Liquid Biopsies Are Viewed as “Liquid Gold” for Diagnostics

  • Invasive Needles and Scalpels Seen as Passé
  • Noninvasive Sampling Advocates Focusing on Circulating Tumor Cells (CTCs) 
  • New Companies are Pursuing the Liquid Biopsy “Gold Rush”

Biopsy Basics

Ultrasound is a real-time procedure that makes it possible to follow the motion of the biopsy needle as it moves through the breast tissue to the region of concern, as discussed elsewhere (taken from via Bing Images).

Ultrasound is a real-time procedure that makes it possible to follow the motion of the biopsy needle as it moves through the breast tissue to the region of concern, as discussed elsewhere (taken from via Bing Images).

As defined in Wikipedia, a biopsy is ‘a medical test commonly performed by a surgeon or an interventional radiologist involving sampling of cells or tissues for examination.’ Biopsies can be excisional (removal of a lump or area), incisional (removal of only a sample of tissue), or a needle aspiration (tissue or fluid removal). Despite the value of these traditional types of biopsies, they are more or less invasive, lack applicability in certain instances, and require accurately “going to the source” of concern, as pictured to the right, for ultrasound-guided breast cancer biopsy. Better methodology is highly desirable and is the topic of this post. By the way, if you want to peruse a lengthy list of scary risks associated with various type of common invasive biopsies, click here to see what I found in Google Scholar by searching “incidence of complications from biopsies.”

Benefits of Liquid Biopsy

Unlike traditional biopsies that invasively “go to the source” of concern, noninvasive liquid biopsies take advantage of relatively recent recognition that some diseases, such as cancer, discharge. Loosely speaking this means cells and/or other characteristic analytes can be readily sampled in blood, urine or even saliva, which are each effectively the “source going to the sample.” This directional flip-flop, if you will, provides liquid biopsies with several important advantages: non-invasiveness that is inherently safer, much wider applicability, and—some argue—better diagnostics. The latter point is expanded upon in an expert review by Tsujiura et al. as follows:

“Surgical and/or biopsy…approaches cannot always be performed because of their invasive characteristics and may fail to reflect current tumor dynamics and drug sensitivities, which may change during the therapeutic process. Therefore, the importance of developing a non-invasive biomarker with the ability to monitor real-time tumor dynamics should be emphasized. This concept, so called ‘liquid biopsy’, would provide an ideal therapeutic strategy for an individual cancer patient and would facilitate the development of ‘tailor-made’ cancer management programs. In the blood of cancer patients, the presence and potent utilities of circulating tumor cells (CTCs) and cell-free nucleic acids (cfNAs) such as DNA, mRNA and microRNA have been recognized, and their clinical relevance is attracting considerable attention.”

In my opinion, the importance of the aforementioned “ability to monitor real-time tumor dynamics” will become increasingly evident as next-generation sequencing costs continue to decrease, thus enabling a patient’s liquid biopsy samples to be analyzed over more time-points, during which viruses or cancer cells can mutate. Following is a real-life story of the latter situation.

Liquid Biopsies for Cancer: A Case Study

Lynn Lewis had her cancer analyzed an easier way: a simple blood test. Credit Michael Nagle for the NY Times.

Lynn Lewis had her cancer analyzed an easier way: a simple blood test. Credit Michael Nagle for the NY Times.

Every personalized medical story has a face and place, and it was Lynn Lewis of Brooklyn who caught my attention in an article by Andrew Pollack in last year’s NY Times. Ms. Lewis had her non-responsive cancer analyzed by means of liquid biopsies, in the form of blood samples, to take advantage of new tests offered by one of many new start-up companies developing safer tests with more actionable information. In a nutshell, here’s Lynn’s story:

Lynn, who at the time of the story was 54 and working as a self-employed lawyer, had already been battling metastatic breast cancer for seven years. In June of 2013, she had a liquid biopsy test of tumor DNA fragments in her blood developed by Guardant Health—then a relatively new start-up company in Redwood City, CA that was investigating use of next-generation sequencing to look at a panel of genes associated with various cancers. The test found a genetic mutation that suggested her disease would be treatable by the Novartis drug Afinitor.

Liquid biopsy blood tests in a lab at Guardant Health. Credit Jim Wilson/the NY Times.

Liquid biopsy blood tests in a lab at Guardant Health. Credit Jim Wilson/the NY Times.

Puzzlingly, Ms. Lewis had been previously treated with Afinitor and it had stopped working. Still, doctors decided to re-administer the drug to her, and again it didn’t work. A second test several months later found additional mutations but did not help determine the best course of treatment.

