St. James's Institute of Oncology, St. James's University Hospital, Leeds, UK
Section of Oncology and Clinical Research, Leeds Institute of Molecular Medicine, University of Leeds, Leeds, UK
Department of Oncology, School of Medicine, Cardiff University, UK
Senior Lecturer/Honorary Consultant in Medical Oncology, Leeds Institute of Molecular Medicine and St. James's Institute of Oncology, Level 4, Bexley Wing, St. James's University Hospital, Beckett Street, Leeds LS9 7TF, United Kingdom
The authors gratefully acknowledge Dr Rob Jones and Dr Simon Chowdhury for critical reading of the manuscript, and Ms Karen Brady for expert secretarial assistance.
Meta-analysis data demonstrate a 5% absolute survival benefit for neoadjuvant chemotherapy (NAC) using cisplatin-based combination regimens in the radical treatment of muscle-invasive bladder cancer (MIBC). However, there are no randomized, controlled trial data on the optimum regimen. Accelerated methotrexate, vinblastine, doxorubicin, and cisplatin (AMVAC) is a dose-intense regimen that has the potential to minimize delays to definitive, potentially curative therapy. A retrospective analysis is presented of the efficacy and toxicity of AMVAC as NAC in patients with MIBC and its impact on the patient pathway.
Eighty consecutive patients with MIBC were treated with AMVAC as NAC by 2 UK multidisciplinary uro-oncology teams. Three or 4 cycles of AMVAC (methotrexate 30 mg/m2, vinblastine 3 mg/m2, doxorubicin 30 mg/m2, and cisplatin 70 mg/m2) were given at 2-week intervals, with granulocyte colony-stimulating factor support, prior to either radical surgery or radical radiotherapy.
All planned cycles of chemotherapy were completed, without dose reduction or delay in 84% of patients. All 80 patients subsequently received their planned definitive therapy. Grade 3/4 toxicities were seen in 26% of the 42% of patients for whom toxicity data are available, including 12% grade 3/4 neutropenia. Pathological complete response to AMVAC was seen in 43% of 60 surgical patients. Objective radiological local response was seen in 83% of 57 evaluable patients. Two-year disease-free and overall survival were 65% and 77%, respectively.
Bladder cancer is the most frequently occurring tumor of the urinary tract and accounts for approximately 70,000 new cancer cases and 14,000 deaths in the United States each year.1, 2 Approximately 25% of patients present with life-threatening muscle-invasive (pT2-4a) transitional cell carcinoma (TCC) of the bladder.3 Radical cystectomy with lymphadenectomy is the curative standard of care worldwide and leads to approximately 50% disease-free survival (DFS) at 5 years.4-6 Radical radiotherapy is widely used as an alternative potentially curative approach for patients with significant comorbidity, and offers the prospect of long-term organ preservation with better preservation of sexual function.7
Adjuvant and neoadjuvant chemotherapy (NAC) have each been investigated as a means of reducing the occurrence of effectively incurable distant metastases.8 Meta-analysis of adjuvant chemotherapy data has shown a trend toward benefit, but not reaching statistical significance,9 whereas individualized patient data for NAC with cisplatin-based combination chemotherapy are more robust, with an absolute overall survival (OS) benefit of 5% (from 45%-50%) at 5 years.10, 11 NAC has therefore come to be regarded by many as standard of care for appropriately selected patients undergoing curative treatment for muscle-invasive bladder cancer (MIBC)12, 13 but is not universally offered to patients. Among the reasons for NAC not being offered to all patients is concern regarding the delay which chemotherapy introduces to radical treatment. As a minor contributor to cure, relative to surgery or radiotherapy, prolonged administration of NAC and/or recovery from toxicity may introduce delays to definitive therapy, with a potential adverse effect on outcome.14-16
Two widely used neoadjuvant combination chemotherapy regimens are methotrexate, vinblastine, doxorubicin (Adriamycin), and cisplatin (MVAC)17 and gemcitabine and cisplatin (GC).18 “Accelerated” MVAC (AMVAC) is a more dose-intense form of “classical” MVAC, which employs the same doses of drugs but is given at shorter, 2-week intervals with granulocyte colony-stimulating factor (GCSF) support. GC and AMVAC have each been compared with classical MVAC (but not with each other) in randomized, controlled trials for patients with advanced urothelial cancer.19, 20 Both regimens result in similar efficacy, but less toxicity compared to MVAC.
