- Top of page
- Conflicts of Interest
Bladder cancer is a commonly diagnosed disease in Western society. In Australia, an estimated 2217 new cases of bladder cancer were diagnosed in 2007, accounting for 2% of all malignancies . About 70% of bladder cancers are superficial or non-muscle invasive and these patients are treated with transurethral resection, sometimes followed by intravesical immunotherapy or chemotherapy. Muscle-invasive bladder cancer, defined as a primary tumour invading the muscularis propria (TNM stage T2 or greater), has a high recurrence rate and metastatic potential with an estimated 5-year survival rate after cystectomy of 33–73% . The high mortality rate is thought to be due to micro-metastatic disease present at the time of radical cystectomy. As a result, optimal management of muscle-invasive bladder cancer includes high quality surgery together with perioperative chemotherapy .
Perioperative chemotherapy entails the administration of neoadjuvant chemotherapy or adjuvant chemotherapy. The rationale for neoadjuvant chemotherapy is its potential to eradicate the primary lesion and to treat micrometastatic foci of disease before the patient is debilitated by the major surgery. Two randomized phase III studies have shown a survival advantage for neoadjuvant chemotherapy in bladder urothelial cancer. The first of these was conducted by the Southwestern Oncology Group (SWOG 8710) and demonstrated an improvement in 5-year survival for patients with clinical stage T2-T4a muscle-invasive bladder cancer treated with three cycles of neoadjuvant MVAC (methotrexate, vinblastine, doxorubicin and cisplatin) followed by radical cystectomy compared to radical cystectomy alone (57% vs. 43%; p = 0.06) . The second study was led by the Medical Research Council (MRC) and found that three cycles of neoadjuvant CMV (cisplatin, methotrexate and vinblastine) prior to RC or primary radiotherapy significantly improved overall survival at 10 years compared to surgery or radiation alone (36% vs. 30%; p = 0.037, HR 0.84) . The findings of these two studies are further supported by a meta-analysis which demonstrated a 5% improvement in overall survival with the addition of neoadjuvant cisplatin-based chemotherapy to radical cystectomy (50% vs. 45%; p = 0.016, HR 0.87) . Gemcitabine and cisplatin (GC) is commonly used based on the favourable toxicity profile and efficacy data similar to MVAC, extrapolated from the metastatic setting . A retrospective analysis examined the benefit of neoadjuvant GC and showed almost similar pT0 rates at cystectomy (26% with GC and 28% with MVAC) . This regimen allowed for timely drug delivery with a median 91% drug delivery for cisplatin and 90% for gemcitabine.
The use of adjuvant chemotherapy in muscle-invasive bladder cancer is more controversial. Individual studies of adjuvant chemotherapy have generally been small, recruited slowly and been relatively underpowered. Three small studies published over 15 years ago demonstrated a significant improvement in progression free survival [9-11], however this did not result in enhanced overall survival in two of these studies [9, 11]. More contemporary studies have also failed to demonstrate an improvement in recurrence free survival with either GC  or MVAC . Nevertheless, a meta-analysis performed in 2006 showed an absolute improvement in overall survival of 9% at three years . Thus, although there is an absence of positive survival data from large, randomized studies, many medical oncologists will administer adjuvant cisplatin based chemotherapy for patients with perivesical tumour extension (≥pT3) or node positive disease .
Despite evidence supporting the use of perioperative chemotherapy in muscle-invasive bladder cancer, anecdotal experience suggests that the uptake of perioperative chemotherapy and especially neoadjuvant chemotherapy has been poor by the Australian oncology community. As a result, we examined the utilisation of perioperative chemotherapy over time in a contemporary cohort of bladder UC patients at our tertiary academic center.
- Top of page
- Conflicts of Interest
In this single institution retrospective study, we examined the utilization over time and consequent oncological impact of perioperative chemotherapy in patients with bladder UC. We found that 17% and 46% of potentially eligible patients received neoadjuvant and adjuvant chemotherapy respectively. Importantly, despite the relatively small number of patients in this study, the use of perioperative chemotherapy significantly reduced the risk of recurrence.
