Cover Story

Immunotherapy, tumor profiling fill ‘gigantic knowledge gap’ in bladder cancer

The standard of care for patients with bladder cancer has remained largely unchanged for nearly 3 decades.

The NCI estimates that 76,960 Americans will be diagnosed with bladder cancer this year, making it one of the most prevalent cancers in the United States.

Still, progress has been slow. Only 4% of patients with metastatic disease survive 5 years.

However, advances in immunotherapy and tumor profiling represent a historic paradigm shift in how the disease is approached, and they offer the potential to dramatically improve outcomes.

In May, the FDA approved atezolizumab (Tecentriq, Roche/Genentech) to treat certain patients with urothelial carcinoma. This represented the first FDA approval of an anti–PD-L1 therapy, as well as the first drug approved to treat bladder cancer since 1998.

Further, data from The Cancer Genome Atlas provided a genomic roadmap for bladder cancer, leading to the exploration of targets such as ERBB2 and FGFR3.

“In bladder cancer, the advances in immunotherapy and targeted therapy are happening parallel to one another, thus filling a gigantic knowledge gap that we have had in the field for years,” Sumanta K. Pal, MD, co-director of City of Hope’s kidney cancer program and head of the kidney and bladder cancer disease team, as well as a HemOnc Today Editorial Board member, told HemOnc Today. “Using genomic profiling and identifying relevant targeted therapies makes more sense than applying cytotoxic agents that have very limited benefit.”

Although immunotherapy is the biggest advance for metastatic bladder cancer in recent years, it is important to remember that only a small percentage of patients will respond, Sumanta K. Pal, MD, said.
Although immunotherapy is the biggest advance for metastatic bladder cancer in recent years, it is important to remember that only a small percentage of patients will respond, Sumanta K. Pal, MD, said.

Photo courtesy of Sumanta K. Pal, MD.

This progress has coincided with improvements in surgical methods and refinements in screening techniques.

Still, barriers remain.

“We have very limited patient resources in bladder cancer compared with other diseases,” Pal said. “It is critical that we emphasize the role of genomic profiling in all patients. At this point, many patients will not receive genomic profiling and, thus, may not qualify for mutation-specific trials.”

HemOnc Today spoke with genitourinary oncologists about the longstanding need for breakthroughs in the treatment of bladder cancer; how immunotherapy, targeted therapies and refined surgical methods stand to change how the disease is treated; and whether early detection through urinary assays can further improve survival.

Diagnosis, staging

The complexity of bladder cancer diagnosis and staging can negate even the most effective treatments for patients.

These challenges in diagnosis are particularly apparent for women.

Gary Steinberg, MD, Bruce and Beth White family professor of surgery and director of urologic oncology at The University of Chicago Medicine, and colleagues compared the time between initial hematuria presentation and bladder cancer diagnosis in men and women.

Results, published in 2014 in Cancer, showed women experienced significantly longer time from hematuria to diagnosis than men (85.4 days vs. 73.6 days; P < 0.001) and were more likely to instead be diagnosed with a urinary tract infection (OR = 2.32, 95% CI, 2.07-2.59). Women also were less likely to undergo pelvic or abdominal imaging (OR = 0.8, 95% CI, 0.71-0.89).

“If a woman who is postmenopausal presents with microscopic hematuria, an internist could assume a diagnosis other than bladder cancer and be correct 99% of the time — if not more,” Steinberg told HemOnc Today. “As a result, bladder cancer is often absent from differential diagnoses.”

Even after a diagnosis is confirmed, disease staging can pose challenges.

“Pathologists can determine the presence of bladder cancer, but they cannot accurately and consistently determine the grade and the stage,” Steinberg said. “Pathologists have done a tremendous job with prostate cancer over the past few decades. However, that effort has not translated into bladder cancer.”

Cystoscopy — which uses a thin tube with a camera to visualize abnormalities in the bladder — is the primary test used to diagnose bladder cancer. This invasive procedure is not always effective, which can lead to inaccurate diagnosis and staging.

If cystoscopy determines the need for biopsy, patients then undergo transurethral resection of bladder tumor (TURBT). This confirms the presence of bladder cancer and determines the grade and whether it has invaded the muscle layer of the bladder wall.

Because as many as 80% of patients with nonmuscle-invasive bladder cancer experience recurrence and 30% progress, better detection methods could improve outcomes and treatment. White light–assisted TURBT is the current standard, but it can miss as many as 20% of tumors, especially flat lesions, such as carcinoma in situ., Steinberg said.

Fluorescence cystoscopy with Cysview — hexaminolevulinate (Photocure) instilled in the bladder 60 minutes prior to using blue light cystoscopy — has been shown to better detect tumors and reduce the recurrence rate of nonmuscle-invasive bladder cancer. An alternative technique — which uses narrow band imaging and enhances blood vessels and possibly neovascularity in the bladder-assisted TURBT — is being studied, Steinberg said.

“There are currently large efforts regarding new imaging techniques,” Eila C. Skinner, MD, Thomas A. Stamey research professor in urology at Stanford University and a HemOnc Today Editorial Board member, said in an interview. “Blue light cystoscopy is being used more widely, which may improve accuracy.”

Eila C. Skinner, MD
Eila C. Skinner

Burger and colleagues conducted a retrospective analysis of 10 prospective studies that compared blue light cystoscopy with white light cystoscopy for Ta/T1 tumors and carcinoma in situ. The analysis included 1,345 patients with known or suspected nonmuscle-invasive bladder cancer.

Results — published in European Urology — showed blue light detected 9.7% to 40.2% additional tumors. Further, 14.7% of Ta tumors were detected only by blue light (OR = 4.89; 95% CI, 1.93-12.39).

Blue light detected an additional 3.6% to 54.5% of T1 tumors, with 10.8% detected only by blue light (OR = 2.25; 95% CI, 0.99-5.08).

Patients whose tumors were detected through blue light cystoscopy had lower recurrence rates through 12 months (34.5% vs. 45.4%; RR = 0.76; 95% CI, 0.62-0.92).

Although better diagnosis and proper staging stand to improve patient outcomes, cystoscopies remain an unpleasant experience for patients. They often lead to hematuria, bladder spasms, painful urination and greater risk for infection.

“I think patients would love to get away from cystoscopy all together.” Skinner said. “There has been tons of work done with urine, looking for markers so as to avoid cystoscopy. But, so far that hasn’t happened.”

Surgical procedures

Patients with stage II or stage III disease, as well as some patients with recurrent nonmuscle-invasive bladder cancer, will proceed to surgery.

However, radical cystectomy is an extremely difficult procedure for patients. Forty-six percent of procedures lead to complications, and 3% lead to mortality.

Despite this, open radical cystectomy remains the standard surgical procedure.

“The intensity of care that is required for patients who require a radical cystectomy is very high,” Steinberg said. “But, there’s no question that there has been a centralization of radical cystectomy, which has led to improvements.”

For example, incidence of deep vein thrombosis after radical cystectomy can be as high as 15%, but some centers have decreased that to 2% to 3%, Steinberg said.

“The procedure should only be done at high-volume centers, because the pre- and postoperative care that is required is very high,” he said. “It requires a dedicated, educated team.”

Two research initiatives are underway to try to improve the procedure, Seth P. Lerner, MD, Beth and Dave Swalm chair in urologic oncology and director of urologic oncology at Baylor College of Medicine in Houston, Texas, told HemOnc Today.