As a dose-intense regimen, AMVAC offers the opportunity to minimize the delay to definitive treatment imposed by more protracted regimens such as GC and MVAC, but there is a lack of data regarding its use in the neoadjuvant setting. We therefore report our experience of AMVAC prior to radical cystectomy or radiotherapy in 80 patients treated at 2 UK cancer centers by specialist multidisciplinary teams.
MATERIALS AND METHODS
We performed a retrospective analysis of patient records for all 80 consecutive patients who received AMVAC chemotherapy in 2 treatment centers in the United Kingdom, at Oxford (44 patients, 55%) and Leeds (36 patients, 45%), between July 2003 and September 2007. Patients were treated in an identical fashion at the 2 centers, except that patients at Leeds received 4 cycles of chemotherapy, whereas patients at Oxford received 3, according to routine local practice. All patients had histologically proven TCC of the bladder with pathological and radiological evidence of T2-4a, N0-2, M0 disease. Patients with no radiological evidence of pelvic (N0) or abdominal nodal involvement received AMVAC as NAC prior to planned radical surgery or radiotherapy with curative intent; those staged as (N1-2) received the same regimen as primary chemotherapy in an attempt to down-stage disease, prior to possible subsequent radical therapy. All patients had a calculated glomerular filtration rate of 55 mL/minute or greater.
The AMVAC chemotherapy regimen consists of methotrexate 30 mg/m2, vinblastine 3 mg/m2, doxorubicin 30 mg/m2, and 4-hour infusion of cisplatin 70 mg/m2, together with hydration before and after administration. Treatment was repeated every 2 weeks, for 3 or 4 cycles, depending on the treating center, as above. Patients were given prophylactic GCSF for 7 days, starting 24 hours after the last dose of cytotoxic drug. All NAC was delivered in the 2 centers.
Staging computed tomography scans were performed within 4 weeks prior to commencing chemotherapy and, in the majority of cases, again within 4 weeks of completion of chemotherapy, prior to definitive treatment. Paired pre- and post-NAC scans were assessed by expert site-specialist uro-oncology radiologists for response to NAC. Where disease was confined to the bladder, scans were not reported to Response Evaluation Criteria In Solid Tumors (RECIST) criteria, because disease manifests almost exclusively as diffuse bladder wall thickening, rather than as discrete, measurable lesions.
After chemotherapy, patients either underwent radical cystectomy or radical radiotherapy as definitive therapy, with curative intent. Patients did not routinely undergo cystoscopic assessment between completion of NAC and definitive treatment. In all cases, at both centers, the definitive treatment modality (surgery or radiotherapy) was determined prior to commencement of chemotherapy. Definitive treatment was planned for approximately 4 weeks after the final dose of chemotherapy, with surgeons and radiotherapists usually being advised of recommended start dates for radical treatment as chemotherapy commenced.
Radical cystectomy routinely included an extended pelvic lymph node dissection, laterally to the genitofemoral nerve, posteriorly to the obturator nerve, and cranially to the aortic bifurcation.
Radical cystectomy specimens were each routinely assessed postoperatively by experienced specialist urological pathologists for evidence of macroscopic and microscopic residual tumor following NAC. Central review was not performed. Pathology reports were retrospectively reviewed to calculate the proportion of patients showing a pathological complete response (pCR or T0) to NAC.
