Chapter 85
Wilms Tumor
John A. Kalapurakal and Patrick R.M. Thomas
Wilms tumor (WT, nephroblastoma) is a highly curable childhood neoplasm. The prognosis of children with WT has improved considerably from a very high mortality rate at the beginning of the 20th century to the current cure rate of >90%.1 The management of WT is a paradigm for successful interdisciplinary treatment of solid tumors of childhood to maximize cure rates and minimize treatment-related complications.

EPIDEMIOLOGY
WT is the most common malignant renal tumor of childhood. It occurs with an annual incidence of 7 cases per million children <15 years of age. Approximately 500 new cases are diagnosed each year in North America. The peak incidence is between 3 and 4 years of age. WT may arise as sporadic or hereditary tumors or in the setting of specific genetic disorders.2 Most WTs are solitary lesions, multifocal within a single kidney in 12% and bilateral in 7%.3 The clinical syndromes associated with WT include WAGR syndrome (WT, Aniridia, Genitourinary malformations, mental Retardation), Denys-Drash syndrome (pseudohermaphroditism, mesangial sclerosis, renal failure, and WT), and overgrowth syndromes like Beckwith-Wiedemann syndrome (somatic gigantism, omphalocele, macroglossia, genitourinary abnormalities, ear creases, hypoglycemia, hemihypertrophy, and a predisposition to WT and other malignancies) and Simpson-Golabi-Behmel syndrome.4,5

BIOLOGY
Among the various genetic changes implicated in the development of WT, the most widely studied involves WT1, which is a tumor suppressor gene at chromosome 11p13 that was isolated from a child with WAGR syndrome.6 WT1 is likely to play a specific role in glomerular and gonadal development.7 WT1 can also act as a dominant negative oncogene resulting in abnormal cell growth such as in Denys-Drash syndrome.8 Germline WT1 mutations are observed in approximately 82% of WT patients who have genitourinary anomalies or renal failure. The frequency of WT1 mutations in sporadic and familial WT is much lower at ∼20% and ∼4%, respectively.9 Beckwith-Wiedemann syndrome maps to chromosome 11p15.5; this locus is also referred to as WT2.10
Patients with loss of heterozygosity (LOH) at 16q and 1p have higher relapse and mortality rates.11 The National Wilms Tumor Study-5 (NWTS-5) prospectively evaluated the prognostic significance of LOH on 16q and 1p. Analysis of these data revealed that the relative risks (RR) for relapse for patients with stages I to IV favorable histology (FH) tumors with LOH stratified by stage were 1.8 for LOH 1p (P <.01) and 1.4 for LOH 16q (P = .05). When the effects of LOH for both 1p and 16q were considered jointly, the RR for relapse in stages I and II FH disease was 2.9 (P = .001) and for stages III and IV FH disease was 2.4 (P = .01). The RR for death for patients with stages I and II FH disease with LOH for both regions was 4.3 (P = .01) and for stages III and IV was 2.7 (P = .04). Based on these results, it was proposed that in future WT trials, the therapy for children with LOH at both 1p and 16q be augmented by the addition of doxorubicin to regimen EE4A (discussed below) for early-stage (stages I and II) tumors and cyclophosphamide/etoposide to regimen DD4A (discussed below) for advanced-stage tumors (stages III and IV).12
A novel Wilms tumor suppressor gene on the X chromosome, WTX, was recently discovered. This gene is inactivated in approximately one-third of sporadic WT cases.13 Anaplastic tumors have shown changes on 17p consistent with TP53 deletion and specific genomic loss or underexpression on 4q and 14q and focal gain of MYCN.14 Rhabdoid tumors are characterized by the genetic loss of the SMARCB1/hSNF5/INI-1 gene located at chromosome 22q11. Global gene expression studies have shown that loss of SMARCB1 results in repression of neural crest development and loss of cyclin-dependent kinase inhibition.15 In children with very low-risk WT treated with just surgery alone, the presence of WT1 mutation and 11p15 loss have been prospectively validated to be an important predictor of relapse. These biomarkers may be used to stratify patients to receive reduced chemotherapy in the future.16