A third liquid biopsy, however, indicated significant levels of the protein Her2, which drives tumor growth, in some of Ms. Lewis’s cancer cells. The increase in Her2 may not have been detected previously because liquid biopsies sample only a tiny bit of the tumor, which is typically heterogeneous. While tumor heterogeneity is—in my opinion—an inherently confounding aspect of cancer, regardless of how biopsies are obtained, Ms. Lewis was successfully switched to Herceptin and Tykerb, drugs that specifically target Her2-positive cancers.

Lynn’s case to me is less an example of cancer treatment being “trial and error” and more an illustration of the importance of liquid biopsies as a safe, repeatable means of sampling inherently heterogeneous and dynamically changing cancers to obtain new data in real-time.

As a side note, I should add that liquid biopsies can also provide a valuable “sneak preview” of cancer before detection by traditional means. For example, a study published in The New England Journal of Medicine showed that in some cases a liquid biopsy could detect the worsening of breast cancer five months before it could be “seen” by conventional—and BTW potentially harmful—CT scans. This could allow an ineffective therapy to be abandoned earlier in favor of an alternative drug.

So, what’s not to like about liquid biopsies? Pollack claims that, despite all the potential of liquid biopsies, developers of such tests for the most part have not yet rigorously established their accuracy and—more importantly—actionable clinical utility. In other words, does blood testing really help patients? In my opinion, this is the biggest problem facing routine adoption of these biopsies. While clinical investigators can fairly quickly use newly developed assays to examine liquid biopsies “for research purposes only,” widespread sales for diagnostics requires FDA approval that is usually a very slow process. This is especially true for DNA/RNA-based assays that have to account for genotypic (i.e. sequence) differences between patients who are nevertheless grouped together as having the “same” cancer. Hence the emergence of the new era of so-called “N of 1” personalized medicine.

The New Gold Rush

If you agree with me that discovering clinical utility of, and extracting commercial value from liquid biopsies are metaphorically a new Gold Rush, then liquid biopsies are effectively “liquid gold.” As proof that indeed a “rush” for discovery is on, a recent review in Clinical Chemistry states that “the detection and molecular characterization of CTCs are one of the most active areas of translational cancer research, with >400 clinical studies having included CTCs as a biomarker.”

While “N of 1” cases like that of Ms. Lewis suggest utility of Guardant’s tests, Pollack’s article asserts that lack of much more and equally compelling evidence for clinical utility has hindered the acceptance of CellSearch®, the first test to detect and count CTCs. Sold by Janssen Diagnostics, a subsidiary of Johnson & Johnson, it was first approved by the Food and Drug Administration in 2004.

Common sense would lead one to accept that having a large number of tumor cells in the blood is cause for worry because of increased probability that the cancer will progress. But it is not always clear what to do with that information, i.e. what is actionable from a clinical perspective? Pollack quotes Dr. Scott Kopetz, an associate professor at the M.D. Anderson Cancer Center in Houston as saying that it’s not useful “to tell someone you have a high chance of a cancer coming back but we don’t know what to do.” Still, some developers of these diagnostic tests claim that detection of these tumor cells could at least provide evidence of ineffective treatments, saving the patient from suffering unnecessary side effects of drug therapies that aren’t working.

Helmy Eltoukh (left), CEO of Guardant Health, with AmirAli Talasaz, President and CTO, in the lab. Credit Jim Wilson/the NY Times.

Helmy Eltoukh (left), CEO of Guardant Health, with AmirAli Talasaz, President and CTO, in the lab. Credit Jim Wilson/the NY Times.

On the other hand, I found a recent study of women diagnosed with breast cancer by traditional mammography, who were also analyzed for CTCs in blood, the latter of which had limited utility. Notwithstanding low diagnostic yield of CTCs such as in this study, there are a growing number of companies working with CTCs, some of which will likely provide actionable data with clinical utility, but you’ll have to stay tuned for those:

Alere                                                     Cynvenio Biosystems

ApoCell                                                 Epic Sciences

Biocept                                                 Fluxion Biosciences

Clearbridge Biomedics                          Rarecells

Creatv MicroTech                                  ScreenCell

SRI International

But some experts say the momentum is shifting to the newer approach of looking for DNA fragments not inside cells but rather released DNA that enters the bloodstream when cancer cells die, and are phenomenologically referred to as circulating tumor DNA (ctDNA). That ctDNA can provide promising biomarkers for noninvasive assessment of cancer, has already been successfully translated into commercial products by Trovagene, which test for ctDNA in urine or blood, and claims to have been the first company to have recognized the diagostic value of ctDNA.

