- Email: [email protected]
GPOH
0022-5347/03/1703-0939/0 THE JOURNAL OF UROLOGY® Copyright © 2003 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 170, 939 –944, September 2003 Printed in U.S.A.
DOI: 10.1097/01.ju.0000073848.33092.c7
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR: RESULTS OF STUDY SIOP 93– 01/GPOH FRANK-MARTIN HAECKER,* DIETRICH
SCHWEINITZ, DIETER HARMS, DIETRICH BUERGER NORBERT GRAF
VON
AND
From the Department of Pediatric Surgery, University Children’s Hospital Basel (F-MH, DvS), Basel, Switzerland, and Central Pediatric Tumor Registry, Department of Pathology, University of Kiel (DH), Kiel, Pediatric Surgery, DRK-Klinik Siegen (DB), Siegen and Pediatric Oncology, University Hospital Homburg/Saar (NG), Homburg/Saar, Germany
ABSTRACT
Purpose: We evaluate results and long-term outcome after partial nephrectomy (PN) for unilateral Wilms tumor (ulWT) in relation to different histological features, performed as initial surgery or after induction chemotherapy (ChT). Materials and Methods: Data from patients with ulWT who had undergone PN in the German Study SIOP 93– 01/GPOH were analyzed for time of surgery in the treatment schedule, postoperative stage, histological features, surgical radicality and outcome. The results were correlated with overall survival and relapse-free survival, and compared with those of patients who had undergone total tumor nephrectomy (TN). Results: A total of 770 patients underwent TN and 37 underwent PN, of which 766 (99.5%) and 36 (97.3%), respectively, were radical procedures. A total of 139 TNs and 15 PNs were performed at primary surgery, 630 TNs and 22 PNs after ChT, and 1 TN after irradiation. Two of 15 patients following primary PN and 1 of 22 following delayed PN had local recurrence. Two patients died of metastatic disease. Tumor stage, overall survival of 93% and relapse-free survival of 88% were equal after PN and TN in analysis of the whole group. This was also true for patients with surgery after ChT. One of 2 patients with histologically unfavorable disease had relapse after PN compared to only 3 of 35 patients with histologically low or intermediate risk disease. Conclusions: PN should be performed only for small, histologically low or intermediate risk tumors after good response to ChT to secure a complete resection. KEY WORDS: nephroblastoma, kidney neoplasms, nephrectomy, antineoplastic agents
Within the last 10 to 15 years the treatment of nephroblastoma or Wilms tumor (WT) has significantly improved the outcome of affected children. But at the same time clinicians have also been faced with the task of treating the late effects of anticancer therapy. The crude 3-year survival rate for children suffering from WT is more than 90%.1 Actual treatment protocols must aim to decrease morbidity associated with treatment without impairing the cure rates. During the postoperative period after total tumor nephrectomy (TN) renal function of the solitary kidney may be influenced by chemotherapy and/or hyperfiltration. Thereafter, more common risks such as trauma, infection and obstruction may cause additional loss of renal function. Children with a single kidney, studied at followup for more than 25 years, present with a significant decrease in the glomerular filtration rate and a significant increase in albuminuria compared to those studied at a shorter followup.2, 3 Additionally, there exists a 2% to 3% risk that a metachronous WT may develop in the contralateral kidney later in life. After successful inauguration of parenchymal sparing surgery in children with bilateral WT recent reports have emphasized conservative surgery for children with unilateral (ul) WT.4 –7 To evaluate the risks and benefits of this procedure for long-term outcome, and to reevaluate indication criteria for partial nephrectomy (PN) in ulWT, we analyzed the data from patients with ulWT treated with PN in com-
parison to TN in the German branch of the study SIOP 93– 01/GPOH. The protocol of the SIOP 93– 01 WT study includes preoperative induction chemotherapy (ChT) followed by tumor nephrectomy, which is in contrast to the protocol of the United States National Wilms Tumor Study (NWTS), with primary surgery followed by chemotherapy. PATIENTS AND METHODS
From April 1994 to December 2001, 865 patients with ulWT were registered in the GPOH branch of the SIOP 93– 01 WT study, which includes patients from the German spoken area of Switzerland, Austria and Germany. The patients were treated according to the SIOP/GPOH study protocol with ChT followed by tumor nephrectomy, or they underwent primary surgery. A total of 807 patients with homogeneous distribution of age and sex with ulWT were further analyzed. Fifty-eight patients were excluded from this study—in 8 patients only a tumor biopsy was performed, in 48 there was no reliable information about the surgical treatment and in 2 the tumor was inoperable. Initial ultrasound and computerized tomography or magnetic resonance imaging were performed in all patients. According to the study protocol, the treatment for patients with ulWT between age 6 months and 16 years consisted of ChT with actinomycin D and vincristine for 4 weeks, as well as additional doxorubicin for patients with stage IV disease for 6 weeks, followed by TN. PN was not part of the SIOP 93– 01 protocol, but was done based on individual choice of the local center in 37 patients, based on criteria such as stage I disease, well-defined tumor margin, absence of invasion into the renal vessels or collecting system and determination that at
Accepted for publication April 4, 2003. Supported by a grant from the German Cancer League (“Deutsche Krebshilfe,” project No. 70 –1899). * Corresponding author: University Children’s Hospital, P.O. Box CH-4005, Basel, Switzerland (telephone: ⫹41/61/685 5239; FAX: ⫹41/61685 5011; e-mail: [email protected]). 939
940
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
least 50% of the remaining kidney was preservable. A frozen section was not done in every case to confirm the absence of capsular extension or invasion. After surgery pathological classification was done by the local pathologist and the Kiel Pediatric Tumor Registry (DH) as the reference institution for pathological evaluation of the SIOP 93– 01/GPOH study.8 Based on the SIOP staging criteria, tumor staging was undertaken postoperatively.9 SIOP staging criteria are equal to those used by the NWTS, except for 2 differences—staging is undertaken after chemotherapy pretreatment, and stage II tumors are distinguished as those with positive lymph nodes (stage II N⫹, corresponding to NWTS stage III) or without positive lymph nodes (stage II N⫺, corresponding to NWTS stage II). The staging system of the new SIOP 2001 protocol is identical to the NWTS staging system. The SIOP histological classification “low risk/intermediate risk” is equivalent to the NWTS “favorable histology,” and the “high risk” SIOP classification is identical to “unfavorable histology” in the NWTS. The clinical and followup data were centrally collected by the SIOP 93– 01/GPOH study office (NG). Data were collected and stored in a relational dBASE (VersaSoft Corp, San Jose, California) database. Statistical analysis was done using the software package SPSS for Windows release 11.0.1 (SPSS, Chicago, Illinois) software, including descriptive functions and life table analysis according to Kaplan-Meier. The log rank test was used for statistical comparison of different factors. RESULTS
Partial nephrectomy. Radical PN resulting in local stage I tumor was performed in 37 patients (4.6%) with a median age of 2.6 years (range 1 month to 27 years) and a sex distribution of 17 females to 20 males. Mean followup was 4 years (range 1 to 7). In 15 patients primary surgery without ChT was performed. Of these 15 patients tumor volume was less than 30 ml in 5, unknown in 5 and 224 to 780 ml in 5. Of the 15 patients 5 were younger than 6 months, 2 had congenital mesoblastic nephroma, and 3 had cystic partial differentiated nephroma and a low risk tumor. A total of 22 patients underwent PN after ChT. In the PN group 76% of the patients had stage I disease, whereas in the TN group only 51% had stage I disease, probably due to smaller tumors in the PN group (table 1). Tumor recurrence was found in 2 patients after primary PN (1 local and 1 combined relapse) as well as in 2 after ChT and subsequent PN (1 local recurrence and 1 metastatic disease). Of these 4 patients 2 had an unfavorable outcome and died, 1 of whom had metastatic disease after primary resection of WT from a horseshoe kidney (histologically blastemic predominant subtype) and 1 had a nonmetastasizing malignant rhabdoid tumor (table 2). This tumor was completely excised and, therefore, stage I. After receiving the result of the histopathological evaluation tumor nephrectomy was not done. The other 2 patients (both stage I) had a recurrent local tumor, which was resected completely at reoperation (table 2). One of these patients again
