- Email: [email protected]
Postchemotherapy Lymphadenectomy in Patients With Metastatic Urothelial Carcinoma: Long-Term Efficacy and Implications for Trial Design
Original Study
Postchemotherapy Lymphadenectomy in Patients With Metastatic Urothelial Carcinoma: Long-Term Efficacy and Implications for Trial Design Andrea Necchi,1 Patrizia Giannatempo,1 Salvatore Lo Vullo,2 Elena Farè,1 Daniele Raggi,1 Nicola Nicolai,3 Luigi Piva,3 Davide Biasoni,3 Tullio Torelli,3 Mario Catanzaro,3 Silvia Stagni,3 Massimo Maffezzini,3 Luigi Mariani,2 Roberto Salvioni3 Abstract There is no clear indication for postchemotherapy surgery in patients with metastatic urothelial cancer (UC). We analyzed the contribution of postchemotherapy lymphadenectomy on survival in patients treated at our center. Twenty-eight patients were identified and results compared with those of a nonsurgically treated cohort. Surgery and response to chemotherapy were prognostic for progression-free (PFS) and overall survival (OS). If confirmed, results might have implications in daily practice and clinical trials. Background: The contribution of postchemotherapy pelvic (PLND) or retroperitoneal lymphadenectomy (RPLND) on survival in patients with advanced and metastatic UC is still unclear. Patients and Methods: Between September 1986 and May 2012, 157 patients with locally advanced or metastatic UC received first-line chemotherapy consisting of mMVAC (modified methotrexate, vinblastine, doxorubicin, and cisplatin), according to our policy. Patients with subdiaphragmatic nodal disease and/or local recurrence only and who experienced at least stable disease (SD) were selected. Fifty-nine patients were identified, 28 of whom underwent surgery, 31 started consolidation chemotherapy with or without radiotherapy or observation. The prognostic effect of candidate factors on survival was evaluated using Cox proportional hazard regression models. Results: A total of 14 PLND and 14 RPLND patients were identified after they had achieved a complete response (CR; n ¼ 7) or a partial response (PR) and SD (n ¼ 21). Median follow-up was 88 months (interquartile range, 24-211 months). Median PFS was 18 (95% confidence interval [CI], 11-not estimated) and 11 (95% CI, 5-19) months, respectively, in favor of the surgical cohort and curves were statistically different (log-rank test, P ¼ .009). In multivariate analysis, postchemotherapy surgery was significantly prognostic for PFS and OS and response to chemotherapy (PR and SD vs. CR) was prognostic for PFS and trended to significance for OS. A model including these 2 factors showed bootstrap-corrected Harrel C statistics for PFS and OS of 0.65 and 0.68, respectively. Conclusion: In well selected patients with UC like those who achieved a clinical benefit with chemotherapy and had nodal metastatic disease, there was a survival advantage in removal of disease residuals. Clinical Genitourinary Cancer, Vol. 13, No. 1, 80-6 ª 2015 Elsevier Inc. All rights reserved. Keywords: Chemotherapy, Lymphadenectomy, Metastases, Postchemotherapy surgery, Transitional cell carcinoma, Urothelial cancer
Presented in part at the Society of Urologic Oncology (SUO) annual meeting, Bethesda, MD, December 4-6, 2013, and the 29th Annual Congress of the European Association of Urology (EAU), Stockholm, Sweden, April 11-15, 2014 1
Department of Medical Oncology 2 Clinical Epidemiology and Trials Organization Unit 3 Department of Surgery, Urology Unit Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
80
-
Clinical Genitourinary Cancer February 2015
Submitted: Mar 6, 2014; Revised: May 3, 2014; Accepted: Jun 3, 2014; Epub: Jun 11, 2014 Address for correspondence: Andrea Necchi, MD, Department of Medical Oncology, Medical Oncology 2 Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Via G. Venezian 1, 20133 Milano, Italy Fax: þ39 02 2390 3150; e-mail contact: [email protected]
1558-7673/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clgc.2014.06.003
Introduction The landscape of therapeutic options for metastatic urothelial cancer (UC) is unchanged since a few decades ago, since the introduction of MVAC (the combination of methotrexate, vinblastine, doxorubicin, and cisplatin) with a reported 65% to 70% complete (CR) or partial response (PR) and a median survival approximating 14 months.1,2 Comparable survival estimates were further achieved with cisplatin and gemcitabine (CG), which became a standard of care based on the results of a single phase III trial reporting a better tolerability, and, because of its easier administration in an outpatient setting.3,4 The picture, however, is that of a relapse or disease progression after response, occurring in most patients except for those (5%-10% at best) who benefit from a durable survival (eg, cure). Attempts to improve the results by modifying the original MVAC schedule or adding taxanes to CG resulted in equal efficacy, indicating the need for a paradigm shift in this disease.5-7 At our institute, patients with unresectable locally advanced disease, metastatic disease, or recurrence after surgery were sequentially administered modified schedules of MVAC (mMVAC), either in or outside clinical trials, and global results suggested an overlapping efficacy with a slightly better tolerability over the original regimen.7 Defining the optimal postchemotherapeutic indication of patients still having disease residuals is still an unmet need that suffers substantial heterogeneity because of a number of factors including the entity of tumor shrinkage, sites of residual disease needing a demanding technical quality, and performance status. As a result of the absence of clinical recommendations on this issue, some patients are indifferently offered either surgery, or observation only, or additional treatment with non-crosse resistant agents, or radiotherapy. Data from surgical series in the postchemotherapy setting are scarce in this disease, but as a matter of fact at least a proportion of patients like those who present with locoregional metastases are at greater risk of relapsing at the sites of response than that of developing distant metastases.8,9 A better understanding of the characteristics of patients who could benefit from an aggressive surgical approach could be useful to improve global outcomes and to better select those who are otherwise best candidates for additional systemic therapy as consolidation or maintenance, in the frame of modern clinical trials. The objective of this study was to assess the contribution of lymphadenectomy after chemotherapy in responding patients with nodal disease or soft tissue recurrence only.
Table 1 Distribution of Clinical Characteristics by Treatment Group Patients, n (%) Characteristic
Study Group (n [ 28)
Median Age, Years (IQR)
Control Group (n [ 31)
59 (50-66)
ECOG PS 0
25 (89.3)
25 (80.7)
1
3 (10.7)
6 (19.3)
Bladder
17 (60.7)
27 (87.1)
Upper tract
11 (39.3)
4 (12.9)
16 (57.2)
12 (38.7)
Tumor Primary
Disease Extent Before mMVAC Regional nodal disease Metastatic nodal disease
3 (10.7)
4 (12.9)
Lymph node or soft tissue relapse after surgerya
9 (32.1)
15 (48.4)
Tumor Burden Single nodal site Multiple nodal sites
9 (32.1)
8 (25.8)
19 (67.9)
23 (74.2)
Response to First-Line mMVAC CR
7 (25.0)
9 (29.0)
PR
17 (60.7)
14 (45.2)
SD
4 (14.3)
8 (25.8)
Data are presented as n (%) except where otherwise stated. Abbreviations: ECOG PS ¼ Eastern Cooperative Oncology Group performance status; IQR ¼ interquartile range; mMVAC ¼ modified methotrexate, vinblastine, doxorubicin, and cisplatin. a Radical cystectomy or nephroureterectomy, including regional lymphadenectomy in all cases.
Metastatic disease was defined as the involvement of lymph nodes outside the true pelvis and above the aortic bifurcation for primary tumors of the bladder, and as disease outside the boundaries of regional nodes for the renal pelvis and ureteral primaries, according
Figure 1 Kaplan Meier Curves of Overall Survival According to Pathologic Response to Methotrexate, Vinblastine, Doxorubicin, and Cisplatin
Patients and Methods Patient Population We reassessed clinical data of the 157 patients with locally advanced or metastatic urothelial carcinoma who received mMVAC as first-line therapy at Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy.7 Of those, we selected patients yielding exclusive subdiaphragmatic, abdominal, or pelvic nodal disease and who achieved at least a stable disease (SD) after 4 to 6 cycles of mMVAC in the period September 1986 to May 2012. Eligibility included patients who presented with either nodal metastases at diagnosis or with a nodal or soft tissue relapse after surgery (radical cystectomy or nephroureterectomy).
