- Email: [email protected]
Does Preoperative Chemotherapy Ease the Surgical Procedure for Wilms Tumor?
Does Preoperative Chemotherapy Ease the Surgical Procedure for Wilms Tumor? Guy A. Bogaert,* Birgit Heremans, Marleen Renard, Liesje Bruninx, Liesbeth De Wever and Hendrik Van Poppel From the University Hospital Gasthuisberg, Gasthuisberg, Belgium
Purpose: A conclusion of Société Internationale d’Oncologie Pédiatrique in 2001 after protocol 2, 5, 6, 9 and 93-01 was that “preoperative chemotherapy would make nephrectomy easier and furthermore, metastases may disappear or become resectable and vascular extension may regress and partial nephrectomy may become possible.” We changed our strategy from preoperative chemotherapy in select cases only to standard 4 to 6-week preoperative chemotherapy in all. However, it appears dubious whether surgical ease or perioperative complications consistently improved after preoperative chemotherapy. We investigated volume changes after chemotherapy and compared all possible variables at a single center using the same surgical team. Materials and Methods: A total of 32 children with Wilms tumor between 1998 and 2007 were included in a retrospective analysis, including 15 without preoperative chemotherapy according to the National Wilms’ Tumor Study Group protocol before 2001 and 17 with preoperative chemotherapy according to the Société Internationale d’Oncologie Pédiatrique 2001 protocol. We studied the change in tumor volume using picture archive and communication systems volume measurement after chemotherapy, surgical procedure parameters (child body mass index, length and type of incision, blood loss, operative time, hospital stay, pain medication and complications) and outcome (histology and staging). Statistical analysis was performed using the t, Fisher and chi-square tests. Results: The 2 groups were comparable in gender, age (mean 5.4 and 3.9 years, respectively) and staging at diagnosis. At diagnosis mean ⫾ SD Wilms tumor volume was 408.93 ⫾ 387.39 and 454.58 ⫾ 236.71 cm3 in the nonchemotherapy and chemotherapy groups, respectively. In the chemotherapy group 12 of 17 tumors decreased a mean of 60.9% ⫾ 24.6% and 5 of 17 increased 42.8% ⫾ 31.3%. No analyzed parameters were different in the 2 groups. Conclusions: Our single center analysis with the same surgical team before and after the preoperative chemotherapy era show that it remains unpredictable whether surgery in a child with Wilms tumor is safer and easier after preoperative chemotherapy.
Abbreviations and Acronyms BMI ⫽ body mass index CT ⫽ computerized tomography NWTSG ⫽ National Wilms’ Tumor Study Group SIOP ⫽ Société Internationale d’Oncologie Pédiatrique * Correspondence: (telephone: ⫹3216346930; FAX: ⫹3216346931; e-mail: guy.bogaert@ uzleuven.be).
Key Words: kidney, nephrectomy, drug therapy, Wilms tumor, preoperative care THE first Wilms tumor was reported in 1814 by Rance1 and the first successful nephrectomy in a 2-year-old child was performed in 1877 by Jessop in Leeds, United Kingdom.2,3 However, in 1899
Wilms was the first to describe treatment in 7 children with nephroblastoma as a mixture of 3 tissues in a monograph on mixed tumors.4 Initially this malignancy was lethal until the
0022-5347/09/1824-1869/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 1869-1875, October 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.03.022
www.jurology.com
1869
1870
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
discovery of tumor radiosensitivity and the introduction of active chemotherapy.2 The largest treatment protocols were described and coordinated by NWTSG in the United States and by the SIOP in Europe. Although these treatment protocols have different approaches, and distinct advantages and disadvantages, each achieves an overall 90% survival rate. Mainly the SIOP approach is to administer 4 to 6 weeks of chemotherapy before surgery (6 weeks in stage IV cases), while NWTSG recommends primary surgery and chemotherapy only in select cases. The strength of the NWTSG protocol is an accurate assessment of histology, genetics and tumor extent. The aimed strength of the SIOP protocol is that preoperative chemotherapy would down stage the tumor, decreasing total chemotherapy and radiotherapy, and resulting in fewer long-term complications.5 As a consequence of down staging, it would be logical that surgery would be safer and easier.6 However, according to SIOP 93-01 results it became clear that approximately 5% of lesions were misdiagnosed and, therefore, not Wilms tumors, while 1.8% were benign.7 Some groups reported advantages for neoadjuvant chemotherapy with less intensive therapy in 20% of patients5,8 but others noted an increased number of complications in the preoperative chemotherapy group, although there was selection bias due to larger tumor dimensions.9 Before 2001 at our institution we followed the NWTSG protocol. However, we changed this policy in 2001 according the SIOP 2001 protocol stating, “Pre-operative chemotherapy as demonstrated by SIOP 2, 5, 6, 9 and 93-01 makes nephrectomy easier and less hazardous. Furthermore, metastases may disappear or become resectable, vascular extension may regress and partial nephrectomy may become possible.” Nevertheless, it was our subjective impression that surgery did not become easier, nor were we able to perform partial nephrectomy. Sometimes we had the impression that peritumor fibrosis rendered surgery more difficult. To assess the surgical ease or possibility of partial nephrectomy we compared tumor volume before chemotherapy and before surgery, perioperative morbidity and surgical parameters under 1 constant variable, that is the same surgical team.
MATERIALS AND METHODS We retrospectively reviewed the records of all children operated on by the same surgical team (GB and HVP) for Wilms tumor between 1998 and 2007. There were 32 children in the study, including 15 treated without preoperative chemotherapy from 1998 to 2001 and 17 treated with preoperative chemotherapy from 2001 to 2007. Study inclusion criteria were the diagnosis of Wilms tumor at age less than 15 years and available CT images on film or on a picture archiving and communication sys-
tem. Study exclusion criteria were Wilms tumor and syndromes during that period, such as Denys-Drash syndrome in 2 patients, Perlman syndrome in 1 and Beckwith-Wiedemann syndrome in 2 as well as bilateral Wilms tumor in 2. Patient variables collected were gender, age at diagnosis, location, BMI at age less than 2 years for percentile comparison curves and age greater than 2 years for age related comparison curves,10 staging according to the SIOP 2001 protocol, recurrence and histopathological risk category, including low—mesoblastic nephroma, cystic and necrotic, intermediate— epithelial, stromal, regressive and focal anaplasia, and high risk— blastema, diffuse anaplasia, clear cell sarcoma and rhabdoid tumor. To determine surgical parameters and feasibility we separately assessed surgery and perioperative morbidity. To evaluate surgery we compared the type of incision, including midline laparotomy, bilateral subcostal inverted V incision (chevron) or unilateral subcostal inverted V incision (hemichevron), blood loss, operative time, the positive surgical margin rate and tumor volume. Morbidity was assessed by length of stay, the total duration of pain medication (tramadol infusion), oral intake time and other complications. Tumor volume was measured preoperatively in the nonchemotherapy group, and before and after chemotherapy in the other group to evaluate tumor volume changes after chemotherapy. Tumor volume can indeed be used to assess surgical difficulty since this is also done to assess the difficulty of liver resection.11 To accurately measure tumor volume CT images were uploaded in MeVisLab® software, enabling us to delineate the exact tumor contours on each slice. The program calculated the surface area on each slice and, thus, demarcated tumor volume because slice thickness was known (figs. 1 and 2). When classic CT images were used, tumor volume was assessed by measuring the 3 largest tumor diameters and calculating volume accordingly. Statistical analysis was performed using the t test to compare mean values the 2 groups. Categorical variables were placed in contingency tables for the Fisher and chisquare tests.