underwent PN because of a cystic nephroma in the contralateral kidney. Total tumor nephrectomy. TN was performed in 770 patients (95.4%) with a median age of 3.2 years (range 1 month to 62 years), of whom 403 were female and 367 male. A total of 139 patients underwent primary surgery without previous ChT. In 630 patients TN was performed after ChT and in 1 patient after irradiation (table 1). Mean followup was 4 years (range 1 to 8). A total of 120 patients had metastatic disease at the time of diagnosis. Of these patients 17 died, as well as 24 patients without initial metastatic disease. Tumor recurrence was observed in 9 patients (6.5%) after primary TN (local recurrence 2, recurrent metastatic disease 6, local recurrence plus recurrent metastatic disease 1) as well as in 69 patients (11%) after ChT with subsequent TN (local recurrence 14, recurrent metastatic disease 48, local recurrence plus recurrent metastatic disease 7). Of the 139 patients undergoing primary TN 6 (4.3%) eventually died of disease, of whom 3 had metastatic disease at diagnosis. Of 630 patients undergoing TN after ChT 35 (5.6%) died, of whom 14 had metastases at diagnosis. Outcome—PN versus TN. Statistical analysis included overall survival (OS) and relapse-free survival (RFS). Comparing OS and RFS, the differences were more obvious for the RFS rates, and so we present the RFS rates herein. The probability of 5-year RFS was 92% for all patients undergoing primary surgery versus 87% for patients receiving ChT and subsequent TN (not significant). Regarding the 5-year RFS rates for PN (primary surgery versus ChT) and TN (primary surgery versus ChT), similarly there were no significant differences. In the PN group RFS was 90% after ChT versus 86% after primary surgery, while in the TN group RFS was 87% after ChT versus 92% after primary surgery (fig. 1). RFS after ChT showed no significant difference between PN (90%) and TN (87%). Similarly, RFS after primary surgery was 92% for TN and 86% for PN (not significant) (fig. 2). In view of the histopathological subtype, there was a significant difference in the outcome between WT with low and standard risk versus high risk histology concerning patients without metastatic disease at diagnosis who underwent TN. There were only 2 patients in the PN group with high risk histology (table 3). The number was too small to demonstrate a difference in outcome between PN and TN. Although in both groups statistical analysis revealed no significant difference (probably due to the small number of patients), the results show a tendency for a poor outcome for patients with high risk histology (table 3). Regarding local recurrence rates, there were 3 recurrences (8.1%) in 37 patients after PN (2 local and 1 combined relapse) versus a 3.1% recurrence rate after TN. In the PN group the recurrence rate was higher in those undergoing primary PN without ChT (13.3%) versus PN after chemotherapy (4.5%). Due to the small numbers, this finding is also only a trend and not a significant difference.
TABLE 1. Postoperative staging of eligible patients No. Partial Nephrectomy Stage
Primary Operation
Operation After Chemotherapy
I II N⫺ II N⫹ III Unknown
13 0 1 1 0
15 6 0 1 0
No. Total Tumor Nephrectomy Total (%)
Primary Operation
Operation After Chemotherapy
Total (%)
28 (76) 6 (16) 1 (3) 2 (5) 0 (0)
67 23 5 38 6
327 160 54 69 20
394 (51) 183 (24) 59 (8) 107 (14) 26 (3)
139
630
769*
Totals 15 22 37 * One patient who received preoperative ChT and irradiation is not included in this table.
941
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR TABLE 2. Partial nephrectomy outcomes Stage I II N⫺ II N⫹ III
No. Primary PN Total (%)
Recurrences
13 (86.6) 0 1§ (6.7) 1 (6.7)
No. PN After Chemotherapy Outcomes
1* 0 1‡ 0
Total (%)
Recurrences
15 (68.2) 6 (27.3)
2† 0
1 (4.5)
0
Outcomes 1 Death‡
1 Death
Totals (%) 15 (100) 2 (13.3)㛳 1 (6.7) * Local recurrence. † One patient had local recurrence and 1 had metastatic disease. ‡ Combined local and metastatic disease. § Metastatic disease at diagnosis. 㛳 One patient had local recurrence, and 1 had combined local and metastatic disease.
22 (100)
2 (9)
1 (4.5)
FIG. 1. A, relapse-free survival after PN—primary surgery versus ChT. B, relapse-free survival after TN—primary surgery versus ChT. ns, Not significant. n, Number of patients. CT, computerized tomography.
FIG. 2. A, relapse-free survival after ChT—PN versus TN. B, relapse-free survival after primary surgery—PN versus TN. ns, Not significant. n, Number of patients.
TABLE 3. Relapse-free survival after partial nephrectomy in view of histological subtype Histology Low risk Intermediate risk High risk Nephroblastomatosis
No. Primary PN
No. PN After Chemotherapy
Total
Relapses
RFS*
Total
Relapses
RFS*
5 10 0 0
0 2 0 0
1.00 0.75
0 17 2 3
0 1 1 0
0.93 0.5 1.0
22
2
Totals 15 * Differences were not significant but series were small.
2
DISCUSSION
Patients with ulWT have a good prognosis (especially with stage I disease), and treatment modifications should not adversely affect the potential for excellent outcomes. Treatment protocols aim to decrease treatment associated morbidity and to guarantee a good quality of life, without decreasing disease-free survival. Considerations regarding nephron sparing surgery are based on the increased survival rate during recent years, heterogeneous syndromes associated with WT that require careful followup examination and bi-
lateral disease with the risk of renal failure in up to 15% of the patients after 15 years.10 On one side of the issue, Baudoin2 and Wikstad3 et al have presented long-term followup studies that demonstrate a significant decrease in the glomerular filtration rate and increase in albuminuria 25 years after unilateral nephrectomy. On the other side, there are reports that do not reveal any significant risk for late renal dysfunction.11 Meticulous and closely controlled examinations of children suffering from syndromes associated with WT, such as the Denys-Drash syndrome and the
942
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
Beckwith-Wiedemann syndrome, may allow early detection of WT so that nephron sparing surgery can be possible.12 Besides surgical complications, children with bilateral WT are at high risk for impaired renal function from radiation therapy to the remaining nephrons and use of nephrotoxic chemotherapeutic agents, and a theoretical risk due to hyperfiltration of the remaining nephrons.13 Additionally, the knowledge of a 2% to 3% risk of metachronous WT in the contralateral kidney and synchronous bilateral disease (in 6% to 8% of all WT cases) has necessitated the development of nephron sparing surgical techniques such as partial nephrectomy and tumor enucleation.14 As for bilateral WT, and also for ulWT, some authors have reported encouraging results.5, 6, 14 –17 Another study has presented bench surgery with autotransplantation as another new surgical technique.18 However, all the studies concerning PN in ulWT were limited to a small number of patients. Our patient groups, including 37 undergoing PN and 770 undergoing TN for ulWT, consist of a remarkable number of children. When we try to draw some conclusions we have to keep in mind that we are presenting data from a retrospective multicenter study, where the protocol was violated on purpose concerning patients who underwent primary surgery and partial nephrectomy. Despite a larger amount of stage I disease in the PN group (76% versus 51% for TN), the 5-year OS rate was 93% for both groups. The surgical results may be influenced by the fact that the 37 patients in the PN group were treated at 26 hospitals. A few centers with a limited number of experienced surgeons could better commit to standardized indication criteria for PN with possibly a better outcome. When justifying PN in patients with ulWT, one always has to balance the proposed benefit (decrease in the late incidence of renal failure) versus the immediate goal (achieving the best survival). Although the difference in local recurrence rates between PN and TN is small, it could have a significant impact. Children who have development of local recurrence after surgery for WT have a markedly decreased survival.19 Despite the fact that statistical analysis for OS and RFS showed no significant difference, our results demonstrate a trend toward a better outcome for patients receiving PN and ChT (RFS rate 90% for ChT versus 86% for primary surgery) (fig. 1, A). This trend also reveals that in selected patients undergoing PN preoperative ChT may be beneficial and lead to a better outcome. Being restrictive in the indication for PN, RFS after PN with ChT is similar to that of TN. Discussing the indication for PN in patients with ulWT, we have to include the histological subtype. There was a difference between low and intermediate versus high risk histology (table 3). Therefore, indication for PN in patients with ulWT should additionally be based on intraoperative biopsy with frozen section. Besides confirming a tumor-free margin, this allows evaluation of the histology at the time of PN and can exclude those patients with high risk histology. Patients who have anaplasia or other unfavorable histology should not be offered PN.
