Treatment and prognosis of Wilms tumor

INTRODUCTION — Wilms tumor is the most common malignancy affecting the kidney(s) in children [1]. Five-year overall survival (OS) rates have dramatically improved with multimodal therapy and now approach 90 percent.

The treatment and outcome of Wilms tumor will be reviewed here. The epidemiology, presentation, diagnosis, and staging of Wilms tumor are discussed separately. (See "Presentation, diagnosis, and staging of Wilms tumor".)

PROGNOSTIC FACTORS — Several prognostic factors at the time of initial diagnosis are associated with an increased risk of tumor recurrence or death and include [2]:

●Tumor histology

●Tumor stage

●Molecular and genetic markers (eg, loss of heterozygosity [LOH] at chromosome 16q, 1p, and 11p15 and 1q gain)

●Age >2 years

These prognostic factors are taken into account when determining the initial treatment regimen. (See 'COG approach' below.)

Tumor histology — Tumor histology is linked to patient outcome. Histologic classification differs somewhat between the two major Wilms tumor research groups (ie, Children's Oncology Group [COG] and the International Society of Paediatric Oncology [SIOP]). (See 'Difference in approaches of COG and SIOP' below.)

Histologic classification systems — In COG protocols, histologic assessment is performed before chemotherapy and tumors are categorized according to the presence or absence of anaplasia:

●Favorable histology (ie, no anaplasia)

●Focal anaplasia

●Diffuse anaplasia

The SIOP system assigns a histologic classification based upon assessment after preoperative chemotherapy (table 1). Tumors are classified as low, intermediate, or high risk based upon the degree of tumor necrosis and the relative proportion of each of the three cell types (epithelial, stromal, or blastemal). Patients with diffuse anaplastic or blastemal-type tumor after chemotherapy are classified as having high-risk histology. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Pathology'.)

Anaplastic histology — The presence of anaplasia is the most important predictor of adverse outcome in children with Wilms tumor [3,4]. Anaplasia is defined as the presence of multipolar polypoid mitotic figures and marked nuclear enlargement with hyperchromasia (picture 1), and it can be further classified as either focal (defined as tumors with anaplasia confined to one or a few discrete loci within the primary tumor, with no anaplasia or marked nuclear atypia elsewhere) or diffuse [3]. Most anaplastic Wilms tumors have TP53 gene mutations [5]. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Genetics'.)

In a report from the National Wilms Tumor Study (NWTS) group, multivariate analysis of 632 patients with nonmetastatic disease at diagnosis demonstrated that anaplastic, rhabdoid, or clear cell sarcoma histology was associated with increased rates of recurrence, metastases, and death [6]. Although diffuse anaplasia was only seen in 10 percent of patients, those with diffuse anaplastic tumor accounted for more than 60 percent of the deaths [7]. In the 165 cases with anaplastic histology, only three relapses and one death occurred among the 39 cases with focal anaplasia. In contrast, 22 of 23 children with stage IV diffuse anaplasia died of causes related to the tumor.

In another report from the NWTS group, diffuse anaplasia was the most significant predictor of a shorter survival time after diagnosis [8]. A subsequent study summarizing the results of the fifth NWTS demonstrated that the presence of focal or diffuse anaplasia in patients with stage I and stage II disease is associated with an inferior four-year overall survival (OS) rate when compared with patients with favorable histology tumors (approximately 83 versus 98 percent) [9]. For patients with diffuse anaplasia who underwent an immediate nephrectomy at diagnosis, patients with stage III or IV disease also fared worse at four years compared with those with favorable histology who were treated on the fifth NWTS (stage III anaplasia 65 percent, stage IV anaplasia 33 percent, and stage III or IV with favorable histology 92 percent).

Blastemal type histology — Blastemal type histology, assessed after preoperative chemotherapy (as in SIOP protocols), also appears to be associated with poor outcome; however, blastemal content has no prognostic significance when histologic categorization is performed before chemotherapy (as in COG protocols).

Results from retrospective analyses of earlier SIOP trials suggest that patients with blastemal type Wilms tumor have five-year event-free survival (EFS) of approximately 60 to 70 percent [10,11]. In particular, the risk of relapse appears to be higher among patients with blastemal type histology compared with other histologic subtypes of nonanaplastic Wilms tumor [12]. Based on these findings, SIOP reclassified this subtype as "high-risk" histology in 2002 [13].

Tumor stage — Staging criteria for Wilms tumor are based on the anatomic extent of the tumor without consideration for genetic, histologic, or molecular markers [14]. Higher stages (stages III through V) are associated with more extensive disease and a poorer outcome when compared with patients with lower stage disease (I and II) [6,15-18]. (See 'Outcome' below.)

There are two major staging systems in use (table 2):

●COG staging – The COG surgical staging occurs prior to administration of chemotherapy and is used throughout the United States and Canada

●SIOP staging – The SIOP surgical staging occurs after four weeks of chemotherapy and is used extensively in Europe

Both staging criteria are discussed in detail separately. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Staging'.)

Molecular markers — Several molecular and genetic markers appear to be predictive of outcome:

●LOH and 1q gain – LOH at chromosomes 1p, 11p15, and 16q and 1q gain in tumor cells are associated with risk of relapse and mortality in some patients with favorable histology Wilms tumor [19-28]. These findings have prompted the incorporation of some of these markers into the assignment of risk-directed therapy in children with Wilms tumor [21,26,27,29,30]. (See 'COG approach' below.)

●Gene expression profiles – Gene expression profiles also have been shown to be predictive of outcome and may ultimately aid in patient stratification and assignment to different treatment regimens, especially identifying patients who may require surgery only [31-34].

