NEW YORK – A new experimental blood test could provide clinicians with information on pancreatic cancer patients who won't respond to traditional treatments such as chemotherapy.
The test, developed by researchers at Van Andel Institute in Grand Rapids, Michigan, measures the levels of a certain glycan, sTRA, to determine which cancer subtype patients have and whether it will react to chemotherapy and/or surgery.
Brian Haab, an author of the study published last month in Clinical Cancer Research, said that the development of the test began with the recognition that pancreatic cancers "are generally resistant to treatment," which typically include surgery and the current standard of care for patients with unresectable disease, chemotherapy. However, half of chemotherapy-treated patients show no benefit at all, and the treatments can be debilitating and high-risk, Haab continued. "They tend to relapse after surgery within a year, and that's the same as if there were no chemotherapy," he said.
There are at least two subtypes of pancreatic cancer, classical and basal or squamous, with classical generally being more resistant to chemotherapy as shown by gene expression, Haab said.
This research was specifically focused on glycans, or carbohydrates, which are biomarkers elevated in around 60 percent of pancreatic cancers and can be found by gene expression. The glycan that the test developed by Haab and his colleagues detects is sTRA, which is particularly useful in combination with another glycan, CA 19-9. The new test was developed from an earlier test with a study published in January 2019 in Clinical Cancer Research that combined a diagnostic test for CA 19-9 with a test detecting sTRA. Classical pancreatic cancers are primarily the ones that make the sTRA glycan biomarker, Haab said.
"The advantage of this biomarker is that we now have a way to detect with a blood specimen the ones that are going to be resistant to chemotherapy and that will not benefit at all from standard chemotherapy," Haab said.
The researchers noted that patients with cancers with glycans have rapid relapse following surgery and chemotherapy, Haab said. There are other glycans with similar structures, and the team used commercially available antibodies to scan combinations of those glycans and found CA 19-9 and sTRA are a good combo for detecting and diagnosing cancer, he said. The two glycans are complementary and if a patient has high levels of either, it is likely they have cancer. Some patients don't make either glycan, so Haab noted they may miss those patients, but with the combination diagnostic they can pick up about 70 percent of all pancreatic cancers. Of that 70 percent, about half of those patients have high levels of sTRA, he added. CA 19-9 is not indicative of subtype, but only provides information on whether a patient has cancer and potentially a higher tumor burden, Haab said.
However, Cesar Castro, director of the cancer program at the Massachusetts General Hospital Center for Systems Biology and an assistant professor of medicine at Harvard Medical School, emphasized caution in depending on these biomarkers. "There are many 'tumor markers' evaluated through blood tests that are almost invariably imperfect and lack the specificity for cancer," he said. "For example, infection, pregnancy, or inflammation can cause elevation of such markers even in the absence of cancer."
Castro, who is not involved in the research Haab and his team are doing, added that "the fact that sTRA can be potentially incorporated into the usual clinical workflow makes it a potentially compelling proposition once validated."
Beyond blood tests, the only option currently available to determine a pancreatic cancer subtype has been to get a sample of the tumor, but getting pancreatic cancer biopsies is high-risk and particularly difficult, Haab added. There is another biomarker, GATA6, that can be detected via biopsy that picks up the same subtype as sTRA, but Haab said it is difficult to get the right cells and there is potential for serious complications.
But the new test is an antibody-based sandwich immunoassay in a microtiter plate that determines sTRA levels from blood samples, bypassing the need for a biopsy. There are two different versions of the immunoassay that utilize two different commercially available capture antibodies – one CA 19-9 antibody, and one against a mucin, either MUC5AC or mucin-16. Once those antibodies capture the proteins, the researchers treat them with an enzyme that exposes the epitope of the detection antibody, which is normally masked by a monosaccharide. This detection antibody is TRA160, a commercially available monoclonal antibody that has been in use since the 1980s, Haab said. "We run it in in a non-standard way that involves the enzyme, epitope exposure step," he added, which is patented by the Van Andel Institute.
The test can run on several different enzyme-linked immunosorbent technologies with color-based readouts, Haab said. It can also be done with fluorescence-based readout on microarrays.
The pre-treatment test hasn't been validated on metastatic patients, Haab was quick to note, but in those patients who are candidates for chemotherapy and surgery that were tested, it had 98 percent specificity for patients who would respond to treatment and can pick up 50 to 60 percent of patients with the resistant kind of cancer. "We don't want to have false positives … so if a patient's going to benefit, we don't want to deny them the treatment," he said.
Those patients are likely receiving both chemotherapy and surgery in combination if they're able, although some aren't healthy enough or have complications from surgery and aren't able to receive chemotherapy, Haab said.
"If they're not going to benefit at all, they should be spared these very toxic treatments," Haab said, especially since pancreatic cancer surgery is very difficult and has a death rate of nearly 2 percent. "If someone's not going to benefit, they could be put onto some sort of alternative treatment, some sort of other plan where they at least would have a better quality of life," he said.
Castro noted some limitations with the study, saying that the general transformation of cancer subtypes with sTRA have the capability to transform into "more mesenchymal features" with the transformation being seen more in aggressive tumors. "So, the actual resistance to chemotherapies may simply be a function of more aggressive properties," he said. "As such, sTRA may be one of various other metrics that would arrive at a similar conclusion of a cancer's potential clinical trajectory."
There are currently two tracks of follow-up research ongoing, with one focused on understanding the biology and metabolism of this subtype that produces glycans to help develop appropriate treatments, and the other developing a clinical assay and preparing to run a prospective clinical study to "further validate the association between high pretreatment levels of the marker and rapid relapse following surgery and chemotherapy," Haab said.
Haab and his team have spent the past year developing the clinical assay and is working with a couple of regional hospitals to get fresh samples for the validation study, he said. The primary goal of the clinical assay is for its use as a diagnostic with both CA 19-9 and sTRA. A secondary goal of the team is to examine outcomes related to sTRA, he continued. The study is in the final stages of planning, with confirmations between hospitals providing samples and some cost measures still under discussion, he noted.
Castro cited the need for broader validation looking at reproducibility and clinical actionability, such as whether it would change clinical behavior by cancer providers. "If it doesn't move the needle, it is not worth pursuing further," he said, but if the test works, it could "guide providers on guard for cancer recurrence to seek broader cancer imaging sooner rather than later."
The team wants to make the test available for doctors for both diagnosis and subtyping, and depending on the results of the prospective study, it will offer the diagnostic test as a laboratory-developed test. Haab said he is hopeful it could become a component of care for pancreatic cancer patients and could run on multiple diagnostic companies' platforms, including those from Beckman Coulter and Abbott.