In a group of sponsored presentations at last week’s annual meeting of the Association for Molecular Pathology, Thermo Fisher Scientific product users reported on their experiences in a variety of clinical areas, including liquid biopsy, immuno-oncology, and pharmacogenomics.
According to Anagh Vora, head of medical affairs for clinical NGS and oncology at Thermo Fisher is eager to strengthen its longstanding position as a technology supplier while increasing the implementation of its tools in clinical practice.
“The ultimate goal for [us] is producing cutting edge, emerging technology that will change clinical paradigms,” Vora said.
“I think the mistakes a lot of companies make, including ourselves in the past, [is to focus on saying ‘this is our box… with all the bells and whistles … but its not just about creating technology … its about educating clinicians on how to implement the technology,” he added in an interview.
In one presentation made via video at the AMP meeting, MD Anderson oncologist Anthony Lucci discussed his and his colleagues’ work using Thermo’s pan-cancer Oncomine ctDNA assay for serial sampling in early-stage breast cancer and melanoma patients.
Although liquid biopsy broke through as a tool in advanced cancer monitoring, clinical researchers are increasingly applying circulating tumor DNA detection technologies to predict and monitor earlier-stage cancers, including colorectal cancer, lung cancers, brain cancers, and particularly the breast and melanoma tumors that MD Anderson has also targeted.
Lucci said that he and his team embraced Thermo Fisher’s Oncomine technology because it offers a standardized, kit-based workflow from DNA isolation through sequencing and interpretation, which helped the group’s studies stay as consistent as possible.
Thermo Fisher offers targeted Oncomine liquid biopsy panels, but the MD Anderson team wanted to be able to use the same tool in both breast cancer and melanoma, so they have adopted the company’s pan-cancer version.
The studies the group is engaged in are aimed at evaluating and hopefully validating, this type of technology in a few different areas: for example, identifying signs of a cancer recurrence in women who have had surgery to remove their breast tumor earlier than would otherwise be possible, or tracking tumor DNA in melanoma patients as an alternative to radiologic imaging, to monitor for recurrences, or to see if they are responding to new treatment.
MD Anderson has been collecting serial blood samples in breast cancer since 2004, which has created a rich cohort to explore whether liquid biopsy can provide earlier detection of cancer recurrences.
Efforts in melanoma began later, but have accrued participants very rapidly. Lucci said the group has over 850 melanoma patients enrolled so far. During the workshop he shared data from an analysis of a small subgroup, showing that the Oncomine panel has been yielding results with very high concordance to patients’ tumor tissue sequencing data.
If the team’s early results hold true, Lucci said that it looks like ctDNA will offer significant benefit to patients, and also to the healthcare system, by providing a cheaper alternative to currently used imaging and biopsy procedures.
“When you look at some of the data, lets say from non-small cell lung cancer, fine need aspiration is not only expensive, but has complications,” Lucci said in his video.
Moreover, CT-guided small needle aspiration can cost $4,000, he added, while the Oncomine pan cancer panel is only $1,000. In the case of melanoma, assuming a patient gets five CT scans over 8 months with some additional PET imaging, the cost could be over $22,000, he said. But using ctDNA over the same time points and time frame would cost less than half of that.
As fervent as the excitement around liquid biopsy has become, the pathology community is also increasingly focused on new methods to identify patients who will respond to cancer immunotherapy — mainly using tumor tissue right now, but potentially also via blood-borne DNA in the future.
In another Thermo Fisher-sponsored presentation, Mark Stewart, vice president of science policy at Friends of Cancer Research, provided an update of the organization’s progress in harmonizing various tumor mutational burden assays or approaches — a program that Thermo, along with a variety of other industry leaders, have joined.
On the heels of Foundation Medicine’s launch of clinical tumor mutational burden testing, various groups, both commercial and academic, have published on the biomarker, creating a complex field of different assays and approaches that need to be compared to one another. For example, just ahead of the AMP meeting Illumina launched a new next-gen sequencing pane, TruSight Oncology 500, which it highlighted as a tool for TMB assessment.
FOCR is working to establish how various TMB cutoff points and measurement strategies relate to one another. Stewart said during the AMP meeting that the group hopes to have the empirical cell line analysis part of its effort complete by the end of this year or early next.
Jin Li, research director of advanced diagnostics at contract research firm Molecular MD, described in a second presentation how his firm evaluated Thermo’s Oncomine Tumor Mutation Load (TML) panel, which the company began offering for clinical research and pharmaceutical development customers this August.
The Oncomine TML assay interrogates 409 cancer-related genes spanning about 1.7 megabases and can be applied to as little as 20 nanograms of formalin-fixed paraffin-embedded tumor DNA in what Thermo Fisher says is a three-day workflow.
The low sample input requirement especially made the product attractive to Molecular MD, Li said, for its potential to be applied to the greatest number of samples.
Li and colleagues were able to confirm that the assay’s assessment of TML highly correlates with tumor mutational burden as determined from whole-exome sequencing data.
As much ballyhoo as there now is around tumor mutation load, or burden, the field of oncology has also recognized that the factors that influence whether a cancer immunotherapy works or not are much more complex than the number of mutations present.
Reflecting this, Thermo Fisher vice president of product management, clinical next generation sequencing and oncology Andy Felton also spoke at the meeting about research the company is conducting using tools it has developed for measuring other aspects of the activity of the immune system in response to a cancer.
“We believe that this is a different way to go,” said Thermo’s Vora. “Its still in the translational stage, but the reason to bring it up in this type of presentation is to tell clinicians that this is the next wave,” he added.
In his presentation, Felton discussed the company’s experiments with its recently launched TCR-beta short-read and long-read assays, highlighting examples of how the company has used it so far, and “how it possibly could be used in the future to predict response to things like checkpoint inhibitors.”
The short-read technology is AmpliSeq-based and runs on the Ion GeneStudio S5 or earlier-generation platforms, Felton said. It works with DNA or RNA, and includes a back-end informatics package that creates visualizations of the immune cell repertoire, including clonality metrics and allele usage.
For the long-read version, AmpliSeq primers are still used, but they provide a view through additional regions that can contribute more information.
In collaboration with MD Anderson and the Roswell Park Cancer Center, Thermo Fisher researchers conducted a study in 55 individuals using the long-read assay to look not only at the classic complementarity determining region-(CDR) 3 region, but also at CDR 2 and CDR 1 areas to try to find a signal that was predictive of adverse events in immunotherapy patients.
The concept, Felton explained, is that as a tumor throws out neoantigens over potentially long periods of time, the immune system is stimulated to produce T-cell receptors that code for the same amino acid, but differ in their nucleotide spacing.
“You have to be particularly careful,” because these are single nucleotide variants … and there is a huge amount of diversity, so any error rate arising from substitution error can masquerade as an increase in the frequency of converging T cell populations.” Because of this, the low substitution error rate of Ion Torrent sequencing is a boon to this type of effort.
In a set of about 20 adenocarcinoma and melanoma patients, the Thermo team was able to distinguish those who had a response to checkpoint inhibitor therapy from those who didn’t, with receiver operating curves similar to currently used tumor mutational burden assays.
This was a relatively small study, so the results are only preliminary, but Felton said the group is now following up with a larger cohort.
“We think this may be the most direct measure of tumor immunogenicity,” he added.