“A minimally invasive approach is being compared to the traditional open approach, and a trial for which I am the principle investigator will examine the anatomic extent of lymph node dissection in open radical cystectomy,” Lerner said.

However, in a letter to the editor published in The New England Journal of Medicine, Bochner and colleagues showed a minimally invasive approach through robot-assisted surgery did not reduce the rate of grade 2 to grade 5 complications within 90 days of surgery compared with an open procedure.

Among 118 patients with Ta–3N0–3M0 bladder cancer, 62% of those assigned robot-assisted cystectomy experienced a grade 2 event or worse compared with 66% of those who underwent open cystectomy. Rates of grade 3 to grade 5 events also were comparable (22% vs. 21%).

“Despite no improvement in 90-day morbidity, we are awaiting completion of trials that compare oncologic efficacy, so the story remains to be told,” Lerner said.

Lerner and colleagues are accruing for a trial that will evaluate PFS in patients with muscle-invasive bladder cancer who will undergo radical cystectomy with extended pelvic lymph node dissection or standard pelvic lymphadenectomy.

“There is a similar German trial with a slightly different design. Both trials will shed light into the anatomic extent of the operation, and our intent is to collaborate with the German group and do a combined analysis,” Lerner said.

Gschwendand colleaguespresented findings from the German trial at the ASCO Annual Meeting in June.

Researchers randomly assigned 375 patients with high-grade invasive urothelial bladder cancer to extended (n = 184) or limited (n = 191) pelvic lymph node dissection.

Patients assigned extended dissection experienced a trend toward improved 5-year RFS (69.3% vs. 62%; HR = 0.8; 95% CI, 0.54-1.19) and 5-year cancer-specific survival (77.5% vs. 66.2%; HR = 0.7; 95% CI, 0.45-1.1), but neither were statistically significant.

Even with these data, whether expanded dissection confers a survival benefit remains unclear, Charles J. Ryan, MD, professor of clinical medicine and urology and clinical program leader for genitourinary medical oncology at UCSF Helen Diller Family Comprehensive Cancer Center, told HemOnc Today after the presentation.

“Either the study was underpowered to detect this difference, the patient outcome was better overall than planned or an extended lymph node dissection is not helpful,” Ryan said.

Accrual for the U.S. study should be completed later this year, Lerner said. The study will include OS as a secondary endpoint. It also will evaluate factors including operative time, 90-day morbidity and mortality, and lymph node counts and density.

“The beauty of this trial, as well as several others, is that it will produce a robust biospecimen repository for asking important molecular and immunologic questions in the future,” Lerner said.

Therapeutic advances

Cisplatin-based chemotherapy is the first-line standard of care for patients with locally advanced or metastatic disease and is associated with a median survival of 12 to 15 months. For patients who are ineligible for cisplatin, carboplatin-based chemotherapy is commonly used, but is associated with a median survival of 9 to 10 months. For patients who progress after platinum-based treatment, median survival is 5 to 7 months.

In May, atezolizumab became the first FDA–approved treatment for bladder cancer in nearly 3 decades. The approval was for second-line therapy for patients with locally advanced or metastatic urothelial carcinoma who had progressed during or after platinum-based chemotherapy.

The FDA based its decision, in part, on data from cohort 2 of IMvigor 210. The single-arm phase 2 study evaluated the safety and efficacy of 1,200 mg IV atezolizumab every 3 weeks for patients with locally advanced or metastatic urothelial cancer who progressed during or following a prior platinum-based chemotherapy regimen.

In the report published this year in The Lancet, Rosenberg and colleagues showed the objective response rate was 15% for all patients treated in the study, which met the primary endpoint of the study when compared with the 10% historical response rate observed with chemotherapy in the second-line setting (15% vs. 10%; P = .0058). The ORR was highest among patients whose tumors overexpressed PD-L1 in infiltrating immune cells (5% PD-L1–positive immune cells [IC 2/3], 26%; P = .0004).

Arjun V. Balar, MD, assistant professor of medicine at New York University Langone Medical Center and director of genitourinary medical oncology at Perlmutter Cancer Center, presented data from cohort 1 of IMvigor 210 at this year’s ASCO Annual Meeting. This cohort consisted of 119 treatment-naive patients who were ineligible for cisplatin-based therapy due to factors such as performance status, renal dysfunction, hearing impairment or peripheral neuropathy.

Results showed the ORR for the entire population was 24% (95% CI, 16-32). Seven percent of patients achieved complete response and 17% achieved partial response.

At a median follow-up of 14.4 months, 21 of 28 (75%) responses were ongoing. Median duration of response had not been reached, with the longest ongoing response reaching 18 months.

The estimated median OS was 14.8 months, and 57% of patients survived 1 year; however, these data are immature as the overall event rate was 47% at the time of analysis.

“This could potentially represent a new standard of care in the first-line setting for metastatic urothelial carcinoma,” Balar told HemOnc Today. “The patients in cohort 1 are truly representative of the patient population seen clinically. For example, the study population was older — the median age is approximately 73 years, and 20% were aged 80 years or older — and two-thirds had visceral sites of metastatic disease.”

Other types of immunotherapy also are under evaluation for bladder cancer. For instance, researchers are evaluating durvalumab (MEDI4736, AstraZeneca) plus tremelimumab (AstraZeneca), as well as nivolumab (Opdivo, Bristol-Myers Squibb) plus ipilimumab (Yervoy, Bristol-Myers Squibb). Phase 2 and phase 3 trials of pembrolizumab (Keytruda, Merck) monotherapy also are underway.

Despite the promise of this class of drugs, not all patients will benefit from immunotherapy.

“There are big gaps in knowledge,” Pal said. “Although we hail immunotherapy as being the biggest advance in metastatic bladder cancer in recent years — and it most certainly is — it is important to note that only a minority of patients are going to respond.”

Genetic understanding

Greater understanding of the genetic landscape in bladder cancer can help researchers identify which patients are most likely to benefit from immunotherapy combinations.

“Imagine a patient with bladder cancer walks through the door and we are able to profile the tumor and assign the most effective therapy, including options that have never been tried for bladder cancer,” Dan Theodorescu, MD, PhD, Paul Bunn professor and director of University of Colorado Cancer Center, told HemOnc Today. “For a specific tumor, we aim to combine the best possible small molecule with the best immunotherapy.”

Dan Theodorescu, MD, PhD
Dan Theodorescu

Although immunotherapy is unlikely to help every patient, targeted treatments may help bridge this gap.

Choudhury and colleagues conducted tumor sequencing in patients with metastatic disease treated with afatinib (Gilotrif, Boehringer Ingelheim). Results, published this year in Journal of Clinical Oncology, showed patients with HER-2 or ERBB3 alterations were more likely to benefit from treatment.

“It is critical to investigate and establish strategies for targeting patients who have multiple mutations that are potentially actionable, such as HER-2, EGFR and multiple others,” Pal said. “We don’t know, however, which mutations drive tumor growth and which are passenger mutations.”

Other trials are evaluating monoclonal antibodies that target tumor antigens.

B-701 (BioClin Therapeutics) — an anti-FGFR3 antibody — is under study in a phase 2 trial for patients with locally advanced or metastatic bladder cancer. MK-6018 (Merck), which targets carcinoembryonic antigen, is being studied in a phase 1 trial for patients with advanced or recurrent cancers, including bladder cancer.