RESULTS Table 1 lists patient variables in the 2 groups. Between the 2 groups with vs without preoperative chemotherapy there was no difference in gender (p ⫽ 015), age at diagnosis (p ⫽ 0.26), location (p ⫽ 012) and BMI for age less than 2 years in percentiles and for age greater than 2 years in age related comparison curves (p ⫽ 0.12). There was also no difference in staging groups according to the SIOP 2001 protocol (p ⫽ 0.28). There was no difference in postoperative recurrence (p ⫽ 0.58) or in the histopathological risk profile (p ⫽ 0.15). For surgical ease there was also no difference in blood loss (p ⫽ 0.25), operative time ( p ⫽ 0.11), incision type ( p ⫽ 0.36) or the incidence of positive surgical margins (p ⫽ 0.45, table 2). Overall no difference was noted in morbidity perioperatively in tramadol infusion duration (p ⫽ 0.42),
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
1871
Figure 1. Axial views show exact tumor contours on each slice to accurately assess tumor volume after uploading CT images in software. A, before chemotherapy. B, after chemotherapy.
time to oral intake (p ⫽ 0.52) or the number and severity of complications. Preoperatively complications were noted in no patients without chemotherapy, and in 17 with chemotherapy, including Epstein-Barr virus, pyelonephritis, stomatitis and catheter sepsis (5), hypertension (3), eye ptosis (1) and hydronephrosis (1). Surgical complications were noted postoperatively in 1 patient with vena caval rupture and in 1 with sympathectomy, and postoperatively in 6 with tumor adhesions with surrounding tissue. There was no tumor spillage in the preoperative chemotherapy or immediate surgery group. Median tumor volume was 408.93 ⫾ 387.39 ml at surgery in the group without preoperative chemotherapy. Histopathology in the 5 nonresponders to chemotherapy was intermediate risk in 2, and regressive, anaplastic and rhabdomyoblastic type in 1 each (table 3). In the preoperative chemotherapy group median tumor volume before vs after chemotherapy was 454.58 ⫾ 236.58 vs 349.65 ⫾ 333.64 ml. The change
Figure 2. Pretreatment coronal views reveal exact tumor contours on each slice to accurately assess tumor volume after uploading CT images in software.
in tumor volume after chemotherapy differed, that is volume decreased by 60.9% ⫾ 24.6% in 12 of 17 lesions and increased by 42.8% ⫾ 31.3% in 5 of 17 despite chemotherapy.
DISCUSSION Under identical surgical team conditions it remained unpredictable whether surgery for Wilms tumors after preoperative chemotherapy was easier or less complicated, or allowed partial nephrectomy, as stated by the SIOP in 2001. The NWTSG and SIOP approaches to Wilms tumor treatment have distinct advantages and disadvantages. The primary strength of the NWTSG ap-
Table 1 Preop Chemotherapy Parameter
Yes
No
No. pts No. boys/girls Mean age (range) No. lt/rt side No. upper pole/lower pole/entire No. BMI percentile: Less than 3 3–97 Greater than 97 No. stage (%): I II III IV V No. histopathological risk (%): Low Intermediate High No. recurrence (%)
15 9/6 3.2 Yrs (1 mo–4 yrs) 12/3 4/6/5
17 8/9 3.9 Yrs (8 mos–15 yrs) 9/9 9/2/7
0 14 1 6 4 2 3 0
0 14 1 2
3 12 0 (40) (26.67) (13.33) (20)
(93.33) (6.67) (13.33)
p Value
0.1253
10 5 0 1 1
(58.82) (29.41)
2 12 4 1
(11.11) (66.67) (22.22) (5.88)
0.2867
(5.88) (5.88)
0.1563
0.5887
1872
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
Table 2 Preop Chemotherapy Parameter
No
Yes
p Value
Mean ⫾ SD blood loss (ml) 193.33 ⫾ 58.26 86.53 ⫾ 47.69 Mean ⫾ SD operative time (mins) 126.3 ⫾ 26.5 152.94 ⫾ 36.5 No. incision laparatomy (%): 1 (6.67) 0 Hemichevron 14 (93.33) 16 (94.12) Chevron 0 1 (5.88) No. section margins (%): Neg 14 (93.33) 18 (100) Pos 1 (6.67) 0 Mean surgery hospital stay (days) 5.6 6.8 Mean tramadol duration (days) 3.1 3.4 Mean time to oral intake (days) 2.5 2.8 No. complications (%): Preop 9 (52.94) Periop 2 (13.3) 6 (35.29) Postop 7 (46.67) 8 (47.06)
0.2522 0.1089 0.3649
0.4545 0.5358 0.4269 0.5247
0.2293 1.000
proach is that preliminary resection allows an accurate assessment of histological diagnosis and tumor extent, in other words correct diagnosis and staging. The primary strength of the SIOP protocol is that preoperative chemotherapy down stages the lesion and can reduce tumor volume,6 decreasing the likelihood of spillage. Also, the tumor response to chemotherapy could provide prognostic information.12 The 2 study groups were comparable in the number of patients, age, gender distribution, and tumor site or stage (table 1). We measured BMI immediately before surgery to avoid the fact that patients would be in a less healthy condition due to chemotherapy before surgery, which could influence surgical morbidity. However, for BMI percentiles there was no difference in the 2 groups. Also, on histological risk
group analysis there was no difference between the groups. The absence of differences between the groups could be due to small study numbers. All patients were confirmed to have Wilms tumor, in accordance with the preoperative or prechemotherapy clinical diagnosis. However, in SIOP 93-01 approximately 5% of the lesions treated with chemotherapy ultimately proved not to be Wilms tumor, including 1.8% benign conditions.7 This is a major concern and a weakness of the SIOP protocol but this problem did not occur in our series. We observed more stage I and II but no stage III cases in the preoperative chemotherapy group. However, the difference vs the no preoperative chemotherapy group was not statistically significant and the trend toward neoplasm down staging is obvious. In addition, mean intraoperative blood loss was higher in the group without preoperative chemotherapy but again the difference was not statistically significant. In the preoperative chemotherapy group the 5 nonresponding tumors that remained unchanged or increased in size during chemotherapy were not all rhabdomyoblastic Wilms tumors as expected (table 3). Two lesions were intermediate risk, and 1 each was regressive, anaplastic and rhabdomyoblastic type. There is no clear explanation or hypothesis why we noted rather heterogeneous histopathology in the group of chemotherapy nonresponders. We never observed intraoperative tumor spillage and, therefore, we could not confirm the relation with tumor size or chemotherapy. However, avoiding tumor spillage is considered a theoretical advantage of preoperative chemotherapy. In the NWTS-4 trial the rate