Age — In the past, children <24 months of age generally had lower relapse rates than older patients, which resulted in a better clinical outcome [35]. With improved therapeutic interventions, the effect of age as a prognostic factor has been reduced [6,36].

Adults with Wilms tumor have the same survival rate as children with similar tumor histology and stage when treated with the same therapeutic regimens [37,38]. However, the rate of treatment-related toxicity appears to be higher in adult patients [38].

MANAGEMENT — Children with suspected Wilms tumor should be referred to a pediatric cancer center for evaluation and treatment. Most children with newly diagnosed Wilms tumor are treated on research protocols, and risk-based therapy is assigned based on results of the initial staging, histologic classification, and molecular studies. These regimens conducted by cooperative groups incorporate multimodal therapy (surgery, chemotherapy, and radiation) and have resulted in dramatic improvements in outcomes in these children, with overall five-year survival rates approaching 90 percent. (See 'Outcome' below and 'Prognostic factors' above.)

Difference in approaches of COG and SIOP — The two major research groups that care for the largest number of patients with Wilms tumor are:

●Children's Oncology Group (COG) – The National Wilms Tumor Study (NWTS) group was established in 1969 and was subsequently incorporated into the COG, which conducts all trials in children with Wilms tumor in North America. The authors of this topic review practice at a COG member institution.

●International Society of Paediatric Oncology (SIOP) – The SIOP group was established in the early 1970s. It primarily includes European pediatric oncology centers, with additional centers located in South America and Australia.

COG and SIOP differ in their management of unilateral Wilms tumor (stages I through IV Wilms tumor), specifically regarding the timing of chemotherapy relative to surgical excision [22]. COG encourages the use of primary surgical resection prior to chemotherapy administration. In contrast, SIOP protocols use a primary chemotherapy approach followed by surgical resection and staging four weeks after the administration of chemotherapy. Both treatment approaches have excellent and comparable clinical outcomes [22]. (See 'Outcome' below.)

The COG approach has the following advantages:

●It allows early accurate assessment of the histologic diagnosis and tumor extent.

●It prevents patients with benign tumors from receiving unnecessary chemotherapy. SIOP trials have reported approximately 1.5 percent of patients treated initially with chemotherapy were found to have benign tumors [39,40].

The SIOP approach has the following advantages:

●Preoperative chemotherapy usually reduces tumor volume and may make surgical excision easier, thus decreasing the likelihood of tumor spillage [41]. This appears to reduce the potential need for local radiation due to spillage of tumor cells [42].

●The SIOP approach incorporates a measure of the in vivo chemosensitivity of each child's tumor into risk stratification, which has allowed reduction of treatment intensity for children with localized stage II or III Wilms tumor [43].

In a study from the United Kingdom Children's Cancer Study (UKCCS) Group, 205 patients with newly diagnosed, potentially resectable kidney tumors were randomized to immediate nephrectomy, or percutaneous kidney biopsy followed by chemotherapy and then nephrectomy [44]. There was improvement in the stage distribution among patients treated with preoperative chemotherapy and delayed surgery compared with those having immediate nephrectomy, which resulted in 20 percent fewer children receiving radiotherapy or doxorubicin. However, there was no difference between the two groups in five-year event-free survival (EFS).

In a subsequent report from the UKCCS group, delayed nephrectomy preceded by preoperative chemotherapy was associated with fewer surgical complications (including tumor rupture and spillage) compared with treatment with immediate nephrectomy (1 versus 20 percent) [45].

Nonetheless, the COG and SIOP regimens have comparable five-year overall and EFS rates. (See 'Survival rate' below.)

COG approach — COG protocols are based upon primary surgical resection as initial treatment for children with unilateral Wilms tumor (table 3) [4,6,16,46-50]. Adjuvant chemotherapy is administered after surgery. The choice of chemotherapeutic agents and the length of therapy are dependent upon the tumor histology and stage (table 2). Radiation therapy is used in patients with stage III or IV tumors. (See 'Tumor histology' above and 'Tumor stage' above.)

Because the authors' institution is a member of COG, they treat patients using COG protocols or use these protocols as guidelines for therapy (table 3) [14]. The approach is based upon the tumor stage as follows:

●Stages I and II – Primary surgical resection is followed by 19 weeks of vincristine and dactinomycin. No radiation therapy is required.

The treatment approach may be modified in patients with any of the following:

•Very low-risk tumors – Patients meeting all of the following criteria are considered to have very low-risk tumors:

-Age <2 years

-Stage I favorable histology

-Tumor weighs <550 g

Most patients with very low-risk tumors can be managed by nephrectomy alone without adjuvant chemotherapy [51-53].

In one study, patients with very low-risk tumors were assigned to either surgery alone or surgery followed by adjunctive chemotherapy (vincristine and dactinomycin) [52,54]. According to the study's stringent stopping rule, the study was closed when the predicted two-year EFS fell below 90 percent. At two years, the rate of relapse among patients treated with surgery alone was 13.5 percent; all patients were alive [54]. In a report of long-term outcomes of the patients in this study, five-year EFS was lower for patients treated with surgery only compared with surgery plus adjuvant chemotherapy (84 versus 97 percent) [52]. One death was observed in each treatment group, and five-year overall survival (OS) did not differ between groups (98 and 99 percent). Investigators observed that the children who relapsed were more successfully retreated compared with similar patients in previous studies, probably because they were naive to both radiation therapy and chemotherapy. The authors concluded that although EFS was lower in the surgery only group, the overall mortality rate did not differ and a majority of the infants in the surgery only group were spared the adverse effects of chemotherapy. This study is limited in that 21 of the 77 patients initially treated with surgery alone were later recalled and treated with chemotherapy and were therefore censored from the analysis.

