A review by AndrÃ© P. Fay, MD, on the data supporting the standard use of neoadjuvant and adjuvant chemotherapy for patients with muscle invasive bladder cancers.
André P. Fay, MD
Dana-Farber Cancer Institute/Lank Center for Genitourinary Oncology
Urothelial carcinoma is the most common histologic subtype of bladder cancer and accounts for 90% of all cases.1 In the United Sates, more than 72,000 new cases of bladder cancer and about 15,000 deaths were estimated in 2013.2
In patients with muscle invasive bladder cancers (MIBC), pathologic stage and nodal status are the most important prognostic factors for progression and overall survival (OS).3 Radical cystectomy alone is associated with a 5-year survival rate of 80% for organ-confined disease without lymph node metastasis.4 However, patients with extravesical disease and patients with lymph node involvement have a 5-year survival rate of approximately 40-50% and 15-35%, respectively.5
Contemporary management of patients with MIBC consists of a multimodality treatment including surgery, cytotoxic systemic chemotherapy, and, for selected patients, chemoradiation with the objective of bladder preservation. While chemotherapy is part of the standard of care in the neoadjuvant setting, studies show it is infrequently used, and while some of the benefits of cytotoxic therapy alternatively may be attained via adjuvant treatment, the effectiveness of that strategy is still under investigation.
Here, we review the data supporting the standard use of neoadjuvant and adjuvant chemotherapy for patients with MIBC.6
Neoadjuvant Treatment Demonstrates Mild Benefit Neoadjuvant chemotherapy (NC) allows the in vivo assessment of the response in the primary tumor, indicating whether the individual’s tumor is responsive to a specific chemotherapy regimen. In addition, the magnitude of the chemotherapy response correlates with long-term remission or survival.7
There can be some drawbacks to neoadjuvant chemotherapy use in this population. For patients without major responses to NC, the delay of definitive therapy may potentially increase the risk of disease progression. Furthermore, pre-surgical staging is not always correctly evaluated before radical cystectomy, and some low-risk patients may unnecessarily receive chemotherapy.3 However, several clinical trials have demonstrated that there is a role for NC in eligible patients with MIBC.
The Nordic I trial randomized 311 patients with cT1G3-T4NxM0 disease to two cycles of cisplatin and doxorubicin versus no chemotherapy. All patients received radiation therapy before radical cystectomy. There was no significant difference in OS and cancer-specific survival (CSS). However, the subgroup of patients who presented with pT3-T4 disease and received NC had a 15% absolute survival benefit.8 The Nordic Cystectomy-2 trial enrolled 309 patients who were randomized to receive three cycles of neoadjuvant cisplatin and methotrexate or surgery alone. No statistically significant differences in OS were found between the two groups (53% vs 46%)9.
While these two studies were underpowered and used what would now be considered suboptimal chemotherapy, a combined analysis of these two trials revealed a trend for better OS favoring NC (OS: 56% vs 48%; P = .049).10
A large prospective trial (EORTC-MRC) evaluated 976 patients with cT2 grade 3-T4N0 MIBC. Patients were randomized to receive three cycles of neoadjuvant cisplatin, methotrexate, and vinblastine (CMV) or no therapy before local definitive therapy.11 In this trial, local therapy was radiotherapy in 42% of patients, radical cystectomy in 50%, and a combination of surgery and radiotherapy in 8%. Updated long-term follow-up of these patients showed a statistically significant benefit in OS (36% vs 30%; HR: 0.84; P = .037).12 Following NC, pathologic complete response (pCR) was observed in 32.5% compared with 12.3% with surgery alone.