Table 3 Pt No.—Gender
Date Birth
Surgery
Side
ml Vol Change (%)
1—F 2—F 3—M 4—M 5—M 6—F 7—M 8—F 9—F
8/12/06 11/04/03 6/15/04 11/18/02 1/20/98 10/18/85 3/7/01 8/22/99 4/02/99
10/07 01/06 02/05 06/04 04/03 03/07 06/05 05/03 12/03
Lt Rt Lt Rt Lt Lt Rt Rt Lt
⫺61 (⫺32) ⫺177 (⫺40) 362 (75) ⫺247 (⫺50) 330 (51) ⫺316 (⫺79) ⫺754 (⫺75) ⫺319 (⫺91) ⫺315 (⫺45)
Mixed Regressive Stromal (intermediate) Mixed (intermediate) Anaplastic Regressive (intermediate) Triphasic ⫹ rhabdomyogenesis
10—F 11—F 12—M 13—F 14—M 15—F 16—M 17—M
8/10/99 7/29/04 1/15/01 8/23/99 11/05/04 1/28/00 12/3/02 6/23/99
05/03 04/07 07/07 07/04 05/07 10/05 11/03 02/05
Rt Lt Rt Lt Lt Lt Lt Rt
⫺135 (⫺31) ⫺175 (⫺70) 129 (21) ⫺310 (⫺91) ⫺108 (⫺32) 0 405 (65) ⫺93 (91)
* Four and 6 weeks of chemotherapy for SIOP 1 and 2, respectively.
Histopathology (risk)
Operative Time (mins)
% Necrosis
Mixed (intermediate)
1 1 1 2
100 120 140 120 120 120 120 140 180
Less than 1 66–99 0 20 75 80 0 100 0
Cystic Mixed (low) Stromal (intermediate) Mixed, triphasic Diffuse anaplastic (high) Blastemal ⫹ focal anaplastic (high) Stromal (intermediate) Mixed (intermediate)
1 2 1 2 2 1 1 1
240 100 120 180 110 180 180 140
25 50 5 50 80 60 66 5
SIOP* 1 1 1 1
Metastasis
Lung Lung Lung Liver Dead of disease 2005
Lung Lung
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
of spillage of favorable histology samples was 20% compared to 6% in the SIOP-5 trial.13 Also, in SIOP results despite induction chemotherapy spillage is a problem because surgery was performed by a heterogeneous group of experts, including general surgeons, urologists, pediatric urologists, pediatric surgeons and oncological surgeons. After tumor spillage the outcome can be improved by intensifying chemotherapy and local or abdominal radiotherapy, which increase toxicity. For similar results the patient must undergo more treatment with more long-term consequences. According to surgical complications in the fourth NWTS-4 study when nephrectomy was performed by a general surgeon, the risk of complications was higher (OR 9.0, 95% CI 1.3– 65, p ⫽ 0.03) than when performed by a pediatric surgeon (referent, OR 1.0) or a pediatric urologist (OR 0.7, 95% CI 0.3–1.8).13 It is obvious that surgeon expertise is the determining prognostic factor for a successful intraoperative outcome. This was also the opinion of Ritchey et al, who concluded after a report from the National Wilms Tumor Study Group that surgeon expertise and tumor dimensions are the most important factors in avoiding surgical complications.13 They also stated that the decision to prescribe or not prescribe preoperative chemotherapy for unilateral nonmetastatic, nonsyndromic Wilms tumor depends on the group research interest, including accurate staging, pathological findings and genetics for those favoring nonpreoperative chemotherapy, and down staging, long-term toxicity and chemotherapy effects for those favoring preoperative chemotherapy. Although we are convinced that in children with preoperative chemotherapy there was more fibrosis or tumor adhesions to surrounding tissues, and although we had the impression that surgery required more time, we could not substantiate any difference in overall mean operative time or blood loss. Quantifying adherences is a subjective issue that cannot be measured. It might also be seen in primary tumors and, therefore, cannot be used for comparison. Mean tumor volume tended to decrease from 454.58 ⫾ 236.58 to 349.65 ⫾ 333.64 ml but this difference was not statistically different. This may be because there was no general trend in volume change, although 2 of 3 tumors decreased by more than 60% ⫾ 24.6%. In 5 of 17 patients (30%) tumor volume even increased by 42.8% ⫾ 31.3%. This may have led to the fact that operative time did not improve in the group with preoperative chemotherapy. This rate is high com-
1873
pared to the observations of Øra et al from the SIOP 93-01 group, who noted that only 5% of tumors showed a volume increase, although they were significantly smaller at diagnosis, and there were more stage III tumors and tumors associated with high risk histopathology.14 In our patients we could not reproduce these observations (table 3). Remarkably although in 12 of 17 patients (60%) the lesion was significantly reduced with results identical to those of Safdar et al,15 tumor size in our study remained so large that we could not safely perform partial nephrectomy. Partial nephrectomy is certainly beneficial in patients with bilateral disease. In patients with unilateral disease partial nephrectomy is not warranted due to the increased risk of local recurrence, which might adversely impact overall survival. Patient morbidity or recovery did not improve with preoperative chemotherapy. There was no difference in the duration of the need for tramadol infusion and no shortening of time to oral intake. Also, overall complications did not differ, although we had the subjective impression that there was a difference (table 2). The lack of a difference may have been due to the limited number of study patients. In 1984 McLorie et al already initiated a protocol at The Hospital for Sick Children in Toronto to stage children radiologically, followed by percutaneous tumor biopsy.16 The United Kingdom Children Cancer Study Group also examined the accuracy of percutaneous needle biopsy for initial diagnosis.17 This attitude showed no success in the larger NWTSG and SIOP study groups, although needle biopsy has minimal complications and carries no increased risk of local recurrence or up staging. With pretherapy biopsy the weakness of the standard SIOP preoperative chemotherapy protocol could be avoided. Compared to the NWTSG protocol the advantage of accurate pathological findings is ensured and the response to chemotherapy may be evaluated.
