NEW YORK – Roughly three years after its launch, the Applied Proteogenomics Organizational Learning and Outcomes (APOLLO) Network has moved beyond its initial focus on lung cancer to add analyses of multiple cancer types including a prospective, multisite study.
The network is also working to incorporate additional proteomic technologies including data-independent acquisition (DIA) mass spectrometry and imaging and top-down mass spec, said Thomas Conrads, director of women's health research at Inova Fairfax Hospital and a researcher on the APOLLO initiative.
Announced in July 2016, the project is a collaboration between the National Cancer Institute's Clinical Proteomic Tumor Analysis Consortium (CPTAC), the US Department of Defense, and the US Department of Veterans Affairs. It aims to use a variety of proteomic and genomic tools to analyze cancer samples from patients in the VA system with the ultimate goal of using proteogenomic information to guide patient therapy.
The initiative consists of six subprojects running concurrently, with APOLLO One through Four focused on analysis of, respectively, lung, ovarian, prostate, and breast cancer, primarily using biobanked samples. APOLLO Six is focused on pancreatic cancer, also using biobanked samples.
The APOLLO Five subproject is a prospective trial enrolling cancer patients at 10 DoD-run medical centers and three VA sites across the country.
That trial will cover all cancers, said Craig Shriver, head of the APOLLO network and director of the John P Murtha Cancer Center at Walter Reed Hospital.
"We want to provide this as a benefit to active duty service members diagnosed with cancer so we can molecularly profile their cancers and potentially identify gene mutations and other actionable targets that can be used therapeutically," he said.
Shriver noted that the patients enrolled in the APOLLO Five trial would receive standard of care but that the data generated via proteogenomic analysis of their tumors could potentially be used to guide additional treatment should it be needed.
"This isn't like NCI-MATCH where we have a specific turnaround time… and they are waiting to find out what therapy they may or may not be eligible for," he said.
Shriver said that APOLLO Five patients would receive germline sequencing in addition to somatic sequencing.
"So if we identify a germline mutation that could be targeted as part of screening or prevention efforts or even targeted therapy including surgeries, then the patient will be notified," he said.
Shriver said the APOLLO Five project has begun enrolling patients at the Palo Alto VA Medical Center and plans to start enrollment at around half the project sites within the next six months.
Proteogenomics has emerged in recent years as an area of significant interest in cancer research, the hope being that integrating proteomic and genomic data from, for instance, a tumor sample, will provide a better understanding of the molecular character of a patient's cancer and help inform treatment.
For instance, past studies have shown that potentially meaningful proteomic changes are not always present at the genomic level, which means that information that could enable better targeted treatments might be missed without proteomic analysis.
Additionally, there is hope that protein-level data could aid interpretation of the large numbers of genomic alterations identified by next-generation sequencing because proteins are the functional molecules that mediate many biological processes, and a genetic mutation that leads to an actual protein change is more likely to be disease-related than one that doesn't.
Conrads cited as an example early results from the APOLLO One project in which researchers identified genomic mutations that led to alterations in lung-specific proteins known to be linked to inflammatory disorders and other lung pathologies.
"Without that [protein profile], you wouldn't necessarily predict that that genotype would lead to a risk of developing cancer," he said. "So I think that is where we will find some very nice and interesting vignettes within these pilot projects that we hope are going to really bear out in APOLLO Five with the large prospective collection."
The NCI's CPTAC project also has a proteogenomic focus, but Shriver said the APOLLO initiative would supplement and expand upon that project's work, as opposed to duplicating it. For instance, he said, that APOLLO would help expand the pool of samples from populations not as well represented in the CPTAC sample sets, many of which were sourced from The Cancer Genome Atlas [TCGA] project, which he noted has been criticized for lacking racial and ethnic diversity.
"TCGA was obviously a great project, but we can perhaps learn even more looking at more diverse sample sets," he said. "And we in the DoD are well positioned to do that. Our existing biobanks contain really enriched [minority] subpopulations."
Conrads also noted that the APOLLO effort is distinguished by its focus on the tumor microenvironment, as opposed to CPTAC's focus on whole tissue digests.
"Everything that we are doing is focused on really disentangling the cancer and the tumor microenvironment," he said. "We are employing laser microdissection to both profile cancer epithelium along with cancer stroma, because … we are not excluding tissues based on tumor purity or cellularity, so we are really looking at a much broader spectrum of tumor phenotypes than has been profiled before in TCGA, for example."
At the recent American Society for Mass Spectrometry annual meeting, researchers presented work from the APOLLO Two project on ovarian cancer. Among their results was the identification of a subset of a tumors exhibiting proteomic changes indicative of increased immune cell movement but decreased inflammatory signaling as well as gene expression signatures suggestive of immune cell infiltration. These patients had 1.33 additional years of progression-free survival compared to patients with tumors without increased immune activity.
Conrads said that as part of its focus on the tumor microenvironment the consortium is looking to add a pair of new researchers —Northwestern University's Neil Kelleher and Vanderbilt University's Richard Caprioli, who specialize, respectively, in top-down proteomics and imaging mass spectrometry.
He said that imaging mass spectrometry could prove particularly useful for analyzing samples with low levels of tumor cellularity to better understand how these cancers remodel their environment.
"In many of these very bad-acting cancers, there is this massive desmoplastic reaction where the tumor microenvironment has a complete remodeling and in fact the tumor makes up a very small portion of that cellularity," he said. "We don't really know a lot about that reactive microenvironment, and so we think imaging mass spectrometry is going to give us a unique view that will augment the laser microdissection we are doing to purify those different cell subsets."
The researchers are also collaborating with Thermo Fisher Scientific and researchers from the International Cancer Moonshot project to assess DIA mass spec methods for tumor analysis. Last year, in an effort led by Conrads, 11 labs in nine countries did high-throughput testing of the performance and reproducibility of DIA mass spec.
Conrads said that the APOLLO researchers now plan to perform DIA mass spec on the biobanked samples previously analyzed using isobaric labeling, which has been the primary mass spec workflow used by the initiative to date.
"With every one of the of the digests that have gone in for [isobaric labeling] we have reserved a small aliquot that we will be running across that same DIA platform [tested in the Thermo Fisher collaboration]," he said.
In addition to the isobaric labeling and DIA analyses, the researchers are also analyzing samples using reverse phase protein arrays and targeted mass spec assays.