CONCLUSION
Besides a good operative technique, the success of a surgical procedure also depends on a correct indication. Summarizing the results of our statistical analysis, we may conclude that PN in ulWT represents a therapeutic option for carefully selected patients. In view of the excellent results after TN, indications for PN must be based on strict criteria. The indications for PN should include stage I disease with a well-defined tumor margin in the kidney, if a complete tumor resection is possible, with low or intermediate risk histology, no invasion of the tumor into renal vessels or collecting system, good response to induction chemotherapy and intra-
operative histological confirmation of tumor-free margin on frozen section.
REFERENCES
1. Tournade, M. F., Com-Nogue, C., Voute, P. A., Lemerle, J., de Kraker, J., Delemarre, J. F. et al: Results of the Sixth International Society of Pediatric Oncology Wilms’ Tumor Trial and Study: a risk-adapted therapeutic approach in Wilms’ tumor. J Clin Oncol, 11: 1014, 1993 2. Baudoin, P., Provoost, A. P. and Molenaar, J. C.: Renal function up to 50 years after unilateral nephrectomy in childhood. Am J Kidney Dis, 21: 603, 1993 3. Wikstad, I., Celsi, G., Larsson, L., Herin, P. and Aperia, A.: Kidney function in adults born with unilateral agenesis or nephrectomized in childhood. Pediatr Nephrol, 2: 177, 1988 4. Horwitz, J. R., Ritchey, M. L., Moksness, J., Breslow, N. E., Smith, G. R., Thomas, P. R. et al: Renal salvage procedures in patients with synchronous bilateral Wilms’ tumors: a report from the National Wilms’ Tumor Study Group. J Pediatr Surg, 31: 1020, 1996 5. Kumar, R., Fitzgerald, R. and Breatnach, F.: Conservative surgical management of bilateral Wilms tumor: results of the United Kingdom Children’s Cancer Study Group. J Urol, 160: 1450, 1998 6. Moorman-Voestermans, C. G., Aronson, D. C., Staalman, C. R., Delemarre, J. F. and de Kraker, J.: Is partial nephrectomy appropriate treatment for unilateral Wilms’ tumor? J Pediatr Surg, 33: 165, 1998 7. Cozzi, F., Schiavetti, A., Bonanni, M., Cozzi, D. A., Matrunola, M. and Castello, M. A.: Enucleative surgery for stage I nephroblastoma with a normal contralateral kidney. J Urol, 156: 1788, 1996 8. Delemarre, J. F., Sandstedt, B., Harms, D., Boccon-Gibod, L. and Vujanic, G. M.: The new SIOP (Stockholm) working classification of renal tumours of childhood. International Society of Paediatric Oncology. Med Pediatr Oncol, 26: 145, 1996 9. Graf, N., Tournade, M. F. and de Kraker, J.: The role of preoperative chemotherapy in the management of Wilms’ tumor. The SIOP studies International Society of Pediatric Oncology. Urol Clin North Am, 27: 443, 2000 10. Blakely, M. L. and Ritchey, M. L.: Controversies in the management of Wilms’ tumor. Semin Pediatr Surg, 10: 127, 2001 11. Bhisitkul, D. M., Morgan, E. R., Vozar, M. A. and Langman, C. B.: Renal functional reserve in long-term survivors of unilateral Wilms tumor. J Pediatr, 118: 698, 1991 12. McNeil, D. E., Langer, J. C., Choyke, P. and DeBaun, M. R.: Feasibility of partial nephrectomy for Wilms’ tumor in children with Beckwith-Wiedemann syndrome who have been screened with abdominal ultrasonography. J Pediatr Surg, 37: 57, 2002 13. Ritchey, M. L., Green, D. M., Thomas, P. R., Smith, G. R., Haase, G., Shochat, S. et al: Renal failure in Wilms’ tumor patients: a report from the National Wilms’ Tumor Study Group. Med Pediatr Oncol, 26: 75, 1996 14. Cooper, C. S., Jaffe, W. I., Huff, D. S., Canning, D. A., Zderic, S. A., Meadows, A. T. et al: The role of renal salvage procedures for bilateral Wilms’ tumor: a 15-year review. J Urol, 163: 265, 2000 15. Paya, K., Horcher, E., Lawrenz, K., Rebhandl, W. and Zoubek, A.: Bilateral Wilms’ tumor—surgical aspects. Eur J Pediatr Surg, 11: 99, 2001 16. Cozzi, D. A., Schiavetti, A., Morini, F., Castello, M. A. and Cozzi, F.: Nephron-sparing surgery for unilateral primary renal tumor in children. J Pediatr Surg, 36: 362, 2001 17. Urban, C. E., Lackner, H., Schwinger, W., Klos, I., Hoellwarth, M., Sauer, H. et al: Partial nephrectomy in well-responding stage I Wilms’ tumors: report of three cases. Pediatr Hematol Oncol, 12: 143, 1995 18. Desai, D., Nicholls, G. and Duffy, P. G.: Bench surgery with autotransplantation for bilateral synchronous Wilms’ tumor: a report of three cases. J Pediatr Surg, 34: 632, 1999 19. Shamberger, R. C., Guthrie, K. A., Ritchey, M. L., Haase, G. M., Takashima, J., Beckwith, J. B. et al: Surgery-related factors and local recurrence of Wilms tumor in National Wilms Tumor Study 4. Ann Surg, 229: 292, 1999
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR EDITORIAL COMMENT The authors have presented a large series of children with unilateral Wilms tumor undergoing partial nephrectomy. This was not part of a prospective study. The individual institutions elected to perform partial nephrectomy if they deemed the procedure to be feasible. The majority of these operations were done after chemotherapy but 15 patients underwent primary partial nephrectomy. They describe this as a “radical partial nephrectomy.” It is not clear whether this approach routinely included removal of lymph nodes or lymph node sampling. The latter is invaluable for adequate staging of the primary tumor (reference 19 in article). There has been interest in performing partial nephrectomy for Wilms tumor for many years. In addition to the citations in the article, others have advocated this approach.1 In most of the early reports partial nephrectomy was performed after preoperative chemotherapy was administered to shrink the tumor. The majority of Wilms tumors are too large for consideration for partial nephrectomy at diagnosis.2 There are 2 reasons frequently cited to justify partial nephrectomy in children with unilateral Wilms tumor. One is the potential for renal insufficiency and the other is the risk of metachronous Wilms tumor development. Several authors have noted an increased risk of proteinuria and decreased creatinine clearance in patients undergoing nephrectomy for Wilms tumor. However, several studies have found no significant alterations in renal function following unilateral nephrectomy (reference 11 in article).3, 4 Most of the studies that have reported some dysfunction noted some alteration in glomerular filtration rate, increased urinary excretion of albumin and occasionally hypertension. The development of frank renal failure is extremely uncommon in patients who have been treated for unilateral Wilms tumor. A review from the National Wilms Tumor Study Group found that the rate of renal failure was 0.25% in this group of patients. Most of the patients with unilateral Wilms tumor in whom renal failure developed had definite risk factors such as the Denys-Drash syndrome. The latter patients are predisposed to renal dysfunction and often present with renal failure at diagnosis (reference 13 in article). A later report from the National Wilms Tumor Study Group showed that patients with aniridia and genitourinary anomalies also have an increased risk of renal failure.5 Breslow et al found a 38% incidence of renal failure that occurred at a median of 14 years from diagnosis.5 Patients who are at increased risk for metachronous tumors include those with overgrowth syndromes, such as the