Liquid biopsy assay exhibits efficacy similar to traditional tumor biopsy

Sumanta K. Pal, MD

The data presented by Philip Mack, PhD, provide some key insights on the feasibility of so-called liquid biopsies. We are increasingly using genomic data in our day-to-day practice, and tests such as this provide a useful alternative to tissue-based testing, acknowledging the fact that many of our patients in the clinic have tumors that are challenging to access. Having a blood-based test to assess genomic profiling is key. Blood-based tests are also a much easier way to continuously access genomic data during the course of treating it is certainly easier than subjecting a patient to repeated biopsies. The researchers have demonstrated not only the feasibility of the test in detecting alterations, but also that the results are often potentially actionable, with half the patients having a result that can be linked to therapy.

One further point: Beyond this test from Guardant, there are several other blood-based tests available. It is really going to be key for us to rigorously define which of these platforms deliver optimal results. Each utilizes distinct methodology and provides distinct results.

Wafik S. El-Deiry, MD, PhD, FACP

Liquid biopsy is emerging as a powerful and perhaps superior test in the oncology clinic with several advantages over conventional biopsy methods and marker assays. The circulating tumor DNA (ctDNA) test performed by Guardant Health, Inc., which funded the study, captures, quantifies and reports driver gene mutations, information on mutation burden, potentially actionable targets for therapy and drug resistance mechanisms. The ctDNA test is much less invasive than a tumor biopsy, results are obtained more rapidly than tumor biopsy and the blood sampling is much more amenable to serial sampling.

It is important to remember that not all tumors show elevated markers such as CEA, AFP, or CA19.9, and this is potentially an area where, in the future, ctDNA will prove to be even more useful.

Serial sampling is important for several reasons, including monitoring tumor burden by assessing levels of the ctDNA in the blood, the potential to detect relapse with greater sensitivity than imaging, and the ability to identify actionable alterations that can arise and change over time as the tumor evolves. Resistance mechanisms such as T790M mutation in EGFR offers patients opportunity for second- or third-generation anti-EGFR therapies that can still work even though EGFR has evolved a resistance mutation to first-line therapy. One feature of ctDNA is that it can capture a snapshot of shed DNA from multiple metastatic lesions. That is not possible with biopsy that largely depends on sampling a single accessible lesion.

It is clear that ctDNA testing has some limitations. For example, while highly sensitive, this particular study finds the test has a sensitivity ranging from 78% to 86% in different tumor types using a sample of over 15,000 patients. ctDNA from some tumors such as glioblastoma was harder to detect in blood. The observed sensitivity means that the test is potentially extremely useful for a large portion of the cancer patient population. Concordance ranged from 87% to 98%, with higher concordance observed when tumor biopsy and ctDNA blood sample were taken closer in time.

It can be predicted that detecting alterations or tumor relapse earlier than by other methods such as imaging may impact patient outcomes. This is a testable hypothesis in the clinic. Certainly, in clinical practice where patients get follow-up scans every 2 to 3 months while on therapy, it is likely that follow-up may be done sooner with clear evidence of disease progression detected by ctDNA much in the same way clinicians would be more suspicious of disease progression with rising tumor markers.

A challenge for the future is to evolve the tests to detect yet-to-be discovered resistance mechanisms for specific tumor types and therapies. However, the rapid progress this year is exciting and should improve patient care.

David M. Nanus, MD

This is a large analysis that demonstrates the ability to detect actionable mutations in a significant proportion of patients, which is exciting. It also supports other studies showing that the genetic abnormalities in cancer can evolve over time.  Examining the primary tumor may not detect a mutation present in a metastasis.  Today we need to perform a biopsy of the metastasis to get tumor tissue for genetic analysis.  The liquid biopsy can detect mutations in peripheral blood and therefore a biopsy may not be necessary.  Similar prior approaches have examined circulating tumor cells to find mutations. But this requires using techniques to isolate circulating tumor cells, processing them and then sequence the DNA.  Here, you sequence directly from plasma isolated from a tube of blood.  Even if actionable mutations are only found in a seemingly small percentage, in this study 6% of patients, that is still a large number of patients that may benefit.  The one caveat is that liquid biopsies may not detect mutations in early-stage patients.  This technique is most effective in patients with a large disease burden.  However, as technologies evolve one would anticipate mutations will be detected in early stage patients.