In a subsequent trial, 116 children with very low-risk tumors (excluding patients with predisposition syndromes and those with inadequate lymph node sampling) were treated with nephrectomy alone followed by close observation over a median follow-up of 80 months [53]. Estimated four-year EFS was 90 percent, and no deaths occurred. Loss of heterozygosity (LOH) at chromosome 11p15 was predictive of relapse. (See 'Molecular markers' above.)

•High-risk molecular markers – As discussed above, certain molecular markers (eg, LOH at chromosomes 1p, 11p15, and 16q and 1q gain) are associated with lower OS and EFS in patients with otherwise favorable histology (see 'Molecular markers' above). Preliminary data suggest that augmentation of therapy for patients with some of these high-risk markers may improve outcomes. For example, in a single-arm study involving 35 patients with standard-risk stage I and II tumors with combined LOH at chromosomes 1p and 16q, the addition of doxorubicin to the chemotherapy regimen was associated with a nonsignificant trend towards improved four-year EFS compared with similar patients enrolled in an earlier study who were treated without doxorubicin (84 percent versus 75 percent, respectively) [29]. It is uncertain whether augmented therapy will improve outcomes for patient with tumors displaying 1q gain. Further studies are needed to address this question.

•Anaplasia – Patients with stage I/II disease and anaplastic histology have a poorer prognosis and are treated with additional chemotherapy and radiation therapy as described below. (See 'Anaplastic histology' below.)

●Stage III – Primary surgical resection is followed by 25 weeks of three-drug chemotherapy with vincristine, dactinomycin, and doxorubicin. Radiation therapy (dose of 10.8 radiation absorbed dose [Gy]) based on lymph node involvement or the extent of the peritoneal contamination is administered to the affected flank, hemiabdomen, or complete abdomen. Most patients with stage III favorable histology have good outcomes with this treatment approach (ie, four-year EFS of 88 percent) [55]. Positive lymph nodes and LOH at either chromosome 1p or 16q are predictors of poor outcome.

Patients with combined LOH at chromosomes 1p and 16q are treated with a more intensive chemotherapy regimen (ie, "regimen M," which includes cyclophosphamide and etoposide). In a single-arm COG study involving 52 patients with stage III/IV favorable histology tumor with 1p and 16q LOH who were treated with regimen M, four-year EFS was superior to that of a similar cohort enrolled in an earlier study and received less intensive therapy (90 percent versus 61 percent, respectively) [29,56].

●Stage IV – Primary surgical resection is followed by 25 weeks of triple-drug chemotherapy of vincristine, dactinomycin, and doxorubicin. As with stage III, patients with combined LOH at chromosomes 1p and 16q are treated with a more intensive chemotherapy (regimen M) [57]. Radiation therapy is administered to the whole lung field only if lung metastases do not completely resolve after six weeks of chemotherapy [57]. (See 'Lung metastases' below.)

In patients with evidence of spread of disease to the regional lymph nodes in the hilum and/or paraaortic nodes, radiation therapy (dose of 10.8 radiation absorbed dose [Gy]) is administered to the locoregional area including the appropriate one-half of the abdomen as indicated. For patients with metastatic sites other than the lung, radiation dose varies according to the site.

●Stage V – Treatment of bilateral (stage V) Wilms tumor is reviewed below. (See 'Bilateral kidney involvement' below.)

SIOP approach — In SIOP protocols (table 4), patients are treated with preoperative chemotherapy for four weeks (six weeks if there is evidence of metastatic disease) followed by surgical resection and postoperative chemotherapy that is based upon the staging and histology assessment performed after initial chemotherapy (table 2) [13,14,39,40,43,58-60].

For patients with nonmetastatic disease, preoperative chemotherapy consists of vincristine and dactinomycin for four weeks. If there is evidence of metastatic disease beyond the abdominal-peritoneal region (eg, lung metastases), triple-drug chemotherapy (vincristine, dactinomycin, and doxorubicin) is given for six weeks preoperatively. Following preoperative chemotherapy, surgical resection and staging are performed.

In some centers, biopsies are used to make a histologic diagnosis prior to the initiation of chemotherapy [61]. In the SIOP Wilms tumor 2001 study, biopsy was allowed without affecting tumor stage [43]. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Staging'.)

Postoperative SIOP regimens are based upon tumor staging and histology as follows (table 1 and table 4) [2,14]:

●Stage I:

•Low risk – Patients with low-risk stage I tumors (ie, no viable tumor after initial preoperative chemotherapy) do not require additional chemotherapy or radiation therapy. These patients are monitored closely for recurrence. (See 'Screening for tumor recurrence' below.)

•Intermediate risk – Treatment consists of postoperative chemotherapy (vincristine and dactinomycin) for four weeks without radiation therapy.

•High risk – Treatment consists of postoperative chemotherapy (vincristine, dactinomycin, and doxorubicin) for 27 weeks without radiation therapy.

●Stage II:

•Low and intermediate risk – Treatment consists of postoperative chemotherapy (vincristine and dactinomycin) for 27 weeks without radiation therapy. Although previously included in SIOP treatment protocols, the elimination of doxorubicin from the postoperative regimen did not significantly worsen EFS and had no effect on five-year OS in patients with intermediate-risk histology [43].

•High risk – Treatment consists of postoperative chemotherapy (doxorubicin, cyclophosphamide, carboplatin, and etoposide) for 34 weeks and radiation therapy (25.2 Gy flank radiotherapy with a 10.8-Gy boost for lymph node involvement or gross disease).

