Management of Invasive Bladder Cancer: A Meticulous Pelvic Node Dissection Can Make a Difference

0022-534 7/82/1281-0034$02.00/0

THE

Vol. 128, July Printe,l in U.S.A.

JOURNAL OF UROLOGY

Copyright© 1982 by The Williams & Wilkins Co.

MANAGEMENT OF INVASIVE BLADDER CANCER: A METICULOUS PELVIC NODE DISSECTION CAN MAKE A DIFFERENCE DONALD G. SKINNER* From the Department of Surgery, Division of Urology, University of California School of Medicine, UCLA Medical Center and University of Southern California School of Medicine, Los Angeles, California

ABSTRACT

During an 8-year period 159 patients with primary epithelial carcinoma of the bladder were operated upon in anticipation of cure. At the operation 6 patients (4 per cent) were found to be inoperable because of extensive disease above the aortic bifurcation. Therefore, 153 patients underwent a meticulous bilateral pelvic iliac lymph node dissection with en bloc radical cystectomy and urinary diversion as a single stage procedure. Of these 153 potentially curable patients 36 had positive nodes histologically. Analysis of these 36 patients revealed that the presence or absence of nodal metastases cannot be predicted accurately on the basis of T or P category of the primary tumor, although the frequency of nodal disease increased with deeper penetration of the bladder wall. The incidence of positive pelvic nodes is 5 to 10 per cent in patients with Pl tumors, 30 to 35 per cent in those with P2 or P3A tumors and 50 to 66 per cent in those with P3B and P4 tumors. The presence of positive pelvic nodes does not mean incurability since the 2, 3 and 5-year survival rates for these patients are 46, 36 and 36 per cent, respectively. In this small series of patients with nodal metastases the extent of the primary tumor (P stage) did not relate to survival. Pelvic recurrence as the first site of failure was noted in only 2 of 22 patients with metastatic disease. Experience indicates that pelvic node dissection does not increase the morbidity associated with cystectomy, can cure some patients with metastatic disease, effectively controls pelvic disease and indicates which patients are at substantial risk for systemic metastatic disease, implying the need for development and use of systemic chemotherapy. The operations were done with a standard technique of en bloc bilateral pelvic iliac lymph node dissection. The operation has been described in detail8 and a film illustrating the technique is available. 9 It is noteworthy that the inoperability rate in patients explored with intent for cure was 3.7 per cent (6 of 159 cases) compared to an inoperability rate of 16 per cent (4 of 25 cases) for patients explored for salvage cystectomy after failure of definitive radiation therapy. Pathologic staging of the 153 patients revealed 36 with nodal metastatic disease, for an over-all incidence of 25 per cent. Analysis of these 29 men and 7 women forms the basis of this report. In 27 cases the planned high dose, short course preoperative radiation therapy was used, while 9 patients underwent the operation without radiation. The official pathology report for all patients was reviewed independently by a pathologist to confirm accuracy of pathologic staging of the cystectomy specimen and numerous histologic features were analyzed. If doubt existed as to stage the original slides were reviewed. Patients also were assigned preoperative clinical stages only on the basis of a review of the biopsy material available immediately before cystectomy. Other parameters, such as the findings on bimanual examination or histology from prior definitive transurethral or segmental resection that preceded the decision for aggressive therapy, an excretory urogram or the impression of the operating surgeon, were not used for preoperative clinical staging. Patients with biopsy showing muscle invasion were considered to have clinical stage B or T2 disease for this review since it was believed impossible to differentiate superficial from deep muscle accurately on the basis of the histologic specimen. Tumor present in perivesical fat was necessary for clinical designation of stage T3 or C disease. The Kaplan-Meier actuarial computations were used for survival determinations. 10 All patients were followed at least 5 years if they were at risk ~5 years after cystectomy, or through December 1980 if at risk <5 years or until death.