The Borealis-2 trial is assessing apatorsen (OGX-427, OncoGenex) — designed to inhibit production of heat shock protein 27, an intracellular protein that protects cancer cells by helping them survive — in combination with docetaxel for patients with metastatic bladder cancer. Final results from that trial are expected this year.

The Southwest Oncology Group’s COXEN study is exploring the possibility of basing treatment off of a genetic understanding of the disease, Theodorescu said. Researchers of that trial are using the COXEN gene expression model to predict the best chemotherapy treatment for patients based on individual tumor characteristics.

“We are realizing now that patients with bladder cancer have a very diverse genomic profile,” Pal said. “This is something we have known for a while, but that had just not been actionable.”

Increased awareness

Challenges in diagnosis and staging, as well as in conducting clinical trials, may stem from the fact that resources historically have been limited.

The influx of potential treatment options appears to have prompted a change.

“Medical oncologists are becoming more interested in bladder cancer,” Steinberg said. “They are recognizing that they have treatments and that they have the ability to be part of the management.”

The experts with whom HemOnc Today spoke each credited The Bladder Cancer Advocacy Network (BCAN) as a driving force behind the effort to accelerate progress in bladder cancer.

BCAN, which celebrated its 10th anniversary last year, announced the launch of the Bladder Cancer Genomics Consortium to improve understanding of the disease’s genetic landscape in hopes of developing targeted therapies.

The project aims to genomically profile an initial cohort of 200 patients with metastatic urothelial carcinoma from eight participating institutions. Physicians will receive the results of these tests, as well as information about relevant clinical trials.

“I predict that, 5 years from now, we will be giving systemic chemotherapy and possibly targeted therapy to patients based on biomarker-specific predictors for the best possible response — true personalized medicine,” Lerner said.

Genomic profiling not only has the potential to alter treatment protocol and predict patient response. It also can improve diagnostics by detecting mutations in urine.

The practical application for urine detection will be in secondary screening of high-risk patients or those with a history of bladder cancer, according to Theodorescu.

“There are always a few cells that are shed,” he said.

Several tests are available to screen for tumor biomarkers in urine. These include UroVysion (Abbott Molecular), which tests for chromosome changes; bladder tumor–associated antigen tests; Immunocyt (DiagnoCure Inc.), which tests for mucin and carcinoembryonic antigen; and NMP22 BladderChek (Alere), which tests for the NMP22 protein.

However, these tests are not 100% accurate, and they may indicate bladder cancer in individuals without the disease.

“If a urine test — which a patient could mail in — could rule out the need for a cystoscopy, the impact would be financially and practically substantial,” Theodorescu said.

Other barriers exist.

Most patients recommended to undergo cystectomy need to travel to an academic center.

“On average, our patients come from 1 to 2 hours away,” Skinner, who is on the BCAN advisory board, told HemOnc Today. “There has been real effort to raise awareness at the primary care and urologist levels regarding all facets of the disease, but there’s still a long way to go.”

Last year, the journal Bladder Cancer — spearheaded by Lerner and Theodorescu — published its first issue, providing another source of information and collaboration for the clinical and scientific community.

“I tell medical students and early career physicians: ‘Publish early, and publish often,’” Steinberg said. “BCAN has done a tremendous job of encouraging research around the calendar, as well as raising awareness at the patient level.

“Years ago, women got together and decided to change breast cancer,” Steinberg added. “They got money, advocacy, awareness and research into that disease. That is what we are seeing in bladder cancer, and what needs to continue.” – by Nick Andrews

References:

Bochner BH, et al. N Engl J Med. 2014;doi:10.1056/NEJMc1405213.

Burger M, et al. Eur Urol. 2013;doi:10.1016/j.eururo.2013.03.059.

Choudhury NJ, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.66.3047.

Johnson D, et al. Abstract 1622. Presented at: American Urologic Association Annual Scientific Meeting; May 4-8, 2013; San Francisco.

Rosenberg JE, et al. Lancet. 2016;doi:10.1016/S0140-6736(16)00561-4.

Steinberg G, et al. Cancer. 2014;doi:10.1002/cncr.28416.

The following were presented at ASCO Annual Meeting; June 3-7, 2016; Chicago:

Balar AV, et al. Abstract LBA4500.

Gschwend JE, et al. Abstract 4503.

For more information:

Arjun V. Balar, MD, can be reached at arjun.balar@nyumc.org.

Seth P. Lerner, MD, can be reached at slerner@bcm.edu.

Sumanta K. Pal, MD, can be reached at spal@coh.org.

Charles J. Ryan, MD, can be reached at ryanc@medicine.ucsf.edu.

Eila C. Skinner, MD, can be reached at skinnere@stanford.edu.

Gary D. Steinberg, MD, can be reached at gsteinbe@surgery.bsd.uchicago.edu.

Dan Theodorescu, MD, PhD, can be reached at dan.theodorescu@ucdenver.edu.

Disclosure: Lerner reports consultant/advisory roles with BioCancell, Ferring, Neucleix, Sitka, Taris, Telesta, UroGen and Vaxiion; research grants or funding from Endo, FKD Therapies, Genome Dx, Roche/Genentech and Viventia; and is co-editor in chief of Bladder Cancer. Pal reports consultant roles with Bristol-Myers Squibb, Genentech, Pfizer and Novartis and honoraria from Genentech. Steinberg reports consultant roles with Ciclo Med, Cold Genesys, Fidia, Heat Biologics, Karl Storz, Photocure, Roche/Genentech and Taris Biomedical. Theodorescu reports equity ownership in Aurora Oncology, as well as intellectual property in and inventor involvement with Myriad Genetics and NantHealth. He also is co-editor in chief of Bladder Cancer. Balar, Ryan and Skinner report no relevant financial disclosures.

 

POINTCOUNTER 

Should adjuvant chemotherapy be standard for patients with muscle-invasive bladder cancer?

POINT

Adjuvant chemotherapy should be offered to fit patients with muscle-invasive bladder cancer who have not received neoadjuvant chemotherapy.

At least three types of evidence should be required for clinicians to adopt a therapy as standard practice, in this case the use of adjuvant chemotherapy for people undergoing local treatment — cystectomy, radiotherapy or both — for transitional cell carcinoma of the bladder.

Ian F. Tannock, MD, PhD
Ian F. Tannock

1. Biological plausibility — ie, that metastatic failure is a common cause of death and that chemotherapy effectively causes remission of metastatic bladder cancer in a substantial proportion of patients;

2. Evidence from randomized controlled trials (RCTs) that adjuvant chemotherapy leads to improved survival. Here, consistency of effect is more important than a single RCT with a P value reaching the arbitrary threshold of less than .05. The American Statistical Association is sufficiently concerned about misinterpretation of P values that they issued a formal statement indicating that P values do not measure the probability that the studied hypothesis is true, nor the probability that the data were produced by random chance alone. They state that scientific conclusions and policy decisions should not be based only on whether a P value passes a specific threshold; and

3. Evidence from well-designed health outcome studies. This is essential to determine whether the proposed treatment leads to improved outcome in the real-world population. Although randomization has the advantage of ensuring similar characteristics between arms, RCTs recruit a highly selected population that does not reflect everyday practice. Health outcome studies are criticized because statistical methods such as propensity analyses effectively correct for known biases — such as the tendency to give adjuvant chemotherapy to those with better performance status — but they cannot guarantee that a difference in outcome is due only to differences in the treatment received. The advantages of randomly assigning selected patients in RCTs and analyzing real-world data from large patient samples in health outcome studies provide complementary evidence. If their results are contradictory, one should question the value of a treatment; if they are consistent, the level of evidence is very high.