●Stage III:

•Low risk – Treatment consists of postoperative chemotherapy (vincristine and dactinomycin) for 27 weeks without radiation therapy.

•Intermediate risk – Treatment consists of postoperative chemotherapy (vincristine and dactinomycin) for 27 weeks and radiation therapy (14.4 Gy flank radiotherapy with a 10.8 Gy boost for lymph node involvement or gross disease). As with stage II tumors, the elimination of doxorubicin from the postoperative regimen did not significantly worsen outcomes in patients with intermediate-risk histology [43].

•High risk – As with stage II high-risk tumors, treatment consists of postoperative chemotherapy (doxorubicin, cyclophosphamide, carboplatin, and etoposide) for 34 weeks and radiation therapy (25.2 Gy flank radiotherapy with a 10.8-Gy boost for lymph node involvement or gross disease).

●Stage IV – Treatment is based on histology and response of lung nodule(s) after the initial six weeks of preoperative chemotherapy (see 'Lung metastases' below):

•Low- and intermediate-risk histology with complete response (ie, resolution or resection of lung nodules) – Treatment consists of postoperative chemotherapy (vincristine, dactinomycin, and doxorubicin) for 27 weeks without lung irradiation. Flank radiotherapy is provided for patients with local stage III tumors.

•Low- and intermediate-risk histology with incomplete response – Treatment consists of postoperative chemotherapy (doxorubicin, cyclophosphamide, carboplatin, and etoposide) for 34 weeks and radiation therapy (15 Gy lung radiotherapy and flank radiotherapy for local stage III tumors).

•High-risk histology – Treatment consists of postoperative chemotherapy (doxorubicin, cyclophosphamide, carboplatin, and etoposide) for 34 weeks and radiation therapy (15 Gy lung radiotherapy and flank radiotherapy for local stage III tumors).

For patients with metastatic sites other than the lung, radiation dose varies according to the site.

●Stage V – Treatment of bilateral (stage V) Wilms tumor is reviewed below. (See 'Bilateral kidney involvement' below.)

Special populations — Patients with bilateral (stage V) Wilms tumor, anaplastic histology, and/or lung metastases generally require more intensive therapy than those with lower-grade tumors with standard-risk histology.

Bilateral kidney involvement — For patients with bilateral Wilms tumor (stage V) (table 2), we suggest preoperative chemotherapy with vincristine, dactinomycin, and doxorubicin rather than other regimens. Following preoperative therapy, we suggest kidney parenchymal-sparing resection rather than more extensive bilateral kidney resection. Postoperative chemotherapy with or without radiation is selected based on treatment response to neoadjuvant chemotherapy, histopathology, and tumor stage. Preoperative combination chemotherapy reduces the risk of end-stage kidney failure by facilitating early nephron-sparing surgery and preserving kidney parenchyma. The treatment approach for children with stage V Wilms tumor is similar in COG and SIOP protocols [2,62-64].

In the COG protocol, which the authors use at their institution, preoperative chemotherapy consists of a six-week regimen of vincristine, dactinomycin, and doxorubicin [63,64]. Postoperative chemotherapy with or without radiation is offered based on a risk-adapted protocol that integrates clinical and histologic responses to preoperative chemotherapy, pathologic classification, and tumor stage [63,64]. Specific treatment details of postoperative therapy may be individualized by an institution and are beyond the scope of this review. However, all treatment should be administered under the guidance of a multidisciplinary team of pediatric oncologists, surgeons, and radiation oncologists with expertise in the management of Wilms tumor.

Although a minority (only 5 to 7 percent) of patients with Wilms tumors present with stage V disease (bilateral kidney involvement), their care is challenging. In these patients, the therapeutic goal is to adequately treat the bilateral tumor loci while trying to preserve kidney function. Historically, the management of patients with bilateral disease included initial nephrectomy of the more involved side combined with partial nephrectomy or excisional resection of the lesion in the contralateral kidney (initial surgical excision). However, the risk of end-stage kidney disease (ESKD) was high in patients treated with this approach [65]. (See 'Late effects' below.)

Subsequent studies demonstrated improved outcomes with the use of intensive chemotherapy followed by a strategy of kidney parenchymal-sparing resection. In a prospective study conducted by the COG (AREN0534), 189 evaluable patients with bilateral Wilms tumor were treated with 6 to 12 weeks of preoperative chemotherapy followed by tumor resection [63,64]. In this study, a majority of patients (84 percent) underwent definitive surgical treatment by 12 weeks after initiation of chemotherapy [63]. Most patients received unilateral total nephrectomy with contralateral partial nephrectomy (48 percent) or bilateral partial nephrectomy (35 percent). In contrast, bilateral total nephrectomies were performed on less than 3 percent of patients. Four-year EFS and OS were 82 and 95 percent, respectively, which were considerably higher than historical controls (eight-year EFS between 40 and 74 percent; eight-year OS between 45 and 89 percent) [62]. Retrospective studies have reported similar results for neoadjuvant chemotherapy, with reduced rates of kidney failure and overall patient survival rates that are comparable with those receiving initial surgical excision [62,66,67].

Patients with diffuse anaplasia (a high-risk tumor) have a poor prognosis despite the use of preoperative chemotherapy. In a subsequent analysis of COG AREN0534, 180 evaluable patients were risk classified (table 1) based on their histopathologic response to preoperative chemotherapy. In this study, four-year EFS and OS was highest for those with low-risk disease and completely necrotic tumors (100 percent each), those with intermediate-risk disease (82 and 97 percent), and those with high-risk blastemal type tumors (79 and 93 percent), respectively. Patients with diffuse anaplasia had the lowest survival rates (four-year EFS and OS of 61 and 72 percent, respectively). [64]. Similar results were seen in another study of 80 children with bilateral tumors where intensive preoperative chemotherapy was not associated with improved OS in patients with diffuse anaplasia or blastemal-type tumors [68]. (See 'Anaplastic histology' below and 'Blastemal type histology' above.)

