Chapter 456. Renal Tumors

Approximately 550 children and adolescents are diagnosed with renal tumors in the United States each year. The vast majority of these (about 500 cases yearly) are nephroblastomas, also known as Wilms tumors (WT). Other types of renal tumors of childhood include rhabdoid tumor of the kidney (RTK), clear cell sarcoma (CCS), renal cell carcinoma (RCC), and congenital mesoblastic nephroma (CMN). The relative incidence of each tumor is presented in eTable 456.1. It is important to note that the incidence of the tumor type varies with the age of the patient. Wilms tumors account for 95% of renal tumors of childhood, whereas adolescents over 15 years of age are more likely to be diagnosed with RCC than WT (incidence RCC .07 versus WT .04 in adolescent patients). A renal mass in an infant less than 3 months of age is likely to be a congenital mesoblastic nephroma, or rhabdoid tumor. Still, the overall incidence of each of these other renal tumors of childhood (CCS, RTK, CMN) is quite low. Therefore, this chapter will focus primarily on WT as the major renal tumor of childhood.

Clinical research in pediatric renal tumors is an exemplary model of how national and international collaborative studies have been able to make marked improvements in outcome over time. Coordination of multimodal therapy for patients with WT has resulted in improvement in overall survival from a nearly uniformly lethal disease in the early 1900s, to approximately 30% in the 1930s, to nearly 90% in the early 21st century. This progress has been largely due to the efforts of several groups, including the National Wilms Tumor Study Group (NWTS) in North America and the International Society of Pediatric Oncology (SIOP) in Europe. Five major studies have been conducted in North America by the NWTS, each providing valuable clinical information to help improve the outcome of future children with WT. NWTS has now been replaced by the Renal Tumors section of the Children’s Oncology Group (COG) which is organizing and implementing the current collaborative trials in North America. These studies involve important cross-disciplinary collaborations between pediatric oncologists, radiation oncologists, pediatric surgeons, pathologists, radiologists, as well as patients and families of patients with Wilms tumor.

Epidemiology, Associated Syndromes, and Populations at Risk

Pediatric renal tumors represent approximately 7% of all childhood cancers; 95% of these are Wilms tumors (WT).1 SEER data identify the peak of incidence of WT as the first 2 years of life, with steadily declining rates with increasing age after that time.2 The majority of WT occur prior to 5 years of age, and WT is very rare in children over 10 years of age. Bilateral tumors tend to present at younger age. Although SEER data from 1975 to 1995 overall showed a slight increased incidence in females and blacks, the more recent period within that of 1990 to 1995 shows no difference in race or sex. A lower incidence in Asian children has been observed both in North America and Europe.3,4

WT has been shown to occur with increased frequency in children with a number of well-described syndromes. These include WAGR (Wilms tumor, aniridia, genitourinary abnormalities, and mental retardation), Denys-Drash, Frasier, Beckwith-Wiedermann, Simpson-Golabi-Behmel, Sotos, and Perlman.5,6 (eTable 456.2) The frequency at which WT occurs in these syndromes ranges from 5% to 90%. A greater incidence of bilateral WT is observed in children with associated syndromes, and age of presentation is overall younger. Several, but not all, of the syndromes are classified as overgrowth syndromes, where macrosomia and often organomegaly are prominent features. Although not termed a syndrome, a clear association of hemihypertrophy and WT is also seen.

All other renal tumors of childhood occur at much lower incidence than WT. Rhabdoid tumors occur almost exclusively in the first 2 years of life, clear cell sarcoma (CCS) sarcoma in the first 3 years, congenital mesoblastic nephroma (CMN) only in infants, and renal cell carcinoma (RCC) nearly always in adolescents and young adults. The overall incidence of these tumors (RCC, CMN, and RT), however, is quite low and has made the clinical study of these tumors challenging.

Pathophysiology

The ontogeny of Wilms tumor (WT) is tightly interlaced with normal kidney development. In 1899, Max Wilms established the now classic description of WT as a mixed tumor, with his meticulous histologic description of the 3 defining components: epithelium, blastema, and stroma.7 He also contributed the important insight that the tumor developed from a common and macroscopically undifferentiated germ cell. The classic triphasic morphology of WT reflects phases of normal nephrogenesis, with the blastemal component resembling condensing nephrogenic mesenchyme and the epithelial portion resembling glomeruli. This link helps explain the correlation between WT and renal abnormalities, as well as the presence of persistent embryonic renal tissue (termed nephrogenic rests) in patients at risk for WT. The study of the genetics and biology of WT overlaps with investigation into normal renal organogenesis and has provided much insight into connections of tumorigenesis and normal development.6

Genetics

The association of Wilms tumor (WT) and several known genetic syndromes has spurred much productive research into the genetics of WT. WT was one of the original tumors employed by Knudsen to develop the “two-hit hypothesis” of oncogenic transformation.8 He postulated that the development of WT required two independent, rate-limiting genetic events. Genetically susceptible children would carry a primary germline mutation, either inherited, or de novo. Tumorigenesis would occur when a second “hit” occurred. The young age of presentation of patients with WT, and the even younger age of presentation of children with bilateral WT and associated syndromes supported his hypothesis. The Knudson model has been confirmed by additional studies in a number of other tumors, which also demonstrate that the genetic loss of 2 alleles of a tumor suppressor gene can result in oncogenesis.

