Skip to main content
Premium Trial:

Request an Annual Quote

Georgetown Researchers Develop IHC Assay to Predict Bladder Cancer Recurrence, Spread


NEW YORK (GenomeWeb) – A international group of researchers led by the Georgetown Lombardi Comprehensive Cancer Center has developed an immunohistochemistry assay for the STAG2 gene, which they believe can act as a biomarker for papillary non-muscle-invasive bladder cancer (NMIBC).

Specifically, the team found that tumors that uniformly express STAG2 are twice as likely to recur and progress to muscle invasion as tumors that exhibit complete or mosaic loss of STAG2 expression.

Senior author Todd Waldman, an oncology professor at Georgetown School of Medicine, explained that his team had identified STAG2 as a potential biomarker in a previous study published in Nature in 2013, but the group needed additional samples to validate the marker.

"The STAG2 gene is mutated in about a third of early-stage bladder cancers, as several innate and unusual parts of the gene make it possible to detect these mutations very reliably at the protein level that the gene makes," Waldman said.

In a more recent study published last week in Clinical Cancer Research, Waldman and his team evaluated the value of STAG2 immunostaining in two cohorts of papillary bladder tumors from Georgetown and Aarhus University in Copenhagen, Denmark. The team searched for prediction of recurrence and progression to muscle invasion in the two groups, whose tumors had previously been removed during initial surgery for STAG2 over-expression.

Waldman explained that his team used antibodies to detect STAG2 expression in the 82 patients as part of the Georgetown cohort to predict tumor recurrence, as bladder cancer frequently appears in the spot that surgeons typically remove the tumor. In contrast, the team used STAG2 in the Denmark cohort to predict which bladder tumors will progress and invade the muscles surrounding the the bladder.

The team stained the tumor tissues with hematoxylin and eosin antibodies in order to detect the presence of truncated STAG2 proteins. The researchers explained that the immunostaining assay is particularly "robust" because the STAG2 protein only exists on the X chromosome, "so only a single mutation is required for complete gene inactivation," and because STAG2 is one of the most prolific proteins in the human proteome.

After immunostaining the cells, the researchers then sliced the cells and prepared them as formalin-fixed, paraffin-embedded tissue samples. In addition to using archived tumor samples provided by Aarhus University, pathologists then examined the tissue samples to determine tumor stage and grade.

In the Georgetown cohort, the team found that STAG2 inactivation occurred in 34 percent of cells, which they said confirmed their previous studies reporting that "32 percent of NMIBCs harbor STAG2 mutations." Meanwhile, 52 percent of of the tumor cells with intact STAG2 expression recurred, whereas only 25 percent of STAG2-deficient tumors recurred.

In order to identify whether STAG2 is an independent predictor of recurrence in the Georgetown cohort, the team performed a multivariable analysis on the tumor samples. The researchers found "that the hazard ratio for recurrence in patients with STAG2-expressing tumors was 2.4 times that of patients with STAG2-negative or mosaic tumors."

In the Aahurs cohort, however, the team found that STAG2 inactivation only occurred in 26 percent of cells. The researchers noted that the discrepancy in the Aarhus cohort may be a result of the samples being enriched for progressors, which they believe are less likely to harbor STAG2 mutations. About 38 percent of the progressor-enriched tumors with intact STAG2 expression progressed within 5 years, versus 16 percent of STAG2-deficient tumors. In addition, the team found that the hazard ratio for the progression in patients was "1.86 times that of patients with STAG2-negative or mosaic tumors."

Multivariable analysis in both cohorts therefore identified intact STAG2 expression as an independent predictor of bladder tumor progression.

According to Waldman, the overall process, from sample prep and assay in the wet lab to examining the sample under the microscope by a pathologist, ideally requires "less than a day using an automated system."

Waldman acknowledged that one of the study's major challenges included the assay's ability to only identify the approximately 85 percent of STAG2-mutant tumors whose mutations are considered truncated. The current assay misclassifies about 15 percent of tumors as the wild-type mutation due to the test's inability to detect missense mutations.

The researchers also acknowledged that the Georgetown cohort included a small sample size and "lacked complete information for several clinicopathologic parameters, including tumor size and multiplicity," which are normally used to stratify patient risk levels.

In order to address the data sparsity issues in the study, Waldman noted that his team is currently performing an additional biomarker validation study with a "prospectively collected cohort of bladder cancer tumors."

In the future, Waldman believes that the assay could be supported by performing DNA sequencing on bladder cancer samples in order to correctly classify the mutant tumors harboring missense mutations, eventually producing an assay with even stronger predictive power. Due to the biomarker's independence of other variables, the team believes that STAG2 could also be used for risk stratification in NMBICs.

"If the tumors were sequenced, you'd be able to more quickly identify STAG2 mutations, and while we haven't started that yet, we could combine the two steps to more effectively identify tumors," Waldman explained. "My only hesitancy is that ... we could add DNA sequencing, but that would make everything much more complicated and expensive. One of the great advantages [of the current assay] is how simple it is, and how it can be easily done by any department of pathology."

However, Waldman noted that as research techniques "move to the point where most tumors are probably going to be sequenced anyway, where that becomes the standard of care, we can potentially extract the STAG2 mutation status from that information."

The researchers also speculate that STAG2 may adjust the developmental state of stem cells that change into bladder epithelial cells. The potential mutated state may cause "aberrant papillary growth pattern but reduced ability to proliferate and progress, as compared with papillary NMIBC tumors" caused by genetic events besides than STAG2 mutation.

While the group does not have any current plans to commercialize the assay, Waldman does envision the test being used in the clinical space once researchers have performed additional validation studies on larger cohorts.

Waldman believes his team could eventually partner with firms to commercialize the assay in order to be accepted by the bladder cancer community as a routine diagnostic tool.

"We are not [currently] sure how the partnership would work practically at the moment ... whether with Georgetown's own office of technology licensing or with interested parties," Waldman said. "However, we are interested in making the assay for bladder cancer more widespread, and [potentially] having hospitals perform the assay in their pathology labs."

Waldman emphasized that the beauty of the test is "its simplicity, consistency with what the pathologists want, and [that] it gives [us] a discriminatory power to figure out who's going to recur and progress using a standard process."