BeckwithWiedemann syndrome, that have a known propensity to develop Wilms tumor. Also, patients with aniridia and children with Wilms tumor and nephrogenic rests, precursor lesions for Wilms tumor, have an increased risk of metachronous tumors.6 Children at risk, such as with the Beckwith-Wiedemann syndrome, are typically followed with ultrasound every 3 months to screen for the development of tumors. These patients are more likely to have small tumors identified that are amenable to partial nephrectomy.7 Included in this report are several patients with cystic partially differentiated nephroblastoma. These patients appear to be at low risk and are good candidates for a nephron sparing approach.8 There are also 2 cases of congenital nephroblastoma included in this report, which is not considered to be Wilms tumor. These patients are also at lower risk for recurrent disease. The principal concern with advocating partial nephrectomy for patients with unilateral Wilms tumor is that there is an increased risk of positive surgical margins after partial nephrectomy in these patients. The greatest experience with partial nephrectomy in Wilms tumor is in children with bilateral Wilms tumor, where a renal sparing approach has long been advocated (reference 4 in article). In children with bilateral Wilms tumor there is a markedly increased risk of renal failure. Therefore, it is accepted that an increased risk of recurrence with nephron sparing surgery with a nephron sparing approach is justified (reference 13 in article). In this report there was an 8.1% incidence of local recurrence after partial nephrectomy. It is not clear whether local disease recurred only in the patients with Wilms tumor. If the kidneys with congenital mesoblastic nephroma and cystic partially differentiated nephroblastoma were excluded, this would suggest a higher rate of local relapse. Although the numbers were too small to demonstrate a significant difference from the patients undergoing complete nephrectomy, it is a cause for real concern. In those undergoing primary partial nephrectomy without chemotherapy the rate of local recurrence was even higher, at 13.3%.
943
Children with local recurrence after nephrectomy for Wilms tumor are clearly at a survival disadvantage. Shamberger et al reported a 43% survival 2 years after local relapse (reference 19 in article). The survivors also require much more intensive therapy with the increased risk of late sequelae. The authors have presented criteria for partial nephrectomy. The lesion should be completely excised with a margin of normal renal parenchyma. These patients should not undergo partial nephrectomy if the tumor cannot be removed as a stage I. They recommend frozen sections to confirm a negative margin, and also to evaluate the histology. Patients with high risk histological patterns such as anaplasia or persistent blastemal predominant tumor after chemotherapy should be treated with complete nephrectomy. These patients have a greater risk of recurrence, as the tumors have been found to be resistant to chemotherapy.9 A nephron sparing approach can probably be best justified for children who are clearly at risk for late renal dysfunction or those patients at increased risk for development of metachronous tumors. This would include patients with aniridia, genitourinary anomalies, and syndromes such as the Beckwith-Wiedemann syndrome and all other overgrowth syndromes. The available data do not support partial nephrectomy for other patients with unilateral Wilms tumor. Given the low rate of renal failure in these children, it would take a fairly large number of partial nephrectomies to prevent even one case of renal failure. This benefit is clearly outweighed by the risk of increased mortality if there is even a 5% increase in local relapse rates. Michael L. Ritchey Division of Urology University of Texas Health Science Center at Houston Houston, Texas 1. McLorie, G. A., McKenna, P. H., Greenberg, M., Babyn, P., Thorner, P., Churchill, B. M. et al: Reduction in tumor burden allowing partial nephrectomy following preoperative chemotherapy in biopsy proved Wilms tumor. J Urol, part 2, 146: 509, 1991 2. Wilimas, J. A., Magill, L., Parham, D. M., Kumar, M. and Douglass, E. C.: The potential for renal salvage in nonmetastatic unilateral Wilms’ tumor. Am J Pediatr Hematol Oncol, 13: 342, 1991 3. Detoledo, J., Rodriguez, G. C., Gallego, M. S., Moraga, F., Catello, F. and Moreno, A.: Renal function in long-term survivors of Wilms tumor. Med Pediatr Oncol, 23: 265, 1994 4. Barrera, M., Roy, L. P. and Stevens, M.: Long-term follow-up after unilateral nephrectomy and radiotherapy for Wilms’ tumour. Pediatr Nephrol, 3: 430, 1989 5. Breslow, N. E., Takashima, J. R., Ritchey, M. L., Strong, L. C. and Green, D. M.: Renal failure in the Denys-Drash and Wilms’ tumor-aniridia syndromes. Cancer Res, 60: 4030, 2000 6. Coppes, M. J., Arnold, M., Beckwith, J. B., Ritchey, M. L., D’Angio, G. J., Green, D. M. et al: Factors affecting the risk of contralateral Wilms tumor development: a report from the National Wilms Tumor Study Group. Cancer, 85: 1616, 1999 7. Choyke, P. L., Siegel, M. J., Craft, A. W., Green, D. M. and DeBaun, M. R.: Screening for Wilms tumor in children with Beckwith-Wiedemann syndrome or idiopathic hemihypertrophy. Med Pediatr Oncol, 32: 196, 1999 8. Blakely, M. L., Shamberger, R. C., Norkool, P., Beckwith, J. B., Green, D. M. and Ritchey, M. L.: Outcome of children with cystic partially differentiated nephroblastoma treated with or without chemotherapy. J Pediatr Surg, 38: 897, 2003 9. Weirich, A., Leuschner, I., Harms, D., Vujanic, G. M., Troger, J., Abel, U. et al: Clinical impact of histologic subtypes in localized non-anaplastic nephroblastoma treated according to the trial and study SIOP-9/GPOH. Ann Oncol, 12: 311, 2001 REPLY BY AUTHORS When performing radical primary partial nephrectomy, lymph node sampling and not only of lymph nodes removal was routinely performed. In the SIOP 2001 protocol it is stated that sampling and histological examination of lymph nodes are imperative for accurate staging and subsequent treatment. Hilar and para-aortic lymph nodes at the origin of the renal artery (regional nodes) and nodes
944
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
below or above this level (extra regional nodes) should be sampled even if not suspicious. Involved or suspicious lymph nodes must be excised without rupture. They must be carefully labeled and sent to the pathologist separately with an accurate description of their position and characteristics. This information affects staging, treatment and outcome. Radical lymph node dissection does not enhance survival and, therefore, is not part of the surgical therapy. In children with syndromes associated with WT, such as the Denis
Drash syndrome or Beckwith-Wiedemann syndrome, early detection of WT may allow nephron sparing surgery. Children with bilateral WT as well as 2% to 3% risk of metachronous WT in the contralateral kidney and synchronous bilateral tumor disease require nephron sparing surgical techniques such as partial nephrectomy. PN should only be performed for small unilateral WT with low or intermediate risk histology. A good response to preoperative ChT is mandatory in these patients.