Importantly, assessment of tumor burden was recently demonstrated for non-small-cell lung cancer by collaborators at Stanford University, who reported an ultrasensitive next-generation sequencing method in Nature Medicine showing that levels of ctDNA were highly correlated with tumor volume and distinguished between residual disease and treatment-related imaging changes. Moreover, measurement of ctDNA levels allowed for earlier response assessment relative to radiographic approaches.

Notwithstanding the relative diagnostic value of “inside v. outside” DNA, various investors are placing their bets, so to speak. Guardant initially raised $10 million from Sequoia Capital, the venture capital firm known for backing Apple, Google and, most recently, WhatsApp. In February, the company raised $50 million for a current total of $100 million, which is obviously a big investment in the potential value of tiny analytes. Other start-ups focusing on ctDNA are Boreal Genomics, and Inostics, which was acquired by Sysmex, a Japanese diagnostics company.


In addition to CTC and ctDNA as sources of potentially actionable diagnostic information, small (30-100nm) exosome vesicles are also being extensively investigated. Interested readers can find extensive details for this approach in Michael D. O’Neill’s BioQuick News summary of the fourth annual meeting of the American Society for Exosomes and Microvesicles.

Here, I’ll simply exemplify this approach by pointing to two cases, the first being the team of Qiagen and Exosome Diagnostics Qiagen that is developing kits to capture and process RNA and DNA from biofluid exosomes and other types of microvesicles. The goal is to provide sample prep technology applicable to patient blood, urine, or cerebrospinal fluid for qPCR or next-generation sequencing.

The second example is a communication by a consortium of Japanese academic, government, and corporate investigators published in 2014 in venerable Nature magazine entitled Ultra-sensitive liquid biopsy of circulating extracellular vesicles involving colorectal cancer. Interested readers will find lots of detail in this report; however, the important “punchline” is that cancer markers were detected in samples with early stage colorectal cancer that are invasive. On the other hand, the authors state that further studies are needed to know whether the reported method reduces colorectal cancer mortality as a screening test.

Tool Providers, Too

“Miner with Pick, Shovel and Pan” ca. 1850. Daguerreotype from the collection of Matthew R. Isenburg (taken from via Bing Images).

“Miner with Pick, Shovel and Pan” ca. 1850. Daguerreotype from the collection of Matthew R. Isenburg (taken from georgetowndivide. via Bing Images).

While all of the aforementioned companies seeking “gold” from liquid biopsies have “staked their claims,” they all need tools with which to “dig.” Like selling picks and axes during the Gold Rush of yesteryear, modern day biotechnology tool providers such as Illumina, Bio-Rad Laboratories, and RainDance Technologies—to mention just a few—each hope their tools will be used to test for tumor DNA.

“Looking at the Sun and Trying to See the Stars”

Technically, analysis of ctDNA is like the proverbial finding “a needle in the haystack” because it is present in very small amounts compared to other cellular components that contain DNA—not to mention the slew of other non-DNA components.

Because of its metaphorical impact, my favorite quote in Pollack’s article regarding liquid biopsies is that of Nitin Sood, CEO of Boreal Genomics, who said “it’s like looking at the sun and trying to see the stars.” Mistakes can easily be made. And fewer genes can typically be analyzed from a liquid biopsy than from a conventional tumor biopsy.

The biggest payoff, so to speak, would come if liquid biopsies could detect cancer in seemingly healthy people, when it is still curable. A recent study by researchers at Johns Hopkins University found that tumor DNA could be detected in the blood of about half the 223 patients with localized cancers that they tested. However, since it was already known that those patients had cancer, the study did not demonstrate that a blood test could detect cancer earlier than other methods.

Ms. Lewis, the breast cancer patient discussed at the beginning of this post, told Pollack that she did not fully understand the results of her tests, but she was sharing them publicly in hopes that they provide some insight to scientists. We should note that her tests where performed as part of a research project and she was not billed for them.

Unfortunately, performing and analyzing such tests may not be up to the doctors as the FDA’s view of approving diagnostic tests ultimately comes into play. Since convincing the FDA of clinical utility is a long and expensive process, it’s understandable that there is growing public support for doctors using unapproved tests of patient samples “for research purposes only.” On the other hand, not all doctors—and I suspect very few—want to operate in a research mode, and all are likely more than concerned about possible lawsuits from disgruntled patients—or their surviving spouse or other relative.