Radical radiotherapy was administered as 55 Gy in 20 fractions (Leeds) or 66 Gy in 33 fractions (Oxford). Gross target volume was defined as whole bladder plus any radiologically detectable tumor; clinical target volume was 0.5 cm around visible tumor (in male patients, the prostatic urethra was included where the bladder base or trigone was involved with disease); planning target volume was 1.5 cm around clinical target volume. Patients treated with radical radiotherapy underwent regular cystoscopic bladder surveillance, to detect in-field local recurrences.
Kaplan-Meier estimates of disease-free survival (DFS) and OS were plotted. DFS was defined as time to recurrence of disease or death.
Eighty patients (64 male, 16 female) with a median age of 60 years were treated with AMVAC as NAC, including 23% aged 70 years or more at the commencement of chemotherapy (Table 1). Sixty-eight patients (85% of the total) were radiologically node-negative (rN0) on prechemotherapy staging by cross-sectional imaging. Definitive local therapy involved radical cystectomy for 60 patients (75%) and radical radiotherapy for 20 patients (25%).
Table 1. Patient Characteristics
No. of Patients (%)
Median age (range), years
(Radiological) nodal stage
No. of chemotherapy cycles
Definitive treatment modality
The median age of patients treated with radiotherapy was significantly greater than in patients treated by cystectomy (age, 64 years vs 59 years; P = .0015), but there was a lower male:female ratio (3:1 vs 4:1; not significant). The distribution of performance status was similar in both groups (PS0:PS1, 63%:37% and 61%:39%, respectively). In the absence of postchemotherapy histological specimens for patients treated with radical radiotherapy, and with the unavoidable uncertainties of radiological T staging, it is not possible to compare the local extent of tumors treated with radical radiotherapy with those treated by cystectomy. However, the proportion of patients with radiologically node-positive disease was similar between the 2 patient groups (13% of the 60 surgical patients; 20% of the 20 radiotherapy patients).
Thirty-eight (63%) of the radical cystectomies were performed at the same 2 centers in which NAC was delivered; the remaining 22 were performed at 1 of 4 other sites. Regardless of surgical location, all cystectomies were performed by specialist urological surgeons with extensive experience of performing cystectomies. The mean number of nodes removed was 17. To date, none of the patients treated with radical radiotherapy have required salvage cystectomy for muscle-invasive recurrence.
There were no deaths within 30 days of the most recent dose of chemotherapy, and only 2 of 60 surgical patients died within 30 days of surgery (perioperative mortality rate of 3.3%), neither of which was attributable to chemotherapy.
Seventy (87%) of the 80 patients completed all planned cycles (89% at Leeds, 4 cycles; 86% at Oxford, 3 cycles), without treatment delays or dose reductions. Six patients (7%) required a dose reduction due to chemotherapy-related toxicity, and a further 4 (5%) patients did not complete the planned number of cycles (3 because of toxicity of chemotherapy and 1 because of a revision of the histological diagnosis to a non-TCC histology). One patient received a total of 5 cycles of chemotherapy rather than the planned 3. The additional 2 cycles were felt to be appropriate to obtain maximal response, following a radiological response in radiologically positive pelvic lymph nodes, while awaiting commencement of radical radiotherapy. All patients who commenced NAC subsequently received their planned definitive treatment.
Duration of Chemotherapy and Time to Definitive Treatment
The median time from diagnosis to definitive treatment, including chemotherapy, for all 80 patients, was 137 days (range, 49-390 days). Eleven patients (18%) experienced significant delays (more than 200 days from diagnosis to definitive treatment). In 2 (2.5%) of these (one due to chest infection and the other to a transient ischemic attack and chest infection), it was impossible to exclude the possibility that delay may have been related to toxicity of chemotherapy. Of the remaining 9 patients, 3 each were due to prolonged waits for radiotherapy, medical events unrelated to treatment, and to clerical error. Median time from diagnosis of MIBC to commencement of NAC was 54 days (Fig. 1). The median time taken for the course of chemotherapy was 28 days (Oxford 3-cycle regimen) and 42 days (Leeds 4-cycle regimen), with an overall median of 34 days. All but 2 patients (97%) completed their course within 56 days. The median time from the start of chemotherapy until definitive treatment was 75 days (range, 40-259 days).