Clinical Genitourinary Cancer February 2015
- 81
Surgery in Metastatic Urothelial Cancer Table 2 Number of Patients Who Relapsed After Postchemotherapy Surgery in the 2 Groups According to Selected Categories Study Group Category Bladder Primary Only Upper Tract Primary Only Metastatic at Diagnosis Relapsed After Surgery Bladder and Metastatic at Diagnosis Bladder and Relapsed After Surgery Upper Tract and Metastatic at Diagnosis Upper Tract and Relapsed After Surgery
Control Group
Total
Relapses, n
%
Total
Relapses, n
%
17 11 19 9 16 1 3 8
8 5 8 5 5 1 1 4
47.1 45.5 42.1 55.6 31.2 100 33.3 50.0
27 4 16 15 16 11 0 4
24 3 15 12 14 9 e 3
88.9 75.0 93.7 80.0 87.5 81.8 e 75.0
to the seventh edition of the American Joint Committee on Cancer Staging Manual.10 Of the 59 identified patients there were 28 cases who underwent surgery after chemotherapy (herein referred to as “study group”) and 31 who did not (control group) because of either achieving a CR to mMVAC (n ¼ 9), receiving consolidation chemotherapy with or without radiotherapy (n ¼ 14), having a history of complications after previous major surgery (n ¼ 4), or for unknown reasons (n ¼ 4). Pelvic lymphadenectomy (PLND) included the common iliac nodes proximally, external iliac nodes to the inguinal ring distally, the obturator fossa, and perivesical nodes medially. Retroperitoneal lymph node dissection (RPLND) included the nodes at the renal vessels cranially, the pre- and periaortic nodes including the right border of the aorta for left-sided cases, the pre- and paracaval nodes including the left border of the inferior vena cava for right-sided cases, and the ipsilateral common iliac nodes distally. After surgery, patients were followed up every 3 months with a thoraxabdomen computed tomography scan alternating with abdominal ultrasound and chest x-ray up to the second year, then every 6 months up to the third year, and annually thereafter.
Statistical Analysis Response Evaluation Criteria in Solid Tumors (RECIST) criteria were used to define either a CR, a PR, or an incomplete response (SD and disease progression) to mMVAC.11 Pathologic CR (pCR) was defined as the absence of residual viable cancer in the radically resected disease. Progression-free survival (PFS) and overall survival (OS) were the primary end points. PFS was defined as the interval from the first dose of mMVAC to the date of the first documented disease progression or death for any reason and OS was defined as
the interval from the first dose of mMVAC to the date of death. Observation time was censored at the date of last contact for alive patients. OS and PFS curves were estimated using the Kaplan-Meier method and compared with the log-rank test. Multivariable Cox regression analyses were conducted to examine the independent prognostic effect of prespecified factors on PFS and OS. The following variables were considered: patient group (study vs. control), site of tumor primary, recurrent disease after surgery (cystectomy or nephroureterectomy and regional lymphadenectomy) or metastatic disease at diagnosis, and response to mMVAC. A prognostic model was built according to these results. The discrimination ability of this model was then quantified using the Harell concordance index, using a bootstrap procedure to correct for overfitting (bias-corrected c-index). The analyses were carried out using the SAS (SAS Institute Inc, Cary, NC) and R software (http:// www.r-project.org/). The results were considered statistically significant whenever a 2-sided P value < .05 was achieved.
Results Patient Characteristics Principal characteristics and distribution of frequencies between patients who did and did not receive surgery are summarized in Table 1. The median age was 59 years (interquartile range [IQR], 50-66), and baseline Eastern Cooperative Oncology Group (ECOG) performance status was 1 in 3 (10.7%) and 6 cases (19.3%) in the 2 groups, respectively, and it was 0 in all of the remaining cases. There were 11 cases (39.3%) of upper tract UC in the study group compared with 4 cases (12.9%) in the control group. Patients needed first-line chemotherapy for these reasons: 9 cases (32.1%) and 15 (48.4%) were histologically-ascertained
Table 3 Cox Model Analysis of Progression-Free Survival (n [ 59) Univariable Analysis
82
-
Multivariable Analysis
Variable
HR
95% CI
P
Postchemotherapy Surgery, Yes Versus No Timing of Nodal Metastases, Recurrent Disease Versus at Diagnosis Response to Chemotherapy, PR/SD Versus CR Tumor Primary, Upper Tract Versus Bladder
0.43 0.80
0.22-0.84 0.43-1.51
1.83 0.56
0.87-3.87 0.26-1.22
Abbreviation: HR ¼ hazard ratio for interquartile range. a P ¼ Wald test P value.
Clinical Genitourinary Cancer February 2015
a
HR
95% CIa
Pa
.014 .498
0.31 0.58
0.14-0.67 0.27-1.24
.003 .159
.113 .144
2.42 1.25
1.07-5.51 0.46-3.43
.035 .660
a
Andrea Necchi et al Table 4 Cox Model Analysis of Overall Survival (n [ 59) Univariable Analysis
Multivariable Analysis
Variable
HR
95% CI
P
Postchemotherapy Surgery, Yes Versus No Timing of Nodal Metastases, Recurrent Disease Versus at Diagnosis Response to Chemotherapy, PR/SD Versus CR Tumor Primary, Upper Tract Versus Bladder
0.37 0.91
0.18-0.77 0.47-1.78
1.85 0.45
0.82-4.14 0.19-1.09
a
HR
95% CIa
Pa
.008 .786
0.30 0.78
0.13-0.70 0.35-1.73
.005 .537
.137 .076
2.29 0.87
0.93-5.65 0.29-2.66
.071 .812
a
Abbreviation: HR ¼ hazard ratio for interquartile range. a P ¼ Wald test P value.
relapses after previous surgery in the 2 cohorts, and the remaining were metastatic patients at diagnosis. Among patients with relapsing disease, all but 1 had an upper tract tumor in the study group compared with 4 of 15 in the control group. Median time from surgery to relapse had been 8 (IQR, 5-11) and 9 months (IQR, 2-16.5), respectively. None had received perioperative chemotherapy and surgery was radical and had provided regional lymphadenectomy in all cases. Seven patients (25%) achieved a CR to mMVAC in the study group compared with 9 (29%) in the control group. Postchemotherapy surgery was radical in all 28 cases and consisted of 14 PLND, 11 RPLND, and 3 pelvic and RPLND. Five patients received a bladder-sparing approach with nodal dissection only (2 PLND and 3 RPLND).
Survival Results After a median follow-up of 88 (IQR, 24-211) months a total of 13 disease progressions occurred in the study group and 27 in the control group. Median PFS was 18 months (IQR, 11-not estimated) and 11 (IQR, 5-19) months in the study and control group, respectively (P ¼ .009; see Supplemental Figure 1A in the online version). Median OS was 37 months (IQR, 20-not estimated) and 19 (IQR, 9-38) months in the 2 groups (P ¼ .005; see Supplemental Figure 1B in the online version). Five-year PFS and OS in the study group were 35.2% (95% confidence interval [CI], 18.1%-68.3%) and 48.7% (95% CI, 29.7%-79.8%), respectively. A difference in PFS was also significant when considering the population of consolidating chemotherapy with or without radiotherapy only in the control arm (P < .001). Eight pCRs were obtained and 6 of those patients remained progression-free after a median of 31.5 (IQR, 15.5-129.2) months. However, pCR did not affect PFS (P ¼ .101) or OS (P ¼ .306; Figure 1). Two of the 5 patients who retained their primary tumor had a relapse; none involving the bladder, and 1 having a recurrence within the surgical field. Additional information on the number of relapses that occurred in each group according to selected categories is provided in Table 2.