Years of additional study have revealed deeper complexity in the story of WT genetics. WT1 was the first gene mutation found in WT and was confirmed to be a tumor suppressor gene. WT1, found at 11p13, has been cloned and found to be a transcription factor of the zinc finger family.9 WT1 also has been found to serve an essential role in renal development in mice10 and is highly expressed in the developing kidney, gonads, spleen, and mesothelium. Variable WT1 mutations occur consistently in several of the WT-associated syndromes, notably in WAGR, Denys-Drash, and Frasier syndromes.11,12 WT1 mutations, however, are found in only 10% of sporadic tumors.

A second WT gene has been long postulated to occur at 11p15.13 Beckwith-Wiedermann syndrome maps to this location, and loss of heterozygosity (LOH) occurs at 11p15 in some cases of sporadic WT as well. Intriguingly, when LOH occurs at 11p15, it is consistently the maternal allele that has been lost, and the paternal allele duplicated.14 Despite the promise of this region, and intensive investigation, no specific WT gene has yet been found at this location. A number of imprinted genes have, however, been identified in this region, including IGF2, H19, and p57kip2. Fifteen percent of Wilms tumors have mutations in the B-catenin gene. In most cases, WT1 is also mutated in tumors that have mutations of B-catenin.15B-catenin is linked to the Wnt signal transduction pathway that is involved in normal nephron development.

WTX is the most recently described WT gene and has been shown to be mutated in almost one third of WT.16 It is an X chromosome gene and, therefore, can be inactivated by a single hit. The function of WTX is yet unknown, but high levels of expression have been identified in the developing lungs, brain, kidney, and spleen of mice.

LOH of 1p and of 16q have been observed in WT. NWTS-5 prospectively analyzed incidence and clinical outcome of patients with these areas of LOH. LOH of 1p and 16q was found to occur together in about 5% of favorable histology WT and was correlated significantly with increased risk of relapse and death.17 This biologic marker is being used in the first Children’s Oncology Group (COG) renal protocols to stratify treatment for patients with WT.

Rhabdoid tumors (RT) occur most commonly in the kidney but can occur in soft tissues throughout the body as well, where they are referred to as extrarenal RT. These tumors can occur in the central nervous system as well, where they are termed atypical teratoid/rhabdoid tumors. Previously grouped by similar histology, the relationship between all these tumors was cemented by the discovery that the majority of these tumors from all anatomic sites have biallelic inactivating mutations in hSNF5.18 Inactivating mutations of hSNF5 have been found in 70% of these tumors, and an additional 25% of the tumors have been found to have absence of expression of the SNF5 protein, even with no identified mutation.19 Germline mutation of one allele is the basis of the majority of “rhabdoid predisposition syndrome,” or familial RT. Patients with RT and this constitutional deletion usually present at a very young age and can present with tumors in more than one location simultaneously.20,21

The cellular or aggressive variant of congenital mesoblastic nephroma (CMN) has been shown to bear the t(12;15)(p13q25) or ETV6-NTRK3 gene fusion. This finding has linked CMN to infantile fibrosarcoma, where this gene fusion was first identified. These tumors are morphologically similar and are now clinically being treated with the same chemotherapy regimen.

Clinical Features and Differential Diagnosis

The classic presentation of a child with Wilms tumor (WT) is the sudden discovery of a large abdominal mass by a parent while bathing or dressing the child. The mass can be quite large at presentation, with obvious abdominal distension, even with a clear history of sudden onset (eFig. 456.1). Children can also present with symptoms of constipation, including pain and distension. Hematuria, hypertension, and anemia are also occasional presenting symptoms. Rarely, spontaneous rupture of WT can occur and can result in presentation with sudden pain and severe anemia due to bleeding in and around the tumor.

The suspicion of a new abdominal mass should prompt appropriate imaging to distinguish mass from other organomegaly (largely hepatomegaly or splenomegaly), to rule out common entities such as constipation, and to identify the origin of the mass, if present. A more complete differential diagnosis of abdominal mass in children is presented in eTable 456.3.

The presence of any features common to genetic predisposition syndromes for WT should increase the clinical suspicion of the likelihood of the diagnosis of WT (see eTable 456.2).