Adjuvant chemotherapy for bladder cancer is supported by all three of the criteria described above.

Biological plausibility is strong. Treatment of people with metastatic bladder cancer with chemotherapy regimens such as cisplatin and gemcitabine, MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) or accelerated MVAC leads to response rates of approximately 50%, with a small proportion of sustained complete responses. This experience is equal or better to that achieved for women with breast cancer, where long-term complete responses are rare. People who relapse after local treatment of bladder cancer usually do so in the first 2 to 3 years, almost certainly from micrometastatic disease present initially, providing a strong rationale for adjuvant chemotherapy.

Several RCTs evaluated modern adjuvant chemotherapy, and although individually underpowered, these trials generally suggest improved survival, with an HR of approximately 0.78.

There are three large health outcome studies of adjuvant chemotherapy applied to real-world populations that provide evidence of similar or better levels of survival benefit.

The evidence in favor of adjuvant chemotherapy for patients with muscle-invasive bladder cancer is very strong. It should be offered to fit patients who have completed local treatment and who have not received neoadjuvant therapy.

 

References:

Austin PC. Multivariate Behav Res. 2011;46:399-424.

Booth CM and Tannock IF. Br J Cancer. 2014;doi:10.1038/bjc.2013.725.

Booth CM and Tannock IF. JAMA Oncol. 2015;doi:10.1001/jamaoncol.2015.1210.

Booth CM, et al. Cancer. 2014;doi:10.1002/cncr.28510.

Galsky MD, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.64.1076.

Ioannidis JP. PLoS Med. 2005;doi.org/10.1371/journal.pmed.0020124.

Leow JJ, et al. Eur Urol. 2014;doi:10.1016/j.eururo.2013.08.033.

Sternberg CN, et al. Lancet Oncol. 2015;doi:10.1016/S1470-2045(14)71160-X.

Svatek RS, et al. Clin Cancer Res. 2010;doi:10.1158/1078-0432.CCR-10-0457.

Wasserstein RL and Lazar NA. Am Stat. 2016;doi:10.1080/00031305.2016.1154108.

Ian F. Tannock, MD, PhD, is professor of medicine and medical biophysics at Princess Margaret Cancer Centre and University of Toronto. He can be reached at ian.tannock@uhn.ca. Disclosure: Tannock reports no relevant financial disclosures.

COUNTER

No strong evidence exists to support use of adjuvant chemotherapy for muscle-invasive bladder cancer.

It’s a daunting task to debate the venerable Professor Ian F. Tannock, MD, PhD, my friend and collaborator of more than 3 decades. Prior to commencing my manifesto, I would refer the reader to our joint publication from 1990, espousing the basic principles of high-quality research that have not changed much in the past quarter century.

Derek Raghavan, MD, PhD, FACP, FASCO
Derek Raghavan

Professor Tannock previously noted that change of treatment requires supporting biology, evidence from appropriately powered randomized trials and data from well-designed outcome studies. When considering why adjuvant chemotherapy has not been proven effective in the management of invasive bladder cancer, the available data do not meet any of his espoused standards.

With regard to biological support, bladder cancer is a markedly heterogeneous malignancy, and it has been shown repeatedly that established paradigms of systemic therapy — such as the MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) or CMV (cisplatin, methotrexate and vinblastine) regimens — have not had equivalent success in advanced disease to the systemic therapy paradigms established in breast cancer, the poster child of classical adjuvant treatment. Tannock and I collaborated in demonstrating the utility of neoadjuvant CMV chemotherapy in a study that began decades ago, showing a true impact on survival that was not an artifact of statistical quirks or chicanery, but this approach placed surgery as definitive therapy following bulk reduction of chemotherapy-sensitive elements within occult metastatic disease.

Sadly, the genitourinary oncology community has disappointed by failing to support the well-powered available trials; but, even the largest reported trial failed to show a pragmatically meaningful survival benefit. Further, the statistical analysis revealed that the most substantial impact of adjuvant chemotherapy was on node-negative disease; this unexpected observation either supports the contention that some of the surgery in this multinational study was suboptimal, or that four cycles of adjuvant chemotherapy is insufficient to control this heterogeneous disease.

It is never wise to ignore the published literature. Thus, I remind the good professor of the Italian study that assessed the gemcitabine–cisplatin regimen as adjuvant chemotherapy after radical cystectomy. It demonstrated — at best — equivalence, or at worst, inferiority of the chemotherapy regimen compared with observation alone. I’m a little surprised that Professor Tannock has previously cited the many flawed meta-analyses of adjuvant chemotherapy for bladder cancer, but simply would remind him that all included the large German studies that tested adjuvant chemotherapy against no chemotherapy (in most patients), which simply confounded each analysis.

This brings me to the final criterion of the Tannock credo — that is, “well-designed outcome studies.” Sadly, the community-based “real population” studies appear to have been led by investigators who believe that propensity matching has similar power as well-powered randomized trials. That simply is not true. The concept of garbage-in-garbage-out is important to this domain, and the problem is relatively simple. Propensity matching simply does not account for all the clinician- and tumor-driven variables that cloud the comparisons listed “in the real world” that lead to huge case- and treatment-selection biases — specifically, selection of the more robust patients with more normal cardiopulmonary and metabolic function for adjuvant treatment, selection “out” of more complex tumors (eg, those with nonurothelial cancer elements), and unstructured variability of pathological diagnosis and follow-up approaches. Thus, the proposition that this type of database study is truly meaningful is ridiculous.

There is no strong evidence biologically, from trials or well-designed outcome studies, to support adjuvant chemotherapy for muscle-invasive bladder cancer. Chemotherapy is expensive and potentially toxic, and using it routinely without clear, supporting level-one data is a bad idea.

 

References:

Booth CM and Tannock IF. Br J Cancer. 2014;doi:10.1038/bjc.2013.725.

Booth CM and Tannock IF. JAMA Oncol. 2015;doi:10.1001/jamaoncol.2015.1210.

Booth CM, et al. Cancer. 2014;doi:10.1002/cncr.28510.

Cognetti F, et al. Ann Oncol. 2011;doi:10.1093/annonc/mdr354.

Galsky MD, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.64.1076.

Leow JJ, et al. Eur Urol. 2014;doi:10.1016/j.eururo.2013.08.033.

Raghavan D and Tannock I. “Clinical trials in genitourinary cancer: What have they achieved?” Combination Therapy in Urologic Malignancy. Ed. Smith PH. London: Springer-Verlag London, 1990. 225-253.

Raghavan D, et al. Ann Oncol. 2014:doi:10.1093/annonc/mdu092.

Sternberg CN, et al. Lancet Oncol. 2015;doi:10.1016/S1470-2045(14)71160-X.

Svatek RS, et al. Clin Cancer Res. 2010;doi:10.1158/1078-0432.CCR-10-0457.

Derek Raghavan, MD, PhD, FACP, FASCO, is HemOnc Today’s Chief Medical Editor for Oncology. He also is president of Levine Cancer Institute at Carolinas HealthCare System. He can be reached at derek.raghavan@carolinashealthcare.org. Disclosure: Raghavan reports no relevant financial disclosures.

