The breakthrough finding that heralded the birth of circulating tumour DNA as a precision medicine diagnostic tool was the presence of signature mutations from a patient’s tumour (1). Armed with this knowledge, a colleague of mine explored the possibility of using allele specific PCR tests that we had developed to detect KRAS mutations in colorectal cancer, to detect mutations in the serum of cancer patients. This first attempt had limited success and it was not until 2007, in collaboration with The CRUK Paterson Institute, UK, and with better access to patient’s samples, that we demonstrated the feasibility of this approach to detect BRAF mutations in the DNA extracted from the serum of melanoma patients (2).

We had a persistent debate about whether serum or plasma would be a better source of circulating tumour DNA. The theory being that serum should contain more circulating tumour DNA but also more contaminating normal DNA through cell lysis prior, whereas plasma would yield a purer circulating DNA sample but with less total circulating DNA. Through numerous studies, we always found the detection rate to be higher in plasma than in serum (2,3).

In parallel, diagnostic and device companies developed DNA extraction technologies specifically targeting circulating DNA which we found increased the yield of DNA extracted. Improvements were made to sample collection, (4), and development of blood collection tubes that stabilised the DNA prior to extraction, which we found further improved yields and mutation detection rates.

Through these incremental enhancements and by conducting circulating tumour DNA utility studies alongside tumour biomarker analysis, we gradually built up the evidence that circulating tumour DNA was viable alternative to tumour tissue testing, especially for patients who were unable to provide biopsies (5).

One of the major issues cited with reliance on circulating DNA is the high false negative rate. However, if you consider that a simple blood draw is required rather than an invasive biopsy, it remains an attractive clinical option.

The FDA has been convinced of the utility of this non-invasive approach to select patients prior to treatment. In 2016, the FDA approved circulating tumour DNA based diagnostics for certain advanced non-small cell lung cancer indications, namely the cobas® EGFR mutation test v2 for selecting patients for treatment with erlotinib and osimertinib. This has paved the way to mainstream adoption of circulating tumour DNA for patient selection.

Even using the most sensitive tests however, not all tumour positive patients are detected, especially in earlier stage disease. Improving diagnostic tests and increasing the amount of DNA added per test may increase the detection rate. However, tumour mutations may not be present in the extracted DNA of all patients, and no improvement to technology will resolve this situation. For these patients, we must continue to explore alternative sources of tumour DNA to ensure safe and easy access to testing is available in this era of precision medicine.

Gillian Ellison, Director
(Gillian.Ellison@DaTAGC.Solutions)
​
1. Stroun M, et al. Oncology. 1989 Jul 1;46(5):318-22.
2. Board RE, et al. . British Journal of Cancer. 2009 Nov 17;101(10):1724-30.
3. Aung KL, et al The Journal of Molecular Diagnostics. 2014 May 31;16(3):343-9.
4. Sherwood JL, et al. PloS one. 2016 Feb 26;11(2):e0150197.
5. Douillard, J.-Y., et al. (2014). Journal of Thoracic Oncology, 9(9), 1345–1353.​