The development of a highly sensitive test for detecting residual disease and recurrence in cancer patients
BY TIM FORSHEW PHD, HEAD OF SCIENCE AND INNOVATION
As cells die in our body, they release DNA into our blood stream. This cell free DNA (cfDNA) can come from many sources, including healthy cells such as white blood cells. In individuals with cancer, DNA from the tumor, known as circulating tumor DNA (ctDNA), will also be present in the bloodstream. In the era of precision medicine, modern methods are now enabling the rapid extraction and analysis of this ctDNA, offering many applications for the improvement of cancer patient care.
One of the critical features of cfDNA and ctDNA is their short half-life, estimated to be somewhere around 1 hour1,2. This short half-life makes ctDNA an extremely powerful tool for assessing if a cancer patient has residual disease or recurrence (RDR) following treatment with curative intent. If the cancer has been successfully and fully removed, any lingering ctDNA will quickly degrade. On the other hand, if small amounts of tumor remain despite efforts to remove it, this short half-life of ctDNA means that extremely sensitive tests are needed to detect it before it breaks down.
This was demonstrated in the 2017 TRACERx study which showed the correlation between tumor size and ctDNA levels in lung cancer3. Whilst ctDNA levels were high in patients with larger tumors, levels detectable from tumors one cubic centimeter in size were measured at just 0.008% variant allele frequency (VAF), or 80ppm, on average. This highlights the need for a highly sensitive test to ensure the accurate detection of ctDNA at extremely low levels, even below 100ppm or 0.01%.
The key challenge is that a typical 10ml blood sample contains approximately 10,000 copies of cfDNA. Analyzing single variants would not yield detection of levels below 100ppm (or 1 in 10,000), posing a challenge for accurate detection in small samples.
Eight years ago, Inivata scientists proposed the solution of sequencing a patient’s tumor then developing a personalized multiplexed assay for high sensitivity ctDNA detection 4. We targeted ten variants found in the patient’s tumor, including drivers and passenger mutations, focusing on those that were high allele fraction in the tumor and therefore likely clonal at the time. By assessing multiple mutations or taking multiple “attempts at ctDNA detection” it is possible to spot if ctDNA is present in low concentration samples, even if there is less than one single copy of the cancer genome in the blood tube. Since then, much has progressed in the field of tumor informed personalized RDR assays.
Inivata CSO, Nitzan Rosenfeld, recently described how either large tumor informed personalized hybrid capture panels targeting up to thousands of mutations or whole-exome sequencing/shallow whole-genome sequencing with personalized analysis could be used for highly sensitive ctDNA detection5. Other recent research in this field has come from teams across the world, including the Dana-Farber Cancer Institute and the Broad Institute, who used duplex sequencing and personalized hybrid capture to study the power of up to 488 variants6. In addition, a group from the Translational Genomics Research Institute (TGen) developed a personalized method called TARDIS7, and another group of researchers used whole genome sequencing (of tumor, normal DNA, pre-treatment and post-treatment plasma)8.
Others personalized products in this space include the Signatera assay, which uses 16 variants found from the tumor to analyze for ctDNA and has reported >95% sensitivity at 0.01% tumor fraction (100ppm)9, and ArcherDX who presented an assay at the recent AACR and tested their ability to detect ctDNA in a dilution down to 30ppm.
Over the course of our work since 2012, it has been clear to us that a more sensitive personalized approach has powerful applications to improving cancer patient care. The key question is how to balance the added sensitivity of personalized assays with scalability, practicality and cost.
AACR 2020 marked an exciting milestone for Inivata where we described the results of an assay designed to meet this need for both sensitivity and practicality. RaDaR™ is a personalized assay that tracks a set of up to 48 tumor-specific variants in a patient. It is built using results from just a single exome run of the tumor DNA with a highly optimized approach for rapid multiplex primer synthesis and testing. Following personalized assay development, each blood sample can be assessed for RDR in under seven days enabling physicians to quickly make decisions on patient treatment.
The first poster we presented at the recent AACR meeting described the analytical development of RaDaR, and a dilution study which shows the ability to detect ctDNA at 20 parts per million (0.002% allele fraction) with 97% sensitivity and 100% specificity.
The second poster described the performance of the assay in a clinical setting. We showed detection of ctDNA down to as little as 6ppm (0.0006% allele fraction). In patients who progressed, ctDNA was detected between 6-12 months ahead of progression in the majority of patients assessed. In addition, patients with detectable ctDNA post-treatment showed an association with lower progression-free survival compared to patients in the ctDNA negative group.
Both posters can be found on the Inivata website here.
Many studies are ongoing that are demonstrating the utility of RaDaR, and we look forward to further discussions as to how this and other ctDNA technologies may be able to enhance clinical studies and patient care.
- Diehl, F. et al. Circulating mutant DNA to assess tumor dynamics. Nat Med 14, 985–990 (2008).
- Lo, Y. M. D. et al. Rapid Clearance of Fetal DNA from Maternal Plasma. Am J Hum Genetics 64, 218–224 (1999).
- Abbosh, C. et al. Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution. Nature 545, 446 (2017).
- Forshew, T. et al. Noninvasive Identification and Monitoring of Cancer Mutations by Targeted Deep Sequencing of Plasma DNA. Sci Transl Med 4, 136ra68-136ra68 (2012).
- Wan, J. C. M. et al. ctDNA monitoring using patient-specific sequencing and integration of variant reads. Sci Transl Med 12, (2020).
- Parsons, H. A. et al. Sensitive detection of minimal residual disease in patients treated for early-stage breast cancer. Clin Cancer Res clincanres.3005.2019 (2020) doi:10.1158/1078-0432.ccr-19-3005.
- McDonald, B. R. et al. Personalized circulating tumor DNA analysis to detect residual disease after neoadjuvant therapy in breast cancer. Sci Transl Med 11, eaax7392 (2019).
- Zviran, A. et al. Genome-wide cell-free DNA mutational integration enables ultra-sensitive cancer monitoring. Nat Med 1–11 (2020) doi:10.1038/s41591-020-0915-3.
- Sethi, H. et al. Abstract 4542: Analytical validation of the Signatera TM RUO assay, a highly sensitive patient-specific multiplex PCR NGS-based noninvasive cancer recurrence detection and therapy monitoring assay. 4542–4542 (2018) doi:10.1158/1538-7445.am2018-4542.