The standard of care for patients with bladder cancer has remained largely unchanged for nearly 3 decades.

The NCI estimates that 76,960 Americans will be diagnosed with bladder cancer this year, making it one of the most prevalent cancers in the United States.

Still, progress has been slow. Only 4% of patients with metastatic disease survive 5 years.

However, advances in immunotherapy and tumor profiling represent a historic paradigm shift in how the disease is approached, and they offer the potential to dramatically improve outcomes.

In May, the FDA approved atezolizumab (Tecentriq, Roche/Genentech) to treat certain patients with urothelial carcinoma. This represented the first FDA approval of an anti–PD-L1 therapy, as well as the first drug approved to treat bladder cancer since 1998.

Further, data from The Cancer Genome Atlas provided a genomic roadmap for bladder cancer, leading to the exploration of targets such as ERBB2 and FGFR3.

“In bladder cancer, the advances in immunotherapy and targeted therapy are happening parallel to one another, thus filling a gigantic knowledge gap that we have had in the field for years,” Sumanta K. Pal, MD, co-director of City of Hope’s kidney cancer program and head of the kidney and bladder cancer disease team, as well as a HemOnc Today Editorial Board member, told HemOnc Today. “Using genomic profiling and identifying relevant targeted therapies makes more sense than applying cytotoxic agents that have very limited benefit.”

Although immunotherapy is the biggest advance for metastatic bladder cancer in recent years, it is important to remember that only a small percentage of patients will respond, Sumanta K. Pal, MD, said.
Although immunotherapy is the biggest advance for metastatic bladder cancer in recent years, it is important to remember that only a small percentage of patients will respond, Sumanta K. Pal, MD, said.

Photo courtesy of Sumanta K. Pal, MD.

This progress has coincided with improvements in surgical methods and refinements in screening techniques.

Still, barriers remain.

“We have very limited patient resources in bladder cancer compared with other diseases,” Pal said. “It is critical that we emphasize the role of genomic profiling in all patients. At this point, many patients will not receive genomic profiling and, thus, may not qualify for mutation-specific trials.”

HemOnc Today spoke with genitourinary oncologists about the longstanding need for breakthroughs in the treatment of bladder cancer; how immunotherapy, targeted therapies and refined surgical methods stand to change how the disease is treated; and whether early detection through urinary assays can further improve survival.

Diagnosis, staging

The complexity of bladder cancer diagnosis and staging can negate even the most effective treatments for patients.

These challenges in diagnosis are particularly apparent for women.

Gary Steinberg, MD, Bruce and Beth White family professor of surgery and director of urologic oncology at The University of Chicago Medicine, and colleagues compared the time between initial hematuria presentation and bladder cancer diagnosis in men and women.

Results, published in 2014 in Cancer, showed women experienced significantly longer time from hematuria to diagnosis than men (85.4 days vs. 73.6 days; P < 0.001) and were more likely to instead be diagnosed with a urinary tract infection (OR = 2.32, 95% CI, 2.07-2.59). Women also were less likely to undergo pelvic or abdominal imaging (OR = 0.8, 95% CI, 0.71-0.89).

“If a woman who is postmenopausal presents with microscopic hematuria, an internist could assume a diagnosis other than bladder cancer and be correct 99% of the time — if not more,” Steinberg told HemOnc Today. “As a result, bladder cancer is often absent from differential diagnoses.”

Even after a diagnosis is confirmed, disease staging can pose challenges.

“Pathologists can determine the presence of bladder cancer, but they cannot accurately and consistently determine the grade and the stage,” Steinberg said. “Pathologists have done a tremendous job with prostate cancer over the past few decades. However, that effort has not translated into bladder cancer.”

PAGE BREAK

Cystoscopy — which uses a thin tube with a camera to visualize abnormalities in the bladder — is the primary test used to diagnose bladder cancer. This invasive procedure is not always effective, which can lead to inaccurate diagnosis and staging.

If cystoscopy determines the need for biopsy, patients then undergo transurethral resection of bladder tumor (TURBT). This confirms the presence of bladder cancer and determines the grade and whether it has invaded the muscle layer of the bladder wall.

Because as many as 80% of patients with nonmuscle-invasive bladder cancer experience recurrence and 30% progress, better detection methods could improve outcomes and treatment. White light–assisted TURBT is the current standard, but it can miss as many as 20% of tumors, especially flat lesions, such as carcinoma in situ., Steinberg said.

Fluorescence cystoscopy with Cysview — hexaminolevulinate (Photocure) instilled in the bladder 60 minutes prior to using blue light cystoscopy — has been shown to better detect tumors and reduce the recurrence rate of nonmuscle-invasive bladder cancer. An alternative technique — which uses narrow band imaging and enhances blood vessels and possibly neovascularity in the bladder-assisted TURBT — is being studied, Steinberg said.

“There are currently large efforts regarding new imaging techniques,” Eila C. Skinner, MD, Thomas A. Stamey research professor in urology at Stanford University and a HemOnc Today Editorial Board member, said in an interview. “Blue light cystoscopy is being used more widely, which may improve accuracy.”

Eila C. Skinner, MD
Eila C. Skinner

Burger and colleagues conducted a retrospective analysis of 10 prospective studies that compared blue light cystoscopy with white light cystoscopy for Ta/T1 tumors and carcinoma in situ. The analysis included 1,345 patients with known or suspected nonmuscle-invasive bladder cancer.

Results — published in European Urology — showed blue light detected 9.7% to 40.2% additional tumors. Further, 14.7% of Ta tumors were detected only by blue light (OR = 4.89; 95% CI, 1.93-12.39).

Blue light detected an additional 3.6% to 54.5% of T1 tumors, with 10.8% detected only by blue light (OR = 2.25; 95% CI, 0.99-5.08).

Patients whose tumors were detected through blue light cystoscopy had lower recurrence rates through 12 months (34.5% vs. 45.4%; RR = 0.76; 95% CI, 0.62-0.92).

Although better diagnosis and proper staging stand to improve patient outcomes, cystoscopies remain an unpleasant experience for patients. They often lead to hematuria, bladder spasms, painful urination and greater risk for infection.

“I think patients would love to get away from cystoscopy all together.” Skinner said. “There has been tons of work done with urine, looking for markers so as to avoid cystoscopy. But, so far that hasn’t happened.”

Surgical procedures

Patients with stage II or stage III disease, as well as some patients with recurrent nonmuscle-invasive bladder cancer, will proceed to surgery.

However, radical cystectomy is an extremely difficult procedure for patients. Forty-six percent of procedures lead to complications, and 3% lead to mortality.

Despite this, open radical cystectomy remains the standard surgical procedure.

“The intensity of care that is required for patients who require a radical cystectomy is very high,” Steinberg said. “But, there’s no question that there has been a centralization of radical cystectomy, which has led to improvements.”

For example, incidence of deep vein thrombosis after radical cystectomy can be as high as 15%, but some centers have decreased that to 2% to 3%, Steinberg said.

“The procedure should only be done at high-volume centers, because the pre- and postoperative care that is required is very high,” he said. “It requires a dedicated, educated team.”

Two research initiatives are underway to try to improve the procedure, Seth P. Lerner, MD, Beth and Dave Swalm chair in urologic oncology and director of urologic oncology at Baylor College of Medicine in Houston, Texas, told HemOnc Today.