As always, your comments are welcomed.


Liquid biopsies as an emerging trend in applied nucleic acids analysis is moving very quickly. Here are three noteworthy items that I came across after completing this post.

(1.) HealthDay News reported that, based on analysis of methylated DNA—see TriLink Epigenetic Reagents—in 10 breast cancer-specific genes in CTCs, a new test is able to detect with more than 90 percent accuracy recurrent breast cancer in women.

Dr. Joanne Mortimer, director of women’s cancer programs at the City of Hope Comprehensive Cancer Center, in Duarte, California is quoted as saying: “There is reason to be optimistic, and to study this further.”

However, Mortimer, who was not involved with the new research, said the study only included a small number of patients—55 healthy women and 57 who had breast cancer that had spread—and more study is needed.

“This is incredibly needed,” she added, as other tests on the market aren’t very reliable.

According to Mortimer, the worst part of being treated for early stage breast cancer for patients is the period after treatment is over. “Then they live with this uncertainty. They may be cured, they may not. And only time will tell.”

(2.) Written in Blood is the title of another recent article worth reading was written by Ed Yong and published in Nature magazine as a News Feature that is freely (yeh!) accessible here. The byline summarizes the gist of this report:

“DNA circulating in the bloodstream could guide cancer treatment — if researchers can work out how best to use it.”

(3.) A conference on Biofluid Biopsies & Companion Diagnostics on October 27-28 had a jam-packed agenda on all manner of discovery and commercialization involving CTCs and ctDNA, as well as exosomes. I presented this downloadable poster on TriLink CleanAmp™ PCR and RT-PCT products. Check out the speakers and presentations by clicking here.

(4.) Coincidentally, I’ve just returned from Circulating Biomarkers World Congress 2015 held in Boston, MA where I presented a poster on Improved Small RNA Library Preparation Workflow for Next-Generation Sequencing that allows automation of processing liquid biopsy samples to find low levels of miRNA biomarkers.

12 thoughts on “Liquid Biopsies Are Viewed as “Liquid Gold” for Diagnostics

  1. Very informative, thank you. Could tell me whether these technologies are developed with conventional detection testing in centralised labs? Is there a drive/demand for POCT? B

  2. Many thanks. All very useful info. Wondering about the application of such technology to POCT? Will there be a drive to decentralise these tests? Or are they currently being worked in combination with conventional detection tests in centralized labs?

    • Bene,
      You’re welcome, and sorry for overlooking your comments and questions.
      I’m sure there is a drive to apply these kinds of test to POCT. If you check Cepheid, you’ll find information about their new small system for PCR-based detection.

    • No, they are very different. Liquid biopsy refers to any liquid form of a patient sample to be analyzed by any method, such as a blood, or urine, or saliva sample to be analyzed byqPCR for DNA, or qRT-PCR for RNA, or sequencing of DNA.

      NGS means “next-generation sequencing” methods, of which there are various types.

  3. Circulating cell-free DNA could be new generation of liquid biopsy for cancer management. However, limitations like large volume of input, low recovery efficiency, labor intensiveness, and difficult for automation had hindered its clinical applications. Now, people are talking about last-generation of cfDNA-based liquid biopsy using a finger-tip volume of blood for tumor monitoring. Check it out at 2015 ASCO abstract by CirculoGene Diagnostics!!

    • Darrell,

      Indeed, all of the points you raise are issues that have to be overcome.

      Thanks for the information about CirculoGene Diagnostics!


  4. Didn’t I read recently there were at least two of examples of cancer found in pregnant women who were undergoing prenatal NGS screening?

    • Curt,
      Perhaps you read something related to this abstract presented at Molecular Med TRI-CON 2015 in February:

      Adventitious Maternal Cancer Detection during Non-Invasive Prenatal Testing of Circulating Fetal DNA
      Nilesh Dharajiya, M.D., Director, Clinical Lab, Sequenom Laboratories
      Non-invasive prenatal testing of cell free DNA (cfDNA) provides a novel tool to detect chromosomal abnormalities prenatally. Maternal cfDNA is also sequenced, and therefore, we may learn about the mother’s genome or other medical conditions. Here, we provide an overview of cfDNA aneuploidy testing and present clinical cases where testing serendipitously detected cell free tumor DNA, resulting in an earlier diagnosis of a neoplastic process.

      Thanks for you comment!


Leave a Reply

Your email address will not be published. Required fields are marked *