Tolerability and Toxicity
Detailed chemotherapy toxicity data are available on 34 of 80 patients (42%) (Fig. 2). Nine patients (26.5%) had 1 or more grade 3 or 4 toxicities. The most frequently encountered grade 3 or 4 toxicities were neutropenia (in 11.8% of assessable patients), nausea and/or vomiting (5.9%), fatigue (5.9%), stomatitis (5.9%), and diarrhea (2.9%). Data were not systematically collected on surgical complication rates following NAC.
Pathological Response Rates
Following NAC, 26 of all 60 surgical patients (43%) were pT0 at cystectomy (pCR to treatment with transurethral resection of bladder tumor [TURBT] plus neoadjuvant chemotherapy); see top of right hand column, Table 2. Twelve (20%) had residual muscle-invasive (pT2 or pT3) disease; none of the cystectomy specimens contained T4 disease. Eleven (19%) of the 60 surgically treated patients had 1 or more pelvic nodes containing viable tumor at cystectomy, despite NAC. There was no significant difference between the proportions of patients with radiologically node-negative (rN0) or node-positive (rN+) (15% and 20%, respectively) prechemotherapy scans who subsequently had pN+ disease at cystectomy.
Table 2. Pathological Outcomes Following Neoadjuvant Chemotherapy and Radical Cystectomya
Pathological Stage (at Cystectomy)
Radiological Stage (Before Chemotherapy)
rN0 (% of all rN0)
rN+ (% of all rN+)
All (% of total)
Histopathological node-negative (pN0) and node-positive (pN+) data are shown according to radiological nodal status (node-negative, rN0, or node-positive, rN+), prior to chemotherapy.
Radiological Response Rates
Histopathological assessment of cystectomy specimens is the best available short-term indicator of the success of NAC. Although this pathological staging information is not available for patients treated with radiotherapy, we sought to establish that radiological responses to NAC were seen both in preradiotherapy and in precystectomy patients. Radiological response after NAC was evaluable in 57 of 80 patients (71.2%) (Table 3). Complete response was seen in 18 (32%) of these 57 patients and partial response in a further 29 patients (51%), for an overall response rate (rCR + rPR) of 83%. There were no statistically significant differences between response rates and disease control rates seen in patients treated with radiotherapy, compared to those seen in cystectomy patients (data not shown). Radiological evidence of disease progression was seen in 3 (5%) of all 57 radiologically evaluable patients. All but 1 of the patients who underwent cystectomy and achieved a complete radiological response also had a pCR.
Table 3. Radiological Responses in Both Cystectomy and Radiotherapy Groupsa
Pre- and post-chemotherapy scans were compared for 57 of the 80 patients. Responses are shown, categorized according to whether patients were radiologically node-negative (rN0) or radiologically node-positive (rN+), prior to chemotherapy.
Objective response rate (CR + PR)
Disease control rate (CR + PR + SD)
All evaluable patients
Detailed follow-up data are available on 78 patients (97%). At a median follow-up of 27.5 months (range, 3-60 months), 32% (25 of 78) of assessable patients have recurred with locoregionally advanced or distant metastatic disease (Table 4). Recurrence rates were much higher in patients who were radiologically pelvic node-positive prior to chemotherapy than in those who were node-negative (75% vs 24%). Higher recurrence rates were seen in both node-negative and node-positive patients treated with radiotherapy (36% and 80%, respectively) than in those treated with surgery (21% and 71%, respectively), but this did not translate into a survival difference (see below).