P ¼ .008). On multivariable analysis, the latter was significantly prognostic for PFS (HR, 0.31; 95% CI, 0.14-0.67; P ¼ .003) and OS (HR, 0.30; 95% CI, 0.13-0.70; P ¼ .005). Response to chemotherapy was prognostic for PFS (HR, 2.42; 95% CI, 1.07-5.51; P ¼ .035) and trended to significance for OS (HR, 2.29; 95% CI, 0.93-5.65; P ¼ .071). Finally, a model was built including CR to chemotherapy alone, disease-free status after surgery, and residual disease. The c-index was 0.65 for PFS and 0.68 for OS and Figure 2 shows the PFS and OS curves according to each category (P ¼ .002 for PFS and P < .001 for OS, respectively). Median PFS was 18 months with CR to mMVAC Figure 2 Progression-Free (A) and Overall Survival (B) According to the Therapeutic Modality and Response: Complete Response or Partial Response-Stable Disease to Chemotherapy Without Surgery Versus Chemotherapy Followed by Surgery (All Patients)
Univariable and Multivariable Cox Regression Analyses Univariable and multivariable Cox analyses on PFS and OS are presented in Tables 3 and 4. None of the variables analyzed had a univariably significant association with PFS and OS except for postchemotherapy surgery (for PFS: hazard ratio [HR], 0.43; 95% CI, 0.22-0.84; P ¼ .014; for OS: HR, 0.37; 95% CI, 0.18-0.77;
Abbreviation: CT ¼ chemotherapy.
Clinical Genitourinary Cancer February 2015
- 83
84
Clinical Response to Chemotherapy
Resected Metastatic Sites
PFS
OS
Notes
NA
CR: 8 PR: 33 SD/PD: 7
Unresectable primary: 4; lymph nodes/abdominal/pelvic mass: 32; viscera: 14
NA
Combination of CR to CT and surgery was associated with the greatest survival estimate
80
Bladder: 80
CR: 24 PR: 44 SD/PD: 12
NA
1996-2001
11
Bladder: 11
CR: 7 PR: 4
Bladder/fixed bulky pelvic mass: 32; pelvic lymph nodes: 31; retroperitoneal lymph nodes: 17 Subdiaphragmatic lymph nodes: 11
16 of 50 alive at 5 years (10 of 33 with viable cancerunresectable tumor vs. 6 of 17 with no residual cancer)a 34 of 80 (42%) survived 9 months to 5 yearsa
Median: 7 months (95% CI, NA)b
Median: 14 months (95% CI, NA)b
de Vries et al17
1987-2007
14
Bladder: 14
CR: 5 PR: 9
Supraregional lymph nodes: 14
10 of 14 patients relapsed
Median: 10.1 months (95% CI, 5.6-13.8)b
Siefker-Radtke et al18
1985-2001
31
Bladder: 24; renal pelvis: 7
NA
Median: 7 months (95% CI, NA)b
Otto et al19
1995-1999
70
Bladder: 70
Lehmann et al21
1991-2008
44
Bladder: 35; upper tract: 9
Refractory to MVAC: 70 NA
Lung metastases: 24; lymph nodes: 4; other viscera: 3 Lymph nodes: 68; viscera: 49 Retroperitoneal lymph nodes: 25; distant lymph nodes: 5; viscera: 19
Median: 14.6 months (95% CI, 6.6-22.5)b
Median: 23 months (95% CI: 19-83)b; 5-year OS: 33% (95% CI, NA)b Median: 7 months (95% CI, NA)b Median: 27.2 months (95% CI, 6.8-47.6)b
Matsuguma et al22 Abe et al23
1990-2005
32
NA
NA
Lung metastases: 32
1989-2012
42
NA
1986-2012
28
Retroperitoneal lymph-nodes: 15; distant lymph nodes: 5; lung: 12; other: 10 Regional nodal disease: 16; metastatic nodal disease: 3; nodal or soft tissue relapse after surgery: 9
5-year OS: 50% (95% CI; NA)b Median: 26 months (IQR, 11-90)b
Present series
Bladder: 21; upper tract: 18; both sites: 3 Bladder: 17; upper tract: 11
5-year PFS rate: 26% (95% CI, NA)b NA
Median: 18 months (95% CI, 11-NE)a
Median OS: 37 months (95% CI, 20-NE)a
Time Frame
Number of Patients
Primary Site
Dodd et al13
1983-1994
50
Herr et al14
1984-1999
Sweeney et al16
Author
CR: 7 PR: 17 SD: 4
NA
Combination of clinical/pathological response to CT and surgery yielded the greatest prognostic effect on survival No correlation between clinical response to CT and survival; 2 positive retroperitoneal lymph nodes were associated with longer PFS Long-term survival was possible in 24% of patients selected from responders to CT and with nodal disease only No clinical/pathological predictor of survival No survival advantage of surgery; possible benefit in QoL indexes No impact of pathologic response and site of metastases on survival; a few patients survived after resection of lung or bone metastases Solitary metastasis 3 cm had the best prognostic effect on survival Resection of solitary lung or nodal metastasis significantly prognostic in UVA Favorable prognostic effect of surgery in patients with subdiaphragmatic nodal metastases after response to CT
Abbreviations: CI ¼ confidence interval; CT ¼ chemotherapy; IQR ¼ interquartile range; MVAC ¼ methotrexate, vinblastine, doxorubicin, and cisplatin; NE ¼ not estimated; OS ¼ overall survival; PFS ¼ progression-free survival; QoL ¼ quality of life; UVA ¼ univariable analysis. a Since the initiation of chemotherapy. b Since metastasectomy.
Surgery in Metastatic Urothelial Cancer
Clinical Genitourinary Cancer February 2015
Table 5 Summary of the Available Outcomes of Postchemotherapy Surgery in Metastatic Urothelial Cancer
Andrea Necchi et al alone (IQR, 15-37) and CR with chemotherapy and surgery (IQR, 11-not estimated), and it was 9.5 month for patients with residuals (IQR, 4-12). Median OS for patients with CR after chemotherapy was 44 months (IQR, 19-196) and it was 37 months for radically resected patients (IQR, 20-not estimated), and 12 months only for those with residuals (IQR, 9-23).
Discussion Postchemotherapy surgery with curative intent is an option in effect since the early 1980s.12-23 In this retrospective analysis of very selected patients with metastatic UC (37.5% of the overall patients receiving first-line mMVAC at our center) we could demonstrate an independent prognostic effect of surgical clearance of disease on PFS and OS. Results aligned with those reported by Dodd and colleagues but in a more homogeneous patient population, because these authors evaluated patients yielding T3-4 N0 disease and those having visceral metastases.13 A few other retrospective series have been subsequently reported on this issue and available findings identified chemosensitive nodal disease as the most favorable prognostic variable (Table 5).13,14,16-19,21-23 Herr and colleagues reported on 207 patients with unresectable or lymph node-positive disease, 80 of whom underwent surgery.14 Results are similar to the present ones; the authors evaluated the role of postchemotherapy surgery in combination with response to chemotherapy. As a difference, the pathologic response was a predictor of outcome in Herr’s series and the combination of clinical and pathologic response had the greatest effect on survival. The site of metastatic disease is a recognized prognostic factor in patients with metastatic UC who undergo first-line chemotherapy and it has been included in the major available prognostic scores.24-26 Results of the surgical series actually underscore this issue, yet a very small fraction of long-term survivors after visceral (ie, pulmonary) metastasectomy should be accounted for. Available and present findings require a prospective validation especially in patients with more extensive residuals, although in our series, most cases were classified as yielding multiple nodal sites, irrespective of the treatment group. Differences in the surgical quality across centers could represent an additional bias when dictating the choice between a surgical approach and observation or alternative treatments; so far, a survival advantage can only be hypothesized. Present results should be interpreted with caution because of the existence of several potential criticisms in the analysis. The retrospective quality of data and the long-dated series might account for a major limitation. Our aim was to discriminate the effect of surgery over recognized confounding variables, particulary because we included bladder and upper tract tumors, and the site of tumor primary did not result as prognostic in the multivariable model. In the absence of a recognized indication, the choice to perform a radical surgical resection of residuals was dictated by a critical evaluation of the features and clinical course of individual cases, and consolidating radiotherapy or chemotherapy, and observation, were equal options. Despite these limitations, the administered treatment was homogeneous (modified mMVAC as per institutional policy) and there were minor differences in baseline characteristics between the 2 groups (eg, distribution of cases with ECOG-performance status
1, although it was < 2 in all cases). Complete CR to chemotherapy was a significantly favorable prognostic factor on multivariable analysis for PFS and trended to significance for OS. Although PR and SD were jointly analyzed to enhance model parsimony, the frequency of SD was slightly different in the 2 groups (14% vs. 25%, respectively), possibly explaining at least in part the difference of survival curves. Remarkably high 5-year PFS and OS rates have been achieved, in the overall series and in the subgroups, confirming that a vast heterogeneity should be accounted for when interpreting the efficacy of first-line chemotherapy. Data showed that a historical 5-year OS rate of 28% with a multimodal approach that included surgery of visceral metastases21 could be improved to > 45% by selecting patients with nodal disease only. Results were not negligible because 9 patients had experienced a relapse after surgery of tumor primary that included regional lymphadenectomy. The relatively stable median PFS estimate of 8 months reported with cisplatin-based regimens could be improved to 18 months in this series. Our numbers did not allow us to dissect a significant survival effect of pathologic response, although there was a trend toward a better outcome in patients who achieved a pCR. Unfortunately, the retrospective quality of the data were not suitable for an analysis of the side effects and recovery after surgery, although no major complications or deaths were observed. Finally, maintenance chemotherapy after response to first-line is a rational approach being increasingly used in clinical trials to test new drugs in an earlier disease setting, and early results are available.27 Data from our analysis suggest that these trials should have more restrictive criteria than those currently used, and exclude candidate patients with the same features of those we presented and for whom an extended radical surgical approach is rather advisable.
