This is Part 3 in a 3-part Q&A blog series with a panel of liquid biopsy experts addressing many of the issues faced in developing and deploying NGS-based liquid biopsy assays for clinical applications in oncology.
At a 2020 liquid biopsy webinar, Dr. Vollbrecht shared a molecular pathologist’s perspective on the current state of liquid biopsy. Laboratory processing and analysis of cfDNA samples is a multi-step process that requires a high degree of precision to achieve consistent results. Her presentation focused on pre-analytics variables, which are often left out of discussions and tend to focus on biochemical manipulation of isolated nucleic acids. Seemingly simple factors at the point of sample collection such as problems with blood test tube filling, storage and labelling are able to affect the cfDNA stability, abundance, and confound the reliability of final interpretation. Variation in sample treatment during laboratory processing, including but not limited to, cfDNA quantification and QC methodology are also amongst the challenges for liquid biopsy.
NGS assays focused on liquid biopsy utilize the high data output from current NGS instrumentation to achieve deep sequencing coverage depth. High depth of read coverage when paired with unique molecular identifiers (UMIs) allows for analysis of low VAFs while mitigating index hopping as well as the error rate inherent to NGS. NGS provides functionally limitless sequencing bandwidth, so the amount cfDNA available to produce a NGS library becomes a major limiting factor for maximizing assay sensitivity. Current mulutplex-PCR-based assays recommend 10 and 20 ng cfDNA as input into library preparation, which is not always achievable in the clinical setting with limited patient material. In order to maximize the amount of cfDNA available for library preparation, using cfDNA from plasma extracted from multiple blood collection tubes is preferable, when possible, to fully scrutinize the patient for markers of cancer. While amplification of isolated DNA can amplify picogram quantities of DNA to produce sufficient DNA to carry out NGS analysis, both the total number of genome equivalents and the efficiency of library incorporation will ultimately determine what the functional VAF limit of detection is for a given sample.
Robust detection of low VAF is of critical importance considering that liquid biopsy analysis is a proverbial needle in the haystack; a standard blood collection tube may contain billions of cfDNA fragments but perhaps only dozens of fragments will contain pathogenic variants at the region(s) of interest. To ensure an optimal patient outcome, verification of assays and the whole workflow with reference material is necessary to determine the function limit of detection of the assay and to build confidence in results. For example, the detection of mutations down to 0.1% VAF is usually required for reimbursement by the EBM (“Einheitlicher Bewerungsmaßstab”, german reimbursement system) standards.
Claudia outlined a few future avenues for liquid biopsy applications to close out the presentation. Increasing the number of variants scrutinized, adding routine PIK3CA testing in breast cancer patients plus expanding the scope to test for other resistance mechanisms is one direction for improvement. Additionally early breast cancer detection by whole exome sequencing of primary tumor followed by specific marker detection in blood for minimal residual disease (MRD) monitoring is another under-utilized area of inquiry. Lastly expanding the scope of specimens used in diagnostic assays to include biomarker detection in exosomes and circulating tumor cells will expand our ability to detect disease early and improve patient outcomes.
Q&A with Dr. Claudia Vollbrecht:
What would be your high-level advice for attendees interesting in adopting ctDNA when moving from tissue to ctDNA testing for the first time?
Dr. Vollbrecht: It is important to be in close contact with the clinicians to address any pre-analytical factors and interpretation of results.
Did you see a rise in the number of requests for liquid biopsy in diagnostics over the past couple of years? Do you feel that the oncologists understand the power of this technology?
Dr. Vollbrecht: We started four years ago and get a constant low number of samples. I think one crucial factor influencing the request for liquid biopsy analysis is reimbursement. At the beginning, liquid biopsy testing in molecular pathology was only reimbursable for EGFR T790M resistance testing in non-small cell lung cancer (NSCLC) patients. Currently EGFR primary testing is also included. It is possible that, with the results from the SOLAR1 trial for the approval of Alpelisib for PIK3CA-mutated breast cancer, the number of requests for liquid biopsy testing will increase.
How do you centrifuge the PAX tubes?
Dr. Vollbrecht: We usually get Cell-Free DNA BCT tubes (Streck) which we centrifuge 2000g at room temperature for 10 minutes to separate the plasma from the other blood components. Afterwards we centrifuge the plasma a 2nd round at 18.000g at room temperature for 10 minutes to separate cell residues from plasma. We used PAX tubes in the beginning and processed them in the same way.
What method was used for extraction from 2 ml of plasma? What was the yield?
Dr. Vollbrecht: We use the Maxwell 48 with the magnetic bead based ccfDNA plasma kit from Promega. With the custom protocol we can extract cfDNA from 1ml to 4ml plasma in one extraction. Average cfDNA concentration is 21ng/ml plasma.
I was told cfDNA tubes may “capture” smaller cfDNA fragments over time. Do you have any experience to share?