The role of pelvic lymphadenectomy in the management of patients with bladder cancer remains controversial despite some reports indicating its efficacy in curing some patients with metastatic foci to a few nodes. 1- 7 I herein review my 8-year experience with meticulous pelvic node dissection on every patient in whom radical cystectomy was considered the primary modality of treatment intended for cure. MATERIALS AND METHODS

Between August 1, 1971 and December 31, 1979, 204 patients with bladder malignancy and no evidence of metastatic disease were explored with the intent of resection for cure. There were 10 patients found to be inoperable at exploration because of the extent of the disease, usually extensive nodal disease above the aortic bifurcation, malignant retroperitoneal fibrosis or unexpected hepatic metastases. Of these patients 21 had been treated previously with definitive high dose radiation therapy (>6,000 rad), in which salvage cystectomy was done without node dissection. An additional 20 patients underwent cystectomy for nonurothelial malignancy, usually a primary sarcoma or primary urothelial carcinoma of the female urethra. The remaining 153 patients with primary bladder cancer underwent a meticulous single stage bilateral pelvic lymph node dissection with en bloc radical cystectomy and urinary diversion intended for cure. Planned preoperative radiation therapy was used in 131 patients, 20 receiving standard fractionation radiation therapy totaling 4,500 rad delivered during a 4 to 6-week period followed by cystectomy 4 to 6 weeks later, and 111 receiving high dose, short course preoperative radiation therapy as part of a protocol completed in 1978 and reported in detail (1,600 rad in 4 days followed by immediate cystectomy). 4 Radiation therapy was not given to 22 patients. Accepted for publication September 18, 1981. Read at annual meeting of American Urological Association, Boston, Massachusetts, May 10-14, 1981. * Requests for reprints: Division of Urology, University of Southern California School of Medicine, 1200 N. State St., Los Angeles, California 90033. 34

35

J~ANAGEI'liENT OF :IN~lASIVE BLADDER. CAN\JER RESULTS

There were no postoperative deaths in this group of 36 patients and no postoperative complications that could be attributed to the pelvic node dissection. Table 1 lists the preoperative clinical stage correlated with pathologic stage of the primary tumor at the time of cystectomy and correlates survival with pathologic stage of the primary tumor. Of the 36 patients 22 have died of metastatic disease and 14 are free of disease for 15 to 84 months. Nine of 27 patients who received preoperative radiation therapy and 5 of 9 who did not receive preoperative radiation therapy survived. Of the 22 deaths 20 occurred within 24 months and 15 occurred within 18 months. No patient died of cancer >33 months postoperatively. The actuarial survival of the entire group of patients, with computed 2, 3 and 5-year survival rates of 46, 36 and 36 per cent, respectively, is shown in the figure. Histologically, all tumors were of transitional cell type, with the exception of 1 well differentiated pure squamous cell carcinoma. All but 1 of the transitional cell carcinomas were grade III or higher. The only patient with low grade transitional cell carcinoma encountered in this group died of metastatic disease 12 months after cystectomy. Table 2 correlates the number of positive nodes with survival. Most survivors had ~3 positive nodes but it is noteworthy that currently there are 3 patients who are well and who had >3 positive nodes. In this series growth patterns such as papillary, solid or ulcerating did not correlate with survival probability. It is noteworthy that only 2 of the 22 patients considered failures of the therapy had tumor 1. Clinical stage correlated with pathologic stage of the primary tumor in 36 patients with positive pelvic nodes

TABLE

Pathologic Stage Pl+ PIS Clinical stage: Tl+ TIS T2 T3 T4 Total pts. Survival (No. pts./total)

1 2

3 3/3

P2

P3A

P3B

P4

2 3

2

14 4

4

2 2/2

6 1/6

1 1 6 2/6

19 5/19

"iOO Oct '71 Dec. '79

75

Survival 36 pis positive pelvic nodes managed by pelvic node dissection radical cysleciomy

12

27 pts 1600 R pre op

-'

"';;;,;;:: 11":

:':I II!

50

IP

25

3

Iii

Hii

2~

I I I I I I I I I I I I I I I I 27 33

39

45

5 'fl

57

63

69 72

TIME- MONTHS

TABLE

2. Survival according to number of positive nodes Pos. Nodes

Survival (No. pts./total) 8/18 5/10

1/4 0/4

recurrence in the disseminated disease and presenting with dence of liver, bone or lung metastases. DISCUSSION