Conclusion To summarize, results of the present retrospective comparison between postchemotherapy surgery and other treatment or observation seemed to favor the first option. The superiority has been confirmed in PFS and OS in very selected patients with advanced UC. These results might have implications for the design of clinical trials, particularly in the maintenance setting after response to firstline chemotherapy.
Clinical Practice Points Present analysis of a single-center series provided additional data
on the role of postchemotherapy surgery for patients with metastatic urothelial carcinoma. Available information is scarce in the field and the close surveillance of disease residuals following first-line chemotherapy is currently the preferred choice in most cases. Present analysis was focused on 28 patients having subdiaphragmatic nodal metastatic disease only and who achieved at least a stable disease after cisplatin-based chemotherapy. The outcomes have been compared to those obtained in a cohort of 31 patients with comparable features who did not undergo surgical resection. Postchemotherapy surgery, as well as response to chemotherapy, were the two significant predictors in the prognostic model including clinical variables.
Clinical Genitourinary Cancer February 2015
- 85
Surgery in Metastatic Urothelial Cancer Results deserve validation in prospective clinical trials and
represent a call for publication of further larger case series on this topic. Most importantly, results could have an implication on the design of postchemotherapy maintenance trials.
Disclosure The authors have stated that they have no conflicts of interest.
Supplemental Data The supplemental figure accompanying this article can be found in the online version at http://dx.doi.org/10.1016/j.clgc. 2014.06.003.
References 1. Gallagher DJ, Milowsky MI, Bajorin DF. Advanced bladder cancer: status of firstline chemotherapy and the search for active agents in the second-line setting. Cancer 2008; 113:1284-93. 2. Sternberg CN, Yagoda A, Scher HI, et al. Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium. Cancer 1989; 64:2448-58. 3. von der Maase H, Hansen SW, Roberts JT, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000; 17:3068-77. 4. von der Maase H, Sengelov L, Roberts JT, et al. Long-term survival results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol 2005; 23:4602-8. 5. Sternberg CN, Mulder PH, Schornagel JH, et al. Randomized phase III trial of high-dose-intensity methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) chemotherapy and recombinant human granulocyte colony-stimulating factor versus classic MVAC in advanced urothelial tract tumors: European Organization for Research and Treatment of Cancer Protocol No. 30924. J Clin Oncol 2001; 19: 2638-46. 6. Bellmunt J, von Der Maase H, Mead GM, et al. Randomized phase III study comparing paclitaxel/cisplatin/gemcitabine and gemcitabine/cisplatin in patients with locally advanced or metastatic urothelial cancer without prior systemic therapy: EORTC Intergroup Study 30987. J Clin Oncol 2012; 30:1107-13. 7. Necchi A, Mariani L, Giannatempo P, et al. Long-term efficacy and safety outcomes of modified (simplified) combinations of methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) as front-line therapy for unresectable or metastatic urothelial cancer. Clin Genitourin Cancer 2014; 12:203-209.e1. 8. Dimopoulos MA, Finn L, Logothetis CJ. Pattern of failure and survival of patients with metastatic urothelial tumors relapsing after cisplatinum-based chemotherapy. J Urol 1994; 151:598-600.
86
-
Clinical Genitourinary Cancer February 2015
9. Svatek RS, Siefker-Radtke AO, Dinney CP. Management of metastatic urothelial cancer: the role of surgery as an adjunct to chemotherapy. Can Urol Assoc J 2009; 3(suppl 4):S228-31. 10. Edge SB, Byrd DR, Compton CC, et al, eds. AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer-Verlag; 2009:491-6. 11. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228-47. 12. Cowles RS, Johnson DE, McMurtrey MJ. Long-term results following thoracotomy for metastatic bladder cancer. Urology 1982; 20:390-2. 13. Dodd PM, McCaffrey JA, Herr H, et al. Outcome of postchemotherapy surgery after treatment with methotrexate, vinblastine, doxorubicin, and cisplatin in patients with unresectable or metastatic transitional cell carcinoma. J Clin Oncol 1999; 17:2546-52. 14. Herr HW, Donat SM, Bajorin DF. Post-chemotherapy surgery in patients with unresectable or regionally metastatic bladder cancer. J Urol 2001; 165:811-4. 15. Abe T, Shinohara N, Harabayashi T, et al. Impact of multimodal treatment on survival in patients with metastatic urothelial cancer. Eur Urol 2007; 52: 1106-13. 16. Sweeney P, Millikan R, Donat M, et al. Is there a therapeutic role for postchemotherapy retroperitoneal lymph node dissection in metastatic transitional cell carcinoma of the bladder? J Urol 2003; 169:2113-7. 17. de Vries RR, Nieuwenhuijzen JA, Meinhardt W, Bais EM, Horenblas S. Longterm survival after combined modality treatment in metastatic bladder cancer patients presenting with supra-regional tumor positive lymph nodes only. Eur J Surg Oncol 2009; 35:352-5. 18. Siefker-Radtke AO, Walsh GL, Pisters LL, et al. Is there a role for surgery in the management of metastatic urothelial cancer? The M.D. Anderson experience. J Urol 2004; 171:145-8. 19. Otto T, Krege S, Suhr J, Rübben H. Impact of surgical resection of bladder cancer metastases refractory to systemic therapy on performance score: a phase II trial. Urology 2001; 57:55-9. 20. Herr HW. Is metastasectomy for urothelial carcinoma worthwhile? Eur Urol 2009; 55:1300-1. 21. Lehmann J, Suttmann H, Albers P, et al. Surgery for metastatic urothelial carcinoma with curative intent: the German experience (AUO AB 30/05). Eur Urol 2009; 55:1293-9. 22. Matsuguma H, Yoshino I, Ito H, et al. Is there a role for pulmonary metastasectomy with a curative intent in patients with metastatic urinary transitional cell carcinoma? Ann Thorac Surg 2011; 92:449-54. 23. Abe T, Kitamura H, Obara W, et al. Outcome of metastasectomy for urothelial carcinoma: a multiinstitutional retrospective study in Japan. J Urol 2014; 191: 932-6. 24. Bajorin DF, Dodd PM, Mazumdar M, et al. Long term survival in metastatic transitional-cell carcinoma and prognostic factors predicting outcome of therapy. J Clin Oncol 1999; 17:3173-81. 25. Apolo AB, Ostrovnaya I, Halabi S, et al. Prognostic model for predicting survival of patients with metastatic urothelial cancer treated with cisplatin-based chemotherapy. J Natl Cancer Inst 2013; 105:499-503. 26. Galsky MD, Moshier E, Krege S, et al. Nomogram for predicting survival in patients with unresectable and/or metastatic urothelial cancer who are treated with cisplatin-based chemotherapy. Cancer 2013; 119:3012-9. 27. Grivas PD, Daignault S, Tagawa ST, et al. Double-blind, randomized, phase 2 trial of maintenance sunitinib versus placebo after response to chemotherapy in patients with advanced urothelial carcinoma. Cancer 2014; 120: 692-701.