Anaplastic histology — Patients with anaplastic histology have a poorer outcome [9]. As a result, trials investigating alternative agents (eg, irinotecan) and using more intensive therapy have been conducted in patients with anaplastic Wilms tumor [69,70]. Despite the more aggressive intervention, the four-year EFS is poor (33 to 70 percent, depending on tumor stage) [9].

Ongoing clinical trials are underway, and future trials are planned in the hope of finding a more effective regimen that will improve EFS in patients with diffuse anaplastic Wilms tumor. An ongoing COG study (AREN1921), is investigating "regimen UH-3" for stages II to IV diffuse anaplastic Wilms tumors [71]. The UH-3 regimen consists of and alternates cycles of vincristine/doxorubicin/cyclophosphamide, cyclophosphamide/carboplatin/etoposide, and vincristine/irinotecan, lasting for a total of 40 weeks.

Lung metastases — Chest imaging for patients with Wilms tumor is an area of controversy. A contrast-enhanced computed tomography (CT) is more sensitive than chest radiography; however, it is unclear how many nodules detected by CT represent malignant disease and whether diagnosing these lesions improves outcome. This issue is discussed in greater detail separately. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Chest imaging'.)

In patients with lung metastases, the optimal management remains uncertain. The approaches of both COG and SIOP attempt to tailor treatment based upon the completeness of lung metastasis response after initial chemotherapy:

●In the COG study, patients with favorable histology and isolated lung metastases who showed complete resolution of lung metastases after six weeks of chemotherapy (vincristine, dactinomycin, and doxorubicin) were continued on the same chemotherapy regimen without lung irradiation [57]. Patients with either an incomplete response after six weeks of chemotherapy or LOH at chromosomes 1p and 16q received whole-lung radiation and the addition of cyclophosphamide and etoposide (regimen M). Of the 292 patients enrolled in the study who had lung metastases reassessed after six weeks of therapy, 46 percent had a complete response to initial chemotherapy and 54 percent had an incomplete response. For patients with a complete response, four-year EFS and OS estimates were 79.5 percent (95% CI, 71.2-87.8 percent) and 96.1 percent (95% CI, 92.1-100.0 percent), respectively. For patients with an incomplete response, four-year EFS and OS estimates were 88.5 percent (95% CI, 81.8-95.3 percent) and 95.4 percent (95% CI, 90.9-99.8 percent), respectively. The pooled four-year EFS and OS estimates for the entire cohort were significantly higher compared with an earlier study in which all patients were treated with lung irradiation (85 versus 73 percent and 96 versus 84 percent, respectively). These results suggest that the strategy of tailoring treatment to the patient's response to initial chemotherapy can maintain excellent outcomes for children with lung metastases while decreasing treatment-related morbidity.

●In the SIOP 93-01 trial, patients with an incomplete response after six weeks of prenephrectomy chemotherapy (vincristine, dactinomycin, and doxorubicin/epirubicin) underwent surgical resection of the lung nodule if feasible [72]. Patients showing complete resolution of lung metastases after six weeks of prenephrectomy chemotherapy and those who underwent successful surgical resection of lung nodules were then treated with postoperative chemotherapy (vincristine, dactinomycin, and doxorubicin/epirubicin for 27 weeks) without lung irradiation. Patients with incompletely resected lung nodules, multiple inoperable metastases, and/or high-risk histology of the primary tumor were treated with a more intense post-nephrectomy chemotherapy regimen (doxorubicin/epirubicin, cyclophosphamide, carboplatin, and etoposide for 34 weeks). If a complete response was obtained after nine weeks, lung irradiation was not given. In children with residual lung nodules and/or high-risk histology, bilateral pulmonary lung radiotherapy was administered. Only 14 percent of patients in the trial required lung irradiation during first-line treatment. Five-year EFS was 73 percent, and OS was 82 percent. Five-year OS was superior among patients with a complete response after six weeks of prenephrectomy chemotherapy and for those who underwent metastasectomy when compared with those who had an incomplete response (88, 92, and 48 percent, respectively).

Salvage therapy for tumor recurrence — Recurrent disease occurs in approximately 15 percent of patients with favorable histology; however, among patients with anaplastic features, the risk of recurrence is considerably higher. (See 'Tumor recurrence' below.)

The optimal salvage therapy regimen for patients with Wilms tumor has not yet been established. Multiagent regimens with cyclophosphamide, ifosfamide, carboplatin, etoposide, and cisplatin have been used to varying degrees of success [73-85].

High-dose chemotherapy with stem cell rescue has been suggested as an option in patients with recurrent tumor, especially in those with adverse prognostic indicators. However, the efficacy of this approach remains unknown [14,86].

OUTCOME

Survival rate — Overall five-year survival rates for Wilms tumor have steadily improved from 20 percent in the late 1960s to >90 percent in current-era reports from the Children's Oncology Group (COG), International Society of Paediatric Oncology (SIOP), and other groups [14,40,48,87].

