Home >> ALL ISSUES >> 2016 Issues >> Bladder cancer preps for its star turn

Bladder cancer preps for its star turn

image_pdfCreate PDF

Karen Titus

February 2016—A streak of sibling rivalry emerges when experts ponder progress in the field of bladder cancer. Whether it’s new markers or therapies, funding or advocacy, advances have come slowly, and the disease has long labored in the shadow of others.

“Urologic malignancies in general lag behind, compared to breast cancer and other tumor types, like colon and lung, where we’ve been envious for a while,” says George Netto, MD, professor of pathology, urology, and oncology and director of surgical pathology molecular diagnostics, Johns Hopkins University School of Medicine.

Now, it appears bladder cancer is making a run at most-favored-child status. Or, as Dr. Netto puts it, “In urologic malignancy, we’re catching up.”

Progress has been made in the field of bladder cancer, particularly in the past year or two, says Dr. Jim Zhai (left), here with Dr. Richard Joseph, who says, “The closest to the clinic is PD-L1.”

Progress has been made in the field of bladder cancer, particularly in the past year or two, says Dr. Jim Zhai (left), here with Dr. Richard Joseph, who says, “The closest to the clinic is PD-L1.”

He pauses, then adds, “Finally.”

“We’ve made a lot of progress, particularly in the last year or two,” agrees Jim Zhai, MD, professor of pathology and laboratory medicine, and consultant pathologist, Mayo Clinic, Jacksonville, Fla. “The bad news is it’s taken 30 years for a breakthrough.”

The most spectacular leap concerns the use of anti-PD-1/PD-L1 immunotherapy to treat patients with metastatic urothelial bladder cancer. Physicians have known for some time that a subset of bladder cancer tumors express programmed death-ligand 1—anywhere from 25 to 50 percent, says Richard Joseph, MD, assistant professor of medicine, Mayo Clinic, Jacksonville. They’ve also known that patients are more likely to develop a recurrence if the primary bladder cancer tumor expresses PD-L1.

Programmed death 1 protein and PD-L1 play a role in one of the so-called checkpoints in the immune system. When tumors learn to express the PD-L1 marker, they can evade immune surveillance, and PD-L1’s “off” switch fails. Not surprisingly, PD-L1 appears to be influential in other cancers as well. A 2012 paper (Brahmer JR, et al. N Engl J Med. 366:2455–2465) reported that an antibody-mediated blockade of PD-L1 led to tumor regression and prolonged stabilization of disease in patients with advanced cancers, including non-small cell lung cancer, melanoma, renal-cell cancer, and ovarian cancer. Roche’s MPDL3280A (anti-PD-L1) has received breakthrough designation from the FDA for both NSCLC and metastatic bladder cancer.

At ASCO’s Genitourinary Cancers Symposium in January, Dr. Joseph and colleagues presented an abstract of the first phase two trial looking at anti-PD-L1 in metastatic bladder cancer. The trial, involving 316 patients, was a single-arm study of patients who had received chemotherapy and whose disease had progressed. Such patients “are really out of options,” says Dr. Joseph, given the lack of proven second-line chemotherapies in metastatic bladder cancer.

The overall response rate was “significantly improved compared to a historical control of 10%,” the authors wrote. Responses were durable and associated with higher PD-L1 expression, though poor prognostic factors did not preclude response. A randomized phase three study is ongoing.

FDA approval of anti-PD-L1 would be a major shift in treatment; it could also spur development of new markers. From a clinical viewpoint, it makes no sense to separate those two strands, says Dr. Joseph. “When there are no good treatments to target these markers, these markers are less important. For example, one of the most commonly mutated genes in all of cancer is p53, and while we know this gene contributes to the pathogenesis of the disease, it is not really a target we can act upon. So to me, as a medical oncologist, rare alterations that we can act upon are more important in the treatment of disease.” An example of this type of marker in bladder cancer could be HER2 (more commonly seen in breast cancers, of course, though there’s evidence to suggest it could also be a player in bladder cancers). The folate growth factor receptors might also predict benefit to anti-FGF drugs. “But the closest to the clinic is PD-L1. That’s here, that’s ready,” says Dr. Joseph.

Adds Dr. Netto: “I believe in the near future—perhaps even this year—this will be something we are regularly asked to do on our patients with metastatic disease.”

Considering the dismal progress in developing clinically useful bladder cancer markers, PD-L1 can be seen as a cause for celebration. One researcher even uses the word “frenzy” to describe the current atmosphere. Then again, experts in this field also sound eerily like Cubs fans—who, let us delicately point out, are counting 108 years since their team’s last World Series championship: This is the year!

One advance, no matter how exciting, can’t solve things overnight. The waters here are muddy, says Dr. Zhai. PD-L1 might be creating a splash—but it’s a splash in a swamp.