A number of studies have previously shown that the majority of patients with bladder UC do not receive perioperative chemotherapy, with rates ranging between 1.2 and 19% for neoadjuvant chemotherapy, and 1.5 and 37% for adjuvant chemotherapy [19-26]. Consistent with these findings, data published from the Surveillance, Epidemiology and End Results (SEER) Medicare database in 2011 showed only 6% of 8719 patients received neoadjuvant chemotherapy . Similarly, data from the National Cancer Database showed that only 6% of bladder UC patients received neoadjuvant chemotherapy in 2003. Interestingly, by 2011 this figure had risen to 13% . In our study, we also saw a temporal increase in the uptake of neoadjuvant chemotherapy from 2007 onwards. This coincided with the introduction of our Uro-Oncology MDM, reflecting the utility of a multi-disciplinary approach to these patients.
There are several possible reasons for the low rates of use of perioperative chemotherapy in bladder UC including patient co-morbidities and potential treatment toxicities. Co-morbid illnesses are commonly found in bladder UC patients, reflecting the strong association with smoking and advanced age at diagnosis. A large United States study of bladder UC patients found that an AACI score >5 correlated with lower rates of adjuvant chemotherapy utilisation, despite the fact that they were more likely to have pT3+ and/or N+ disease . In contrast, we found that the AACI score did not appear to impact on administration of perioperative chemotherapy, although smaller numbers may have limited the power of our study.
Among our patients, clinical stage was the main determinant for the use of neoadjuvant chemotherapy, consistent with prior reports . However, it is well recognized that clinical staging is imprecise with risks of both under-staging and over-staging. Transurethral resection is susceptible to under-staging due to sampling error, and clinical examination is also often inaccurate in assigning clinical T-stage . Conversely, computed tomography, especially if it is carried out after trans-urethral resection, is prone to overstaging. Nodal staging by imaging can be inaccurate, with high rates of both false positive and false negative results . Positron emission tomography (PET) may provide greater accuracy , but funding and availability currently limit its use.
In regards to treatment toxicities from chemotherapy, the retrospective design of this study did not allow us to accurately assess this. However, patients prescribed neoadjuvant and adjuvant chemotherapy received 85% and 67%, respectively of their planned cycles, indicating that treatment was relatively well tolerated. Moreover, neoadjuvant chemotherapy did not delay cystectomy with the median time from completion of chemotherapy to surgery being less than six weeks. This is an important indicator since delays in the definitive surgical treatment are associated with worse outcomes in bladder UC patients .
Nevertheless, certain chemotherapy regimens for bladder UC can be associated with high rates of toxicity. The SWOG 8710 study reported grade 3 or 4 adverse event in over 50 % of patients treated with neoadjuvant MVAC. In the adjuvant setting, grade 3 or 4 toxicity rates of 20% and 15 % have been reported with PGC (paclitaxel, gemcitabine and cisplatin) and GC respectively . Concerns about toxicity have perhaps led to lower rates of use of MVAC or CMV in Australia. In our study, the most commonly used perioperative chemotherapy regimens combined a platinum agent with gemcitabine. The use of such regimens may be associated with lower efficacy, with pathological complete response rates ranging from 7–26% [8, 33-35] for GC in comparison to a pathological complete response rate of 38% for MVAC in the SWOG 8710 study. In contrast, a recent retrospective study reported comparable pathological complete response rates for GC and MVAC (25% vs. 31%) and no difference in overall survival .
We acknowledge that this study has several limitations. Being a retrospective study we were not always able to ascertain the specific reasons why perioperative chemotherapy was not offered in particular patients. In addition, the number of patients was relatively small and thus there may have been insufficient power for some of the analysis. The duration of follow-up was shorter for some of the patients treated with chemotherapy, although this did not seem to have a confounding effect on outcomes.
In conclusion, we have found that perioperative chemotherapy remains relatively under-utilised at our center despite recent increases in the use of neoadjuvant treatment. Patients were more likely to receive perioperative chemotherapy if they had adverse clinical or pathological features, and chemotherapy appears to reduce their risk of recurrence. Improving outcomes for bladder UC patients remains a major challenge and our study highlights the need for further prospective studies to establish the optimal application of perioperative chemotherapy and simultaneously the development of strategies for increased utilisation of perioperative chemotherapy for those who may benefit from it.