Southwest Oncology Group (SWOG) 8710 was a prospective randomized controlled trial of 317 patients with cT2-T4aN0M0 MIBC to compare three cycles of MVAC chemotherapy (methotrexate,vinblastine, doxorubicin, and cisplatin) preceding radical cystectomy against cystectomy alone.13 The pCR rate with MVAC was 38% compared with 12% with surgery alone. Although NC did not provide a statistically significant benefit in OS (57% vs 43%; P = 0.06), the original planned analysis was defined with one-sided P < 0.05, and this goal was achieved. In addition, NC did not increase surgeryrelated complications, and median OS was poor for patients with residual muscle-invasive disease or lymph node-positive disease: 3.4 and 2.4 years, respectively.14
Before the publication of the updated positive results of the EORTC-MRC trial,12 a meta-analysis of 11 randomized trials of NC included 3005 patients, and an absolute OS benefit of 5% (HR: 0.86; 95% CI, 0.77—0.95; P = .003) favoring platinum-based combinations was reported. Cisplatin was the platinum agent used in the vast majority of patients (>90%), with carboplatin used in 6-7% of patients. Additionally, platinum-based combinations appeared to be superior to platinum monotherapy.15
Due to the low rates of adoption of neoadjuvant chemotherapy, clinicians are often faced with the decision of whether or not to recommend adjuvant chemotherapy (AC) for many moderate- to high-risk patients.7
In a randomized trial, 140 patients were assigned to receive two neoadjuvant cycles followed by three adjuvant cycles after surgery compared to five adjuvant cycles of MVAC. This study was unable to demonstrate a survival difference, but there was a suggestion that NC may be more feasible than AC. Only 54 of 70 patients (77%) received two or more cycles of MVAC after surgery, whereas 68 of 70 (97%) received at least two cycles when assigned to initial neoadjuvant MVAC.16
Larger but incomplete contemporary studies of AC have also failed to provide definitive supportive evidence for routinely recommending AC. The prospective Italian Multicenter Trial of 194 patients was underpowered to demonstrate a survival difference in patients receiving four cycles of adjuvant gemcitabine and cisplatin (GC) (P = 0.24; HR: 1.29; 95% CI, 0.84—1.99).17 The Spanish Oncology Genitourinary Group (SOGUG) trial was designed to randomize 340 patients with T3-T4 or node-positive disease to treatment using four cycles of paclitaxel, gemcitabine, and cisplatin (PGC) or observation. However, the trial was prematurely closed after only 142 patients were enrolled. At a median follow-up of 51 months, adjuvant PGC resulted in a significant increase in OS compared to no chemotherapy (60% versus 30%, HR: 0.44).18
The largest study of AC was designed by EORTC to compare deferred therapy at the time of recurrence with adjuvant therapy according to physician choice of GC, MVAC, or dose-dense MVAC.
This trial had a goal of 660 patients but was closed after enrollment of only 242 due to slow accrual. Follow-up results are expected.
Meanwhile, an innovative biomarker-driven trial in patients with organ-confined disease (pT1-2, N0M0) randomized patients with altered p53 levels to three cycles of MVAC versus observation. There was no difference in relapse-free survival and no impact on any of the outcome measures based on p53 status.19
Recently, an updated meta-analysis of nine randomized AC trials including 945 patients reported a disease-free survival (DFS) and OS benefit (HR: 0.77; CI, 0.59—0.99; P = .049) in MIBC patients who received adjuvant chemotherapy after radical cystectomy compared with those who underwent surgery alone.20 Additionally, lymph node-positive patients appear to have a greater DFS benefit than lymph node-negative patients. There is still insufficient evidence to support the routine use of adjuvant chemotherapy, and NC is the standard. Nonetheless, many physicians consider AC to be reasonable in patients who have not received NC, who have pathologic extravesical and/or node-positive disease, and who are eligible to receive cisplatin-based combination therapy.20
Physicians are sometimes reluctant to pursue NC because they fear that it may increase the incidence of perioperative morbidity.21
However, when neoadjuvant and adjuvant chemotherapy have been compared, NC did not increase perioperative morbidity.16
Cisplatin-based NC is an established standard, improving overall survival in MIBC. However, less than 20% of patients undergoing radical cystectomy actually receive this modality of treatment, and its standard implementation needs to be pursued.22
At the same time, we must look for ways to improve the efficacy of these treatments. Dose intensification of MVAC has been studied in several prospective studies with the aim of increasing pCR rates while minimizing local treatment delay and avoiding the shortcomings of a potential curative surgery in non-responders.23,24
The addition of new agents to cisplatin-based chemotherapies may improve the pCR rate beyond the 30% observed in patients with cT2-4aN0M0 who are treated with cisplatin-based chemotherapy alone. Some attempts to add bevacizumab have been reported.25 However, it is still uncertain whether increasing complete response with dose intensification or the addition of targeted therapies will result in better long-term outcomes.
In the genomic era, understanding MIBC biology may lead to improvements in therapeutics. The Cancer Genome Atlas has allowed the molecular characterization of a large number of tumors, opening new avenues in the treatment of urothelial tumors. This may help us identify new targets for therapeutic interventions and biomarkers that can predict response to therapies, making it easier for us to appropriately select patients for the therapeutic strategies that are on the horizon.