For all his enthusiasm, even Dr. Joseph is careful neither to oversell nor undersell his study’s findings. “The study demonstrated clinical benefit in about one-third of patients who received the drug, with very little toxicity,” he says. Some had complete remissions, though they were the minority; most patients had either partial remissions or disease reduction that didn’t quite qualify as a partial remission. The drug was most effective in those whose tumors were PD-L1 positive.

Should anti-PD-1/PD-L1 therapy prove to be worthwhile clinically, pathologists will still have high hurdles to clear.

“Defining PD-L1 expression has been a big-time challenge for pathologists,” says Dr. Joseph. Many antibodies are currently in use, each with its own sensitivity and specificity. “And each can show different levels of expression even on the same tissues,” he says. In the aforementioned phase two trial, the researchers used a proprietary antibody. (Roche/Genentech sponsored the study.) “I don’t know if it’s going to become a companion diagnostic test or not, but I hope that this antibody at least becomes publicly available.”

Clinical trials for anti-PD-1/PD-L1 therapy have used different IHC antibodies, Dr. Zhai says. “So here the waters become muddy again.” Based on conversations he’s had with a pharmaceutical company VP, Dr. Zhai says, with a laugh, “The FDA fully realizes this mess.”

He also points to an editorial in Archives of Pathology & Laboratory Medicine (Cagle PT, et al. 2015;139:1329–1330). While the piece focuses on PD-L1 in the context of lung cancer, Dr. Zhai says it gives a framework for thinking about new markers in immune-oncology, chiefly immune checkpoint manipulation. “The news about pembrolizumab [the monoclonal antibody] is only the tip of the iceberg,” the authors write. With several others also in clinical trials, “There are different proposed IHC companion diagnostics for each of these drugs.”

This will need to be addressed, though Dr. Zhai says the current nodus is not paralyzing. “It’s not the end of the world.” In patients with positive tumors, 30 percent will respond to the treatment; among those with negative tumors, 10 percent will respond. “But we should have a guideline, based on data. I think we will be able to develop acceptable criteria eventually.”

Another area of concern, says Dr. Joseph, is that some people believe that tumor expression is the most important aspect of PD-L1; others contend it’s the tumor microenvironment—perhaps immune infiltrates, rather than the tumor itself, are expressing PD-L1. In his study, he says, the immune infiltrates appeared to be more important.

Right now PD-L1 “is the most exciting development in bladder cancer,” says Dr. Joseph. But as in any cancer, bladder’s potential markers are, like a to-do list, seemingly inexhaustible.

Case in point: Dr. Zhai is the coauthor (along with Jae Y. Ro, MD, PhD) of the recently published book Advances in Surgical Pathology: Bladder Cancer (Wolters Kluwer, 2016), which includes a chapter devoted to molecular pathology of urinary bladder neoplasms. The chapter (which Dr. Joseph helped write) contains an extensive overview of potential biomarkers, including p53, p21, p15, p16, p63, FGFR3, EGFR, VEGF, HER2, RB gene, PTEN, estrogen beta, EZH2, p27Kip1, and Cyclin D1. And those are just the tissue-based biomarkers.

As the Republican presidential campaign is demonstrating, a long list of candidates is no guarantee of, well, anything. Markers can be a disobliging bunch. “Up to this day, we don’t do any markers on our bladder cancer biopsies in a routine fashion,” says Dr. Netto.

“The field is unusually crowded right now,” says David McConkey, PhD, professor of urology and cancer biology, and director of urological research, University of Texas MD Anderson Cancer Center. He attributes that, in part, to large genomics projects such as The Cancer Genome Atlas, which has generated not only excitement but also funding.

Likewise, he says, exuberance over PD-1/PD-L1 checkpoint blockade/immunotherapy appears to have spread to other areas of bladder cancer research. It’s not news that activating mutations in FGFR3 are common in both nonmuscle-invasive and muscle-invasive bladder cancers; moreover, he says, plenty of companies have developed small molecules that inhibit FGFR3. “But industry hasn’t been all that interested in doing clinical trials in bladder cancer. They’ve been more focused on breast and multiple myeloma and other disease types. But now all of a sudden we’re flooded with companies that want to work with us on FGFR3,” says Dr. McConkey.

Here again, years of neglect have taken their toll. Even PD-L1, which was cloned nearly 20 years ago, had a hard time generating interest, says Dr. Zhai. “At the time, people didn’t pay attention to it,” given limited treatment options. The trio of surgery, chemotherapy, and radiation barely budged in decades. And in looking for therapeutic targets, researchers tended to focus on tumor cells alone, he says. “We found a lot of molecular markers, a lot of changes on tumor, but the environment was kind of ignored.” That meant potential immune therapy was overlooked, too.

Lack of research funds has also delayed progress, says Dr. Netto. “Bladder cancer is, in that regard, an orphan disease. And bladder cancer patients for the longest time did not have huge advocacy groups,” he adds, though that’s changing with the establishment of the Bladder Cancer Advocacy Network, or BCAN, which is lobbying not only for philanthropic support but looking to secure funds from government agencies to sponsor clinical trials.