PAGE BREAK

“A minimally invasive approach is being compared to the traditional open approach, and a trial for which I am the principle investigator will examine the anatomic extent of lymph node dissection in open radical cystectomy,” Lerner said.

However, in a letter to the editor published in The New England Journal of Medicine, Bochner and colleagues showed a minimally invasive approach through robot-assisted surgery did not reduce the rate of grade 2 to grade 5 complications within 90 days of surgery compared with an open procedure.

Among 118 patients with Ta–3N0–3M0 bladder cancer, 62% of those assigned robot-assisted cystectomy experienced a grade 2 event or worse compared with 66% of those who underwent open cystectomy. Rates of grade 3 to grade 5 events also were comparable (22% vs. 21%).

“Despite no improvement in 90-day morbidity, we are awaiting completion of trials that compare oncologic efficacy, so the story remains to be told,” Lerner said.

Lerner and colleagues are accruing for a trial that will evaluate PFS in patients with muscle-invasive bladder cancer who will undergo radical cystectomy with extended pelvic lymph node dissection or standard pelvic lymphadenectomy.

“There is a similar German trial with a slightly different design. Both trials will shed light into the anatomic extent of the operation, and our intent is to collaborate with the German group and do a combined analysis,” Lerner said.

Gschwendand colleaguespresented findings from the German trial at the ASCO Annual Meeting in June.

Researchers randomly assigned 375 patients with high-grade invasive urothelial bladder cancer to extended (n = 184) or limited (n = 191) pelvic lymph node dissection.

Patients assigned extended dissection experienced a trend toward improved 5-year RFS (69.3% vs. 62%; HR = 0.8; 95% CI, 0.54-1.19) and 5-year cancer-specific survival (77.5% vs. 66.2%; HR = 0.7; 95% CI, 0.45-1.1), but neither were statistically significant.

Even with these data, whether expanded dissection confers a survival benefit remains unclear, Charles J. Ryan, MD, professor of clinical medicine and urology and clinical program leader for genitourinary medical oncology at UCSF Helen Diller Family Comprehensive Cancer Center, told HemOnc Today after the presentation.

“Either the study was underpowered to detect this difference, the patient outcome was better overall than planned or an extended lymph node dissection is not helpful,” Ryan said.

Accrual for the U.S. study should be completed later this year, Lerner said. The study will include OS as a secondary endpoint. It also will evaluate factors including operative time, 90-day morbidity and mortality, and lymph node counts and density.

“The beauty of this trial, as well as several others, is that it will produce a robust biospecimen repository for asking important molecular and immunologic questions in the future,” Lerner said.

Therapeutic advances

Cisplatin-based chemotherapy is the first-line standard of care for patients with locally advanced or metastatic disease and is associated with a median survival of 12 to 15 months. For patients who are ineligible for cisplatin, carboplatin-based chemotherapy is commonly used, but is associated with a median survival of 9 to 10 months. For patients who progress after platinum-based treatment, median survival is 5 to 7 months.

In May, atezolizumab became the first FDA–approved treatment for bladder cancer in nearly 3 decades. The approval was for second-line therapy for patients with locally advanced or metastatic urothelial carcinoma who had progressed during or after platinum-based chemotherapy.

The FDA based its decision, in part, on data from cohort 2 of IMvigor 210. The single-arm phase 2 study evaluated the safety and efficacy of 1,200 mg IV atezolizumab every 3 weeks for patients with locally advanced or metastatic urothelial cancer who progressed during or following a prior platinum-based chemotherapy regimen.

In the report published this year in The Lancet, Rosenberg and colleagues showed the objective response rate was 15% for all patients treated in the study, which met the primary endpoint of the study when compared with the 10% historical response rate observed with chemotherapy in the second-line setting (15% vs. 10%; P = .0058). The ORR was highest among patients whose tumors overexpressed PD-L1 in infiltrating immune cells (5% PD-L1–positive immune cells [IC 2/3], 26%; P = .0004).

PAGE BREAK

Arjun V. Balar, MD, assistant professor of medicine at New York University Langone Medical Center and director of genitourinary medical oncology at Perlmutter Cancer Center, presented data from cohort 1 of IMvigor 210 at this year’s ASCO Annual Meeting. This cohort consisted of 119 treatment-naive patients who were ineligible for cisplatin-based therapy due to factors such as performance status, renal dysfunction, hearing impairment or peripheral neuropathy.

Results showed the ORR for the entire population was 24% (95% CI, 16-32). Seven percent of patients achieved complete response and 17% achieved partial response.

At a median follow-up of 14.4 months, 21 of 28 (75%) responses were ongoing. Median duration of response had not been reached, with the longest ongoing response reaching 18 months.

The estimated median OS was 14.8 months, and 57% of patients survived 1 year; however, these data are immature as the overall event rate was 47% at the time of analysis.

“This could potentially represent a new standard of care in the first-line setting for metastatic urothelial carcinoma,” Balar told HemOnc Today. “The patients in cohort 1 are truly representative of the patient population seen clinically. For example, the study population was older — the median age is approximately 73 years, and 20% were aged 80 years or older — and two-thirds had visceral sites of metastatic disease.”

Other types of immunotherapy also are under evaluation for bladder cancer. For instance, researchers are evaluating durvalumab (MEDI4736, AstraZeneca) plus tremelimumab (AstraZeneca), as well as nivolumab (Opdivo, Bristol-Myers Squibb) plus ipilimumab (Yervoy, Bristol-Myers Squibb). Phase 2 and phase 3 trials of pembrolizumab (Keytruda, Merck) monotherapy also are underway.

Despite the promise of this class of drugs, not all patients will benefit from immunotherapy.

“There are big gaps in knowledge,” Pal said. “Although we hail immunotherapy as being the biggest advance in metastatic bladder cancer in recent years — and it most certainly is — it is important to note that only a minority of patients are going to respond.”

Genetic understanding

Greater understanding of the genetic landscape in bladder cancer can help researchers identify which patients are most likely to benefit from immunotherapy combinations.

“Imagine a patient with bladder cancer walks through the door and we are able to profile the tumor and assign the most effective therapy, including options that have never been tried for bladder cancer,” Dan Theodorescu, MD, PhD, Paul Bunn professor and director of University of Colorado Cancer Center, told HemOnc Today. “For a specific tumor, we aim to combine the best possible small molecule with the best immunotherapy.”

Dan Theodorescu, MD, PhD
Dan Theodorescu

Although immunotherapy is unlikely to help every patient, targeted treatments may help bridge this gap.

Choudhury and colleagues conducted tumor sequencing in patients with metastatic disease treated with afatinib (Gilotrif, Boehringer Ingelheim). Results, published this year in Journal of Clinical Oncology, showed patients with HER-2 or ERBB3 alterations were more likely to benefit from treatment.

“It is critical to investigate and establish strategies for targeting patients who have multiple mutations that are potentially actionable, such as HER-2, EGFR and multiple others,” Pal said. “We don’t know, however, which mutations drive tumor growth and which are passenger mutations.”

Other trials are evaluating monoclonal antibodies that target tumor antigens.

B-701 (BioClin Therapeutics) — an anti-FGFR3 antibody — is under study in a phase 2 trial for patients with locally advanced or metastatic bladder cancer. MK-6018 (Merck), which targets carcinoembryonic antigen, is being studied in a phase 1 trial for patients with advanced or recurrent cancers, including bladder cancer.