Dr. Vollbrecht: Quite interesting point. We have no experience with that.
What is your preferred approach for ongoing liquid biopsy-based monitoring for cancer recurrence?
Dr. Vollbrecht: I think monitoring of cancer specific mutations in blood after, for example whole-exome analysis of the primary tumor, by a very sensitive approach like digital PCR makes sense. There are already promising approaches: TRACERx study for NSCLC patients1 and TARDIS-Test2.
Could you comment on the clinical utility of liquid biopsies for the clinical detection of ctRNA and gene fusions? What are your experiences?
Dr. Vollbrecht: I think in general it makes sense to use ctRNA approaches to detect therapeutic relevant fusions. We already tested an Oncomine Lung cfTNA NGS assay, which worked well. Still, we have the issue with missing reimbursement and furthermore analysing cfRNA would mean to use different blood collection tubes, which also complicates the process.
Guidelines are in place for the use of liquid biopsy at recurrence as an alternative for tissue biopsy. Do you see liquid biopsy replacing tissue biopsy at point of diagnosis?
Dr. Vollbrecht: Currently we still need a histology-based diagnosis for therapy decision in solid tumors, but liquid biopsy can be a helpful tool for molecular testing in patients in conditions too bad to undergo a tissue biopsy.
What do you see as the next steps in liquid biopsy studies?
Dr. Vollbrecht: There are many interesting and promising studies out there to check out, such as:
- A broader scope for NGS approaches3
- Tumor Mutational Burden (TMB) assessment to guide immunotherapy: POPLAR and OAK studies for atezolizumab in NSCLC showed association of ctDNA TMB with clinical outcome for TMB >16mut/mb4,5
- Cancer cells tend to accumulate mutations over time, and after the cancer reaches a certain density of mutations per megabase of genome sequence it will be more likely to be susceptible to checkpoint inhibitor therapy
- Actionable TMB scores appear to be specific to cancer subtypes
- MRD monitoring and longitudinal testing: TRACERx study of NSCLC patients using a patient-specific multiplex PCR panel1
- As called out in the introductory comments, after a primary tumor is excised and analysed by WES, a subset of mutations specific to the patient can be monitored in order to monitor return of the cancer
- Reducing the number of targets to a smaller panel allows for very high coverage to become economically feasible such that clinicians can observe cancer emergence at sub 0.1% VAF in a cfDNA sample
- Methylation profiling to indicate tumor site origin6,7
- Many DNA mutations are well characterized with various cancers, but a challenge with liquid biopsy is that cfDNA could potentially originate from any part of the body
- If a tumor has been excised and the signature mutations become detectable in a blood draw, it is unclear from DNA sequence alone if the cancer has returned at the site of the primary lesion or has metastasized
- Analysis of the pattern of methylated base pairs appears to be sufficiently predictive of tissue of origin to greatly improve the utility of liquid biopsy for recurrence monitoring and potentially early screening as well
Download the video to replay or watch the webinar The Promise of Liquid Biopsy for Cancer Diagnostics and Therapeutic Monitoring: Are We There Yet?
References:
- Abbosh C, et al”.Early stage NSCLC - challenges to implementing ctDNA-based screening and MRD detection”. Nat Rev Clin Oncol. 2018;15(9):577-586. https://doi.org/10.1038/s41571-018-0058-3
- McDonald BR et al, “Personalized circulating tumor DNA analysis to detect residual disease after neoadjuvant therapy in breast cancer”. Sci Transl Med (2019): Vol. 11, Issue 504, https://doi.org/10.1126/scitranslmed.aax7392
- Rothwell, D.G. et al. Utility of ctDNA to support patient selection for early phase clinical trials: the TARGET study. Nat Med 25, 738–743 (2019). https://doi.org/10.1038/s41591-019-0380-z
- Gandara, D.R. et al. “An exploratory analysis of on-treatment ctDNA measurement as a potential surrogate for overall survival for atezolizumab benefit in the OAK study”. Annals of Oncology, Volume 30, v642 - v643 https://doi.org/10.1093/annonc/mdx380
- Yu, Y et al. “Association of survival and blood-based genomic signature with atezolizumab for patients with second-line and third-line EGFR wild-type non-small cell lung cancer: Pooled analysis of individual patient data from the POPLAR and OAK trials”. Annals of Oncology, Volume 30, ix107 https://doi.org/10.1093/annonc/mdz438
- Shen, S.Y. et al. “Sensitive tumour detection and classification using plasma cell-free DNA methylomes”. Nature 563, 579–583 (2018). https://doi.org/10.1038/s41586-018-0703-0
- Ulz, P. et al. „Inference of transcription factor binding from cell-free DNA enables tumor subtype prediction and early detection”. Nat Commun 10, 4666 (2019). https://doi.org/10.1038/s41467-019-12714-4