Table 4. Recurrence and Disease-Specific Mortality Outcome Data for 78 of the 80 Patients, at a Median Follow-Up of 27.5 Months
Definitive Treatment Modality
Nodal Status Prior to Chemotherapy
Disease Status, No. (% of Total)
Cause-Specific Death, No. (%)
Surgery (n = 58)
Radiotherapy (n = 20)
Overall (n = 78)
Of the 78 patients for whom detailed follow-up data are available, 65% remained recurrence-free at 2 years. Projected median DFS is more than 60 months (Fig. 3A). There is a statistically significant difference (P = .01) in DFS for the node-negative (greater than 5 years) versus node-positive patients (13 months) (Fig. 3A), but no statistically significant difference in DFS between groups treated with radiotherapy and surgery (P = .32).
Fifteen patients (19%) have died, in 11 (14%) of whom death was attributable to bladder cancer (Table 4). The 2-year OS was 77% and projected median OS is more than 5 years (Fig. 3B). There was no statistically significant difference in OS between the node-positive, node-negative, and all-patient groups, nor between the surgery and radiotherapy groups.
Meta-analysis data10, 11, 21 support the routine use of neoadjuvant cisplatin-based combination chemotherapy as part of multimodality treatment with curative intent for patients with MIBC. However, the trials on which this is based include a variety of cisplatin-based regimens, resulting in a lack of consensus on the most appropriate regimen. We adopted AMVAC as our standard NAC regimen for MIBC, on the basis of its impressive response rates and DFS in the metastatic setting,19 and we present a retrospective analysis of outcomes. Meta-analysis data and the recent publication of mature data from the largest single trial of NAC13 demonstrate similar benefits for NAC, regardless of whether radical surgery or radical radiotherapy was used to achieve local control. We therefore analyzed data from all our NAC patients treated with radical intent, whether local treatment involved surgery or radiotherapy.
Our toxicity data, although retrospective and incomplete, and therefore subject to selection and ascertainment bias, suggest that AMVAC is well-tolerated and easily deliverable. The relative timings of endoscopic diagnosis, NAC, and definitive therapy suggest that delays to radical surgery or radiotherapy are acceptable.
The median time from diagnosis until definitive treatment of 137 days is considerably less than the 208 days reported in a recently published contemporary series employing predominantly GC chemotherapy.22 Intervals between endoscopic diagnosis of MIBC and commencement of NAC, and between completion of NAC and definitive therapy, were almost entirely due to “real world” administrative issues including multidisciplinary team discussion, second referral, radiotherapy, or operating theater waiting lists, rather than arising from clinical issues such as complications of treatment. Both the median and range of times from initiation of NAC to definitive treatment (10.7 weeks and 5.7-37 weeks, respectively) compare favorably with those reported recently in a retrospective analysis of 153 patients treated with NAC prior to cystectomy (16.6 and 6.3-195.6 weeks, respectively).23 The same study has suggested that the timing of cystectomy relative to NAC has little or no effect upon survival.23 Nevertheless, 70% of our patients waited less than 3 months from the start of chemotherapy to definitive treatment. This may be partly attributable to close multidisciplinary cooperation wherein the surgeon or radiotherapist was given a projected earliest date for definitive therapy as soon as the chemotherapy start date was determined, permitting advance planning of surgery or radiotherapy dates.
Concerns have previously been raised that surgery after NAC might result in a larger proportion of postoperative complications. The primary focus of this analysis is upon the outcomes of NAC, not upon surgical issues. However, blood loss, operating time, transfusion rates, and inpatient stay were felt by the specialist urological surgeons involved to be broadly comparable with patients undergoing radical surgery without preceding NAC. A separate formal study of surgical outcomes would be required to confirm this, but this impression is supported by mature data from an international study of NAC in 976 patients, including 428 patients undergoing cystectomy, which found that there was no evidence to suggest that the addition of NAC made cystectomy more dangerous.13
There is continuing controversy on the appropriate approach to patients with enlarged pelvic nodes, as determined radiologically, and whether chemotherapy should be used as the primary treatment modality in these patients. Results presented here show a statistically significant difference in DFS between node-positive and node-negative patients. We, and others, see cures in this group. In a series of 30 patients with initially unresectable or minimally metastatic disease, Ghadjar et al24 obtained a pT0 rate of 30% and observed a radiological complete response in enlarged pelvic lymph nodes in 16 of 21 rN+ patients after “induction” platinum-based chemotherapy, resulting in a projected DFS of 42%. However, it seems likely that the benefit of chemotherapy is much lower in the rN+ group, and chemotherapy used as primary therapy for these patients should not be regarded as genuinely neoadjuvant.