Leadbetter and Cooper were early proponents of a meticulous pelvic iliac lymph node dissection and radical cystectomy. 11 In 1973 Dretler and associates reported on 35 patients with positive nodes who underwent the operation for cure and who were followed >5 years. These investigators indicated that some patients with a few positive nodes could be cured with node dissection without increasing the morbidity or the mortality associated with the operation. 2 Of their 35 patients 6 (17 per cent) survived >5 years and 33 per cent of the patients with involvement of only 1 or 2 nodes survived. Whitmore and Marshall reported that 2 of 13 patients (16 per cent) who had metastasis of only 1 or 2 pelvic nodes and no invasion of adjacent organs survived >5 years. 12 Laplante and Brice reported a 13 per cent survival rate among 39 patients with positive lymph nodes at or below the iliac bifurcation but without invasion of adjacent pelvic structures. 13 This is a similar cure rate to that in patients with stage C disease without nodal involvement. Reid and associates used the protocol of high dose, short course preoperative radiation therapy with immediate single stage radical cystectomy and node dissection, and reported on 5 of 24 patients (21 per cent) with positive nodes who survived >5 years. 3 Other reports of anecdotal long-term survival of patients with nodal metastases treated aggressively 1 and the 36 per cent survival rate reported herein coupled with the uniform short survival of those patients with nodal involvement in whom plans for cystectomy are abandoned because of the histological demonstration of nodal disease, regardless of subsequent use of radiation therapy, justify the continued use of aggressive surgery whenever feasible. 6 The increased cure rates reported in this series may reflect an important trend in the management of bladder cancer, primarily patient selection with earlier treatment of invasive disease. The over-all incidence of positive pelvic nodes (25 per cent) in patients operated on for cure is not dissimilar to other contemporary reports, such as those by Reid (19 per cent) 3 and Whitmore (19 per cent) 5 and their associates, who treated similar groups of patients with similar protocols of preoperative radiation. It seems probable that the incidence of positive nodes found at the time of cystectomy has decreased compared to previous reports in which patients often were referred late and suxgical treatment was withheld until later in the natural course of the disease. In 1968 Whitmore and associates reported that the incidence of invasion of adjacent organs the primary tumor or positive pelvic nodes was 35 per cent 138 patients treated with a protocol using 4,000 rad of preoperative radiation followed by cystectomy 4 to 6 weeks later. 6 Another factor responsible for improved survival may be a more critical pathologic assessment of the resected pelvic nodes with better and more standard fixation techniques, including bladder inflation with formalin before sectioning. In addition, development of an en bloc meticulous pelvic node dissection performed by the same surgeon on a frequent basis also may be another factor leading to improved survival. The available data relative to nodal disease were analyzed. Approximately 20 to 35 per cent of the patients believed to be candidates for radical cystectomy for management of bladder cancer will have metastasis to the pelvic nodes. The incidence of positive pelvic nodes relates directly to the pathologic stage of the primary tumor and the most critical factors associated with nodal metastases are muscle invasion and penetration into the perivesical fat. The incidence of pelvic nodal metastatic disease for stages Pl, P2, P3A and P3B tumors is 6, 30, 31 and 65 per cent, respectively. 4 The inaccuracy of clinical staging before the operation implies the need to treat the nodes when-

36

SKINNER

ever cystectomy is indicated. Positive nodes were seen at the time of cystectomy in 17 per cent of the patients with histologic evidence of stage Tl disease and 37 per cent with T2. 4 Patients who probably benefit the most from a meticulous dissection are those with clinically undetectable micrometastases to a few nodes. More than 35 per cent of these patients can be cured, while it is rare for patients with multiple or macroscopic nodal involvement to survive >3 years, regardless of current therapy. The low morbidity and mortality associated with a meticulous pelvic node dissection, together with an expected cure rate of 15 to 35 per cent for patients with positive nodes, justifies its continued and routine use when cystectomy seems justified. Despite the curability of some patients with an operation or a combination of preoperative radiation therapy and an operation lymph node involvement generally indicates systemic metastatic disease and the need for adjuvant systemic therapy. It seems probable that pelvic nodal disease has a prominent role in selecting patients for aggressive adjuvant chemotherapy once clinical trials demonstrate an efficacy to adjuvant chemotherapy and drugs truly effective against bladder carcinoma are identified. In conclusion, the µumber of patients reported herein is small and may not be comparable to other surgical therapy experience. Nonetheless, these data together with those reported by others indicate that pelvic node dissection can cure some patients with bladder cancer metastatic to a few pelvic nodes and that its performance does not increase the morbidity or mortality associated with cystectomy. REFERENCES