CONCLUSIONS Due to preoperative chemotherapy there is a trend toward down staging and decreased blood loss in patients with Wilms tumor. However, it remains doubtful whether surgery in a child indeed becomes safer and easier because of chemotherapy.
REFERENCES 1. Rance TF: Case of fungus haematodes of the kidnies. Med Phys J 1814; 32: 19.
3. Willetts IE: Jessop and the Wilms’ tumor. J Pediatr Surg 2003; 38: 1496.
2. Metzger ML and Dome JS: Current therapy for Wilms’ tumor. Oncologist 2005; 10: 815.
4. Wilms M: Die Mischgeschwülste der Niere. Leipzig: Verlag von Arthur Georgi 1899; pp 1–90.
5. Mitchell C, Pritchard-Jones K, Shannon R et al: Immediate nephrectomy versus preoperative chemotherapy in the management of non-metastatic Wilms’ tumour: results of a randomised trial
1874
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
(UKW3) by the UK Children’s Cancer Study Group. Eur J Cancer 2006; 42: 2554. 6. Beckwith JB: Wilms’ tumor and other renal tumors of childhood: a selective review from the National Wilms’ Tumor Study Pathology Center. Hum Pathol 1983; 14: 481. 7. de Kraker J, Graf N, van Tinteren H et al: Reduction of postoperative chemotherapy in children with stage I intermediate-risk and anaplastic Wilms’ tumour (SIOP 93-01 trial): a randomised controlled trial. Lancet 2004; 364: 1229. 8. Stehr M, Deilmann K, Haas RJ et al: Surgical complications in the treatment of Wilms’ tumor. Eur J Pediatr Surg 2005; 15: 414. 9. Zugor V, Krot D and Schott GE: Risk factors for intra- and postoperative complications in Wilms’ tumor surgery. Urologe A 2007; 46: 274.
10. Roelants M and Hauspie R: Growth Charts Flanders 2004. Brussels: Laboratory for Anthropogenesis, Vreije Universiteit Brussels 2004. 11. Beller S, Eulenstein S, Lange T et al: A new measure to assess the difficulty of liver resection. Eur J Surg Oncol 2008; 35: 59. 12. Boccon-Gibod L, Rey A, Sandstedt B et al: Complete necrosis induced by preoperative chemotherapy in Wilms tumor as an indicator of low risk: report of the International Society of Paediatric Oncology (SIOP) Nephroblastoma Trial and Study 9. Med Pediatr Oncol 2000; 34: 183. 13. Ritchey ML, Shamberger RC, Haase G et al: Surgical complications after primary nephrectomy for Wilms’ tumor: report from the National Wilms’ Tumor Study Group. J Am Coll Surg 2001; 192: 63.
14. Øra I, van Tinteren H, Bergeron C et al: Progression of localised Wilms’ tumour during preoperative chemotherapy is an independent prognostic factor: a report from the SIOP 93-01 nephroblastoma trial and study. Eur J Cancer 2007; 43: 131. 15. Safdar CA, Aslam M, Awan SH et al: Wilms’ tumour: a comparison of surgical aspects in patients with or without pre-operative chemotherapy. J Coll Physicians Surg Pak 2006; 16: 521. 16. McLorie GA, Khoury AE, Weitzman SS et al: Preoperative chemotherapy in management of Wilms’ tumor. Urology 1996; 47: 792. 17. Vujanic GM, Kelsey A, Mitchell C et al: The role of biopsy in the diagnosis of renal tumors of childhood: results of the UKCCSG Wilms tumor study 3. Med Pediatr Oncol 2003; 40: 18.
EDITORIAL COMMENTS These authors report a single institution experience with children with Wilms tumor. They compared their initial experience using primary surgery to their more recent approach of preoperative chemotherapy before surgery. What is notable is that their subjective impression was that surgery is not easier after preoperative chemotherapy. Intuitively one would expect that resection would be simpler after preoperative chemotherapy. A fairly high percent of this small patient series did not show decreased tumor volume after treatment. This may have influenced the results. In the United States most experience with preoperative chemotherapy is in children with massive tumors that are inoperable or bilateral tumors. In these circumstances preoperative treatment certainly facilitates the ability to remove a large tumor that was initially unresectable or perform renal sparing surgery for bilateral tumors. I have noted some difficult resections after chemotherapy but generally there had been prior exploration with open biopsy. Although the authors state that there was no difference in surgical complications, their data may not support that conclusion. They report tumor adhesion as a complication in children with preoperative chemotherapy. This may be a consequence of preoperative chemotherapy but is it a complication of surgery? The cohort that underwent primary surgery had some significant complications, including inferior vena caval rupture and sympathectomy. Published data suggest that there is a decrease in
surgical complications with the preoperative chemotherapy approach. It is clear from study comparisons that there was no difference in operative time, length of stay or recovery from surgery. If chemotherapy had facilitated surgery, one would expect to see a difference. Since the latter group of patients was treated more recently, one would expect a better outcome in this cohort since the authors were more experienced with managing renal tumors. They had excellent results in each group with no tumor spillage and overall low blood loss. The final chapter remains to be written on which approach (primary surgery vs preoperative chemotherapy) provides the greatest advantage in children with Wilms tumor. There does not appear to be any difference in terms of survival between the approaches. It boils down to which approach decreases the morbidity of therapy. This is usually viewed in terms of the risk of adjuvant therapy, eg trying to avoid radiation therapy and its long-term complications. However, surgeons must keep in mind that surgical morbidity can be an issue and these authors put the differences between these approaches in perspective.
Previous large SIOP studies show that preoperative chemotherapy down stages Wilms tumor and significantly decreases the rate of intraoperative tumor rupture from 27% to less than 5%.1 It was also reported
that up to 30% of Wilms tumors become amenable to partial nephrectomy after chemotherapy, notwithstanding subjective impression of these authors to the contrary. The local recurrence rate after nephron spar-
Michael Ritchey Department of Urology Mayo Clinic Arizona Scottsdale, Arizona
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
ing surgery is comparable to that of renal cell carcinoma in adults, in whom partial nephrectomy is increasingly performed. Although they are undefined, the long-term risks of renal insufficiency and secondary renal malignancy in patients with Wilms tumor who undergo total nephrectomy might justify the risk of local recurrence. Preoperative chemotherapy clearly offers some surgical advantages. Unfortunately while the authors present a large single center experience,
1875
this study is inadequately powered to illustrate these advantages or disprove them. Whether the surgical benefits outweigh the potential risks of neoadjuvant chemotherapy remains an open question. Jonathan H. Ross Department of Pediatric Urology Rainbow Babies and Children’s Hospital Cleveland, Ohio
REFERENCE 1. Graf N, Tournade MF and de Kraker J: The role of preoperative chemotherapy in the management of Wilms’ tumor. Urol Clin North Am 2000; 27: 443.