Andrea Necchi et al Supplemental Figure 1
Kaplan Meier Curves of ProgressionFree (A) and Overall Survival (B) According to Treatment Group
Clinical Genitourinary Cancer February 2015
- 86.e1
Postchemotherapy Lymphadenectomy in Patients With Metastatic Urothelial Carcinoma: Long-Term Efficacy and Implications for Trial Design Andrea Necchi,1 Patrizia Giannatempo,1 Salvatore Lo Vullo,2 Elena Farè,1 Daniele Raggi,1 Nicola Nicolai,3 Luigi Piva,3 Davide Biasoni,3 Tullio Torelli,3 Mario Catanzaro,3 Silvia Stagni,3 Massimo Maffezzini,3 Luigi Mariani,2 Roberto Salvioni3 Abstract There is no clear indication for postchemotherapy surgery in patients with metastatic urothelial cancer (UC). We analyzed the contribution of postchemotherapy lymphadenectomy on survival in patients treated at our center. Twenty-eight patients were identified and results compared with those of a nonsurgically treated cohort. Surgery and response to chemotherapy were prognostic for progression-free (PFS) and overall survival (OS). If confirmed, results might have implications in daily practice and clinical trials. Background: The contribution of postchemotherapy pelvic (PLND) or retroperitoneal lymphadenectomy (RPLND) on survival in patients with advanced and metastatic UC is still unclear. Patients and Methods: Between September 1986 and May 2012, 157 patients with locally advanced or metastatic UC received first-line chemotherapy consisting of mMVAC (modified methotrexate, vinblastine, doxorubicin, and cisplatin), according to our policy. Patients with subdiaphragmatic nodal disease and/or local recurrence only and who experienced at least stable disease (SD) were selected. Fifty-nine patients were identified, 28 of whom underwent surgery, 31 started consolidation chemotherapy with or without radiotherapy or observation. The prognostic effect of candidate factors on survival was evaluated using Cox proportional hazard regression models. Results: A total of 14 PLND and 14 RPLND patients were identified after they had achieved a complete response (CR; n ¼ 7) or a partial response (PR) and SD (n ¼ 21). Median follow-up was 88 months (interquartile range, 24-211 months). Median PFS was 18 (95% confidence interval [CI], 11-not estimated) and 11 (95% CI, 5-19) months, respectively, in favor of the surgical cohort and curves were statistically different (log-rank test, P ¼ .009). In multivariate analysis, postchemotherapy surgery was significantly prognostic for PFS and OS and response to chemotherapy (PR and SD vs. CR) was prognostic for PFS and trended to significance for OS. A model including these 2 factors showed bootstrap-corrected Harrel C statistics for PFS and OS of 0.65 and 0.68, respectively. Conclusion: In well selected patients with UC like those who achieved a clinical benefit with chemotherapy and had nodal metastatic disease, there was a survival advantage in removal of disease residuals. Clinical Genitourinary Cancer, Vol. 13, No. 1, 80-6 ª 2015 Elsevier Inc. All rights reserved. Keywords: Chemotherapy, Lymphadenectomy, Metastases, Postchemotherapy surgery, Transitional cell carcinoma, Urothelial cancer
Presented in part at the Society of Urologic Oncology (SUO) annual meeting, Bethesda, MD, December 4-6, 2013, and the 29th Annual Congress of the European Association of Urology (EAU), Stockholm, Sweden, April 11-15, 2014 1
Department of Medical Oncology 2 Clinical Epidemiology and Trials Organization Unit 3 Department of Surgery, Urology Unit Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
80
-
Clinical Genitourinary Cancer February 2015
Submitted: Mar 6, 2014; Revised: May 3, 2014; Accepted: Jun 3, 2014; Epub: Jun 11, 2014 Address for correspondence: Andrea Necchi, MD, Department of Medical Oncology, Medical Oncology 2 Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Via G. Venezian 1, 20133 Milano, Italy Fax: þ39 02 2390 3150; e-mail contact: [email protected]
1558-7673/$ - see frontmatter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clgc.2014.06.003
Introduction The landscape of therapeutic options for metastatic urothelial cancer (UC) is unchanged since a few decades ago, since the introduction of MVAC (the combination of methotrexate, vinblastine, doxorubicin, and cisplatin) with a reported 65% to 70% complete (CR) or partial response (PR) and a median survival approximating 14 months.1,2 Comparable survival estimates were further achieved with cisplatin and gemcitabine (CG), which became a standard of care based on the results of a single phase III trial reporting a better tolerability, and, because of its easier administration in an outpatient setting.3,4 The picture, however, is that of a relapse or disease progression after response, occurring in most patients except for those (5%-10% at best) who benefit from a durable survival (eg, cure). Attempts to improve the results by modifying the original MVAC schedule or adding taxanes to CG resulted in equal efficacy, indicating the need for a paradigm shift in this disease.5-7 At our institute, patients with unresectable locally advanced disease, metastatic disease, or recurrence after surgery were sequentially administered modified schedules of MVAC (mMVAC), either in or outside clinical trials, and global results suggested an overlapping efficacy with a slightly better tolerability over the original regimen.7 Defining the optimal postchemotherapeutic indication of patients still having disease residuals is still an unmet need that suffers substantial heterogeneity because of a number of factors including the entity of tumor shrinkage, sites of residual disease needing a demanding technical quality, and performance status. As a result of the absence of clinical recommendations on this issue, some patients are indifferently offered either surgery, or observation only, or additional treatment with non-crosse resistant agents, or radiotherapy. Data from surgical series in the postchemotherapy setting are scarce in this disease, but as a matter of fact at least a proportion of patients like those who present with locoregional metastases are at greater risk of relapsing at the sites of response than that of developing distant metastases.8,9 A better understanding of the characteristics of patients who could benefit from an aggressive surgical approach could be useful to improve global outcomes and to better select those who are otherwise best candidates for additional systemic therapy as consolidation or maintenance, in the frame of modern clinical trials. The objective of this study was to assess the contribution of lymphadenectomy after chemotherapy in responding patients with nodal disease or soft tissue recurrence only.
Table 1 Distribution of Clinical Characteristics by Treatment Group Patients, n (%) Characteristic
Study Group (n [ 28)
Median Age, Years (IQR)
Control Group (n [ 31)
59 (50-66)
ECOG PS 0
25 (89.3)
25 (80.7)
1
3 (10.7)
6 (19.3)
Bladder
17 (60.7)
27 (87.1)
Upper tract
11 (39.3)
4 (12.9)
16 (57.2)
12 (38.7)
Tumor Primary
Disease Extent Before mMVAC Regional nodal disease Metastatic nodal disease
3 (10.7)
4 (12.9)
Lymph node or soft tissue relapse after surgerya
9 (32.1)
15 (48.4)
Tumor Burden Single nodal site Multiple nodal sites
9 (32.1)
8 (25.8)
19 (67.9)
23 (74.2)
Response to First-Line mMVAC CR
7 (25.0)
9 (29.0)
PR
17 (60.7)
14 (45.2)
SD
4 (14.3)
8 (25.8)
Data are presented as n (%) except where otherwise stated. Abbreviations: ECOG PS ¼ Eastern Cooperative Oncology Group performance status; IQR ¼ interquartile range; mMVAC ¼ modified methotrexate, vinblastine, doxorubicin, and cisplatin. a Radical cystectomy or nephroureterectomy, including regional lymphadenectomy in all cases.
Metastatic disease was defined as the involvement of lymph nodes outside the true pelvis and above the aortic bifurcation for primary tumors of the bladder, and as disease outside the boundaries of regional nodes for the renal pelvis and ureteral primaries, according
Figure 1 Kaplan Meier Curves of Overall Survival According to Pathologic Response to Methotrexate, Vinblastine, Doxorubicin, and Cisplatin
Patients and Methods Patient Population We reassessed clinical data of the 157 patients with locally advanced or metastatic urothelial carcinoma who received mMVAC as first-line therapy at Fondazione IRCCS Istituto Nazionale dei Tumori, Milano, Italy.7 Of those, we selected patients yielding exclusive subdiaphragmatic, abdominal, or pelvic nodal disease and who achieved at least a stable disease (SD) after 4 to 6 cycles of mMVAC in the period September 1986 to May 2012. Eligibility included patients who presented with either nodal metastases at diagnosis or with a nodal or soft tissue relapse after surgery (radical cystectomy or nephroureterectomy).
