Muscle-Invasive Bladder Cancer

Adjuvant Chemotherapy and Overall Survival in Muscle-Invasive Bladder Cancer: Still Climbing the Mountain

By Maha Hussain, MD, FACP, FASCO
April 10, 2015, Volume 6, Issue 6



Meta-analysis indicates a favorable effect of adjuvant cisplatin-based therapy—however, that does not equal level 1 evidence.

—Maha Hussain, MD, FACP, FASCO
Muscle-invasive bladder cancer can be a lethal disease despite curative intent local therapy, with 5-year survival that can be as low as 30% based on the extent of T status and/or lymph node involvement. The use of neoadjuvant chemotherapy with MVAC (methotrexate, vinblastine, doxorubicin, and cisplatin) or CMV (cisplatin, methotrexate, and vinblastine) has been shown to significantly improve overall survival, based on the results of two randomized phase III trials.1,2 Yet adopting this approach in the general patient population has been less than optimal in the United States.

Clinical Trials in High-Risk Patients

The desire to optimize the risk/benefit ratio based on better patient selection and to avoid delay of potentially curative local therapy has fueled several adjuvant clinical trials in high-risk patients, particularly those with pT3-4 disease or those with node-positive disease post radical cystectomy and pelvic lymph node dissection. Historically, several phase II trials have been conducted, but they had multiple design and conduct limitations.

Since 2001, three randomized phase III trials were attempted in high-risk patients based on pT or N+ status, testing the survival impact of adjuvant cisplatin-based chemotherapy post radical cystectomy and pelvic lymph node dissection. One of these trials was negative,3 whereas the other was reportedly positive, although no manuscript has been published to date.4 The third trial (European Organisation for Research and Treatment of Cancer [EORTC] 30994) was recently reported by Dr. Sternberg and colleagues in ­Lancet Oncology5 and is reviewed in this issue of The ASCO Post.

Trial Similarities and Differences

There are several common features across these three trials, including the key eligibility criteria and type of chemotherapy. Gemcitabine-cisplatin was used in one trial, and the combination of gemcitabine, cisplatin, and paclitaxel was used in another. EORTC 30994 allowed the institution’s choice of gemcitabine-cisplatin, MVAC, or high-dose MVAC, although most of the patients (108 of 128) received gemcitabine-cisplatin. Unfortunately, all of these trials were closed due to slow accrual and were thus underpowered for the primary endpoint.


However, relative to the other trials, EORTC 30994 had a larger number of patients recruited: 284 of a planned 660 patients. Despite significant improvements with immediate treatment in 5-year progression-free survival (47.6% vs 31.8%) and median progression-free survival (3.11 vs 0.99 years, hazard ratio = 0.54, P < .0001), there was no “statistically” significant difference in overall survival. There was, however, a trend in favor of the immediate treatment group: 5-year overall survival was 53.6% (95% confidence interval [CI] = 44.5%–61.8%) vs 47.7% (95% CI = 39.1%–55.8%), and median overall survival was 6.74 years (95% CI = 3.85 years to not reached) vs 4.60 years (2.15–6.25 years), with a hazard ratio of 0.78 (95% CI = 0.56–1.08, P = .13). Technically, this study is negative for a confirmed overall survival advantage; however, it is rather difficult to ignore the significant difference in progression-free survival in a disease in which relapse invariably predicts death and the trends in median and 5-year overall survival all favoring adjuvant chemotherapy. It is also difficult to resist the speculation of “what if”—specifically, could these results have been statistically significant if the study had reached full accrual? The genitourinary medical oncology community in general has adopted gemcitabine-cisplatin for managing advanced bladder cancer based on what is deemed to be clinically comparable outcomes, better toxicity profile, and better feasibility in metastatic disease; yet to date, the only positive prospective level 1 evidence for overall survival in the perioperative setting has been with MVAC or CMV, and no combination chemotherapy in metastatic disease has eclipsed MVAC. So “what if” the three adjuvant trials had been conducted with MVAC? Could the results have been different despite the small sample size? “Absence of proof” is not “proof of absence,” so when there are no definitive data from prospective randomized trials, large data pools can serve to provide some guidance. In the setting of adjuvant chemotherapy, a recent updated systematic review and meta-analysis of randomized trials indicated a pooled hazard ratio in favor of adjuvant cisplatin-based chemotherapy of 0.77 (95% CI = 0.59–0.99, P = .049), translating into a 23% relative decrease in risk of death with adjuvant chemotherapy vs control.
Weighing the Pros and Cos


So, with the totality of the data from perioperative chemotherapy trials to date, what should the standards be for patients with muscle-invasive bladder cancer in 2015? I would argue that the best available level 1 evidence to date with regard to survival, feasibility, safety, and tolerance supports the use of “neoadjuvant” cisplatin-based chemotherapy when possible. Adjuvant chemotherapy-based trials have not been feasible to conduct successfully. Meta-analysis indicates a favorable effect of adjuvant cisplatin-based therapy—however, that does not equal level 1 evidence.

Managing patients requires both the “science and art” of medicine. When the scientific evidence is not conclusive, informed joint decisions based on considering the pros and cons of adjuvant chemotherapy are certainly warranted in the setting of high-risk patients with resected muscle-invasive bladder cancer who did not receive neoadjuvant chemotherapy but who are medical candidates for chemotherapy.

New Treatment Strategies

Moving forward, it is critical that new treatment strategies/paradigms be expeditiously investigated to overcome the stagnation and current limitations of our therapies. With the exciting preliminary data from the PD-L1 (programmed death-ligand 1) inhibitors, including their overall promising safety profile, and other biologic discoveries in this disease come the promise of improving outcomes for patients with advanced urothelial carcinoma in general and muscle-invasive bladder cancer in the perioperative setting in particular—where cure is possible. ■

Disclosure: Dr. Hussain reported no potential conflicts of interest.