Purpose: A conclusion of Société Internationale d’Oncologie Pédiatrique in 2001 after protocol 2, 5, 6, 9 and 93-01 was that “preoperative chemotherapy would make nephrectomy easier and furthermore, metastases may disappear or become resectable and vascular extension may regress and partial nephrectomy may become possible.” We changed our strategy from preoperative chemotherapy in select cases only to standard 4 to 6-week preoperative chemotherapy in all. However, it appears dubious whether surgical ease or perioperative complications consistently improved after preoperative chemotherapy. We investigated volume changes after chemotherapy and compared all possible variables at a single center using the same surgical team. Materials and Methods: A total of 32 children with Wilms tumor between 1998 and 2007 were included in a retrospective analysis, including 15 without preoperative chemotherapy according to the National Wilms’ Tumor Study Group protocol before 2001 and 17 with preoperative chemotherapy according to the Société Internationale d’Oncologie Pédiatrique 2001 protocol. We studied the change in tumor volume using picture archive and communication systems volume measurement after chemotherapy, surgical procedure parameters (child body mass index, length and type of incision, blood loss, operative time, hospital stay, pain medication and complications) and outcome (histology and staging). Statistical analysis was performed using the t, Fisher and chi-square tests. Results: The 2 groups were comparable in gender, age (mean 5.4 and 3.9 years, respectively) and staging at diagnosis. At diagnosis mean ⫾ SD Wilms tumor volume was 408.93 ⫾ 387.39 and 454.58 ⫾ 236.71 cm3 in the nonchemotherapy and chemotherapy groups, respectively. In the chemotherapy group 12 of 17 tumors decreased a mean of 60.9% ⫾ 24.6% and 5 of 17 increased 42.8% ⫾ 31.3%. No analyzed parameters were different in the 2 groups. Conclusions: Our single center analysis with the same surgical team before and after the preoperative chemotherapy era show that it remains unpredictable whether surgery in a child with Wilms tumor is safer and easier after preoperative chemotherapy.
Abbreviations and Acronyms BMI ⫽ body mass index CT ⫽ computerized tomography NWTSG ⫽ National Wilms’ Tumor Study Group SIOP ⫽ Société Internationale d’Oncologie Pédiatrique * Correspondence: (telephone: ⫹3216346930; FAX: ⫹3216346931; e-mail: guy.bogaert@ uzleuven.be).
Key Words: kidney, nephrectomy, drug therapy, Wilms tumor, preoperative care THE first Wilms tumor was reported in 1814 by Rance1 and the first successful nephrectomy in a 2-year-old child was performed in 1877 by Jessop in Leeds, United Kingdom.2,3 However, in 1899
Wilms was the first to describe treatment in 7 children with nephroblastoma as a mixture of 3 tissues in a monograph on mixed tumors.4 Initially this malignancy was lethal until the
0022-5347/09/1824-1869/0 THE JOURNAL OF UROLOGY® Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 182, 1869-1875, October 2009 Printed in U.S.A. DOI:10.1016/j.juro.2009.03.022
www.jurology.com
1869
1870
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
discovery of tumor radiosensitivity and the introduction of active chemotherapy.2 The largest treatment protocols were described and coordinated by NWTSG in the United States and by the SIOP in Europe. Although these treatment protocols have different approaches, and distinct advantages and disadvantages, each achieves an overall 90% survival rate. Mainly the SIOP approach is to administer 4 to 6 weeks of chemotherapy before surgery (6 weeks in stage IV cases), while NWTSG recommends primary surgery and chemotherapy only in select cases. The strength of the NWTSG protocol is an accurate assessment of histology, genetics and tumor extent. The aimed strength of the SIOP protocol is that preoperative chemotherapy would down stage the tumor, decreasing total chemotherapy and radiotherapy, and resulting in fewer long-term complications.5 As a consequence of down staging, it would be logical that surgery would be safer and easier.6 However, according to SIOP 93-01 results it became clear that approximately 5% of lesions were misdiagnosed and, therefore, not Wilms tumors, while 1.8% were benign.7 Some groups reported advantages for neoadjuvant chemotherapy with less intensive therapy in 20% of patients5,8 but others noted an increased number of complications in the preoperative chemotherapy group, although there was selection bias due to larger tumor dimensions.9 Before 2001 at our institution we followed the NWTSG protocol. However, we changed this policy in 2001 according the SIOP 2001 protocol stating, “Pre-operative chemotherapy as demonstrated by SIOP 2, 5, 6, 9 and 93-01 makes nephrectomy easier and less hazardous. Furthermore, metastases may disappear or become resectable, vascular extension may regress and partial nephrectomy may become possible.” Nevertheless, it was our subjective impression that surgery did not become easier, nor were we able to perform partial nephrectomy. Sometimes we had the impression that peritumor fibrosis rendered surgery more difficult. To assess the surgical ease or possibility of partial nephrectomy we compared tumor volume before chemotherapy and before surgery, perioperative morbidity and surgical parameters under 1 constant variable, that is the same surgical team.
MATERIALS AND METHODS We retrospectively reviewed the records of all children operated on by the same surgical team (GB and HVP) for Wilms tumor between 1998 and 2007. There were 32 children in the study, including 15 treated without preoperative chemotherapy from 1998 to 2001 and 17 treated with preoperative chemotherapy from 2001 to 2007. Study inclusion criteria were the diagnosis of Wilms tumor at age less than 15 years and available CT images on film or on a picture archiving and communication sys-
tem. Study exclusion criteria were Wilms tumor and syndromes during that period, such as Denys-Drash syndrome in 2 patients, Perlman syndrome in 1 and Beckwith-Wiedemann syndrome in 2 as well as bilateral Wilms tumor in 2. Patient variables collected were gender, age at diagnosis, location, BMI at age less than 2 years for percentile comparison curves and age greater than 2 years for age related comparison curves,10 staging according to the SIOP 2001 protocol, recurrence and histopathological risk category, including low—mesoblastic nephroma, cystic and necrotic, intermediate— epithelial, stromal, regressive and focal anaplasia, and high risk— blastema, diffuse anaplasia, clear cell sarcoma and rhabdoid tumor. To determine surgical parameters and feasibility we separately assessed surgery and perioperative morbidity. To evaluate surgery we compared the type of incision, including midline laparotomy, bilateral subcostal inverted V incision (chevron) or unilateral subcostal inverted V incision (hemichevron), blood loss, operative time, the positive surgical margin rate and tumor volume. Morbidity was assessed by length of stay, the total duration of pain medication (tramadol infusion), oral intake time and other complications. Tumor volume was measured preoperatively in the nonchemotherapy group, and before and after chemotherapy in the other group to evaluate tumor volume changes after chemotherapy. Tumor volume can indeed be used to assess surgical difficulty since this is also done to assess the difficulty of liver resection.11 To accurately measure tumor volume CT images were uploaded in MeVisLab® software, enabling us to delineate the exact tumor contours on each slice. The program calculated the surface area on each slice and, thus, demarcated tumor volume because slice thickness was known (figs. 1 and 2). When classic CT images were used, tumor volume was assessed by measuring the 3 largest tumor diameters and calculating volume accordingly. Statistical analysis was performed using the t test to compare mean values the 2 groups. Categorical variables were placed in contingency tables for the Fisher and chisquare tests.