Clinical Genitourinary Cancer February 2015
- 81
Surgery in Metastatic Urothelial Cancer Table 2 Number of Patients Who Relapsed After Postchemotherapy Surgery in the 2 Groups According to Selected Categories Study Group Category Bladder Primary Only Upper Tract Primary Only Metastatic at Diagnosis Relapsed After Surgery Bladder and Metastatic at Diagnosis Bladder and Relapsed After Surgery Upper Tract and Metastatic at Diagnosis Upper Tract and Relapsed After Surgery
Control Group
Total
Relapses, n
%
Total
Relapses, n
%
17 11 19 9 16 1 3 8
8 5 8 5 5 1 1 4
47.1 45.5 42.1 55.6 31.2 100 33.3 50.0
27 4 16 15 16 11 0 4
24 3 15 12 14 9 e 3
88.9 75.0 93.7 80.0 87.5 81.8 e 75.0
to the seventh edition of the American Joint Committee on Cancer Staging Manual.10 Of the 59 identified patients there were 28 cases who underwent surgery after chemotherapy (herein referred to as “study group”) and 31 who did not (control group) because of either achieving a CR to mMVAC (n ¼ 9), receiving consolidation chemotherapy with or without radiotherapy (n ¼ 14), having a history of complications after previous major surgery (n ¼ 4), or for unknown reasons (n ¼ 4). Pelvic lymphadenectomy (PLND) included the common iliac nodes proximally, external iliac nodes to the inguinal ring distally, the obturator fossa, and perivesical nodes medially. Retroperitoneal lymph node dissection (RPLND) included the nodes at the renal vessels cranially, the pre- and periaortic nodes including the right border of the aorta for left-sided cases, the pre- and paracaval nodes including the left border of the inferior vena cava for right-sided cases, and the ipsilateral common iliac nodes distally. After surgery, patients were followed up every 3 months with a thoraxabdomen computed tomography scan alternating with abdominal ultrasound and chest x-ray up to the second year, then every 6 months up to the third year, and annually thereafter.
Statistical Analysis Response Evaluation Criteria in Solid Tumors (RECIST) criteria were used to define either a CR, a PR, or an incomplete response (SD and disease progression) to mMVAC.11 Pathologic CR (pCR) was defined as the absence of residual viable cancer in the radically resected disease. Progression-free survival (PFS) and overall survival (OS) were the primary end points. PFS was defined as the interval from the first dose of mMVAC to the date of the first documented disease progression or death for any reason and OS was defined as
the interval from the first dose of mMVAC to the date of death. Observation time was censored at the date of last contact for alive patients. OS and PFS curves were estimated using the Kaplan-Meier method and compared with the log-rank test. Multivariable Cox regression analyses were conducted to examine the independent prognostic effect of prespecified factors on PFS and OS. The following variables were considered: patient group (study vs. control), site of tumor primary, recurrent disease after surgery (cystectomy or nephroureterectomy and regional lymphadenectomy) or metastatic disease at diagnosis, and response to mMVAC. A prognostic model was built according to these results. The discrimination ability of this model was then quantified using the Harell concordance index, using a bootstrap procedure to correct for overfitting (bias-corrected c-index). The analyses were carried out using the SAS (SAS Institute Inc, Cary, NC) and R software (http:// www.r-project.org/). The results were considered statistically significant whenever a 2-sided P value < .05 was achieved.
Results Patient Characteristics Principal characteristics and distribution of frequencies between patients who did and did not receive surgery are summarized in Table 1. The median age was 59 years (interquartile range [IQR], 50-66), and baseline Eastern Cooperative Oncology Group (ECOG) performance status was 1 in 3 (10.7%) and 6 cases (19.3%) in the 2 groups, respectively, and it was 0 in all of the remaining cases. There were 11 cases (39.3%) of upper tract UC in the study group compared with 4 cases (12.9%) in the control group. Patients needed first-line chemotherapy for these reasons: 9 cases (32.1%) and 15 (48.4%) were histologically-ascertained
Table 3 Cox Model Analysis of Progression-Free Survival (n [ 59) Univariable Analysis
82
-
Multivariable Analysis
Variable
HR
95% CI
P
Postchemotherapy Surgery, Yes Versus No Timing of Nodal Metastases, Recurrent Disease Versus at Diagnosis Response to Chemotherapy, PR/SD Versus CR Tumor Primary, Upper Tract Versus Bladder
0.43 0.80
0.22-0.84 0.43-1.51
1.83 0.56
0.87-3.87 0.26-1.22
Abbreviation: HR ¼ hazard ratio for interquartile range. a P ¼ Wald test P value.
Clinical Genitourinary Cancer February 2015
a
HR
95% CIa
Pa
.014 .498
0.31 0.58
0.14-0.67 0.27-1.24
.003 .159
.113 .144
2.42 1.25
1.07-5.51 0.46-3.43
.035 .660
a
Andrea Necchi et al Table 4 Cox Model Analysis of Overall Survival (n [ 59) Univariable Analysis
Multivariable Analysis
Variable
HR
95% CI
P
Postchemotherapy Surgery, Yes Versus No Timing of Nodal Metastases, Recurrent Disease Versus at Diagnosis Response to Chemotherapy, PR/SD Versus CR Tumor Primary, Upper Tract Versus Bladder
0.37 0.91
0.18-0.77 0.47-1.78
1.85 0.45
0.82-4.14 0.19-1.09
a
HR
95% CIa
Pa
.008 .786
0.30 0.78
0.13-0.70 0.35-1.73
.005 .537
.137 .076
2.29 0.87
0.93-5.65 0.29-2.66
.071 .812
a
Abbreviation: HR ¼ hazard ratio for interquartile range. a P ¼ Wald test P value.
relapses after previous surgery in the 2 cohorts, and the remaining were metastatic patients at diagnosis. Among patients with relapsing disease, all but 1 had an upper tract tumor in the study group compared with 4 of 15 in the control group. Median time from surgery to relapse had been 8 (IQR, 5-11) and 9 months (IQR, 2-16.5), respectively. None had received perioperative chemotherapy and surgery was radical and had provided regional lymphadenectomy in all cases. Seven patients (25%) achieved a CR to mMVAC in the study group compared with 9 (29%) in the control group. Postchemotherapy surgery was radical in all 28 cases and consisted of 14 PLND, 11 RPLND, and 3 pelvic and RPLND. Five patients received a bladder-sparing approach with nodal dissection only (2 PLND and 3 RPLND).
Survival Results After a median follow-up of 88 (IQR, 24-211) months a total of 13 disease progressions occurred in the study group and 27 in the control group. Median PFS was 18 months (IQR, 11-not estimated) and 11 (IQR, 5-19) months in the study and control group, respectively (P ¼ .009; see Supplemental Figure 1A in the online version). Median OS was 37 months (IQR, 20-not estimated) and 19 (IQR, 9-38) months in the 2 groups (P ¼ .005; see Supplemental Figure 1B in the online version). Five-year PFS and OS in the study group were 35.2% (95% confidence interval [CI], 18.1%-68.3%) and 48.7% (95% CI, 29.7%-79.8%), respectively. A difference in PFS was also significant when considering the population of consolidating chemotherapy with or without radiotherapy only in the control arm (P < .001). Eight pCRs were obtained and 6 of those patients remained progression-free after a median of 31.5 (IQR, 15.5-129.2) months. However, pCR did not affect PFS (P ¼ .101) or OS (P ¼ .306; Figure 1). Two of the 5 patients who retained their primary tumor had a relapse; none involving the bladder, and 1 having a recurrence within the surgical field. Additional information on the number of relapses that occurred in each group according to selected categories is provided in Table 2.
P ¼ .008). On multivariable analysis, the latter was significantly prognostic for PFS (HR, 0.31; 95% CI, 0.14-0.67; P ¼ .003) and OS (HR, 0.30; 95% CI, 0.13-0.70; P ¼ .005). Response to chemotherapy was prognostic for PFS (HR, 2.42; 95% CI, 1.07-5.51; P ¼ .035) and trended to significance for OS (HR, 2.29; 95% CI, 0.93-5.65; P ¼ .071). Finally, a model was built including CR to chemotherapy alone, disease-free status after surgery, and residual disease. The c-index was 0.65 for PFS and 0.68 for OS and Figure 2 shows the PFS and OS curves according to each category (P ¼ .002 for PFS and P < .001 for OS, respectively). Median PFS was 18 months with CR to mMVAC Figure 2 Progression-Free (A) and Overall Survival (B) According to the Therapeutic Modality and Response: Complete Response or Partial Response-Stable Disease to Chemotherapy Without Surgery Versus Chemotherapy Followed by Surgery (All Patients)
Univariable and Multivariable Cox Regression Analyses Univariable and multivariable Cox analyses on PFS and OS are presented in Tables 3 and 4. None of the variables analyzed had a univariably significant association with PFS and OS except for postchemotherapy surgery (for PFS: hazard ratio [HR], 0.43; 95% CI, 0.22-0.84; P ¼ .014; for OS: HR, 0.37; 95% CI, 0.18-0.77;
Abbreviation: CT ¼ chemotherapy.
