Published online May 16, 2012 by a GenomeWeb staff reporter
A new Science Translational Medicine study suggests high-throughput sequencing is effective for finding post-treatment leukemia cells that can lead to patient relapse.
Researchers from Fred Hutchinson Cancer Research Center, the University of Washington, and elsewhere used high-throughput sequencing to test matched pre- and post-chemotherapy samples from 43 individuals with T-lineage acute lymphoblastic leukemia, looking at how well their high-throughput sequencing method detected lingering leukemia cells compared with flow cytometry.
Indeed, the sequencing method tracked down minimal residual disease in 25 of the cases tested, including several missed by flow cytometry, prompting study authors to suggest that the approach may eventually help in not only tracking leukemia treatment and relapse but also for diagnosing and/or classifying the disease.
“The ability to predict disease relapse sooner with high-throughput sequencing would give hematologists the option to treat cancer recurrence earlier, offering a greater chance of survival,” corresponding author Harlan Robins, a computational biologist at the Fred Hutchinson Cancer Research Center, said in statement.
“Longer term,” he added, “this technology potentially also could be used to initially diagnose leukemia and lymphoma much earlier than we can today.”
Past studies have shown that minimal residual disease following chemotherapy can be used to help classify T-ALL patients and predict relapse risk, researchers explained. At the moment, they added, leukemia cells remaining after treatment are primarily detected using flow cytometry-based methods or real-time, quantitative PCR with primers designed specifically for each patient.
In an effort to come up with a cheaper and more streamlined strategy for finding residual leukemia, Robins and colleagues came up with a method that relies on high-throughput sequencing to find telltale immune cell receptor patterns that coincide with the presence of disease.
To compare this method with flow cytometry, they started with matched pre- and post-treatment samples from 43 individuals collected at the University of Washington as part of the Children’s Oncology Group trial. For each of the participants, the team amplified targeted stretches of DNA from around 67,000 peripheral blood or bone marrow cells taken prior to chemotherapy as well as 200,000 bone marrow cells collected almost a month after treatment.
Using Illumina HiSeq 2000 sequencing, they then generated at least five-fold coverage of amplified DNA, which represented variable regions of the adaptive immune genes TCRB and TCRG, which code for T-cell antigen receptors.
From pre-treatment samples, researchers defined individuals’ specific leukemia-related TCRB and/or TCRG sequences. The frequency with which the same sequences turned up in samples taken 29 days after chemotherapy was then used to gauge leukemia persistence.
For instance, the team determined characteristic clonal sequences for so-called complementarity-determining region 3, or CDR3, of TCRB 31 of the 43 individuals using samples taken prior to treatment.
In four of the remaining 12 samples, the researchers relied on clonal TCRG rearrangements to track leukemia.
Another eight samples could not be tested further using the sequencing-based method owing to a lack of detectable pre-treatment clonal TCRB or TCRG sequence patterns.
For the 35 individuals who did have discernible TCRB or TCRG sequences, the researchers went on to look at how many minimal residual disease cases they could detect by sequencing compared to flow cytometry, a method that’s currently used for clinical minimal residual disease testing.
In the group of patients with defined pre-treatment TCRB CDR3 clonal sequences, the researchers saw a dozen minimal residual disease cases by both sequencing and flow cytometry as well as nine cases showing no residual disease by either method. For another 10 individuals, though, sequencing identified minimal residual disease that had been missed by flow cytometry.
Moreover, when researchers narrowed in on samples from individuals without distinguishable pre-treatment TCRB CDR3 clonal sequences, they found an over-representation of T-ALLs from a particularly aggressive sub-type of the disease known as ETP T-ALL as well as T-ALLs with some but not all of the ETP features.
Meanwhile, three of the four patients with distinguishable pre-treatment TCRG clonal sequence patterns had residual disease found by sequencing but not flow cytometry, while one case was found by flow cytometry but not sequencing.
Through follow-up experiments on additional T-ALL samples, the team found three false positive minimal residual disease cases in 1,344 individuals evaluated using TCRB sequences and 17 false positives in 1,512 patients tracked with TCRG sequences.
Authors of the study argued that the sequencing method decreases sample preparation time and complexity, making it a cost-effective way to detect T-ALL and perhaps other types of leukemia in a scalable manner.
Still, the researchers cautioned that there are limitations to the approach, which requires sufficient levels of DNA and currently hinges on the presence of certain clonal rearrangements in the immune genes targeted. In addition, they noted that further research is needed to verify that the sequencing-based test meets appropriate clinical standards.
The Fred Hutchinson Cancer Research Center has applied for patents related to the methods employed in the study and the Seattle biotech firm Adaptive Biotechnologies, which was co-founded by Robins, has reportedly secured an exclusive license for the core technologies.