RESULTS Table 1 lists patient variables in the 2 groups. Between the 2 groups with vs without preoperative chemotherapy there was no difference in gender (p ⫽ 015), age at diagnosis (p ⫽ 0.26), location (p ⫽ 012) and BMI for age less than 2 years in percentiles and for age greater than 2 years in age related comparison curves (p ⫽ 0.12). There was also no difference in staging groups according to the SIOP 2001 protocol (p ⫽ 0.28). There was no difference in postoperative recurrence (p ⫽ 0.58) or in the histopathological risk profile (p ⫽ 0.15). For surgical ease there was also no difference in blood loss (p ⫽ 0.25), operative time ( p ⫽ 0.11), incision type ( p ⫽ 0.36) or the incidence of positive surgical margins (p ⫽ 0.45, table 2). Overall no difference was noted in morbidity perioperatively in tramadol infusion duration (p ⫽ 0.42),
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
1871
Figure 1. Axial views show exact tumor contours on each slice to accurately assess tumor volume after uploading CT images in software. A, before chemotherapy. B, after chemotherapy.
time to oral intake (p ⫽ 0.52) or the number and severity of complications. Preoperatively complications were noted in no patients without chemotherapy, and in 17 with chemotherapy, including Epstein-Barr virus, pyelonephritis, stomatitis and catheter sepsis (5), hypertension (3), eye ptosis (1) and hydronephrosis (1). Surgical complications were noted postoperatively in 1 patient with vena caval rupture and in 1 with sympathectomy, and postoperatively in 6 with tumor adhesions with surrounding tissue. There was no tumor spillage in the preoperative chemotherapy or immediate surgery group. Median tumor volume was 408.93 ⫾ 387.39 ml at surgery in the group without preoperative chemotherapy. Histopathology in the 5 nonresponders to chemotherapy was intermediate risk in 2, and regressive, anaplastic and rhabdomyoblastic type in 1 each (table 3). In the preoperative chemotherapy group median tumor volume before vs after chemotherapy was 454.58 ⫾ 236.58 vs 349.65 ⫾ 333.64 ml. The change
Figure 2. Pretreatment coronal views reveal exact tumor contours on each slice to accurately assess tumor volume after uploading CT images in software.
in tumor volume after chemotherapy differed, that is volume decreased by 60.9% ⫾ 24.6% in 12 of 17 lesions and increased by 42.8% ⫾ 31.3% in 5 of 17 despite chemotherapy.
DISCUSSION Under identical surgical team conditions it remained unpredictable whether surgery for Wilms tumors after preoperative chemotherapy was easier or less complicated, or allowed partial nephrectomy, as stated by the SIOP in 2001. The NWTSG and SIOP approaches to Wilms tumor treatment have distinct advantages and disadvantages. The primary strength of the NWTSG ap-
Table 1 Preop Chemotherapy Parameter
Yes
No
No. pts No. boys/girls Mean age (range) No. lt/rt side No. upper pole/lower pole/entire No. BMI percentile: Less than 3 3–97 Greater than 97 No. stage (%): I II III IV V No. histopathological risk (%): Low Intermediate High No. recurrence (%)
15 9/6 3.2 Yrs (1 mo–4 yrs) 12/3 4/6/5
17 8/9 3.9 Yrs (8 mos–15 yrs) 9/9 9/2/7
0 14 1 6 4 2 3 0
0 14 1 2
3 12 0 (40) (26.67) (13.33) (20)
(93.33) (6.67) (13.33)
p Value
0.1253
10 5 0 1 1
(58.82) (29.41)
2 12 4 1
(11.11) (66.67) (22.22) (5.88)
0.2867
(5.88) (5.88)
0.1563
0.5887
1872
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
Table 2 Preop Chemotherapy Parameter
No
Yes
p Value
Mean ⫾ SD blood loss (ml) 193.33 ⫾ 58.26 86.53 ⫾ 47.69 Mean ⫾ SD operative time (mins) 126.3 ⫾ 26.5 152.94 ⫾ 36.5 No. incision laparatomy (%): 1 (6.67) 0 Hemichevron 14 (93.33) 16 (94.12) Chevron 0 1 (5.88) No. section margins (%): Neg 14 (93.33) 18 (100) Pos 1 (6.67) 0 Mean surgery hospital stay (days) 5.6 6.8 Mean tramadol duration (days) 3.1 3.4 Mean time to oral intake (days) 2.5 2.8 No. complications (%): Preop 9 (52.94) Periop 2 (13.3) 6 (35.29) Postop 7 (46.67) 8 (47.06)
0.2522 0.1089 0.3649
0.4545 0.5358 0.4269 0.5247
0.2293 1.000
proach is that preliminary resection allows an accurate assessment of histological diagnosis and tumor extent, in other words correct diagnosis and staging. The primary strength of the SIOP protocol is that preoperative chemotherapy down stages the lesion and can reduce tumor volume,6 decreasing the likelihood of spillage. Also, the tumor response to chemotherapy could provide prognostic information.12 The 2 study groups were comparable in the number of patients, age, gender distribution, and tumor site or stage (table 1). We measured BMI immediately before surgery to avoid the fact that patients would be in a less healthy condition due to chemotherapy before surgery, which could influence surgical morbidity. However, for BMI percentiles there was no difference in the 2 groups. Also, on histological risk
group analysis there was no difference between the groups. The absence of differences between the groups could be due to small study numbers. All patients were confirmed to have Wilms tumor, in accordance with the preoperative or prechemotherapy clinical diagnosis. However, in SIOP 93-01 approximately 5% of the lesions treated with chemotherapy ultimately proved not to be Wilms tumor, including 1.8% benign conditions.7 This is a major concern and a weakness of the SIOP protocol but this problem did not occur in our series. We observed more stage I and II but no stage III cases in the preoperative chemotherapy group. However, the difference vs the no preoperative chemotherapy group was not statistically significant and the trend toward neoplasm down staging is obvious. In addition, mean intraoperative blood loss was higher in the group without preoperative chemotherapy but again the difference was not statistically significant. In the preoperative chemotherapy group the 5 nonresponding tumors that remained unchanged or increased in size during chemotherapy were not all rhabdomyoblastic Wilms tumors as expected (table 3). Two lesions were intermediate risk, and 1 each was regressive, anaplastic and rhabdomyoblastic type. There is no clear explanation or hypothesis why we noted rather heterogeneous histopathology in the group of chemotherapy nonresponders. We never observed intraoperative tumor spillage and, therefore, we could not confirm the relation with tumor size or chemotherapy. However, avoiding tumor spillage is considered a theoretical advantage of preoperative chemotherapy. In the NWTS-4 trial the rate