Vol. 170, 939 –944, September 2003 Printed in U.S.A.
DOI: 10.1097/01.ju.0000073848.33092.c7
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR: RESULTS OF STUDY SIOP 93– 01/GPOH FRANK-MARTIN HAECKER,* DIETRICH
SCHWEINITZ, DIETER HARMS, DIETRICH BUERGER NORBERT GRAF
VON
AND
From the Department of Pediatric Surgery, University Children’s Hospital Basel (F-MH, DvS), Basel, Switzerland, and Central Pediatric Tumor Registry, Department of Pathology, University of Kiel (DH), Kiel, Pediatric Surgery, DRK-Klinik Siegen (DB), Siegen and Pediatric Oncology, University Hospital Homburg/Saar (NG), Homburg/Saar, Germany
ABSTRACT
Purpose: We evaluate results and long-term outcome after partial nephrectomy (PN) for unilateral Wilms tumor (ulWT) in relation to different histological features, performed as initial surgery or after induction chemotherapy (ChT). Materials and Methods: Data from patients with ulWT who had undergone PN in the German Study SIOP 93– 01/GPOH were analyzed for time of surgery in the treatment schedule, postoperative stage, histological features, surgical radicality and outcome. The results were correlated with overall survival and relapse-free survival, and compared with those of patients who had undergone total tumor nephrectomy (TN). Results: A total of 770 patients underwent TN and 37 underwent PN, of which 766 (99.5%) and 36 (97.3%), respectively, were radical procedures. A total of 139 TNs and 15 PNs were performed at primary surgery, 630 TNs and 22 PNs after ChT, and 1 TN after irradiation. Two of 15 patients following primary PN and 1 of 22 following delayed PN had local recurrence. Two patients died of metastatic disease. Tumor stage, overall survival of 93% and relapse-free survival of 88% were equal after PN and TN in analysis of the whole group. This was also true for patients with surgery after ChT. One of 2 patients with histologically unfavorable disease had relapse after PN compared to only 3 of 35 patients with histologically low or intermediate risk disease. Conclusions: PN should be performed only for small, histologically low or intermediate risk tumors after good response to ChT to secure a complete resection. KEY WORDS: nephroblastoma, kidney neoplasms, nephrectomy, antineoplastic agents
Within the last 10 to 15 years the treatment of nephroblastoma or Wilms tumor (WT) has significantly improved the outcome of affected children. But at the same time clinicians have also been faced with the task of treating the late effects of anticancer therapy. The crude 3-year survival rate for children suffering from WT is more than 90%.1 Actual treatment protocols must aim to decrease morbidity associated with treatment without impairing the cure rates. During the postoperative period after total tumor nephrectomy (TN) renal function of the solitary kidney may be influenced by chemotherapy and/or hyperfiltration. Thereafter, more common risks such as trauma, infection and obstruction may cause additional loss of renal function. Children with a single kidney, studied at followup for more than 25 years, present with a significant decrease in the glomerular filtration rate and a significant increase in albuminuria compared to those studied at a shorter followup.2, 3 Additionally, there exists a 2% to 3% risk that a metachronous WT may develop in the contralateral kidney later in life. After successful inauguration of parenchymal sparing surgery in children with bilateral WT recent reports have emphasized conservative surgery for children with unilateral (ul) WT.4 –7 To evaluate the risks and benefits of this procedure for long-term outcome, and to reevaluate indication criteria for partial nephrectomy (PN) in ulWT, we analyzed the data from patients with ulWT treated with PN in com-
parison to TN in the German branch of the study SIOP 93– 01/GPOH. The protocol of the SIOP 93– 01 WT study includes preoperative induction chemotherapy (ChT) followed by tumor nephrectomy, which is in contrast to the protocol of the United States National Wilms Tumor Study (NWTS), with primary surgery followed by chemotherapy. PATIENTS AND METHODS
From April 1994 to December 2001, 865 patients with ulWT were registered in the GPOH branch of the SIOP 93– 01 WT study, which includes patients from the German spoken area of Switzerland, Austria and Germany. The patients were treated according to the SIOP/GPOH study protocol with ChT followed by tumor nephrectomy, or they underwent primary surgery. A total of 807 patients with homogeneous distribution of age and sex with ulWT were further analyzed. Fifty-eight patients were excluded from this study—in 8 patients only a tumor biopsy was performed, in 48 there was no reliable information about the surgical treatment and in 2 the tumor was inoperable. Initial ultrasound and computerized tomography or magnetic resonance imaging were performed in all patients. According to the study protocol, the treatment for patients with ulWT between age 6 months and 16 years consisted of ChT with actinomycin D and vincristine for 4 weeks, as well as additional doxorubicin for patients with stage IV disease for 6 weeks, followed by TN. PN was not part of the SIOP 93– 01 protocol, but was done based on individual choice of the local center in 37 patients, based on criteria such as stage I disease, well-defined tumor margin, absence of invasion into the renal vessels or collecting system and determination that at
Accepted for publication April 4, 2003. Supported by a grant from the German Cancer League (“Deutsche Krebshilfe,” project No. 70 –1899). * Corresponding author: University Children’s Hospital, P.O. Box CH-4005, Basel, Switzerland (telephone: ⫹41/61/685 5239; FAX: ⫹41/61685 5011; e-mail: [email protected]). 939
940
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
least 50% of the remaining kidney was preservable. A frozen section was not done in every case to confirm the absence of capsular extension or invasion. After surgery pathological classification was done by the local pathologist and the Kiel Pediatric Tumor Registry (DH) as the reference institution for pathological evaluation of the SIOP 93– 01/GPOH study.8 Based on the SIOP staging criteria, tumor staging was undertaken postoperatively.9 SIOP staging criteria are equal to those used by the NWTS, except for 2 differences—staging is undertaken after chemotherapy pretreatment, and stage II tumors are distinguished as those with positive lymph nodes (stage II N⫹, corresponding to NWTS stage III) or without positive lymph nodes (stage II N⫺, corresponding to NWTS stage II). The staging system of the new SIOP 2001 protocol is identical to the NWTS staging system. The SIOP histological classification “low risk/intermediate risk” is equivalent to the NWTS “favorable histology,” and the “high risk” SIOP classification is identical to “unfavorable histology” in the NWTS. The clinical and followup data were centrally collected by the SIOP 93– 01/GPOH study office (NG). Data were collected and stored in a relational dBASE (VersaSoft Corp, San Jose, California) database. Statistical analysis was done using the software package SPSS for Windows release 11.0.1 (SPSS, Chicago, Illinois) software, including descriptive functions and life table analysis according to Kaplan-Meier. The log rank test was used for statistical comparison of different factors. RESULTS
Partial nephrectomy. Radical PN resulting in local stage I tumor was performed in 37 patients (4.6%) with a median age of 2.6 years (range 1 month to 27 years) and a sex distribution of 17 females to 20 males. Mean followup was 4 years (range 1 to 7). In 15 patients primary surgery without ChT was performed. Of these 15 patients tumor volume was less than 30 ml in 5, unknown in 5 and 224 to 780 ml in 5. Of the 15 patients 5 were younger than 6 months, 2 had congenital mesoblastic nephroma, and 3 had cystic partial differentiated nephroma and a low risk tumor. A total of 22 patients underwent PN after ChT. In the PN group 76% of the patients had stage I disease, whereas in the TN group only 51% had stage I disease, probably due to smaller tumors in the PN group (table 1). Tumor recurrence was found in 2 patients after primary PN (1 local and 1 combined relapse) as well as in 2 after ChT and subsequent PN (1 local recurrence and 1 metastatic disease). Of these 4 patients 2 had an unfavorable outcome and died, 1 of whom had metastatic disease after primary resection of WT from a horseshoe kidney (histologically blastemic predominant subtype) and 1 had a nonmetastasizing malignant rhabdoid tumor (table 2). This tumor was completely excised and, therefore, stage I. After receiving the result of the histopathological evaluation tumor nephrectomy was not done. The other 2 patients (both stage I) had a recurrent local tumor, which was resected completely at reoperation (table 2). One of these patients again