Clinical Genitourinary Cancer February 2015
- 83
84
Clinical Response to Chemotherapy
Resected Metastatic Sites
PFS
OS
Notes
NA
CR: 8 PR: 33 SD/PD: 7
Unresectable primary: 4; lymph nodes/abdominal/pelvic mass: 32; viscera: 14
NA
Combination of CR to CT and surgery was associated with the greatest survival estimate
80
Bladder: 80
CR: 24 PR: 44 SD/PD: 12
NA
1996-2001
11
Bladder: 11
CR: 7 PR: 4
Bladder/fixed bulky pelvic mass: 32; pelvic lymph nodes: 31; retroperitoneal lymph nodes: 17 Subdiaphragmatic lymph nodes: 11
16 of 50 alive at 5 years (10 of 33 with viable cancerunresectable tumor vs. 6 of 17 with no residual cancer)a 34 of 80 (42%) survived 9 months to 5 yearsa
Median: 7 months (95% CI, NA)b
Median: 14 months (95% CI, NA)b
de Vries et al17
1987-2007
14
Bladder: 14
CR: 5 PR: 9
Supraregional lymph nodes: 14
10 of 14 patients relapsed
Median: 10.1 months (95% CI, 5.6-13.8)b
Siefker-Radtke et al18
1985-2001
31
Bladder: 24; renal pelvis: 7
NA
Median: 7 months (95% CI, NA)b
Otto et al19
1995-1999
70
Bladder: 70
Lehmann et al21
1991-2008
44
Bladder: 35; upper tract: 9
Refractory to MVAC: 70 NA
Lung metastases: 24; lymph nodes: 4; other viscera: 3 Lymph nodes: 68; viscera: 49 Retroperitoneal lymph nodes: 25; distant lymph nodes: 5; viscera: 19
Median: 14.6 months (95% CI, 6.6-22.5)b
Median: 23 months (95% CI: 19-83)b; 5-year OS: 33% (95% CI, NA)b Median: 7 months (95% CI, NA)b Median: 27.2 months (95% CI, 6.8-47.6)b
Matsuguma et al22 Abe et al23
1990-2005
32
NA
NA
Lung metastases: 32
1989-2012
42
NA
1986-2012
28
Retroperitoneal lymph-nodes: 15; distant lymph nodes: 5; lung: 12; other: 10 Regional nodal disease: 16; metastatic nodal disease: 3; nodal or soft tissue relapse after surgery: 9
5-year OS: 50% (95% CI; NA)b Median: 26 months (IQR, 11-90)b
Present series
Bladder: 21; upper tract: 18; both sites: 3 Bladder: 17; upper tract: 11
5-year PFS rate: 26% (95% CI, NA)b NA
Median: 18 months (95% CI, 11-NE)a
Median OS: 37 months (95% CI, 20-NE)a
Time Frame
Number of Patients
Primary Site
Dodd et al13
1983-1994
50
Herr et al14
1984-1999
Sweeney et al16
Author
CR: 7 PR: 17 SD: 4
NA
Combination of clinical/pathological response to CT and surgery yielded the greatest prognostic effect on survival No correlation between clinical response to CT and survival; 2 positive retroperitoneal lymph nodes were associated with longer PFS Long-term survival was possible in 24% of patients selected from responders to CT and with nodal disease only No clinical/pathological predictor of survival No survival advantage of surgery; possible benefit in QoL indexes No impact of pathologic response and site of metastases on survival; a few patients survived after resection of lung or bone metastases Solitary metastasis 3 cm had the best prognostic effect on survival Resection of solitary lung or nodal metastasis significantly prognostic in UVA Favorable prognostic effect of surgery in patients with subdiaphragmatic nodal metastases after response to CT
Abbreviations: CI ¼ confidence interval; CT ¼ chemotherapy; IQR ¼ interquartile range; MVAC ¼ methotrexate, vinblastine, doxorubicin, and cisplatin; NE ¼ not estimated; OS ¼ overall survival; PFS ¼ progression-free survival; QoL ¼ quality of life; UVA ¼ univariable analysis. a Since the initiation of chemotherapy. b Since metastasectomy.
Surgery in Metastatic Urothelial Cancer
Clinical Genitourinary Cancer February 2015
Table 5 Summary of the Available Outcomes of Postchemotherapy Surgery in Metastatic Urothelial Cancer
Andrea Necchi et al alone (IQR, 15-37) and CR with chemotherapy and surgery (IQR, 11-not estimated), and it was 9.5 month for patients with residuals (IQR, 4-12). Median OS for patients with CR after chemotherapy was 44 months (IQR, 19-196) and it was 37 months for radically resected patients (IQR, 20-not estimated), and 12 months only for those with residuals (IQR, 9-23).
Discussion Postchemotherapy surgery with curative intent is an option in effect since the early 1980s.12-23 In this retrospective analysis of very selected patients with metastatic UC (37.5% of the overall patients receiving first-line mMVAC at our center) we could demonstrate an independent prognostic effect of surgical clearance of disease on PFS and OS. Results aligned with those reported by Dodd and colleagues but in a more homogeneous patient population, because these authors evaluated patients yielding T3-4 N0 disease and those having visceral metastases.13 A few other retrospective series have been subsequently reported on this issue and available findings identified chemosensitive nodal disease as the most favorable prognostic variable (Table 5).13,14,16-19,21-23 Herr and colleagues reported on 207 patients with unresectable or lymph node-positive disease, 80 of whom underwent surgery.14 Results are similar to the present ones; the authors evaluated the role of postchemotherapy surgery in combination with response to chemotherapy. As a difference, the pathologic response was a predictor of outcome in Herr’s series and the combination of clinical and pathologic response had the greatest effect on survival. The site of metastatic disease is a recognized prognostic factor in patients with metastatic UC who undergo first-line chemotherapy and it has been included in the major available prognostic scores.24-26 Results of the surgical series actually underscore this issue, yet a very small fraction of long-term survivors after visceral (ie, pulmonary) metastasectomy should be accounted for. Available and present findings require a prospective validation especially in patients with more extensive residuals, although in our series, most cases were classified as yielding multiple nodal sites, irrespective of the treatment group. Differences in the surgical quality across centers could represent an additional bias when dictating the choice between a surgical approach and observation or alternative treatments; so far, a survival advantage can only be hypothesized. Present results should be interpreted with caution because of the existence of several potential criticisms in the analysis. The retrospective quality of data and the long-dated series might account for a major limitation. Our aim was to discriminate the effect of surgery over recognized confounding variables, particulary because we included bladder and upper tract tumors, and the site of tumor primary did not result as prognostic in the multivariable model. In the absence of a recognized indication, the choice to perform a radical surgical resection of residuals was dictated by a critical evaluation of the features and clinical course of individual cases, and consolidating radiotherapy or chemotherapy, and observation, were equal options. Despite these limitations, the administered treatment was homogeneous (modified mMVAC as per institutional policy) and there were minor differences in baseline characteristics between the 2 groups (eg, distribution of cases with ECOG-performance status
1, although it was < 2 in all cases). Complete CR to chemotherapy was a significantly favorable prognostic factor on multivariable analysis for PFS and trended to significance for OS. Although PR and SD were jointly analyzed to enhance model parsimony, the frequency of SD was slightly different in the 2 groups (14% vs. 25%, respectively), possibly explaining at least in part the difference of survival curves. Remarkably high 5-year PFS and OS rates have been achieved, in the overall series and in the subgroups, confirming that a vast heterogeneity should be accounted for when interpreting the efficacy of first-line chemotherapy. Data showed that a historical 5-year OS rate of 28% with a multimodal approach that included surgery of visceral metastases21 could be improved to > 45% by selecting patients with nodal disease only. Results were not negligible because 9 patients had experienced a relapse after surgery of tumor primary that included regional lymphadenectomy. The relatively stable median PFS estimate of 8 months reported with cisplatin-based regimens could be improved to 18 months in this series. Our numbers did not allow us to dissect a significant survival effect of pathologic response, although there was a trend toward a better outcome in patients who achieved a pCR. Unfortunately, the retrospective quality of the data were not suitable for an analysis of the side effects and recovery after surgery, although no major complications or deaths were observed. Finally, maintenance chemotherapy after response to first-line is a rational approach being increasingly used in clinical trials to test new drugs in an earlier disease setting, and early results are available.27 Data from our analysis suggest that these trials should have more restrictive criteria than those currently used, and exclude candidate patients with the same features of those we presented and for whom an extended radical surgical approach is rather advisable.