PATHOLOGIC CLASSIFICATION
Although histopathologists had attempted to relate appearance to prognosis, no generally acceptable classification was available until the report of Beckwith and Palmer17 from the National Wilms Tumor Study-1 (NWTS-1). The NWTS classifies all tumors as having either FH or unfavorable histology (UH). The UH tumors include anaplastic tumors, clear cell sarcoma, and rhabdoid tumor of kidney. Of 1,465 patients randomly assigned on NWTS-3, 163 (11.1%) had UH.18 WTs are usually sharply demarcated, spherical masses with a “pushing” border and a surrounding distinct intrarenal pseudocapsule. Histologically, WT reflects the development of the normal kidney, consisting of three components: blastemal, epithelial (tubules), and stromal elements, in varying proportions.17 The proportion of the different components has prognostic significance.19 Nephrogenic rests consist of embryonal nephroblastic tissue and are found in 35% of kidneys with unilateral WT and in nearly 100% of kidneys with bilateral WT.20 Nephrogenic rests may be intralobar or perilobar based on their location within the kidney.21 Most nephrogenic rests undergo spontaneous regression and only a small proportion (1% to 5%) transform into WT.22 The histologic feature of greatest clinical significance in WT is anaplasia.23 Anaplasia may be focal (FA) or diffuse (DA). The definitions of FA and DA have been revised to reflect the distribution of anaplastic cells in the tumor rather than their quantitative density. These revised definitions are of prognostic significance. The 4-year survival rates for patients with stages II, III, and IV FA were 90%, 100%, and 100%, compared with 55%, 45%, and 4%, respectively, for patients with similar stage DA WT.24
Clear cell sarcoma of kidney (CCSK) and malignant rhabdoid tumor of kidney (RTK) are no longer considered true WT, but they have been included in NWTS protocols.17 CCSK has a propensity to metastasize to bone, and a skeletal survey and bone scan should be performed. RTK is the most lethal renal neoplasm in children. Primitive neuroepithelial tumors of the cerebellum or pineal region may be seen in 10% to 15% of patients with RTK.25

CLINICAL PRESENTATION
The classic presentation for WT is that of a healthy child in whom abdominal swelling is discovered by the child’s mother or by a physician during a routine physical examination. A smooth, firm, nontender mass on one side of the abdomen is felt. Gross hematuria occurs in as many as 25% of these cases.26 The child may be hypertensive or have nonspecific symptoms such as malaise or fever.27 Only rarely does a patient present with symptomatic metastases.

DIAGNOSTIC WORK-UP

Figure 85.1  Computed tomography scan of a 4-year-old girl with a large right…
The differential diagnosis of WT includes other malignant ­childhood lesions of the kidney, neuroblastoma, and benign conditions such as hydronephrosis, polycystic disease, and splenomegaly in left-sided tumors. Plain films of the abdomen may demonstrate calcifications, which occur in 60% to 70% of neuroblastomas but in only 5% to 10% of WT. Excretory urography (intravenous pyelography) was once the mainstay of imaging in WT and now has largely been replaced by ultrasonography and computed tomography (CT) scanning. Ultrasonography is very useful because it is readily available and is cost-effective.28 A specific advantage of ultrasonography is its ability to assess vessels for flow and tumor thrombus with duplex and color Doppler.29 Routine use of Doppler sonography after abdominal CT scans was not found to be useful in detecting cavoatrial thrombus in a Children’s Oncology Group (COG) study.30 Abdominal CT scans can demonstrate gross extrarenal spread, lymph node involvement, liver metastases, and the status of the opposite kidney (Fig. 85.1).31 Magnetic resonance imaging (MRI) has several advantages over CT scans, especially in identifying renal origin and vascular extension of the tumor.32 CT and MRI are useful in the detection and follow-up of patients with nephrogenic rests.33 Clinical and imaging impressions do not, however, obviate the need for inspection at laparotomy.34 Plain chest radiography and chest CT are also essential because asymptomatic pulmonary metastases are common.35 A complete blood cell count and urinalysis should be performed. Patients with WT can be anemic from hematuria. Serum blood urea nitrogen and creatinine levels and liver function tests are routine. If neuroblastoma is not ruled out, a test for urinary catecholamines should be performed. Table 85.1 outlines the pretreatment investigations recommended by the COG.