underwent PN because of a cystic nephroma in the contralateral kidney. Total tumor nephrectomy. TN was performed in 770 patients (95.4%) with a median age of 3.2 years (range 1 month to 62 years), of whom 403 were female and 367 male. A total of 139 patients underwent primary surgery without previous ChT. In 630 patients TN was performed after ChT and in 1 patient after irradiation (table 1). Mean followup was 4 years (range 1 to 8). A total of 120 patients had metastatic disease at the time of diagnosis. Of these patients 17 died, as well as 24 patients without initial metastatic disease. Tumor recurrence was observed in 9 patients (6.5%) after primary TN (local recurrence 2, recurrent metastatic disease 6, local recurrence plus recurrent metastatic disease 1) as well as in 69 patients (11%) after ChT with subsequent TN (local recurrence 14, recurrent metastatic disease 48, local recurrence plus recurrent metastatic disease 7). Of the 139 patients undergoing primary TN 6 (4.3%) eventually died of disease, of whom 3 had metastatic disease at diagnosis. Of 630 patients undergoing TN after ChT 35 (5.6%) died, of whom 14 had metastases at diagnosis. Outcome—PN versus TN. Statistical analysis included overall survival (OS) and relapse-free survival (RFS). Comparing OS and RFS, the differences were more obvious for the RFS rates, and so we present the RFS rates herein. The probability of 5-year RFS was 92% for all patients undergoing primary surgery versus 87% for patients receiving ChT and subsequent TN (not significant). Regarding the 5-year RFS rates for PN (primary surgery versus ChT) and TN (primary surgery versus ChT), similarly there were no significant differences. In the PN group RFS was 90% after ChT versus 86% after primary surgery, while in the TN group RFS was 87% after ChT versus 92% after primary surgery (fig. 1). RFS after ChT showed no significant difference between PN (90%) and TN (87%). Similarly, RFS after primary surgery was 92% for TN and 86% for PN (not significant) (fig. 2). In view of the histopathological subtype, there was a significant difference in the outcome between WT with low and standard risk versus high risk histology concerning patients without metastatic disease at diagnosis who underwent TN. There were only 2 patients in the PN group with high risk histology (table 3). The number was too small to demonstrate a difference in outcome between PN and TN. Although in both groups statistical analysis revealed no significant difference (probably due to the small number of patients), the results show a tendency for a poor outcome for patients with high risk histology (table 3). Regarding local recurrence rates, there were 3 recurrences (8.1%) in 37 patients after PN (2 local and 1 combined relapse) versus a 3.1% recurrence rate after TN. In the PN group the recurrence rate was higher in those undergoing primary PN without ChT (13.3%) versus PN after chemotherapy (4.5%). Due to the small numbers, this finding is also only a trend and not a significant difference.
TABLE 1. Postoperative staging of eligible patients No. Partial Nephrectomy Stage
Primary Operation
Operation After Chemotherapy
I II N⫺ II N⫹ III Unknown
13 0 1 1 0
15 6 0 1 0
No. Total Tumor Nephrectomy Total (%)
Primary Operation
Operation After Chemotherapy
Total (%)
28 (76) 6 (16) 1 (3) 2 (5) 0 (0)
67 23 5 38 6
327 160 54 69 20
394 (51) 183 (24) 59 (8) 107 (14) 26 (3)
139
630
769*
Totals 15 22 37 * One patient who received preoperative ChT and irradiation is not included in this table.
941
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR TABLE 2. Partial nephrectomy outcomes Stage I II N⫺ II N⫹ III
No. Primary PN Total (%)
Recurrences
13 (86.6) 0 1§ (6.7) 1 (6.7)
No. PN After Chemotherapy Outcomes
1* 0 1‡ 0
Total (%)
Recurrences
15 (68.2) 6 (27.3)
2† 0
1 (4.5)
0
Outcomes 1 Death‡
1 Death
Totals (%) 15 (100) 2 (13.3)㛳 1 (6.7) * Local recurrence. † One patient had local recurrence and 1 had metastatic disease. ‡ Combined local and metastatic disease. § Metastatic disease at diagnosis. 㛳 One patient had local recurrence, and 1 had combined local and metastatic disease.
22 (100)
2 (9)
1 (4.5)
FIG. 1. A, relapse-free survival after PN—primary surgery versus ChT. B, relapse-free survival after TN—primary surgery versus ChT. ns, Not significant. n, Number of patients. CT, computerized tomography.
FIG. 2. A, relapse-free survival after ChT—PN versus TN. B, relapse-free survival after primary surgery—PN versus TN. ns, Not significant. n, Number of patients.
TABLE 3. Relapse-free survival after partial nephrectomy in view of histological subtype Histology Low risk Intermediate risk High risk Nephroblastomatosis
No. Primary PN
No. PN After Chemotherapy
Total
Relapses
RFS*
Total
Relapses
RFS*
5 10 0 0
0 2 0 0
1.00 0.75
0 17 2 3
0 1 1 0
0.93 0.5 1.0
22
2
Totals 15 * Differences were not significant but series were small.
2
DISCUSSION
Patients with ulWT have a good prognosis (especially with stage I disease), and treatment modifications should not adversely affect the potential for excellent outcomes. Treatment protocols aim to decrease treatment associated morbidity and to guarantee a good quality of life, without decreasing disease-free survival. Considerations regarding nephron sparing surgery are based on the increased survival rate during recent years, heterogeneous syndromes associated with WT that require careful followup examination and bi-
lateral disease with the risk of renal failure in up to 15% of the patients after 15 years.10 On one side of the issue, Baudoin2 and Wikstad3 et al have presented long-term followup studies that demonstrate a significant decrease in the glomerular filtration rate and increase in albuminuria 25 years after unilateral nephrectomy. On the other side, there are reports that do not reveal any significant risk for late renal dysfunction.11 Meticulous and closely controlled examinations of children suffering from syndromes associated with WT, such as the Denys-Drash syndrome and the
942
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
Beckwith-Wiedemann syndrome, may allow early detection of WT so that nephron sparing surgery can be possible.12 Besides surgical complications, children with bilateral WT are at high risk for impaired renal function from radiation therapy to the remaining nephrons and use of nephrotoxic chemotherapeutic agents, and a theoretical risk due to hyperfiltration of the remaining nephrons.13 Additionally, the knowledge of a 2% to 3% risk of metachronous WT in the contralateral kidney and synchronous bilateral disease (in 6% to 8% of all WT cases) has necessitated the development of nephron sparing surgical techniques such as partial nephrectomy and tumor enucleation.14 As for bilateral WT, and also for ulWT, some authors have reported encouraging results.5, 6, 14 –17 Another study has presented bench surgery with autotransplantation as another new surgical technique.18 However, all the studies concerning PN in ulWT were limited to a small number of patients. Our patient groups, including 37 undergoing PN and 770 undergoing TN for ulWT, consist of a remarkable number of children. When we try to draw some conclusions we have to keep in mind that we are presenting data from a retrospective multicenter study, where the protocol was violated on purpose concerning patients who underwent primary surgery and partial nephrectomy. Despite a larger amount of stage I disease in the PN group (76% versus 51% for TN), the 5-year OS rate was 93% for both groups. The surgical results may be influenced by the fact that the 37 patients in the PN group were treated at 26 hospitals. A few centers with a limited number of experienced surgeons could better commit to standardized indication criteria for PN with possibly a better outcome. When justifying PN in patients with ulWT, one always has to balance the proposed benefit (decrease in the late incidence of renal failure) versus the immediate goal (achieving the best survival). Although the difference in local recurrence rates between PN and TN is small, it could have a significant impact. Children who have development of local recurrence after surgery for WT have a markedly decreased survival.19 Despite the fact that statistical analysis for OS and RFS showed no significant difference, our results demonstrate a trend toward a better outcome for patients receiving PN and ChT (RFS rate 90% for ChT versus 86% for primary surgery) (fig. 1, A). This trend also reveals that in selected patients undergoing PN preoperative ChT may be beneficial and lead to a better outcome. Being restrictive in the indication for PN, RFS after PN with ChT is similar to that of TN. Discussing the indication for PN in patients with ulWT, we have to include the histological subtype. There was a difference between low and intermediate versus high risk histology (table 3). Therefore, indication for PN in patients with ulWT should additionally be based on intraoperative biopsy with frozen section. Besides confirming a tumor-free margin, this allows evaluation of the histology at the time of PN and can exclude those patients with high risk histology. Patients who have anaplasia or other unfavorable histology should not be offered PN.