The survival advantage of NAC seen in meta-analysis is presumed to arise from the elimination of micro-metastatic disease, rather than from improved local disease control. However, rates of pCR (pT0), which predicts for improved survival,25, 26 can be used as a surrogate marker of systemic efficacy. Previously, pCR rates as high as 33% to 40% have been reported, compared to 6% to 15% in patients treated with TURBT alone (summarized in Weight et al).22 The pCR rate of 43% reported here is at least as good as any previously reported in this setting, and is similar to the 38% figure reported in a previous randomized study that included 126 patients treated with neoadjuvant “classical” MVAC, followed by cystectomy.17 Our principal interest has been in the efficacy and toxicity of NAC, not in drawing comparisons between local treatment approaches. Nevertheless, our data support the need for further trials to investigate an organ-sparing, multimodality approach (including AMVAC as NAC and radical radiotherapy) for patients with MIBC who achieve a complete clinical (radiological plus cystoscopic) response to TURBT plus NAC. Our high pCR figure is in contrast to the recently published figure of 7% in a contemporary series of 29 patients treated predominantly with GC as NAC.22
Radiological complete response rates (32%) underestimated the pCR rate (43%). It is important to note that quantitative assessment of bladder wall lesions by cross-sectional imaging is particularly difficult,27 because disease is often seen as a diffuse thickening over a significant area of urothelium, rather than as discrete, measurable lesions. A further caveat is that the assessment method used here is anatomical, rather than functional. Areas of abnormality on cross-sectional imaging (including enlarged nodes) may or may not represent viable tumor: we saw several rN0/pN+ as well as rN+/pN0 patients. Functional imaging, such as positron emission tomography–computed tomography28 or diffusion-weighted magnetic resonance29 scans may therefore be more appropriate for guiding potential bladder-conserving therapy, which includes NAC, or for identifying early in a course of chemotherapy those patients who are not responding, and who would be better served by discontinuing chemotherapy early, and proceeding straight to definitive treatment.
The 5% absolute survival benefit conferred by NAC in meta-analysis means that systemic treatment makes the difference between cure and no cure in only 1 patient in 20. The other 19 would either survive without chemotherapy or will die despite it. Correlation of functional imaging data with molecular biomarkers predictive of response to, or toxicity from, NAC may allow us to distinguish, on the basis of the molecular characteristics of transurethral resection specimens alone, between those patients whose best interests are not served by NAC and those with the most to gain from it.
To the best of our knowledge, this is the first published report of the use of AMVAC as NAC for treatment of MIBC. Although the retrospective analysis presented here is of a relatively small cohort of patients with short follow-up and incomplete toxicity data, our results suggest that AMVAC has many of the characteristics of an ideal NAC regimen. It is safe, well-tolerated, and easily deliverable, with high pathological and radiological response rates. Radiological disease progression is rare, and delays to radical therapy are acceptable. No patients were deprived of definitive therapy by receiving NAC. AMVAC therefore represents an acceptable regimen for use on a routine, off-trial basis and is an appropriate comparator for future randomized trials of NAC for MIBC.
J.D.C. and V.M. are supported by Cancer Research UK. A PhD studentship for P.H. (unrelated) is funded by Roche pharmaceuticals. The study itself has no specific funding.