Improved Liquid Biopsies With Combined Digital PCR and Next-Generation Sequencing

The discovery that circulating tumor DNA (ctDNA) can be found in the blood, urine and other bodily fluids of cancer patients has led to a new type of biopsy that bypasses surgery—liquid biopsy. However, the tiny fraction of ctDNA within a liquid biopsy is difficult to measure amidst the background of non-tumor circulating cell-free DNA (cfDNA) sloughed from normal cells. It was recently shown that the cell-free liquid biopsy, especially of plasma cfDNA, is a valid assay for cancer genotyping and has potential to direct cancer treatment plans.¹

Table 1 – Cancer management using NGS and dPCR

Through the combination of next-generation sequencing (NGS) and digital PCR (dPCR), liquid biopsy is poised to become the standard in cancer management (see Table 1). These technologies together provide quantitative detection and sequencing of small amounts of DNA required to link cancer genotype with available cancer treatments.

NGS and digital PCR together provide a complete picture of the cancer genome

Because cancers vary in type and complexity, effective mutational analysis tools are needed for research and care management. The wide-angle panoramic view of whole-genome sequencing afforded by NGS combined with dPCR’s zoomed-in precision detection of DNA provide a comprehensive picture of a cancer’s genetic makeup. By applying these complementary techniques at the appropriate time based on the disease type and stage, doctors can treat cancers more quickly, precisely and cost-effectively.

Through gene panels or whole genome sequencing, NGS can reveal unpredictable genomic mutations and rearrangements critical for making informed treatment decisions. The information achievable with NGS also allows researchers to identify non-driver (“passenger”) mutations arising early in the development of the tumor that may assist in tracking tumor recurrence. Unfortunately, the NGS process is slow, which can lead to delays in obtaining patient results and is economically unattractive as an assay for serial monitoring.

Genotyping a smaller set of known gene markers is more sensitive, costeffective and faster with the Droplet Digital PCR (ddPCR) QX200 system (Bio-Rad Laboratories, Hercules, Calif.). Through multiplexing, a single ddPCR reaction can interrogate multiple genes and at least 5–10 mutations simultaneously. ddPCR identifies and precisely quantifies target DNA within a liquid biopsy, and thus is often used to validate NGS results. ddPCR provides results that specify (absolute) copies per milliliter of plasma, both of the cancerous mutation and its wild-type counterpart.

This nonrelative reporting of ctDNA values is less subject to physiological variations in background cfDNA levels, and may allow healthcare providers to more accurately evaluate cancer growth longitudinally, as well as a patient’s response to treatment. NGS, on the other hand, typically provides only a ratio of mutant to wild-type DNA copies due to the need for DNA library construction and preamplification prior to sequencing.

Monitoring driver mutations, resistance and recurrence

While some cancers have more elusive driver mutations, cancers such as melanoma have a small set of defining mutations that can be predicted in the majority of tumors.² In these cases, a tissue or plasma genotyping assay using ddPCR multiplexing can quickly determine if any of the most probable mutations in the cancer are present. The physician can then design and execute a treatment plan specifically for the patient based on these results. This genotyping tool is also useful for devising treatment plans for non-small cell lung cancer (NSCLC) patients, the majority of whom have mutations in either the epidermal growth factor receptor (EGFR) or KRAS.³

When initial ddPCR plasma genotyping does not identify common drivers, or for tumors with a large set of defining mutations, NGS is critical for charting a personalized treatment path. The technique provides a global view of driver mutations as well as additional genomic information about the patient. To confirm NGS results, a more targeted and sensitive assay like ddPCR is often used.

A key benefit for physicians is the serial use of digital PCR to monitor one or more driver mutations throughout treatment to determine response and recurrence since the level of mutated DNA found in liquid biopsies has been found to reflect the size of the tumor(s). It is also possible to use NGS to track progress; however, in these cases it is less practical due to cost and longer turnaround time.

In a clinical study conducted by Garcia-Murillas et al., early-stage breast cancer patients were initially genotyped by targeted NGS of DNA from tissue biopsy. Patient-specific ddPCR assays then identified minimal residual disease and tracked disease progression by analyzing the presence of ctDNA in liquid biopsies after conventional treatment (Figure 1).⁴ The likelihood of disease recurrence was predicted almost eight months earlier than would have been possible using traditional clinical diagnostics.

 Figure 1 – Droplet Digital PCR enables liquid biopsy for monitoring cancer treatment.

Siravegna et al. examined the liquid biopsies of patients with colorectal cancers (CRCs) to determine mutational dynamics of CRCs treated with a therapeutic antibody specific to the EGFR.⁵

The researchers used ddPCR to investigate the mutational status of seven genes critical in CRC growth and resistance. This study uncovered how the CRC genome adapts to intermittent drug schedules, allowing a drug to which the tumor had previously grown insensitive to again be effective. Such “real-time” monitoring of ctDNA levels for driver and resistance mutations will ultimately help point to the best course of treatment for CRC patients. When resistance mutations cannot be predicted, NGS is needed to give a global view of the genetic landscape and to identify the important new mutations driving cancer growth.

The zoomed-in view of ddPCR for detecting and tracking actionable mutations with very high sensitivity complements the zoomed-out view provided by NGS to assess larger numbers of possible driver and resistance mutations.