1. DeCenzo, J. M. and Leadbetter, G. W., Jr.: Survival with stage D bladder carcinoma. Surgical improvement. Urology, 3: 221, 1974. 2. Dretler, S. P., Ragsdale, B. D. and Leadbetter, W. F.: The value of pelvic lymphadenectomy in the surgical treatment of bladder cancer. J. Urol., 109: 414, 1973. 3. Reid, E. C., Oliver, J. A. and Fishman, I. J.: Preoperative irradiation and cystectomy in 135 cases of bladder cancer. Urology, 8: 247, 1976. 4. Skinner, D. G., Tift, J. P. and Kaufman, J. J.: High dose, short course preoperative radiation therapy and immediate single stage radical cystectomy with pelvic node dissection in the management of bladder cancer. J. Urol., 127: 671, 1982. 5. Whitmore, W. F., Jr., Batata, M. A., Hilaris, B. S., Reddy, G. N., Unal, A., Ghoneim, M. A., Grabstald, H. A. and Chu, F.: A comparative study of two preoperative radiation regimens with cystectomy for bladder cancer. Cancer, 40: 1077, 1977. 6. Whitmore, W. F., Jr., Grabstald, H. A., MacKenzie, A. R., Ishwariah, G. and Phillips, R.: Preoperative irradiation and cystectomy in the management of bladder cancer. Amer. J. Roentgen., 102: 570, 1968.

7. Smith, J. A., Jr. and Whitmore, W. F., Jr.: Regional lymph node metastasis from bladder cancer. Read at annual meeting of American Association of Genito-Urinary Surgeons, Phoenix, Arizona, February 26-28, 1981. 8. Skinner, D. G.: Technique of radical cystectomy. Urol. Clin. N. Amer., 8: 353, 1981. 9. Skinner, D. G.: Technique of radical cystectomy: a movie. Film available from Eaton Laboratories Film Library, Eaton Laboratories, Norwich, New York. 10. Kaplan, E. L. and Meier, P.: Nonparametric estimation from incomplete observations. J. Amer. Stat. Ass., 53: 457, 1958. 11. Leadbetter, W. F. and Cooper, J. F.: Regional gland dissection for carcinoma of the bladder: a technique for one-stage cystectomy, gland dissection, and bilateral uretero-enterostomy. J. Urol., 63: 242, 1950. 12. Whitmore, W. F., Jr. and Marshall, V. F.: Radical total cystectomy for cancer of the bladder: 230 consecutive cases five years later. J. Urol., 87: 853, 1962. 13. Laplante, M. and Brice, M., II: The upper limits of hopeful application of radical cystectomy for vesical carcinoma: does nodal metastasis always indicate incurability? J. Urol., 109: 261, 1973. EDITORIAL COMMENT These results with radical cystectomy, with or without preoperative irradiation, and pelvic lymph node dissection for patients with bladder cancer and regional node metastases are commendable. Most patients in this series seem beyond risk of recurrence and the low rate of pelvic recurrence is notable. This experience supports that of others in demonstrating the therapeutic usefulness of lymph node dissection and its value in defining high risk groups for potential adjuvant therapy. The over-all incidence oflymph node metastases in patients undergoing radical cystectomy for bladder cancer at the Memorial SloanKettering Cancer Center during the last 30 years has varied around 20 per cent and the occurrence of lymph node metastasis in these patients has correlated with the T category of the tumor: Tl-5 per cent, T213 per cent, T3-18 per cent and T4-44 per cent. That muscle infiltration is associated with later evidence of distant dissemination in approximately half of the patients treated by radical cystectomy with preoperative irradiation and that lymph node metastases are not identified in many such patients suggest that direct blood vessel invasion by the primary tumor contributes to distant failures. As was indicated, the extent of lymph node metastases (N category) also is relevant to the success of lymph node dissection, an observation consistent with other experiences. Thus, the T or P category of the primary tumor and the N category may be pertinent factors in determining treatment failures. The efforts of the pathologist in seeking lymph node metastases are a more or less unquantified variable in studies of regional node metastasis in bladder cancer and may contribute importantly to apparent differences in end results after lymph node dissection. Willet F. Whitmore, Jr. Department of Surgery Memorial Sloan-Kettering Cancer Center New York, New York