NEW YORK (GenomeWeb News) – In Science Translational Medicine, researchers from the UK and the US have demonstrated the feasibility of doing blood-based 'liquid biopsies' using a targeted deep sequencing strategy that can uncover mutations in several genes using tumor DNA that is circulating in the bloodstream.
The team tested their approach — called tagged-amplicon deep sequencing, or Tam-Seq — first using preserved tumor samples and then using blood samples from women with advanced ovarian cancer, looking for low-frequency, cancer-related mutations in almost 6,000 bases of sequence representing cancer-related genes.
As they reported today, the investigators were able to detect mutant alleles present at frequencies as low as 2 percent with more than 97 percent sensitivity and specificity.
Based on their success in using Tam-Seq to non-invasively profile mutations in individuals with ovarian and other cancer types, those involved in the study are touting the technique's potential for cancer detection, characterization, and treatment tracking.
"This sequencing approach allowed us to monitor changes in tumor burden by sampling only patient plasma over time," co-corresponding author Nitzan Rosenfeld, an oncology researcher with the Li Ka Shing Centre and the University of Cambridge, and colleagues wrote. "Combined with faster, more accurate sequencing technologies or rare allele amplification strategies, this approach could potentially be used for personalized medicine at point of care."
While several groups have taken a crack at testing for specific mutations or sets of mutations in circulating tumor DNA, Rosenfeld and his colleagues explained, characterizing all of the mutations within numerous genes of interest using circulating tumor DNA has been more difficult.
"Various methods have been optimized to detect extremely rare alleles, and can assay for predefined or hotspot mutations," they wrote. "These methods, however, interrogate individual or few loci and have limited ability to identify mutations in genes that lack mutation hotspots, such as the TP53 or PTEN tumor suppressor genes."
In an effort to get around such issues — while at once developing a technique that is sensitive enough to pick up even low-frequency mutations — the team came up with a strategy for doing deep sequencing on tagged amplicons that tile across sequences of interest. The method involves a multi-step amplification that relies, in part, on Fluidigm's Access Array microfluidic system.
Researchers pre-amplify numerous regions of interest using pooled primers before doing more focused single-plex PCR on these sequences, they explained. From there, a barcoding PCR step tags the amplicons and adds on sequencing adaptors so that the barcoded amplicons can be multiplexed and deep sequenced.
For the current study, the team used 48 primer pairs to target 5,995 bases of sequence from coding regions of the TP53 and PTEN genes and specific sections of the EGFR, BRAF, KRAS, and PIK3CA genes. After generating tagged amplicons, these samples were deep sequenced by multiplex, single-end sequencing on the Illumina GAIIx.
To initially try out the Tam-Seq approach, researchers started with several dozen formalin-fixed paraffin-embedded tumor samples. There, they found that the Tam-Seq approach detected all of the TP53 mutations found by Sanger sequencing along with a few additional alterations in the gene.
From there, the team turned their attention to blood samples — first testing TAm-Seq's ability to distinguish between SNPs from two healthy individuals whose blood samples were mixed together and then using it to look for cancer-related mutations in blood samples from women with high-grade serous ovarian carcinoma.
For example, in blood samples from seven women with advanced ovarian cancer who had high levels of circulating tumor DNA and known TP53 mutations, researchers found all of the TP53 changes with TAm-Seq. The method also unearthed a previously undetected EGFR mutation in a blood sample from one woman experiencing relapse.
When the investigators compared Tam-Seq to digital PCR for finding mutations in 62 blood samples from a larger set of women with advanced ovarian cancer, meanwhile, they detected 40 point mutations by PCR and 39 by Tam-Seq.
At an average sequencing depth of 650 times, they reported, Tam-Seq uncovered mutations present at around 1 percent to 2 percent allele frequency.
Moreover, using blood samples collected regularly from two more ovarian cancer patients over the course of their treatment, follow-up, and relapse, the researchers showed that they could trace relapse-related mutations back to specific samples biopsied during initial surgery for one of the women. In the other woman, they used Tam-Seq to track mutation frequencies over the course of treatment and subsequent relapse.
The team's subsequent experiments suggest it's possible to tweak TAm-Seq to find genes that are mutated in other types of cancers and to determine the origin of metastases in individuals with primary tumors at multiple sites in the body.