Using Genetics to Improve Treatment Selection for Muscle-Invasive Bladder Cancers

THE TOPLINE

  • An accelerated neoadjuvant chemotherapy regimen is tolerable and effective for patients with muscle-invasive bladder cancer (MIBC)
  • Three genes can predict which MIBC patients are likely to respond to cisplatin-based chemotherapy
  • There are three distinct varieties of MIBC, which have biomarkers analogous to similar subtypes in basal and luminal breast cancers

Research by Fox Chase medical oncologist Elizabeth R. Plimack, MD, and colleagues is helping find appropriate treatments for patients with muscle-invasive bladder cancer (MIBC) by identifying different subtypes of the cancer and determining which patients are likely to respond to chemotherapy prior to surgical bladder removal.

Although radical cystectomy is the most common treatment for MIBC, cisplatin-based chemotherapy before surgery improves the prognosis for many patients by helping reduce the risk of relapse and shrinking tumor size. While 30 to 50 percent of patients benefit from such neoadjuvant chemotherapy (NAC), the rest do not. Plimack and colleagues are investigating ways to predict whether a patient will respond to NAC, so that patients are not unnecessarily exposed to chemotherapy or delays in surgical treatment.

One way to minimize time to surgery is to condense chemotherapy. The researchers conducted a trial with an accelerated neoadjuvant regimen of methotrexate, vinblastine, doxorubicin, and cisplatin (AMVAC) to examine its effectiveness and tolerability. They found that providing three cycles of the therapy in half the standard time (six vs. 12 weeks) yielded a significant patient response—15 of the 44 patients showed no sign of remaining cancer at the conclusion of the trial, and had few serious side effects. This finding, published in the Journal of Clinical Oncology, suggests that AMVAC works and can expedite patients’ time to surgery.

MIBC patients in the AMVAC trial agreed to donate their stored tumor tissue for research. Using these samples, Plimack and her colleagues sequenced the tumor DNA and found that patients who responded to AMVAC had a mutation in one of three genes associated with a tumor’s ability to repair its DNA. Responders also had higher numbers of mutations overall, suggesting that the cancer’s inability to repair its DNA leads to both accumulation of gene mutations and sensitivity to chemotherapy.

“If this test is confirmed to be predictive, we can test patients when they are diagnosed, identify patients who won’t benefit from NAC, and send those patients directly to surgery to save time,” says Plimack. “But if they carry at least one of these mutations, we can treat them with NAC knowing they are likely to respond.”

Experimental structure of the retinoblastoma-associated protein, Rb, with three positions marked that, when mutated, are associated with response to the AMVAC treatment. Each domain of the protein is shown in a different color. Photo by: Roland Dunbrack

Experimental structure of the retinoblastoma-associated protein, Rb, with three positions marked that, when mutated, are associated with response to the AMVAC treatment. Each domain of the protein is shown in a different color. Image by Roland Dunbrack

For another study that examined gene expression in MIBC, Plimack and her team shared samples of tumors from their AMVAC study with David McConkey, PhD, and his team of researchers at MD Anderson. Using RNA expression profiling, they identified three distinct varieties of MIBC: luminal, basal and “p53-like.” Like basal breast cancers, the basal MIBC subtype was characterized by p63 gene activation, squamous differentiation, and more aggressive disease at presentation. The luminal subtype showed active peroxisome proliferator-activated receptor gamma (PPARγ) and estrogen receptor transcription, and these tumors were enriched with activating fibroblast growth factor receptor 3 (FGFR3) mutations, suggesting they might be susceptible to FGFR inhibitors.

The p53-like subtype had biomarkers usually associated with the wild-type version of the p53 tumor suppressor protein—even though these tumors did not always express wild-type p53—and were highly resistant to cisplatin-based chemotherapy. Tumors of all subtypes that did not respond to NAC ended up expressing a p53-like signature after chemotherapy, indicating that this gene expression pathway may play an important role in chemoresistance.

Elizabeth R. Plimack, MDPlimack’s co-authors on the AMVAC study include Roland Dunbrack, PhD, Tim Brennan, MS, Yan Zhou, MSE, PhD, Mark Andrake, PhD, Roman Yelensky, PhD, Ilya Serebriiskii, PhD, Jean Hoffman-Censits, MD, Alexander Kutikov, MD, FACS, Katherine Alpaugh, PhD, Essel Dulaimi Al-Saleem, MD, Rosalia Viterbo, MD, FACS, Richard E. Greenberg, MD, FACS, David Chen, MD, FACS, Costas D. Lallas, MD, Yu-Ning Wong, MD, MSCE, Edouard J. Trabulsi, MD, Norma Palma, PhD, Vincent A. Miller, MD, Erica Golemis, PhD, and Eric Ross, PhD.