Table 3 Pt No.—Gender
Date Birth
Surgery
Side
ml Vol Change (%)
1—F 2—F 3—M 4—M 5—M 6—F 7—M 8—F 9—F
8/12/06 11/04/03 6/15/04 11/18/02 1/20/98 10/18/85 3/7/01 8/22/99 4/02/99
10/07 01/06 02/05 06/04 04/03 03/07 06/05 05/03 12/03
Lt Rt Lt Rt Lt Lt Rt Rt Lt
⫺61 (⫺32) ⫺177 (⫺40) 362 (75) ⫺247 (⫺50) 330 (51) ⫺316 (⫺79) ⫺754 (⫺75) ⫺319 (⫺91) ⫺315 (⫺45)
Mixed Regressive Stromal (intermediate) Mixed (intermediate) Anaplastic Regressive (intermediate) Triphasic ⫹ rhabdomyogenesis
10—F 11—F 12—M 13—F 14—M 15—F 16—M 17—M
8/10/99 7/29/04 1/15/01 8/23/99 11/05/04 1/28/00 12/3/02 6/23/99
05/03 04/07 07/07 07/04 05/07 10/05 11/03 02/05
Rt Lt Rt Lt Lt Lt Lt Rt
⫺135 (⫺31) ⫺175 (⫺70) 129 (21) ⫺310 (⫺91) ⫺108 (⫺32) 0 405 (65) ⫺93 (91)
* Four and 6 weeks of chemotherapy for SIOP 1 and 2, respectively.
Histopathology (risk)
Operative Time (mins)
% Necrosis
Mixed (intermediate)
1 1 1 2
100 120 140 120 120 120 120 140 180
Less than 1 66–99 0 20 75 80 0 100 0
Cystic Mixed (low) Stromal (intermediate) Mixed, triphasic Diffuse anaplastic (high) Blastemal ⫹ focal anaplastic (high) Stromal (intermediate) Mixed (intermediate)
1 2 1 2 2 1 1 1
240 100 120 180 110 180 180 140
25 50 5 50 80 60 66 5
SIOP* 1 1 1 1
Metastasis
Lung Lung Lung Liver Dead of disease 2005
Lung Lung
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
of spillage of favorable histology samples was 20% compared to 6% in the SIOP-5 trial.13 Also, in SIOP results despite induction chemotherapy spillage is a problem because surgery was performed by a heterogeneous group of experts, including general surgeons, urologists, pediatric urologists, pediatric surgeons and oncological surgeons. After tumor spillage the outcome can be improved by intensifying chemotherapy and local or abdominal radiotherapy, which increase toxicity. For similar results the patient must undergo more treatment with more long-term consequences. According to surgical complications in the fourth NWTS-4 study when nephrectomy was performed by a general surgeon, the risk of complications was higher (OR 9.0, 95% CI 1.3– 65, p ⫽ 0.03) than when performed by a pediatric surgeon (referent, OR 1.0) or a pediatric urologist (OR 0.7, 95% CI 0.3–1.8).13 It is obvious that surgeon expertise is the determining prognostic factor for a successful intraoperative outcome. This was also the opinion of Ritchey et al, who concluded after a report from the National Wilms Tumor Study Group that surgeon expertise and tumor dimensions are the most important factors in avoiding surgical complications.13 They also stated that the decision to prescribe or not prescribe preoperative chemotherapy for unilateral nonmetastatic, nonsyndromic Wilms tumor depends on the group research interest, including accurate staging, pathological findings and genetics for those favoring nonpreoperative chemotherapy, and down staging, long-term toxicity and chemotherapy effects for those favoring preoperative chemotherapy. Although we are convinced that in children with preoperative chemotherapy there was more fibrosis or tumor adhesions to surrounding tissues, and although we had the impression that surgery required more time, we could not substantiate any difference in overall mean operative time or blood loss. Quantifying adherences is a subjective issue that cannot be measured. It might also be seen in primary tumors and, therefore, cannot be used for comparison. Mean tumor volume tended to decrease from 454.58 ⫾ 236.58 to 349.65 ⫾ 333.64 ml but this difference was not statistically different. This may be because there was no general trend in volume change, although 2 of 3 tumors decreased by more than 60% ⫾ 24.6%. In 5 of 17 patients (30%) tumor volume even increased by 42.8% ⫾ 31.3%. This may have led to the fact that operative time did not improve in the group with preoperative chemotherapy. This rate is high com-
1873
pared to the observations of Øra et al from the SIOP 93-01 group, who noted that only 5% of tumors showed a volume increase, although they were significantly smaller at diagnosis, and there were more stage III tumors and tumors associated with high risk histopathology.14 In our patients we could not reproduce these observations (table 3). Remarkably although in 12 of 17 patients (60%) the lesion was significantly reduced with results identical to those of Safdar et al,15 tumor size in our study remained so large that we could not safely perform partial nephrectomy. Partial nephrectomy is certainly beneficial in patients with bilateral disease. In patients with unilateral disease partial nephrectomy is not warranted due to the increased risk of local recurrence, which might adversely impact overall survival. Patient morbidity or recovery did not improve with preoperative chemotherapy. There was no difference in the duration of the need for tramadol infusion and no shortening of time to oral intake. Also, overall complications did not differ, although we had the subjective impression that there was a difference (table 2). The lack of a difference may have been due to the limited number of study patients. In 1984 McLorie et al already initiated a protocol at The Hospital for Sick Children in Toronto to stage children radiologically, followed by percutaneous tumor biopsy.16 The United Kingdom Children Cancer Study Group also examined the accuracy of percutaneous needle biopsy for initial diagnosis.17 This attitude showed no success in the larger NWTSG and SIOP study groups, although needle biopsy has minimal complications and carries no increased risk of local recurrence or up staging. With pretherapy biopsy the weakness of the standard SIOP preoperative chemotherapy protocol could be avoided. Compared to the NWTSG protocol the advantage of accurate pathological findings is ensured and the response to chemotherapy may be evaluated.
CONCLUSIONS Due to preoperative chemotherapy there is a trend toward down staging and decreased blood loss in patients with Wilms tumor. However, it remains doubtful whether surgery in a child indeed becomes safer and easier because of chemotherapy.
