Challenges in circulating tumor DNA analysis for cancer diagnosis

People knew that the DNA molecule existed outside of cell even before finding out its famous double helix structure. Mandel and colleagues identified DNA molecule, termed as cell-free nucleic acids (cfDNA) later, in human bloodstream in as early as 1948.1 However, at that time no people realized how these DNA molecules associate with human diseases. Thing started turning around until 1964, DNA was found being released into sera for certain systemic lupus erythematosus patients.2 Since then, many clinical studies were carried out, more evidences demonstrated the strong correlation between cell free DNA and human diseases, especially for cancer.3,4 It was observed that even DNA could be isolated from blood of healthy people, but the amount of DNA significantly increased in the blood sample from patients with serious tumor. Particularly, as the earliest research, DNA fragments from mutant Kras gene were found in blood of pancreatic carcinoma patients5 and mutant N-ras gene fragment for myelodysplastic syndrome patients.6 These studies successfully demonstrated the direct correlation between circulating DNA and tumor. Recently it has been widely accepted that the levels of circulating nucleic acids strongly connected with tumor burden and malignant progression.7–11 For not being confused with cell-free DNA in healthy people, tumor cell related DNA circulating in human cardiovascular system were specially termed as circulating tumor DNA, ctDNA. Generally, it is widely considered that most DNA in circulation system is the debris of dead tumor cells. However, due to the complexity of cancer development, more fundamental studies are required to investigate questions, such as which processes contribute to ctDNA release from tumor cells12 and how the release process change the state of ctDNA in the circulation system. Besides being the debris left behind by dead cells, DNA is the key component of neutrophil extracellular traps (NETs), a host immune defense system against invading pathogens. Recently, increasing studies have demonstrated that NETs got involved in cancer development at every stages.13–15 With the development of molecular oncology, more and more tumor specific gene mutations were identified,16 and detail information about relevant tumor specific mutation could be found in systematically organized database, such as My Cancer Genome (www. mycancergenome.org). Up to now, ctDNA has been investigated with numerous of prevalent tumors, including Breast,17,18 Colorectal,7,19 Hepatocellular carcinoma,20,21 lung,22–24 Melanoma,25,26 Ovarian,27 Pancreatic28 and so on. In comparison with other biomarkers, e.g. protein, ctDNA is more informative with more precise analysis methods.29 Due to its nature, ctDNA is becoming a remarkable clinical tool. Especially, the convenience in collecting blood sample grant the liquid biopsy application great potential through ctDNA analysis in cancer diagnosis. However, precise analysis of ctDNA is still a challenge for some technique and biophysical reasons. For becoming a solid tool for tumor diagnosis, more clinical researches are still necessary to address some crucial questions about the physiological mechanism and analysis technical issues. In this review, we put the spotlight on the crucial issues in ctDNA based cancer diagnosis, especially the experimental issues which have led to contradictory results in different studies.