In the Journals Plus

STAG2 identified as biomarker for bladder cancer recurrence, progression

Show Citation

June 28, 2018

Photo of Todd Waldman
Todd Waldman

STAG2 mutations served as an independent predictor of progression and recurrence of papillary nonmuscle-invasive bladder cancer, according to results of a validation study.

“Nonmuscle-invasive bladder cancer is extremely common and is extremely expensive to treat,” Todd Waldman, MD, PhD, professor of oncology at Georgetown Lombardi Comprehensive Cancer Center and director of the Georgetown MD/PhD Dual Degree Program, told HemOnc Today. “This new biomarker could help make it possible to better identify those patients who would and would not benefit from extensive postresection surveillance by cystoscopy and adjuvant therapy.”

Despite surgical resection and outpatient cystoscopy-based procedures, 70% of nonmuscle-invasive bladder cancers recur locally, of which nearly 20% progress to muscle invasion. It has remained unknown which nonmuscle-invasive bladder cancers have a greater likelihood of recurrence or progression; clinicians have relied on intensive postresection surveillance.

STAG2 occurs in approximately 35% of nonmuscle-invasive bladder cancers.

A pilot study conducted by Waldman and colleagues in a cohort of 34 patients from The University of Texas MD Anderson Cancer Center showed one of eight (12%) STAG2-deficient tumors recurred, compared with 15 of 26 (58%) STAG2-expressing tumors (P = .05).

Based on those findings, Waldman and colleagues hypothesized that nonmuscle-invasive bladder cancers with deficient STAG2 genes may be less likely to recur and progress to muscle invasion than those with STAG2 expression.

In the current validation study, researchers used the immunohistochemistry-based assay to identify tumors in two cohorts of papillary nonmuscle-invasive bladder cancers tumors that harbored STAG2 mutations: a cohort of 82 tumors known as the Georgetown cohort, and a cohort of 253 tumors known as the Aarhus cohort.

All patients in the Georgetown cohort had been treated at Medstar Georgetown University Hospital before December 2012. Sixty-nine patients received no adjuvant therapy, nine received adjuvant Bacillus Calmette-Guerin immunotherapy, three received mitomycin C and one was treated with thiotepa.

Among the Aarhus cohort, 66 patients received adjuvant Bacillus Calmette-Guerin immunotherapy, three received intravesical chemotherapy and one received both.

All tumor sample studies came from the initial diagnosis without any sign of recurrence.

Among the Georgetown cohort, 52% of tumors with STAG2 expression recurred compared with 25% of STAG2-deficient tumors (P = .02).

Multivariable analysis showed STAG2 expression independently predicted recurrence (HR = 2.4; 95% CI, 1-5.81).

However, researchers noted these multivariate analyses were limited because all clinical and pathologic characteristics could not be included. Further, the Georgetown cohort lacked statistical power to assess the prognostic value of STAG2 expression on progression because only 6% of the patients progressed to muscle invasion during the follow-up period.

As a result, researchers used the Aarhus cohort — which was enriched for nonmuscle-invasive bladder cancer that progressed to muscle invasion — to evaluate the prognostic value of STAG2 expression on progression.

Among the tumors, 186 (74%) were STAG2 positive; 61 (24%) were STAG2 negative and six (2%) were mosaic.

Thirty-eight percent of tumors with STAG2 expression progressed within 5 years compared with 16% of STAG2-deficient tumors (P < .01).

Independent predictors of progression included STAG2 expression (HR = 1.86; 95% CI, 1-3.43), as well as increasing age at diagnosis (P = .03) and high-grade disease (P = .01), according to multivariable analysis.

STAG2 has a more powerful ability to predict which patients with resected nonmuscle invasive bladder cancer will recur and progress to muscle invasion,” Waldman said. “STAG2 is a new biomarker for predicting the clinical behavior of early-stage bladder cancers.”

Waldman noted that the design of the study — a retrospective biomarker validation study — was the biggest limitation in its findings.

“A prospective study will happen next,” Waldman said. – by Melinda Stevens

For more information:

Todd Waldman, MD, PhD, can be reached at Georgetown University School of Medicine, NRB E304, 3970 Reservoir Road NW, Washington, DC 20057; email:

Disclosures: The authors report no relevant financial disclosures.

itj+ Infographic

itj+ Perspective