NATURAL HISTORY
The disease is often localized at diagnosis, as evidenced by the fact that surgery and radiation therapy is curative in almost 50% of cases.36 The first signs of local tumor spread beyond the pseudocapsule are invasion into the renal sinus or the intrarenal blood and lymphatic vessels. Spread throughout the peritoneal cavity may also occur, especially if there has been preoperative rupture or the disease has been spilled at surgery.37,38 The most common sites of metastases of WT are in the lungs, lymph nodes, and liver. Among patients with stage IV disease, lungs were the only metastatic site in approximately 80% and 15% have liver metastases.39 The NWTS-2 study demonstrated the prognostic importance of lymph node involvement. The 2-year relapse-free survival (RFS) with and without lymph node involvement was 54% and 82%, respectively.37

STAGING
Tumor staging is performed after examining the radiologic, operative, and histopathologic findings.38,39 In NWTS-1 and NWTS-2, a tumor grouping system was used for staging and treatment stratification. After analyzing the prognostic significance of several clinicopathologic factors in NWTS-1 and NWTS-2, a new staging system was adopted in NWTS-3. The presence of lymph node involvement was upstaged to stage III instead of group II, and local tumor spill was downstaged from group III to stage II.38 In NWTS-5, the most significant change was the distinction between stages I and II. The criteria for stage I was revised to accommodate an important subset of WT that is being managed by nephrectomy alone. Before NWTS-5, the distinction between stages I and II in the renal sinus was established by the hilar plane, which was an imaginary plane connecting the most medial aspects of the upper and lower poles of the kidney. This criterion was difficult to apply because of tumor distortion, and thus the hilar plane criterion has been replaced with renal sinus vascular or lymphatic invasion. This definition includes not only the involvement of vessels within the hilar soft tissue, but also the vessels located in the radial extensions of the renal sinus into the renal parenchyma.40,41 The COG staging guidelines for WT are shown in Table 85.2. The major change from NWTS-5 is that children with tumor spillage are upstaged from stage II to stage III because of the higher risk for relapse with two-drug chemotherapy alone.42 The COG risk group classification for treatment assignment in the new generation of WT protocols is shown in Table 85.3. In addition to tumor stage, this classification will also consider the patient’s age, tumor weight, presence or absence of LOH at 1p and 16q, and response to chemotherapy in children with FH tumors and lung metastases.

GENERAL MANAGEMENT
The diagnosis of WT is usually made before surgery and confirmed at surgery. A transverse transabdominal, transperitoneal incision is recommended for adequate exposure and thorough abdominal exploration.43 The surgeon must excise all tumors without spillage, if possible. Lymph node sampling from the para-aortic, celiac, and iliac areas must be performed. The use of titanium clips to identify residual tumor and margins of resection is also recommended. Routine exploration of the contralateral kidney was mandated in the past, but it is no longer recommended due to better imaging of the contralateral kidney with CT and MRI scans.
The chemotherapy and radiation therapy (RT) regimens for WT in the COG protocols are outlined in Tables 85.4 and 85.5.