In a review of 6185 patients enrolled in the National Wilms Tumor Study (NWTS) between 1969 and 1995, the overall survival (OS) rate was 84 percent through 2002 [88]. Causes of death included:

●Tumor-related – 86 percent

●Late effects of therapy – 9 percent

●Nontreatment or nondisease-related – 5 percent

●Unknown – 1 percent

Ninety-one percent of deaths occurred early within the first five years of diagnosis (n = 819) and were primarily due to the original tumor (94 percent). In contrast, the causes of the 159 late deaths were evenly distributed between the late effects of therapy (39 percent) and tumor-related mortality (40 percent).

OS and event-free survival (EFS) rates vary among individuals and are dependent upon the tumor histology, stage, and size and age of the patient at diagnosis. Patients with very low-risk tumors (ie, patients <24 months old with stage I favorable histology tumors <550 g) have the best prognosis, with five-year OS rates ≥98 percent [52]. Patients with diffuse anaplastic tumor have the poorest prognosis, with four-year EFS estimates of 83, 65, and 33 percent for stages II, III, and IV, respectively [9].

A summary of the outcomes of patients enrolled in COG studies is available through the National Cancer Institute website.

Survivors of childhood Wilms tumor are at increased risk of premature death during adulthood due to secondary neoplasms and other late complications (see 'Late effects' below). In the British Childhood Cancer Survivor Study, which included 1441 five-year survivors of Wilms tumor, the cumulative risk of death at 30 and 50 years from diagnosis was 5 and 23 percent, respectively [89]. Three-quarters of excess late deaths in this cohort were attributable to secondary neoplasms; cardiovascular disease accounted for the remainder.

Tumor recurrence — Recurrent disease occurs in approximately 15 percent of patients with favorable histology; however, among patients with anaplastic features, the risk of recurrence is nearly 50 percent [9]. Unfavorable histology is the most powerful prognostic factor associated with tumor recurrence [50,90,91]. Other risk factors include tumor size, tumor stage, age, and presence of certain molecular markers. (See 'Prognostic factors' above.)

For patients with stage III disease, the risk of recurrence does not appear to differ according to stage III subtype [92]. Biopsy is thought to increase the risk of local recurrence by introducing tumor cells into the abdominal cavity. However, a study investigating this found that biopsy is not clearly associated with increased risk of local recurrence [90].

The majority of recurrences occur within the first two years of therapy. Recurrences most often involve the lung. Recurrence in the central nervous system is very rare [93].

Approximately 1 percent of children develop a tumor in the contralateral kidney within six years of the initial diagnosis of Wilms tumor, with 90 percent occurring within the first two years [16,48]. The presence of nephrogenic rests (foci of persistent metanephric cells) places a child at increased risk for contralateral kidney recurrence [94]. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'Pathogenesis'.)

Patients who have recurrence of their tumor have post-relapse four-year survival rates of 50 to 80 percent [95,96]. In a study of children who relapsed after receiving two chemotherapeutic agents (vincristine and actinomycin D), salvage therapy (which included doxorubicin, cyclophosphamide, and radiation therapy) was associated with 82 percent OS and 71 percent EFS rates at four years [95]. In a study of children who relapsed after receiving initial therapy with vincristine, actinomycin D, doxorubicin, and radiation therapy, salvage therapy using an aggressive chemotherapy regimen (including cyclophosphamide and carboplatin along with surgery and radiation therapy) was associated with 48 percent OS and 42 percent EFS rates at four years [96].

In patients with recurrent tumor, the following prognostic factors are associated with favorable response to salvage therapy and improved outcome [14,97-99]:

●Relapse occurs more than 12 months after initial diagnosis

●Favorable histology of the initial tumor

●Low tumor stage of initial disease

●Initial treatment with only dactinomycin and vincristine

●Few pulmonary nodules

●No previous radiation to tumor bed

●Complete resection of original tumor

COMPLICATIONS

Early complications — Early complications are generally related to therapy and include adverse effects of chemotherapeutic drugs and surgical complications such as bowel obstruction, hemorrhage, and wound infection [100].

Late effects — Survivors are at increased risk for long-term sequelae [101]. Late adverse effects are primarily dependent upon the type and intensity of therapy used to treat patients. Patients who received more intense chemotherapy and radiation therapy are more likely to have late complications. These include kidney impairment; cardiotoxicity; hepatotoxicity; orthopedic, growth, and pulmonary problems; infertility; and secondary malignancies [102-104]. All patients should undergo routine screening for therapy-related complications.

●Kidney impairment – Kidney impairment may be caused by surgical loss of kidney parenchyma as well as kidney injury from chemotherapy and radiation therapy. Patients with bilateral kidney involvement are at increased risk for kidney impairment, especially if they also received radiation therapy [18,65,105-107]. This was illustrated from a study from the National Wilms Tumor Study (NWTS) group of 81 patients with stage V tumor who received radiation therapy [106]. One-third of patients had elevated serum creatinine concentrations, 18 patients had moderate kidney insufficiency, and 10 had severe kidney insufficiency.

In contrast, the risk of severe kidney impairment is low in patients with unilateral Wilms tumor treated without nephrotoxic chemotherapy or ionizing radiation [108,109]. In a long-term (median 19.6 years) follow-up study of 75 survivors of nonsyndromic unilateral Wilms tumor who underwent unilateral nephrectomy without receiving nephrotoxic chemotherapy or ionizing radiation, 21 percent had mild kidney dysfunction (ie, stage II chronic kidney disease [CKD] (table 5)); no patients had stage III or more severe CKD [109]. In another study of 37 survivors of childhood Wilms tumor (95 percent had unilateral disease; 60 percent had received radiation therapy) who were followed for a median of 22.5 years, 53 percent developed stage II CKD, one patient developed stage III CKD, and no patients had end-stage kidney disease (ESKD) [110]. Hypertension was noted in 40 percent of patients.