REFERENCES 1. Rance TF: Case of fungus haematodes of the kidnies. Med Phys J 1814; 32: 19.
3. Willetts IE: Jessop and the Wilms’ tumor. J Pediatr Surg 2003; 38: 1496.
2. Metzger ML and Dome JS: Current therapy for Wilms’ tumor. Oncologist 2005; 10: 815.
4. Wilms M: Die Mischgeschwülste der Niere. Leipzig: Verlag von Arthur Georgi 1899; pp 1–90.
5. Mitchell C, Pritchard-Jones K, Shannon R et al: Immediate nephrectomy versus preoperative chemotherapy in the management of non-metastatic Wilms’ tumour: results of a randomised trial
1874
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
(UKW3) by the UK Children’s Cancer Study Group. Eur J Cancer 2006; 42: 2554. 6. Beckwith JB: Wilms’ tumor and other renal tumors of childhood: a selective review from the National Wilms’ Tumor Study Pathology Center. Hum Pathol 1983; 14: 481. 7. de Kraker J, Graf N, van Tinteren H et al: Reduction of postoperative chemotherapy in children with stage I intermediate-risk and anaplastic Wilms’ tumour (SIOP 93-01 trial): a randomised controlled trial. Lancet 2004; 364: 1229. 8. Stehr M, Deilmann K, Haas RJ et al: Surgical complications in the treatment of Wilms’ tumor. Eur J Pediatr Surg 2005; 15: 414. 9. Zugor V, Krot D and Schott GE: Risk factors for intra- and postoperative complications in Wilms’ tumor surgery. Urologe A 2007; 46: 274.
10. Roelants M and Hauspie R: Growth Charts Flanders 2004. Brussels: Laboratory for Anthropogenesis, Vreije Universiteit Brussels 2004. 11. Beller S, Eulenstein S, Lange T et al: A new measure to assess the difficulty of liver resection. Eur J Surg Oncol 2008; 35: 59. 12. Boccon-Gibod L, Rey A, Sandstedt B et al: Complete necrosis induced by preoperative chemotherapy in Wilms tumor as an indicator of low risk: report of the International Society of Paediatric Oncology (SIOP) Nephroblastoma Trial and Study 9. Med Pediatr Oncol 2000; 34: 183. 13. Ritchey ML, Shamberger RC, Haase G et al: Surgical complications after primary nephrectomy for Wilms’ tumor: report from the National Wilms’ Tumor Study Group. J Am Coll Surg 2001; 192: 63.
14. Øra I, van Tinteren H, Bergeron C et al: Progression of localised Wilms’ tumour during preoperative chemotherapy is an independent prognostic factor: a report from the SIOP 93-01 nephroblastoma trial and study. Eur J Cancer 2007; 43: 131. 15. Safdar CA, Aslam M, Awan SH et al: Wilms’ tumour: a comparison of surgical aspects in patients with or without pre-operative chemotherapy. J Coll Physicians Surg Pak 2006; 16: 521. 16. McLorie GA, Khoury AE, Weitzman SS et al: Preoperative chemotherapy in management of Wilms’ tumor. Urology 1996; 47: 792. 17. Vujanic GM, Kelsey A, Mitchell C et al: The role of biopsy in the diagnosis of renal tumors of childhood: results of the UKCCSG Wilms tumor study 3. Med Pediatr Oncol 2003; 40: 18.
EDITORIAL COMMENTS These authors report a single institution experience with children with Wilms tumor. They compared their initial experience using primary surgery to their more recent approach of preoperative chemotherapy before surgery. What is notable is that their subjective impression was that surgery is not easier after preoperative chemotherapy. Intuitively one would expect that resection would be simpler after preoperative chemotherapy. A fairly high percent of this small patient series did not show decreased tumor volume after treatment. This may have influenced the results. In the United States most experience with preoperative chemotherapy is in children with massive tumors that are inoperable or bilateral tumors. In these circumstances preoperative treatment certainly facilitates the ability to remove a large tumor that was initially unresectable or perform renal sparing surgery for bilateral tumors. I have noted some difficult resections after chemotherapy but generally there had been prior exploration with open biopsy. Although the authors state that there was no difference in surgical complications, their data may not support that conclusion. They report tumor adhesion as a complication in children with preoperative chemotherapy. This may be a consequence of preoperative chemotherapy but is it a complication of surgery? The cohort that underwent primary surgery had some significant complications, including inferior vena caval rupture and sympathectomy. Published data suggest that there is a decrease in
surgical complications with the preoperative chemotherapy approach. It is clear from study comparisons that there was no difference in operative time, length of stay or recovery from surgery. If chemotherapy had facilitated surgery, one would expect to see a difference. Since the latter group of patients was treated more recently, one would expect a better outcome in this cohort since the authors were more experienced with managing renal tumors. They had excellent results in each group with no tumor spillage and overall low blood loss. The final chapter remains to be written on which approach (primary surgery vs preoperative chemotherapy) provides the greatest advantage in children with Wilms tumor. There does not appear to be any difference in terms of survival between the approaches. It boils down to which approach decreases the morbidity of therapy. This is usually viewed in terms of the risk of adjuvant therapy, eg trying to avoid radiation therapy and its long-term complications. However, surgeons must keep in mind that surgical morbidity can be an issue and these authors put the differences between these approaches in perspective.
Previous large SIOP studies show that preoperative chemotherapy down stages Wilms tumor and significantly decreases the rate of intraoperative tumor rupture from 27% to less than 5%.1 It was also reported
that up to 30% of Wilms tumors become amenable to partial nephrectomy after chemotherapy, notwithstanding subjective impression of these authors to the contrary. The local recurrence rate after nephron spar-
Michael Ritchey Department of Urology Mayo Clinic Arizona Scottsdale, Arizona
PREOPERATIVE CHEMOTHERAPY FOR WILMS TUMOR
ing surgery is comparable to that of renal cell carcinoma in adults, in whom partial nephrectomy is increasingly performed. Although they are undefined, the long-term risks of renal insufficiency and secondary renal malignancy in patients with Wilms tumor who undergo total nephrectomy might justify the risk of local recurrence. Preoperative chemotherapy clearly offers some surgical advantages. Unfortunately while the authors present a large single center experience,
1875
this study is inadequately powered to illustrate these advantages or disprove them. Whether the surgical benefits outweigh the potential risks of neoadjuvant chemotherapy remains an open question. Jonathan H. Ross Department of Pediatric Urology Rainbow Babies and Children’s Hospital Cleveland, Ohio
REFERENCE 1. Graf N, Tournade MF and de Kraker J: The role of preoperative chemotherapy in the management of Wilms’ tumor. Urol Clin North Am 2000; 27: 443.