Monitoring chromosomal rearrangements in heterogeneous tumors

Due to the heterogeneous nature of tumors and their evolving subclones, it is important to find genetic markers present in all cells within the tumor. Lao Saal, Ph.D., assistant professor and head of the Translational Oncogenomics Unit in the Division of Oncology and Pathology at Lund University Cancer Center, describes why chromosomal rearrangements are key markers for monitoring cancer. “Chromosomal rearrangements are often early events in cancer development; thus they are shared across the subclones, and rearrangements are present in essentially all cancers,” he said. “Therefore, chromosomal rearrangements can be used to monitor all patients, whereas the gene-based cell-free DNA tests won’t be informative in cases that don’t have the typical mutations.”

Chromosomal fusions (also called structural variants) found in tumors can serve as straightforward biomarkers to identify and monitor cancer progression, whereas any given single nucleotide polymorphism (SNP) or insertion-deletion mutation might not be representative of the whole tumor. NGS enables researchers to cast a wide net to identify chromosomal rearrangements that mark—but do not necessarily drive the cancer’s growth—and allow monitoring of potentially all of a tumor’s cells. With tumor-specific rearrangements identified, Saal can convert these patient-specific fusion markers into highly sensitive ddPCR assays for a fast, cost-effective, real-time biomarker assay through liquid biopsy.

Olsson et al. studied early detection of metastasis in breast cancer. Metastasis caught too late cannot be effectively treated. By monitoring about a half dozen of these structural variants using liquid biopsies with ddPCR analysis, the researchers were able to detect metastasis as much as three years earlier (with an average of 11 months) than conventional techniques. They also demonstrated that the quantity of ctDNA was predictive of the likelihood of recurrence and death.⁶

This approach is also being studied by other laboratories and with other cancers. Reinert et al. analyzed ctDNA in patients with CRC following surgery.⁷ Colorectal cancers are difficult to biopsy, which makes their liquid biopsy especially advantageous for serial monitoring. Reinert’s team first analyzed the tumor tissue with NGS to identify somatic structural variants specific to each patient. Similar to Olsson’s group, Reinert was able to detect cancer recurrence an average of 10 months in advance of traditional methods by converting these rearrangements into easily monitored ddPCR assays.

Clinical applications of liquid biopsy and genomic biomarkers

Liquid biopsy for solid cancer is beginning to gain traction outside of the research laboratory. In 2015, Biodesix (Boulder, Col.) commercialized the GeneStrat liquid biopsy test for monitoring patients with non-small cell lung cancer for driver mutations in the EGFR, KRAS and BRAF oncogenes. Cancer diagnosis can be made within 72 hours of drawing a sample.

The clinical utility of using NGS to uncover genome-wide data combined with a practical reflex assay such as ddPCR to monitor specific biomarkers depends on the ability to find clinically actionable events. Hanlee P. Ji, Ph.D., assistant professor of Medicine at the Stanford School of Medicine, said, “Our research group is very focused on trying to find genetic determinants that are indicators of metastatic recurrence…. And that’s very practical because that leads to potential risk assessment for individuals who are more likely to have metastatic recurrent disease, and thus you’d be more likely to treat them aggressively up front [even] if they are lower stage disease.”

Sequencing information obtained from NGS permits researchers to identify possible genomic indicators not only for metastatic recurrence, but also tumor aggressiveness and other central properties of the disease. Once genetic biomarkers are defined, a simplified, more cost-effective ddPCR assay can be developed to enable faster decisions in diagnosis and treatment.

References

1. Bettegowda, C.; Sausen, M. et al. Detection of circulating tumor DNA in early- and latestage human malignancies. Sci. Transl. Med. 2014, 6(224), 224ra24.
2. Chang, G.A.; Tadepalli, J.S. et al. Sensitivity of plasma BRAF^mutant and NRAS^mutant cell-free DNA assays to detect metastatic melanoma in patients with low RECIST scores and non- RECIST disease progression. Molec. Oncol. 2015, 10(1),157–65.
3. Naidoo, J.; Drilon, A. KRAS-mutant lung cancers in the era of targeted therapy. Advances in Experimental Medicine and Biology 2016, 155–78.
4. Garcia-Murillas, I.; Schiavon, G. et al. Mutation tracking in circulating tumor DNA predicts relapse in early breast cancer. Sci. Transl. Med. 2015, 7(302), 302ra133.
5. Siravegna, G.; Mussolin, B. et al. Clonal evolution and resistance to EGFR blockade in the blood and colorectal cancer patients. Nature Med. 2015, 21, 795–801.
6. Olsson, E.; Winter, C. et al. Serial monitoring of circulating tumor DNA in patients with primary breast cancer for detection of occult metastatic disease. EMBO Molec. Med. 2015, 7(12), 1034–47.
7. Reinert, T.; Schøler, L.V. et al. Analysis of circulating tumour DNA to monitor disease burden following colorectal cancer surgery. Gut 2015, 0:1–10. doi: 10.1136/gutjnl-2014-308859.

George Karlin-Neumann, Ph.D., is director of scientific affairs at Bio-Rad Laboratories, Digital Biology Center, 5731 W. Las Positas Blvd., Pleasanton, Calif. 94588, U.S.A.; tel.: 925-474-9072; e-mail: GeorgeKN@bio-rad. com; www.bio-rad.com