Liquid biopsies are becoming increasingly popular in cancer research. We take a look at their potential impact in the clinic and how current technologies can help us keep pushing the boundaries of precision medicine.
What is a liquid biopsy?
A liquid biopsy is sequencing of DNA from a person's blood, to detect, diagnose or monitor diseases such as cancer.
Most tumors release some of their DNA into circulating blood, this could be during necrosis, apoptosis or immune-system-related tumor cell eradication - this is circulating tumor DNA (ctDNA). ctDNA carries the same mutations as the tumor, so it can be used to detect the presence of cancer after surgery as well as monitoring progression.
Analysis of a patient's ctDNA can be used to quantify disease burden and provide the genomic information simultaneously. This is achieved by quantifying the mutated alleles in the sample and comparing it to "normal" DNA, calculating the mutant allele fraction (or % allele frequency).
ctDNA - an overview
ctDNA is a specific form of circulating cell-free DNA (cfDNA), which comes from a tumor. In healthy individuals, most of the cfDNA present in blood is released from hematopoietic cells. cfDNA has a half-life of under 2.5 hours, which makes it very suitable for monitoring changes over a rapid time scale. It's around 166 bp in length - perfect for short-read next-generation sequencing (NGS) systems.
We have been aware of ctDNA for over 20 years. First reports came from bladder cancer patients with TP53 mutations and pancreatic cancer patients with KRAS mutations. However, these initial reports were made using PCR assays with limited detection sensitivity.
Technological developments have made the detection of ctDNA much easier and more robust. Researchers have published plenty of papers describing the use of ctDNA in virtually all cancers. In 2014, a paper in Science Translational Medicine showed that ctDNA was detectable in nearly all cancers and that the levels significantly correlated with tumor stage1.
The application of liquid biopsies in precision medicine
There are three key areas where liquid biopsies are positively impacting growth in precision medicine. These all have the potential to not only improve patient well-being and outcome but decrease time and cost.
1. Targeted therapy
With the introduction of projects like the International Cancer Gene Consortium, we have seen a surge in molecular therapies for cancer over the last decade. Liquid biopsies can be used to detect targetable mutations in cancer patients.
This has many benefits over traditional biopsies, not only in cost and patient well-being, but also the sample quality. With a tumor biopsy, there is a risk of sampling from a non-mutated cancer clone, due to intra-tumor heterogeneity (where cells from the same tumor can have different gene expression), nullifying the sample test.
2. Early patient diagnosis
The earlier a cancer is diagnosed; the better patient prognosis can be. For example: in lung cancer, early detection means the disease is operable, which has higher long-term survival rates compared to inoperable tumors.
3. Monitoring response and minimal residual disease in patients
After surgery, when the tumor that is shedding DNA is removed, ctDNA levels drop sharply. If a patient presents elevated levels of ctDNA post-surgery, it could indicate metastatic disease and further treatments identified.
Additionally, by monitoring ctDNA over time, we can measure the effectiveness of treatment. This is currently done using imaging (CT or MRI) or ultrasound, which are all expensive in-patient procedures.
Finally, liquid biopsy monitoring can also detect the evolution of tumor resistance to therapy, which could mean alternative treatment options are considered earlier.
Liquid biopsy analysis - the technology
NGS - (Next-generation sequencing) Most liquid biopsy tests in development make use of NGS of specific cancer genes, from PCR-amplification or exome-capture. Sequencing of these panels allows high-depth analysis of each locus. This means even allele frequencies below 5% can be detected. Their small size means they can be cost-effective too.
dPCR - (Digital-PCR) dPCR is the most sensitive method of ctDNA analysis. It allows accurate identification and absolute quantification of mutant molecules in a normal background. However, its use is limited to known mutations and it is not amenable to multiplexing. This makes it very useful for specific mutations (such as EGFR T790M), but less so for genes that can be mutated anywhere along their coding sequence (like TP53). It is challenging to use for patient monitoring, as a panel of assays is required for each patient - this makes it relatively slow and expensive.
Alternative technologies such as ICE-COLD PCR and Boreal Genomics OnTarget can be used to enrich for mutant alleles, while not amplifying the wild-type alleles. This can increase detection sensitivity, although the impact on absolute quantification is currently not well understood.
Things to consider
The most important consideration is the collection and processing of blood. Leukocytes will release DNA into the blood after collection if the sample is not handled carefully. Most studies recommend a rapid ultra-centrifugation of blood to separate plasma within only four hours of collection. Recommendations are also being made for using blood collection tubes specifically designed to limit the lysis of white blood cells (e.g. Streck tubes).
The simplest method for increasing the robustness of a liquid biopsy is replication. By comparing duplicate liquid biopsy results, and using mutations called in both samples, false positives can be reduced. This can significantly decrease the number of variants called in exome analysis of ctDNA.
Using unique molecular identifiers (UMIs) improves the accuracy of mutant allele detection by reducing errors from PCR and sequencing. By comparing reads with the same UMI, a consensus call can be made. When comparing across UMIs, errors in PCR amplification can be removed.
Controls, such as Horizon's Structural Multiplex Reference Standard, are recommended to validate the limit of detection in any assay. Controls need to mimic the biological sample and have validated mutations at significantly lower than the desired MAF.
Combining all of these methods for limiting errors can allow liquid biopsies to detect mutations at below 0.1% MAF, so that there is a greater chance of detection and diagnosis.
Summary
The development of liquid biopsies has opened new doors for biological research into cancer, which will soon turn into the improvement of clinical management of cancer. Proof-of-concept studies have shown the potential for molecular profiling of patients as well as the monitoring and prognosis of their disease.
Improving the ability to detect lower and lower mutant allele frequencies may, one day, mean that liquid biopsies can be used in early-stage and even pre-symptomatic patients.
References:
- Bettegowda, C et al. Detection of Circulating Tumor DNA in Early- and Late-Stage Human Malignancies. Transl. Med. 24, (2014).