes, which can differ in different individuals,” observed Alan Lambowitz, Ph.D., a professor in the Institute for Cellular and Molecular Biology. “That's one reason a cancer-causing mutation can have different effects and respond differently to treatment, depending on the individual, a key consideration for personalized medicine.”

By contrast, “Monitoring cellular RNAs provides a snapshot of exactly what is happening in diseased tissue, such as a tumor, at a particular time,” Lambowitz said. “The method can be used to monitor day-to-day progression of the disease and response to treatment and to predict how different individuals with the same cancer will respond to different treatments.”

Lambowitz envisions a liquid biopsy that, in combination with other tools, would provide health professionals with all of this information. The group of enzymes he studies and that he believes can help are called TGIRTs, short for thermostable group II intron reverse transcriptases. TGIRTs find strands of RNA and create complementary strands of DNA that encode the same information and can be sequenced to provide diagnostic information. Because they are able to accurately make DNA copies of almost any type of RNA from very small amounts of starting material, they would do a better job of catching biomarkers for disease than anything currently available in an RNA-based liquid biopsy, according to Lambowitz.

Reference standards

One of the main challenges for liquid biopsy-based assays is detecting mutations at low allelic frequency with small amounts of DNA. Assays need to be sensitive enough to detect mutations at allelic frequencies as low as 0.1-0.01%.

Reference standards are designed specifically to address the challenges associated with liquid biopsies. Horizon Discovery’s Reference Standards are derived from human cell lines and fragmented to an average size of 160 base pairs; they cover a range of clinically relevant mutations at 5%, 1%, and 0.1% allelic frequency in order to closely resemble cfDNA extracted from real human plasma. “By providing cfDNA in synthetic plasma, our products can move through an entire sample workflow, from DNA extraction all the way through to analysis,” commented Eva-Maria Surmann, product manager at Horizon Discovery.

Where are we now?

Despite the incredible promise, copious research, and significant commercial investment, liquid biopsies cannot entirely replace tissue biopsies now or even in the near future. “There is nothing on the horizon that appears ready to replace [tissue] biopsy since histopathologic confirmation of cancer is considered biological truth and most/all treatment decisions rely upon establishing the true nature of the offending lesion,” said Luderer.

Sharma and Surmann are more optimistic. “We expect that over the next three years the majority of drug monitoring and disease measurement in solid tumors will move to liquid or blood-based molecular profiling approaches,” Sharma said. And, according to Surmann, “The more we learn about the molecular makeup of tumor, and its diagnostic, predictive, and prognostic value, the less we might need to rely on traditional tissue biopsies in the future.”

Image courtesy of Dreamstime Images.