ESKD requiring kidney replacement therapy is rare following treatment for Wilms tumor, and, when it occurs, it is usually associated with bilateral tumor involvement. Other causes of ESKD include Denys-Drash syndrome, radiation nephritis, or WAGR syndrome (Wilms tumor, aniridia, genitourinary anomalies, and intellectual disability [mental retardation]) [111-113]. (See "Presentation, diagnosis, and staging of Wilms tumor", section on 'WAGR syndrome' and "Presentation, diagnosis, and staging of Wilms tumor", section on 'Denys-Drash syndrome'.)

●Cardiotoxicity – Treatment with anthracyclines (eg, doxorubicin) may cause cardiovascular complications including heart failure and increased left ventricular afterload [114,115]. Risk of developing cardiac abnormalities is dependent upon the cumulative dose and dose intensity used. In a report of 97 patients with Wilms tumor who received doxorubicin (mean cumulative dose 303 mg/m²), 25 percent of patients had evidence of increased left ventricular afterload documented by echocardiography [115]. (See "Clinical manifestations, diagnosis, and treatment of anthracycline-induced cardiotoxicity" and "Risk and prevention of anthracycline cardiotoxicity".)

●Hepatotoxicity – Chemotherapy (vincristine and dactinomycin) and hepatic radiation therapy can cause severe hepatopathy [116-118]. This includes sinusoidal obstruction syndrome (SOS), which has been observed in patients receiving treatment for pediatric tumors that does not include hematopoietic cell transplantation [118]. (See "Hepatic sinusoidal obstruction syndrome (veno-occlusive disease) in children".)

The incidence of hepatopathy and/or SOS is variable in patients with Wilms tumor, with estimates ranging from 1 to 14 percent [116-118]. Some studies suggest that the incidence may be related to chemotherapy dosing [117]. (See "Hepatotoxicity of chemotherapy and other cytotoxic agents".)

Most patients with mild or moderate treatment-related hepatopathy can subsequently receive chemotherapy, following treatment protocol guidelines. Limited data also suggest that most patients with Wilms tumor who develop severe hepatopathy may also be able to safely continue or restart chemotherapy. In one observational study of 8862 patients treated for Wilms tumor on the National Wilms Tumor Study protocols 3 to 5, 71 patients (0.8 percent) were diagnosed with severe hepatopathy; this excluded those with WHO grade 1 or II transaminase elevations without signs or symptoms of sinusoidal obstructive syndrome [118]. Severe hepatopathy most commonly occurred after actinomycin D and vincristine courses (80 percent) with a median onset of 51 days. Among this cohort with severe hepatopathy, most patients (85 percent) were able to either continue (14 patients) or restart chemotherapy after a treatment delay (45 patients). For those patients who restarted chemotherapy after a delay, 14 did so at full dose and 12 were alive and disease-free at five years. Among the 10 patients who were unable to restart chemotherapy, four patients died from severe hepatopathy and six patients were alive and disease-free at follow-up.

●Fertility – Patients with Wilms tumor generally do not have infertility problems. However, females who received high doses of abdominal radiation have higher risk of infertility, pregnancy complications, perinatal mortality, and low birth weight babies [119-121].

●Second malignancies – Patients with Wilms tumors are at increased risk for developing second malignancies. In a report from the NWTS group that reviewed the late outcome of 5278 patients treated between 1969 and 1991, there were 43 cases of second neoplasms [122]. The overall risk was estimated to be 1.6 percent at 15 years, and the highest risk was among patients who received radiation therapy and doxorubicin. All reported second solid tumors occurred within the radiation field and were detected at a mean of 16.1 years post-therapy for the original Wilms tumor.

In a report from the Childhood Cancer Survivor Study, 3 percent of survivors developed second malignancies 25 years after initial diagnosis of Wilms tumor [101]. In another study of a large international cohort of patients diagnosed with Wilms tumor before 15 years of age during 1960 to 2004, 174 solid tumors and 28 leukemias were detected in 13,357 patients after observation through 2005 [123].

In the previously described British Childhood Cancer Study, secondary malignancies accounted for three-quarters of excess late deaths among five-year survivors of Wilms tumor [89]. Bowel and breast cancers were most frequent, and radiation therapy was an important predictor.

●Other adverse effects from radiation – Other complications resulting from radiation therapy include muscle atrophy, short stature, and scoliosis. Children who receive 10 radiation absorbed dose (Gy) or more at an age younger than one year are most likely to have growth (height) impairment. Stature loss is due to the extension of standard flank radiation over the spinal column [124,125].

Radiation to the chest for lung metastases may result in thyroid disease and damage mammary tissue in young patients. Higher doses of radiation to the liver are associated with increased risk of portal hypertension [126].

FOLLOW-UP — Children with Wilms tumor should be examined on a regular basis by a clinician who is familiar with the natural history of the disease and the complications of therapy. Follow-up care includes ongoing surveillance for tumor recurrence and late treatment-related complications.

The primary care provider can often provide routine care with coordinated scheduled visits to the oncologist, especially if the patient's family lives some distance from the oncology treatment center. Long-term follow-up guidelines for children who have received treatment for cancer in childhood are available at the Children's Oncology Group (COG) website.

Screening for tumor recurrence — Ongoing surveillance usually alternates chest radiography and abdominal ultrasound with chest, abdomen, and pelvic computed tomography (CT). Imaging studies are performed every six to eight weeks during therapy, then every three months for two years, followed by every six months for an additional two years. We perform more frequent abdominal ultrasounds in selected patients, as discussed below.