Conclusion To summarize, results of the present retrospective comparison between postchemotherapy surgery and other treatment or observation seemed to favor the first option. The superiority has been confirmed in PFS and OS in very selected patients with advanced UC. These results might have implications for the design of clinical trials, particularly in the maintenance setting after response to firstline chemotherapy.
Clinical Practice Points Present analysis of a single-center series provided additional data
on the role of postchemotherapy surgery for patients with metastatic urothelial carcinoma. Available information is scarce in the field and the close surveillance of disease residuals following first-line chemotherapy is currently the preferred choice in most cases. Present analysis was focused on 28 patients having subdiaphragmatic nodal metastatic disease only and who achieved at least a stable disease after cisplatin-based chemotherapy. The outcomes have been compared to those obtained in a cohort of 31 patients with comparable features who did not undergo surgical resection. Postchemotherapy surgery, as well as response to chemotherapy, were the two significant predictors in the prognostic model including clinical variables.
Clinical Genitourinary Cancer February 2015
- 85
Surgery in Metastatic Urothelial Cancer Results deserve validation in prospective clinical trials and
represent a call for publication of further larger case series on this topic. Most importantly, results could have an implication on the design of postchemotherapy maintenance trials.
Disclosure The authors have stated that they have no conflicts of interest.
Supplemental Data The supplemental figure accompanying this article can be found in the online version at http://dx.doi.org/10.1016/j.clgc. 2014.06.003.
References 1. Gallagher DJ, Milowsky MI, Bajorin DF. Advanced bladder cancer: status of firstline chemotherapy and the search for active agents in the second-line setting. Cancer 2008; 113:1284-93. 2. Sternberg CN, Yagoda A, Scher HI, et al. Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium. Cancer 1989; 64:2448-58. 3. von der Maase H, Hansen SW, Roberts JT, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 2000; 17:3068-77. 4. von der Maase H, Sengelov L, Roberts JT, et al. Long-term survival results of a randomized trial comparing gemcitabine plus cisplatin, with methotrexate, vinblastine, doxorubicin, plus cisplatin in patients with bladder cancer. J Clin Oncol 2005; 23:4602-8. 5. Sternberg CN, Mulder PH, Schornagel JH, et al. Randomized phase III trial of high-dose-intensity methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) chemotherapy and recombinant human granulocyte colony-stimulating factor versus classic MVAC in advanced urothelial tract tumors: European Organization for Research and Treatment of Cancer Protocol No. 30924. J Clin Oncol 2001; 19: 2638-46. 6. Bellmunt J, von Der Maase H, Mead GM, et al. Randomized phase III study comparing paclitaxel/cisplatin/gemcitabine and gemcitabine/cisplatin in patients with locally advanced or metastatic urothelial cancer without prior systemic therapy: EORTC Intergroup Study 30987. J Clin Oncol 2012; 30:1107-13. 7. Necchi A, Mariani L, Giannatempo P, et al. Long-term efficacy and safety outcomes of modified (simplified) combinations of methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) as front-line therapy for unresectable or metastatic urothelial cancer. Clin Genitourin Cancer 2014; 12:203-209.e1. 8. Dimopoulos MA, Finn L, Logothetis CJ. Pattern of failure and survival of patients with metastatic urothelial tumors relapsing after cisplatinum-based chemotherapy. J Urol 1994; 151:598-600.
86
-
Clinical Genitourinary Cancer February 2015
9. Svatek RS, Siefker-Radtke AO, Dinney CP. Management of metastatic urothelial cancer: the role of surgery as an adjunct to chemotherapy. Can Urol Assoc J 2009; 3(suppl 4):S228-31. 10. Edge SB, Byrd DR, Compton CC, et al, eds. AJCC Cancer Staging Manual. 7th ed. New York, NY: Springer-Verlag; 2009:491-6. 11. Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228-47. 12. Cowles RS, Johnson DE, McMurtrey MJ. Long-term results following thoracotomy for metastatic bladder cancer. Urology 1982; 20:390-2. 13. Dodd PM, McCaffrey JA, Herr H, et al. Outcome of postchemotherapy surgery after treatment with methotrexate, vinblastine, doxorubicin, and cisplatin in patients with unresectable or metastatic transitional cell carcinoma. J Clin Oncol 1999; 17:2546-52. 14. Herr HW, Donat SM, Bajorin DF. Post-chemotherapy surgery in patients with unresectable or regionally metastatic bladder cancer. J Urol 2001; 165:811-4. 15. Abe T, Shinohara N, Harabayashi T, et al. Impact of multimodal treatment on survival in patients with metastatic urothelial cancer. Eur Urol 2007; 52: 1106-13. 16. Sweeney P, Millikan R, Donat M, et al. Is there a therapeutic role for postchemotherapy retroperitoneal lymph node dissection in metastatic transitional cell carcinoma of the bladder? J Urol 2003; 169:2113-7. 17. de Vries RR, Nieuwenhuijzen JA, Meinhardt W, Bais EM, Horenblas S. Longterm survival after combined modality treatment in metastatic bladder cancer patients presenting with supra-regional tumor positive lymph nodes only. Eur J Surg Oncol 2009; 35:352-5. 18. Siefker-Radtke AO, Walsh GL, Pisters LL, et al. Is there a role for surgery in the management of metastatic urothelial cancer? The M.D. Anderson experience. J Urol 2004; 171:145-8. 19. Otto T, Krege S, Suhr J, Rübben H. Impact of surgical resection of bladder cancer metastases refractory to systemic therapy on performance score: a phase II trial. Urology 2001; 57:55-9. 20. Herr HW. Is metastasectomy for urothelial carcinoma worthwhile? Eur Urol 2009; 55:1300-1. 21. Lehmann J, Suttmann H, Albers P, et al. Surgery for metastatic urothelial carcinoma with curative intent: the German experience (AUO AB 30/05). Eur Urol 2009; 55:1293-9. 22. Matsuguma H, Yoshino I, Ito H, et al. Is there a role for pulmonary metastasectomy with a curative intent in patients with metastatic urinary transitional cell carcinoma? Ann Thorac Surg 2011; 92:449-54. 23. Abe T, Kitamura H, Obara W, et al. Outcome of metastasectomy for urothelial carcinoma: a multiinstitutional retrospective study in Japan. J Urol 2014; 191: 932-6. 24. Bajorin DF, Dodd PM, Mazumdar M, et al. Long term survival in metastatic transitional-cell carcinoma and prognostic factors predicting outcome of therapy. J Clin Oncol 1999; 17:3173-81. 25. Apolo AB, Ostrovnaya I, Halabi S, et al. Prognostic model for predicting survival of patients with metastatic urothelial cancer treated with cisplatin-based chemotherapy. J Natl Cancer Inst 2013; 105:499-503. 26. Galsky MD, Moshier E, Krege S, et al. Nomogram for predicting survival in patients with unresectable and/or metastatic urothelial cancer who are treated with cisplatin-based chemotherapy. Cancer 2013; 119:3012-9. 27. Grivas PD, Daignault S, Tagawa ST, et al. Double-blind, randomized, phase 2 trial of maintenance sunitinib versus placebo after response to chemotherapy in patients with advanced urothelial carcinoma. Cancer 2014; 120: 692-701.
Andrea Necchi et al Supplemental Figure 1
Kaplan Meier Curves of ProgressionFree (A) and Overall Survival (B) According to Treatment Group
Clinical Genitourinary Cancer February 2015
- 86.e1