The Borealis-2 trial is assessing apatorsen (OGX-427, OncoGenex) — designed to inhibit production of heat shock protein 27, an intracellular protein that protects cancer cells by helping them survive — in combination with docetaxel for patients with metastatic bladder cancer. Final results from that trial are expected this year.

PAGE BREAK

The Southwest Oncology Group’s COXEN study is exploring the possibility of basing treatment off of a genetic understanding of the disease, Theodorescu said. Researchers of that trial are using the COXEN gene expression model to predict the best chemotherapy treatment for patients based on individual tumor characteristics.

“We are realizing now that patients with bladder cancer have a very diverse genomic profile,” Pal said. “This is something we have known for a while, but that had just not been actionable.”

Increased awareness

Challenges in diagnosis and staging, as well as in conducting clinical trials, may stem from the fact that resources historically have been limited.

The influx of potential treatment options appears to have prompted a change.

“Medical oncologists are becoming more interested in bladder cancer,” Steinberg said. “They are recognizing that they have treatments and that they have the ability to be part of the management.”

The experts with whom HemOnc Today spoke each credited The Bladder Cancer Advocacy Network (BCAN) as a driving force behind the effort to accelerate progress in bladder cancer.

BCAN, which celebrated its 10th anniversary last year, announced the launch of the Bladder Cancer Genomics Consortium to improve understanding of the disease’s genetic landscape in hopes of developing targeted therapies.

The project aims to genomically profile an initial cohort of 200 patients with metastatic urothelial carcinoma from eight participating institutions. Physicians will receive the results of these tests, as well as information about relevant clinical trials.

“I predict that, 5 years from now, we will be giving systemic chemotherapy and possibly targeted therapy to patients based on biomarker-specific predictors for the best possible response — true personalized medicine,” Lerner said.

Genomic profiling not only has the potential to alter treatment protocol and predict patient response. It also can improve diagnostics by detecting mutations in urine.

The practical application for urine detection will be in secondary screening of high-risk patients or those with a history of bladder cancer, according to Theodorescu.

“There are always a few cells that are shed,” he said.

Several tests are available to screen for tumor biomarkers in urine. These include UroVysion (Abbott Molecular), which tests for chromosome changes; bladder tumor–associated antigen tests; Immunocyt (DiagnoCure Inc.), which tests for mucin and carcinoembryonic antigen; and NMP22 BladderChek (Alere), which tests for the NMP22 protein.

However, these tests are not 100% accurate, and they may indicate bladder cancer in individuals without the disease.

“If a urine test — which a patient could mail in — could rule out the need for a cystoscopy, the impact would be financially and practically substantial,” Theodorescu said.

Other barriers exist.

Most patients recommended to undergo cystectomy need to travel to an academic center.

“On average, our patients come from 1 to 2 hours away,” Skinner, who is on the BCAN advisory board, told HemOnc Today. “There has been real effort to raise awareness at the primary care and urologist levels regarding all facets of the disease, but there’s still a long way to go.”

Last year, the journal Bladder Cancer — spearheaded by Lerner and Theodorescu — published its first issue, providing another source of information and collaboration for the clinical and scientific community.

“I tell medical students and early career physicians: ‘Publish early, and publish often,’” Steinberg said. “BCAN has done a tremendous job of encouraging research around the calendar, as well as raising awareness at the patient level.

“Years ago, women got together and decided to change breast cancer,” Steinberg added. “They got money, advocacy, awareness and research into that disease. That is what we are seeing in bladder cancer, and what needs to continue.” – by Nick Andrews

References:

Bochner BH, et al. N Engl J Med. 2014;doi:10.1056/NEJMc1405213.

Burger M, et al. Eur Urol. 2013;doi:10.1016/j.eururo.2013.03.059.

Choudhury NJ, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.66.3047.

PAGE BREAK

Johnson D, et al. Abstract 1622. Presented at: American Urologic Association Annual Scientific Meeting; May 4-8, 2013; San Francisco.

Rosenberg JE, et al. Lancet. 2016;doi:10.1016/S0140-6736(16)00561-4.

Steinberg G, et al. Cancer. 2014;doi:10.1002/cncr.28416.

The following were presented at ASCO Annual Meeting; June 3-7, 2016; Chicago:

Balar AV, et al. Abstract LBA4500.

Gschwend JE, et al. Abstract 4503.

For more information:

Arjun V. Balar, MD, can be reached at arjun.balar@nyumc.org.

Seth P. Lerner, MD, can be reached at slerner@bcm.edu.

Sumanta K. Pal, MD, can be reached at spal@coh.org.

Charles J. Ryan, MD, can be reached at ryanc@medicine.ucsf.edu.

Eila C. Skinner, MD, can be reached at skinnere@stanford.edu.

Gary D. Steinberg, MD, can be reached at gsteinbe@surgery.bsd.uchicago.edu.

Dan Theodorescu, MD, PhD, can be reached at dan.theodorescu@ucdenver.edu.

Disclosure: Lerner reports consultant/advisory roles with BioCancell, Ferring, Neucleix, Sitka, Taris, Telesta, UroGen and Vaxiion; research grants or funding from Endo, FKD Therapies, Genome Dx, Roche/Genentech and Viventia; and is co-editor in chief of Bladder Cancer. Pal reports consultant roles with Bristol-Myers Squibb, Genentech, Pfizer and Novartis and honoraria from Genentech. Steinberg reports consultant roles with Ciclo Med, Cold Genesys, Fidia, Heat Biologics, Karl Storz, Photocure, Roche/Genentech and Taris Biomedical. Theodorescu reports equity ownership in Aurora Oncology, as well as intellectual property in and inventor involvement with Myriad Genetics and NantHealth. He also is co-editor in chief of Bladder Cancer. Balar, Ryan and Skinner report no relevant financial disclosures.

 

POINTCOUNTER 

Should adjuvant chemotherapy be standard for patients with muscle-invasive bladder cancer?

POINT

Adjuvant chemotherapy should be offered to fit patients with muscle-invasive bladder cancer who have not received neoadjuvant chemotherapy.

At least three types of evidence should be required for clinicians to adopt a therapy as standard practice, in this case the use of adjuvant chemotherapy for people undergoing local treatment — cystectomy, radiotherapy or both — for transitional cell carcinoma of the bladder.

Ian F. Tannock, MD, PhD
Ian F. Tannock

1. Biological plausibility — ie, that metastatic failure is a common cause of death and that chemotherapy effectively causes remission of metastatic bladder cancer in a substantial proportion of patients;

2. Evidence from randomized controlled trials (RCTs) that adjuvant chemotherapy leads to improved survival. Here, consistency of effect is more important than a single RCT with a P value reaching the arbitrary threshold of less than .05. The American Statistical Association is sufficiently concerned about misinterpretation of P values that they issued a formal statement indicating that P values do not measure the probability that the studied hypothesis is true, nor the probability that the data were produced by random chance alone. They state that scientific conclusions and policy decisions should not be based only on whether a P value passes a specific threshold; and

3. Evidence from well-designed health outcome studies. This is essential to determine whether the proposed treatment leads to improved outcome in the real-world population. Although randomization has the advantage of ensuring similar characteristics between arms, RCTs recruit a highly selected population that does not reflect everyday practice. Health outcome studies are criticized because statistical methods such as propensity analyses effectively correct for known biases — such as the tendency to give adjuvant chemotherapy to those with better performance status — but they cannot guarantee that a difference in outcome is due only to differences in the treatment received. The advantages of randomly assigning selected patients in RCTs and analyzing real-world data from large patient samples in health outcome studies provide complementary evidence. If their results are contradictory, one should question the value of a treatment; if they are consistent, the level of evidence is very high.