CONCLUSION
Besides a good operative technique, the success of a surgical procedure also depends on a correct indication. Summarizing the results of our statistical analysis, we may conclude that PN in ulWT represents a therapeutic option for carefully selected patients. In view of the excellent results after TN, indications for PN must be based on strict criteria. The indications for PN should include stage I disease with a well-defined tumor margin in the kidney, if a complete tumor resection is possible, with low or intermediate risk histology, no invasion of the tumor into renal vessels or collecting system, good response to induction chemotherapy and intra-
operative histological confirmation of tumor-free margin on frozen section.
REFERENCES
1. Tournade, M. F., Com-Nogue, C., Voute, P. A., Lemerle, J., de Kraker, J., Delemarre, J. F. et al: Results of the Sixth International Society of Pediatric Oncology Wilms’ Tumor Trial and Study: a risk-adapted therapeutic approach in Wilms’ tumor. J Clin Oncol, 11: 1014, 1993 2. Baudoin, P., Provoost, A. P. and Molenaar, J. C.: Renal function up to 50 years after unilateral nephrectomy in childhood. Am J Kidney Dis, 21: 603, 1993 3. Wikstad, I., Celsi, G., Larsson, L., Herin, P. and Aperia, A.: Kidney function in adults born with unilateral agenesis or nephrectomized in childhood. Pediatr Nephrol, 2: 177, 1988 4. Horwitz, J. R., Ritchey, M. L., Moksness, J., Breslow, N. E., Smith, G. R., Thomas, P. R. et al: Renal salvage procedures in patients with synchronous bilateral Wilms’ tumors: a report from the National Wilms’ Tumor Study Group. J Pediatr Surg, 31: 1020, 1996 5. Kumar, R., Fitzgerald, R. and Breatnach, F.: Conservative surgical management of bilateral Wilms tumor: results of the United Kingdom Children’s Cancer Study Group. J Urol, 160: 1450, 1998 6. Moorman-Voestermans, C. G., Aronson, D. C., Staalman, C. R., Delemarre, J. F. and de Kraker, J.: Is partial nephrectomy appropriate treatment for unilateral Wilms’ tumor? J Pediatr Surg, 33: 165, 1998 7. Cozzi, F., Schiavetti, A., Bonanni, M., Cozzi, D. A., Matrunola, M. and Castello, M. A.: Enucleative surgery for stage I nephroblastoma with a normal contralateral kidney. J Urol, 156: 1788, 1996 8. Delemarre, J. F., Sandstedt, B., Harms, D., Boccon-Gibod, L. and Vujanic, G. M.: The new SIOP (Stockholm) working classification of renal tumours of childhood. International Society of Paediatric Oncology. Med Pediatr Oncol, 26: 145, 1996 9. Graf, N., Tournade, M. F. and de Kraker, J.: The role of preoperative chemotherapy in the management of Wilms’ tumor. The SIOP studies International Society of Pediatric Oncology. Urol Clin North Am, 27: 443, 2000 10. Blakely, M. L. and Ritchey, M. L.: Controversies in the management of Wilms’ tumor. Semin Pediatr Surg, 10: 127, 2001 11. Bhisitkul, D. M., Morgan, E. R., Vozar, M. A. and Langman, C. B.: Renal functional reserve in long-term survivors of unilateral Wilms tumor. J Pediatr, 118: 698, 1991 12. McNeil, D. E., Langer, J. C., Choyke, P. and DeBaun, M. R.: Feasibility of partial nephrectomy for Wilms’ tumor in children with Beckwith-Wiedemann syndrome who have been screened with abdominal ultrasonography. J Pediatr Surg, 37: 57, 2002 13. Ritchey, M. L., Green, D. M., Thomas, P. R., Smith, G. R., Haase, G., Shochat, S. et al: Renal failure in Wilms’ tumor patients: a report from the National Wilms’ Tumor Study Group. Med Pediatr Oncol, 26: 75, 1996 14. Cooper, C. S., Jaffe, W. I., Huff, D. S., Canning, D. A., Zderic, S. A., Meadows, A. T. et al: The role of renal salvage procedures for bilateral Wilms’ tumor: a 15-year review. J Urol, 163: 265, 2000 15. Paya, K., Horcher, E., Lawrenz, K., Rebhandl, W. and Zoubek, A.: Bilateral Wilms’ tumor—surgical aspects. Eur J Pediatr Surg, 11: 99, 2001 16. Cozzi, D. A., Schiavetti, A., Morini, F., Castello, M. A. and Cozzi, F.: Nephron-sparing surgery for unilateral primary renal tumor in children. J Pediatr Surg, 36: 362, 2001 17. Urban, C. E., Lackner, H., Schwinger, W., Klos, I., Hoellwarth, M., Sauer, H. et al: Partial nephrectomy in well-responding stage I Wilms’ tumors: report of three cases. Pediatr Hematol Oncol, 12: 143, 1995 18. Desai, D., Nicholls, G. and Duffy, P. G.: Bench surgery with autotransplantation for bilateral synchronous Wilms’ tumor: a report of three cases. J Pediatr Surg, 34: 632, 1999 19. Shamberger, R. C., Guthrie, K. A., Ritchey, M. L., Haase, G. M., Takashima, J., Beckwith, J. B. et al: Surgery-related factors and local recurrence of Wilms tumor in National Wilms Tumor Study 4. Ann Surg, 229: 292, 1999
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR EDITORIAL COMMENT The authors have presented a large series of children with unilateral Wilms tumor undergoing partial nephrectomy. This was not part of a prospective study. The individual institutions elected to perform partial nephrectomy if they deemed the procedure to be feasible. The majority of these operations were done after chemotherapy but 15 patients underwent primary partial nephrectomy. They describe this as a “radical partial nephrectomy.” It is not clear whether this approach routinely included removal of lymph nodes or lymph node sampling. The latter is invaluable for adequate staging of the primary tumor (reference 19 in article). There has been interest in performing partial nephrectomy for Wilms tumor for many years. In addition to the citations in the article, others have advocated this approach.1 In most of the early reports partial nephrectomy was performed after preoperative chemotherapy was administered to shrink the tumor. The majority of Wilms tumors are too large for consideration for partial nephrectomy at diagnosis.2 There are 2 reasons frequently cited to justify partial nephrectomy in children with unilateral Wilms tumor. One is the potential for renal insufficiency and the other is the risk of metachronous Wilms tumor development. Several authors have noted an increased risk of proteinuria and decreased creatinine clearance in patients undergoing nephrectomy for Wilms tumor. However, several studies have found no significant alterations in renal function following unilateral nephrectomy (reference 11 in article).3, 4 Most of the studies that have reported some dysfunction noted some alteration in glomerular filtration rate, increased urinary excretion of albumin and occasionally hypertension. The development of frank renal failure is extremely uncommon in patients who have been treated for unilateral Wilms tumor. A review from the National Wilms Tumor Study Group found that the rate of renal failure was 0.25% in this group of patients. Most of the patients with unilateral Wilms tumor in whom renal failure developed had definite risk factors such as the Denys-Drash syndrome. The latter patients are predisposed to renal dysfunction and often present with renal failure at diagnosis (reference 13 in article). A later report from the National Wilms Tumor Study Group showed that patients with aniridia and genitourinary anomalies also have an increased risk of renal failure.5 Breslow et al found a 38% incidence of renal failure that occurred at a median of 14 years from diagnosis.5 Patients who are at increased risk for metachronous tumors include those with overgrowth syndromes, such as the