The tide began to turn with the 2012 NEJM publication. With anti-PD-L1 therapy “melting” tumor cells in the metastatic setting, says Dr. Zhai, “people got very excited.” Suddenly, bladder cancer became a hoverboard. Pharmaceutical companies saw the future, Dr. Zhai says, “and they started jumping on this.”

The need for markers in bladder cancer is strong. Among solid tumors, it’s the most expensive cancer per patient, says Dr. Netto. The majority of patients present with superficial disease that typically recurs, thus incurring costs related to follow-up, including repeat cystoscopies and biopsies. “Finding markers that could change the follow-up approach from invasive procedures for surveillance to utilizing molecular urine markers, be it cell-free DNA or DNA from tumor cells that are shed in the urine, for detecting cancer recurrence, or lack of recurrence, will be very valuable,” he says.

Better markers of surveillance is only one need. Clinicians would also like to see better prognostic markers, which might include small IHC panels that have their roots in gene expression signatures. Current research is hinting at signatures of aggressive and nonaggressive disease. “I think this is one of the hottest areas in bladder cancer research,” says Dr. Netto, “where people are talking about bladder cancer types similar to what has been shown in breast cancer: a luminal type, a basal type, and a p53 wild-type.” At this point, says Dr. Netto, the signatures are based on hundreds of genes, but he’s reasonably confident they can be whittled down to a more manageable size using IHC-based signatures or surrogates.

“I think these signatures will finally translate into routine testing,” Dr. Netto continues. Markers could include different cytokeratins, such as cytokeratin 20 and high-molecular-weight cytokeratins 5 and 6, along with CD44, HER2, and FGFR3. “We could build this small panel that would classify tumors into one of these three, or four, intrinsic genetic signatures, based on such methodology.”

Signatures might also be used theranostically, Dr. Netto says, particularly related to use of neoadjuvant chemotherapy in metastatic disease. Those who respond to such therapy enjoy good survival rates, he says. But those who don’t certainly won’t benefit from a three-month delay in cystectomy. Assuming signatures of this type are reimbursed, says Dr. Netto, there could be a financial advantage. He says he’s more optimistic about an IHC panel being reimbursed rather than a gene panel assay. “If we can identify a neoadjuvant signature, you can make sure you’re putting your chemotherapy cost [toward] the right patients.”

Delving more deeply into current research efforts shows just how complex this work is.
Dr. Netto’s own area of interest lies in early detection markers. He and his colleagues have been sequencing tumors for several markers that are altered and detectable in tumor cells that are shed in urine.

TERT promoter mutations look promising, he says. Activating mutations in this gene occur in 66 percent of muscle-invasive urothelial carcinomas. In a pilot study, he and his colleagues in the Bert Vogelstein Laboratory at the Ludwig Cancer Research Center at Johns Hopkins looked at TERT promoter mutations in urine samples in patients with prior bladder cancer and noted that every patient who had a mutation detected in his or her urine sample had a recurrence, and those without the mutation did not (Kinde I, et al. Cancer Res. 2013;73:7162–7167). In subsequent studies they looked at more patients and expanded their panel to include 11 additional genes. Based on findings from The Cancer Genome Atlas and other studies, “We know how we can design an assay of around 10 or 12 genes. We’re beginning to prove that genetic alterations in urine samples tightly mirror those in associated bladder tumors, and we’re finding some encouraging results in terms of developing a very specific assay for a bladder cancer detection screen,” says Dr. Netto.

Their panel of urine genetic markers is also applicable to the surveillance setting, where the need is huge, Dr. Netto says. And he’d love to see automated next-generation-sequencing–based panels replace costly and labor-intensive FISH-based assays that have been in use.

“We also need markers of BCG response.” With Bacillus Calmette-Guérin immunotherapy as the mainstay treatment in patients with superficial disease, he says, “We need to identify patients whose tumor will resist BCG and progress to muscle-invasive disease.”

Dr. McConkey takes it a step further. “BCG is very effective, but in some sense this is an albatross around our necks—it’s effective but nondurable.” As a result, patients need to be followed their whole lives, and ultimately many eventually require cystectomies. “We’re not reducing the cost burden with BCG. So what we really need is a replacement for it. Again, the immune checkpoint blockade frenzy has spilled over into the nonmuscle-invasive cancers.” At Johns Hopkins, he says, Noah Hahn, MD, is leading a multi-institutional effort to develop novel immunotherapy approaches for this type of disease.

Nonetheless, BCG is not departing anytime soon. With that reality in mind, Dr. McConkey notes that colleagues at MD Anderson have developed a cytokine biomarker panel to measure patient response post-BCG but before recurrence.

CAP TODAY
X