Adjuvant chemotherapy for bladder cancer is supported by all three of the criteria described above.

Biological plausibility is strong. Treatment of people with metastatic bladder cancer with chemotherapy regimens such as cisplatin and gemcitabine, MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) or accelerated MVAC leads to response rates of approximately 50%, with a small proportion of sustained complete responses. This experience is equal or better to that achieved for women with breast cancer, where long-term complete responses are rare. People who relapse after local treatment of bladder cancer usually do so in the first 2 to 3 years, almost certainly from micrometastatic disease present initially, providing a strong rationale for adjuvant chemotherapy.

PAGE BREAK

Several RCTs evaluated modern adjuvant chemotherapy, and although individually underpowered, these trials generally suggest improved survival, with an HR of approximately 0.78.

There are three large health outcome studies of adjuvant chemotherapy applied to real-world populations that provide evidence of similar or better levels of survival benefit.

The evidence in favor of adjuvant chemotherapy for patients with muscle-invasive bladder cancer is very strong. It should be offered to fit patients who have completed local treatment and who have not received neoadjuvant therapy.

 

References:

Austin PC. Multivariate Behav Res. 2011;46:399-424.

Booth CM and Tannock IF. Br J Cancer. 2014;doi:10.1038/bjc.2013.725.

Booth CM and Tannock IF. JAMA Oncol. 2015;doi:10.1001/jamaoncol.2015.1210.

Booth CM, et al. Cancer. 2014;doi:10.1002/cncr.28510.

Galsky MD, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.64.1076.

Ioannidis JP. PLoS Med. 2005;doi.org/10.1371/journal.pmed.0020124.

Leow JJ, et al. Eur Urol. 2014;doi:10.1016/j.eururo.2013.08.033.

Sternberg CN, et al. Lancet Oncol. 2015;doi:10.1016/S1470-2045(14)71160-X.

Svatek RS, et al. Clin Cancer Res. 2010;doi:10.1158/1078-0432.CCR-10-0457.

Wasserstein RL and Lazar NA. Am Stat. 2016;doi:10.1080/00031305.2016.1154108.

Ian F. Tannock, MD, PhD, is professor of medicine and medical biophysics at Princess Margaret Cancer Centre and University of Toronto. He can be reached at ian.tannock@uhn.ca. Disclosure: Tannock reports no relevant financial disclosures.

COUNTER

No strong evidence exists to support use of adjuvant chemotherapy for muscle-invasive bladder cancer.

It’s a daunting task to debate the venerable Professor Ian F. Tannock, MD, PhD, my friend and collaborator of more than 3 decades. Prior to commencing my manifesto, I would refer the reader to our joint publication from 1990, espousing the basic principles of high-quality research that have not changed much in the past quarter century.

Derek Raghavan, MD, PhD, FACP, FASCO
Derek Raghavan

Professor Tannock previously noted that change of treatment requires supporting biology, evidence from appropriately powered randomized trials and data from well-designed outcome studies. When considering why adjuvant chemotherapy has not been proven effective in the management of invasive bladder cancer, the available data do not meet any of his espoused standards.

With regard to biological support, bladder cancer is a markedly heterogeneous malignancy, and it has been shown repeatedly that established paradigms of systemic therapy — such as the MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) or CMV (cisplatin, methotrexate and vinblastine) regimens — have not had equivalent success in advanced disease to the systemic therapy paradigms established in breast cancer, the poster child of classical adjuvant treatment. Tannock and I collaborated in demonstrating the utility of neoadjuvant CMV chemotherapy in a study that began decades ago, showing a true impact on survival that was not an artifact of statistical quirks or chicanery, but this approach placed surgery as definitive therapy following bulk reduction of chemotherapy-sensitive elements within occult metastatic disease.

Sadly, the genitourinary oncology community has disappointed by failing to support the well-powered available trials; but, even the largest reported trial failed to show a pragmatically meaningful survival benefit. Further, the statistical analysis revealed that the most substantial impact of adjuvant chemotherapy was on node-negative disease; this unexpected observation either supports the contention that some of the surgery in this multinational study was suboptimal, or that four cycles of adjuvant chemotherapy is insufficient to control this heterogeneous disease.

It is never wise to ignore the published literature. Thus, I remind the good professor of the Italian study that assessed the gemcitabine–cisplatin regimen as adjuvant chemotherapy after radical cystectomy. It demonstrated — at best — equivalence, or at worst, inferiority of the chemotherapy regimen compared with observation alone. I’m a little surprised that Professor Tannock has previously cited the many flawed meta-analyses of adjuvant chemotherapy for bladder cancer, but simply would remind him that all included the large German studies that tested adjuvant chemotherapy against no chemotherapy (in most patients), which simply confounded each analysis.

This brings me to the final criterion of the Tannock credo — that is, “well-designed outcome studies.” Sadly, the community-based “real population” studies appear to have been led by investigators who believe that propensity matching has similar power as well-powered randomized trials. That simply is not true. The concept of garbage-in-garbage-out is important to this domain, and the problem is relatively simple. Propensity matching simply does not account for all the clinician- and tumor-driven variables that cloud the comparisons listed “in the real world” that lead to huge case- and treatment-selection biases — specifically, selection of the more robust patients with more normal cardiopulmonary and metabolic function for adjuvant treatment, selection “out” of more complex tumors (eg, those with nonurothelial cancer elements), and unstructured variability of pathological diagnosis and follow-up approaches. Thus, the proposition that this type of database study is truly meaningful is ridiculous.

PAGE BREAK

There is no strong evidence biologically, from trials or well-designed outcome studies, to support adjuvant chemotherapy for muscle-invasive bladder cancer. Chemotherapy is expensive and potentially toxic, and using it routinely without clear, supporting level-one data is a bad idea.

 

References:

Booth CM and Tannock IF. Br J Cancer. 2014;doi:10.1038/bjc.2013.725.

Booth CM and Tannock IF. JAMA Oncol. 2015;doi:10.1001/jamaoncol.2015.1210.

Booth CM, et al. Cancer. 2014;doi:10.1002/cncr.28510.

Cognetti F, et al. Ann Oncol. 2011;doi:10.1093/annonc/mdr354.

Galsky MD, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.64.1076.

Leow JJ, et al. Eur Urol. 2014;doi:10.1016/j.eururo.2013.08.033.

Raghavan D and Tannock I. “Clinical trials in genitourinary cancer: What have they achieved?” Combination Therapy in Urologic Malignancy. Ed. Smith PH. London: Springer-Verlag London, 1990. 225-253.

Raghavan D, et al. Ann Oncol. 2014:doi:10.1093/annonc/mdu092.

Sternberg CN, et al. Lancet Oncol. 2015;doi:10.1016/S1470-2045(14)71160-X.

Svatek RS, et al. Clin Cancer Res. 2010;doi:10.1158/1078-0432.CCR-10-0457.

Derek Raghavan, MD, PhD, FACP, FASCO, is HemOnc Today’s Chief Medical Editor for Oncology. He also is president of Levine Cancer Institute at Carolinas HealthCare System. He can be reached at derek.raghavan@carolinashealthcare.org. Disclosure: Raghavan reports no relevant financial disclosures.