BeckwithWiedemann syndrome, that have a known propensity to develop Wilms tumor. Also, patients with aniridia and children with Wilms tumor and nephrogenic rests, precursor lesions for Wilms tumor, have an increased risk of metachronous tumors.6 Children at risk, such as with the Beckwith-Wiedemann syndrome, are typically followed with ultrasound every 3 months to screen for the development of tumors. These patients are more likely to have small tumors identified that are amenable to partial nephrectomy.7 Included in this report are several patients with cystic partially differentiated nephroblastoma. These patients appear to be at low risk and are good candidates for a nephron sparing approach.8 There are also 2 cases of congenital nephroblastoma included in this report, which is not considered to be Wilms tumor. These patients are also at lower risk for recurrent disease. The principal concern with advocating partial nephrectomy for patients with unilateral Wilms tumor is that there is an increased risk of positive surgical margins after partial nephrectomy in these patients. The greatest experience with partial nephrectomy in Wilms tumor is in children with bilateral Wilms tumor, where a renal sparing approach has long been advocated (reference 4 in article). In children with bilateral Wilms tumor there is a markedly increased risk of renal failure. Therefore, it is accepted that an increased risk of recurrence with nephron sparing surgery with a nephron sparing approach is justified (reference 13 in article). In this report there was an 8.1% incidence of local recurrence after partial nephrectomy. It is not clear whether local disease recurred only in the patients with Wilms tumor. If the kidneys with congenital mesoblastic nephroma and cystic partially differentiated nephroblastoma were excluded, this would suggest a higher rate of local relapse. Although the numbers were too small to demonstrate a significant difference from the patients undergoing complete nephrectomy, it is a cause for real concern. In those undergoing primary partial nephrectomy without chemotherapy the rate of local recurrence was even higher, at 13.3%.
943
Children with local recurrence after nephrectomy for Wilms tumor are clearly at a survival disadvantage. Shamberger et al reported a 43% survival 2 years after local relapse (reference 19 in article). The survivors also require much more intensive therapy with the increased risk of late sequelae. The authors have presented criteria for partial nephrectomy. The lesion should be completely excised with a margin of normal renal parenchyma. These patients should not undergo partial nephrectomy if the tumor cannot be removed as a stage I. They recommend frozen sections to confirm a negative margin, and also to evaluate the histology. Patients with high risk histological patterns such as anaplasia or persistent blastemal predominant tumor after chemotherapy should be treated with complete nephrectomy. These patients have a greater risk of recurrence, as the tumors have been found to be resistant to chemotherapy.9 A nephron sparing approach can probably be best justified for children who are clearly at risk for late renal dysfunction or those patients at increased risk for development of metachronous tumors. This would include patients with aniridia, genitourinary anomalies, and syndromes such as the Beckwith-Wiedemann syndrome and all other overgrowth syndromes. The available data do not support partial nephrectomy for other patients with unilateral Wilms tumor. Given the low rate of renal failure in these children, it would take a fairly large number of partial nephrectomies to prevent even one case of renal failure. This benefit is clearly outweighed by the risk of increased mortality if there is even a 5% increase in local relapse rates. Michael L. Ritchey Division of Urology University of Texas Health Science Center at Houston Houston, Texas 1. McLorie, G. A., McKenna, P. H., Greenberg, M., Babyn, P., Thorner, P., Churchill, B. M. et al: Reduction in tumor burden allowing partial nephrectomy following preoperative chemotherapy in biopsy proved Wilms tumor. J Urol, part 2, 146: 509, 1991 2. Wilimas, J. A., Magill, L., Parham, D. M., Kumar, M. and Douglass, E. C.: The potential for renal salvage in nonmetastatic unilateral Wilms’ tumor. Am J Pediatr Hematol Oncol, 13: 342, 1991 3. Detoledo, J., Rodriguez, G. C., Gallego, M. S., Moraga, F., Catello, F. and Moreno, A.: Renal function in long-term survivors of Wilms tumor. Med Pediatr Oncol, 23: 265, 1994 4. Barrera, M., Roy, L. P. and Stevens, M.: Long-term follow-up after unilateral nephrectomy and radiotherapy for Wilms’ tumour. Pediatr Nephrol, 3: 430, 1989 5. Breslow, N. E., Takashima, J. R., Ritchey, M. L., Strong, L. C. and Green, D. M.: Renal failure in the Denys-Drash and Wilms’ tumor-aniridia syndromes. Cancer Res, 60: 4030, 2000 6. Coppes, M. J., Arnold, M., Beckwith, J. B., Ritchey, M. L., D’Angio, G. J., Green, D. M. et al: Factors affecting the risk of contralateral Wilms tumor development: a report from the National Wilms Tumor Study Group. Cancer, 85: 1616, 1999 7. Choyke, P. L., Siegel, M. J., Craft, A. W., Green, D. M. and DeBaun, M. R.: Screening for Wilms tumor in children with Beckwith-Wiedemann syndrome or idiopathic hemihypertrophy. Med Pediatr Oncol, 32: 196, 1999 8. Blakely, M. L., Shamberger, R. C., Norkool, P., Beckwith, J. B., Green, D. M. and Ritchey, M. L.: Outcome of children with cystic partially differentiated nephroblastoma treated with or without chemotherapy. J Pediatr Surg, 38: 897, 2003 9. Weirich, A., Leuschner, I., Harms, D., Vujanic, G. M., Troger, J., Abel, U. et al: Clinical impact of histologic subtypes in localized non-anaplastic nephroblastoma treated according to the trial and study SIOP-9/GPOH. Ann Oncol, 12: 311, 2001 REPLY BY AUTHORS When performing radical primary partial nephrectomy, lymph node sampling and not only of lymph nodes removal was routinely performed. In the SIOP 2001 protocol it is stated that sampling and histological examination of lymph nodes are imperative for accurate staging and subsequent treatment. Hilar and para-aortic lymph nodes at the origin of the renal artery (regional nodes) and nodes
944
PARTIAL NEPHRECTOMY FOR UNILATERAL WILMS TUMOR
below or above this level (extra regional nodes) should be sampled even if not suspicious. Involved or suspicious lymph nodes must be excised without rupture. They must be carefully labeled and sent to the pathologist separately with an accurate description of their position and characteristics. This information affects staging, treatment and outcome. Radical lymph node dissection does not enhance survival and, therefore, is not part of the surgical therapy. In children with syndromes associated with WT, such as the Denis
Drash syndrome or Beckwith-Wiedemann syndrome, early detection of WT may allow nephron sparing surgery. Children with bilateral WT as well as 2% to 3% risk of metachronous WT in the contralateral kidney and synchronous bilateral tumor disease require nephron sparing surgical techniques such as partial nephrectomy. PN should only be performed for small unilateral WT with low or intermediate risk histology. A good response to preoperative ChT is mandatory in these patients.