●Pulmonary surveillance – The lung is the most frequent first site of recurrence. Although tumor recurrence is best detected by lung CT, this modality is associated with increased radiation exposure compared with chest radiography. As a result, there is no consensus regarding the best modality for screening or the frequency of screening. As noted above, we alternate modalities.

●Abdominal surveillance – Abdominal ultrasonography is performed to detect tumor recurrence in residual kidney tissue or another abdominal location. Abdominal ultrasound can detect early infradiaphragmatic relapses, which may be seen in patients with stage II and III tumor. The frequency of ultrasonography is dependent on whether nephrogenic rests (foci of persistent metanephric cells) were identified in the kidney(s) at the time of initial tumor diagnosis. Patients with nephrogenic rests are at increased risk for relapse. (See 'Tumor recurrence' above and "Presentation, diagnosis, and staging of Wilms tumor", section on 'Pathogenesis'.)

At our institution, we utilize the following schedule for abdominal ultrasonography [127]:

•In patients with nephrogenic rests or those with tumor-predisposing syndromes (WAGR, Denys-Drash, or Beckwith-Wiedemann syndromes), examinations are performed every three months until the child is eight years old.

•In patients without nephrogenic rests or tumor-predisposing syndromes, examinations are performed every three months for two years and then every six months for an additional two years.

Screening for therapy-related complications — Screening for late complications is based upon the expected likelihood of adverse events related to the therapeutic regimens used. Many of these complications, as previously described, can be detected by history, physical examination, assessment of serum creatinine and urinalysis, and echocardiogram. (See 'Late effects' above.)

SUMMARY AND RECOMMENDATIONS

●General principles of management

•Children with suspected Wilms tumor should be referred to a pediatric cancer center for evaluation and treatment. Most children with newly diagnosed Wilms tumor are treated on research protocols, and treatment generally involves multimodal therapy (surgery, chemotherapy, and radiation). (See 'Management' above.)

•The two major research groups that care for the largest number of patients with Wilms tumor are the Children's Oncology Group (COG) and the International Society of Paediatric Oncology (SIOP). Both COG and SIOP use risk-directed therapeutic regimens that are assigned based upon results of the initial staging, histologic classification, and molecular studies (table 3 and table 4). (See 'COG approach' above and 'SIOP approach' above.)

●Prognostic factors – Several prognostic factors at the time of initial diagnosis are associated with an increased risk of tumor recurrence or death. These include:

•Tumor histology (see 'Tumor histology' above)

•Tumor stage (table 2) (see 'Tumor stage' above and "Presentation, diagnosis, and staging of Wilms tumor", section on 'Staging')

•Molecular and genetic markers (see 'Molecular markers' above)

•Age >2 years (see 'Age' above)

●Treatment modalities – Multimodal therapy for patients with unilateral Wilms tumor (stages I through IV) includes (table 3 and table 4):

•Surgical excision for all patients with resectable tumors

•Chemotherapy for all patients, except those with very low-risk tumors

•Radiation therapy as indicated by stage and/or histology (table 3 and table 4)

●Unilateral Wilms tumor – Because the authors' institution is a member of the COG group, they treat patients with unilateral Wilms tumor with initial surgical resection, followed by chemotherapy (table 3). An alternate approach (used by the SIOP group) is to treat patients preoperatively with chemotherapy prior to surgical resection (table 4). Patients treated by either approach have excellent and comparable clinical outcomes. (See 'Difference in approaches of COG and SIOP' above.)

●Bilateral Wilms tumor – For patients with bilateral Wilms tumor (stage V) (table 2), we suggest preoperative chemotherapy with vincristine, dactinomycin, and doxorubicin rather than other regimens (Grade 2C). Following preoperative therapy, we suggest kidney parenchymal-sparing resection rather than more extensive bilateral kidney surgery (Grade 2C). Postoperative chemotherapy with or without radiation is selected based on treatment response to neoadjuvant chemotherapy, histopathology, and tumor stage. This approach reduces the risk of end-stage kidney failure by facilitating early nephron-sparing surgery and preserving kidney parenchyma. (See 'Bilateral kidney involvement' above.)

●Anaplastic histology – Patients with anaplastic histology (table 1) have a poor outcome. Intensive treatment regimens and new agents are under investigation in this population. (See 'Anaplastic histology' above.)

●Lung metastases – For patients with lung metastases, treatment is tailored based upon the completeness of lung metastasis response after initial chemotherapy. For patients with complete resolution of pulmonary nodules after initial chemotherapy, lung irradiation is generally not needed. (See 'Lung metastases' above.)

●Prognosis – Overall five-year survival for Wilms tumor is 90 percent. Event-free survival (EFS) rates vary among individuals and are primarily dependent upon tumor histology and stage. (See 'Outcome' above.)

●Recurrent disease – The risk of tumor recurrence is dependent on tumor histology. Recurrent disease develops in 15 percent of children with favorable histology and in 50 percent of those with anaplastic features. Post-relapse four-year survival rates are 50 to 80 percent. (See 'Tumor recurrence' above.)

●Complications – Long-term complications are related to the type and intensity of the treatment regimen. Patients with diffuse anaplastic tumors or higher-stage tumors receive more aggressive chemotherapy and radiation therapy and are at greater risk for both early and late adverse effects. (See 'Complications' above.)

●Follow-up – Long-term follow-up is necessary for survivors of Wilms tumor for ongoing surveillance for tumor recurrence and late treatment-related complications. This includes scheduled imaging of the chest (chest computed tomography [CT] or chest radiography) and abdomen (abdominal ultrasonography or CT). (See 'Follow-up' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Jodi A Muscal, MD, who contributed to an earlier version of this topic review.