RADIATION THERAPY TECHNIQUES
RT guidelines used for primary and recurrent WT in the COG protocols are shown in Table 85.5.
Timing of Radiation Therapy
The NWTS has shown that although RT does not need to be given immediately after surgery,36 a delay of ≥10 days after surgery was associated with a significantly higher abdominal relapse rate, particularly among patients with UH tumors.44,45,46,47 Because the pathologist cannot always rule out UH quickly, all patients with WT should be scheduled to start RT no later than day 9, the day of surgery being day 0. Although most patients may not be irradiated, it is easier to cancel than to make arrangements to start RT for a small child on short notice. The influence of RT delay on abdominal tumor recurrence in patients with FH tumors treated on NWTS-3 and NWTS-4 has been reported. The mean RT delay was 10.9 days. Although univariate and multivariate analysis did not reveal RT delay of ≥10 days to adversely influence flank and abdominal recurrence, it is important to note that in 59% of children the RT delay ranged from 8 to 12 days.48 For the COG protocols, it is recommended that RT be given preferably by day 9 but no later than day 14 after surgery.
Radiation Therapy Dose
In NWTS-1 and NWTS-2 RT dosages to the operative bed were given according to the age of the patient, however, no significant dose–response association was detected.45,47 In NWTS-3, there was a randomization for patients with FH tumors that resulted in elimination of RT for stage II FH, and a reduction of dose to 10 Gy for stage III patients.46 NWTS-3 and NWTS-4 data showed no RT dose response for CCSK and anaplastic tumors.49 Therefore, it was decided to treat all abdominal disease with 10 Gy. In the COG protocols, the dose is 10 Gy for most indications except for stage III DA and stages I to III RTK, where a higher dose of 19.8 Gy is recommended (Table 85.5).50,51
Radiation Therapy Volume

Figure 85.2 Anteroposterior flank irradiation portal in a 2…
Parallel-opposed fields using 4 or 6 MV photons are preferred. The flank RT field is determined by the CT or MRI scan performed at diagnosis before any chemotherapy is administered. The planning target volume is the tumor bed (outline of the kidney and associated tumor on the initial CT or MRI) with a 1-cm margin. The medial border must cross the midline to include the entire width of the vertebrae so as to minimize growth disturbances. An example of a flank RT portal is shown in Figure 85.2.44 When whole-abdomen RT is administered, the femoral heads and acetabulum must be shielded (Fig. 85.3). Whole-lung Irradiation (WLI) portals are shown in Figure 85.4. If the lungs and either the flank or whole abdomen have to be treated simultaneously, it is preferable to include them in one treatment portal.

SUMMARY OF CLINICAL TRIALS
No tumor has been studied by clinical trials as thoroughly and effectively as WT. The NWTS has been active in North America since 1969. There have also been successful studies run by the International Society for Pediatric Oncology (SIOP). The long-term results of NWTS-3 and NWTS-4 are shown in Table 85.6.

First National Wilms Tumor Study (1969–1974)
NWTS-1 showed that postoperative RT was not necessary for children younger than 2 years of age with group I tumors, and that combined dactinomycin and vincristine for irradiated patients with group II and III tumors was better than therapy with either agent alone. The RFS with and without RT among patients with group I tumors younger than 2 years of age was 90% and 88%, respectively.44
Second National Wilms Tumor Study (1974–1979)
NWTS-2 showed that in patients with group I tumors there was no survival difference between 6 months or 15 months of dactinomycin plus vincristine. Patients with groups II to IV tumors had a superior 2-year RFS of 77% with doxorubicin, dactinomycin, and vincristine compared with 63% with dactinomycin and vincristine alone.37
Third National Wilms Tumor Study (1979–1985)
The overall objective of NWTS-3 was to reduce therapy for low-risk patients (stages I to III FH) and to intensify treatment by adding a fourth drug, cyclophosphamide, for stage IV tumors with FH and all UH tumors. The results of this study demonstrated that RT and doxorubicin could be eliminated in children with stage II FH tumors. Patients with stage III FH tumors who received doxorubicin or 20 Gy had fewer abdominal relapses than those receiving 10 Gy without doxorubicin.46 The addition of cyclophosphamide in high-risk patients did not improve outcomes.49

Fourth National Wilms Tumor Study (1986–1994)
By the conclusion of NWTS-3, it was clear that the treatment of WT had been refined for the majority of patients; 62% of patients with WT have stage I or II FH disease and therefore require neither flank RT nor the potentially cardiotoxic doxorubicin. NWTS-4 was designed with cost containment in mind. The results proved that the survival was similar among patients who received standard-course (5 days) or single-dose, pulse-intensive dactinomycin chemotherapy. Further, pulse-intensive therapy was associated with less hematologic toxicity and marked reduction of treatment costs.52,53