Diagnostic Evaluation

History and Physical Exam

Patients with a suspected renal mass should be asked a thorough history, including any family history of cancer predisposition, congenital anomalies or urogenital defects, as well as birth and developmental history of the child. A physical exam should include notation of hypertension, any anomalous features, and careful plotting on a growth chart. Typical physical exam finding is of a palpable, firm, nontender, smooth abdominal mass that rarely crosses the midline, and usually does not move with respiration, helping to differentiate from splenomegaly or neuroblastoma.

Imaging

All patients with suspected renal tumors should undergo radiologic imaging. Abdominal radiographs can be obtained quickly and can often identify the presence of a mass. Calcifications, seen in some cases of neuroblastoma and germ cell tumor, can also be seen on plain films. Once the presence of a mass is confirmed, ultrasound (US) can be used for further characterization. Cystic lesions are very well defined by US, as is vascular involvement. Thorough examination of the renal vasculature and inferior vena cava (IVC) is important for the identification of presence and extent of intravascular tumor thrombus. The contralateral kidney can be examined for presence and function, evidence of bilateral tumor, or nephrogenic rests. US has the advantage of avoiding radiation exposure and the need for sedation in young children.

Identification of a solid mass by ultrasound prompts further imaging by computerized tomography (CT). Both abdomen and lungs should be imaged by CT if a malignant process is suspected. Studies of pulmonary lesions in Wilms tumor (WT) that are not seen on chest x-ray, but are seen on CT (known as CT-only lung mets) have demonstrated the clinical importance of identifying these lesions. Patients with CT-only mets have been found to have a higher rate of relapse than patients with no CT chest lesions.22eFigure 456.2 demonstrates CT findings of a patient with Stage IV favorable histology (FH) WT, presenting with a large renal mass and multiple large pulmonary nodules.

Magnetic resonance imaging (MRI) has been established as useful in some patients with WT.23 MRI can be critical in defining vascular involvement, particularly the level of involvement of tumor thrombus within the inferior vena cava. MRI is also being employed in patients with bilateral nephrogenic rests and WT. MRI may be more sensitive in following and differentiating WT and nephrogenic rests. eFigure 456.3 demonstrates MRI findings of a patient with bilateral anaplastic WT and bilateral nephrogenic rests.

The use of positive emission tomography (PET) scan is an area of active study in WT. PET is emerging as a potentially useful modality in WT and is anticipated to have a role in tumor staging, assessment of response, and distinguishing between benign and malignant lesions.24

Patients with suspected or confirmed clear cell sarcoma should undergo a radionucleotide bone scan as well as brain imaging, as this tumor has a known high rate of bone and brain metastasis. Patients with suspected or confirmed rhabdoid tumor of the kidney should have brain imaging (usually MRI) at diagnosis and at scheduled intervals throughout treatment, as both synchronous and metachronous second tumors can occur.

Laboratory Evaluation

Patients with a suspected renal mass should have a complete blood count and differential, a full electrolyte panel, kidney function testing (blood urea nitrogen and creatinine)25,26 and liver function tests. WT has been associated with acquired von Willebrand disease, and coagulation studies as well as a type and cross should be done prior to any planned surgical procedure. Urinalysis may reveal hematuria. Urine catecholamines will be negative in WT and positive in neuroblastoma, which can present similarly.

Additional Diagnostic Studies

Once a diagnosis has been made and a treatment plan established, a patient may require additional prechemotherapy testing, such as echocardiogram and glomerular filtraton rate (GFR). Patients with WT have been noted to have co-incidence of cardiac defects.

Histology

Although the classic description of Wilms tumor (WT) is of triphasic morphology, including blastemal, stromal, and epithelial elements (eFig. 456.4), tremendous histologic diversity can be seen in these tumors. A variety of cell types and aggregation patterns that are seen in normal developing kidney can be found in WT samples, as well as tissues not normally present in the kidney, such as skeletal muscle, cartilage, and squamous epithelium. It is believed that these varied cell types may arise through the developmental potential of the primitive metanephric blastema. Subtypes of WT also include biphasic tumors (with exclusively blastemal and stromal cells) and uniphasic tumors of epithelial subtype.

Grossly, tumors are often described as friable. Areas of tumor within the kidney are visualized as well-circumscribed or lobular masses, gray or variegated pink in color. Multiple nodules of variable size can often be found. Cystic changes, necrosis, and hemorrhage are commonly seen. An example of a gross specimen of FH WT with several of these features is presented in eFigure 456.5.

A WT is identified by visualization of 3 major histologic components—undifferentiated blastema, mesenchymal stroma (which can contain heterologous elements, skeletal muscle, and cartilage), and an epithelial component. The typical WT is triphasic, with each of these components, but can be either biphasic or monophasic. WT is defined as favorable histology (FH) if no areas meeting the criteria of anaplasia are identified. Anaplasia can be focal or diffuse and is heralded by the presence of gigantic (> 3 × diameter of adjacent cells) hyperchromatic nuclei and polypoid or bizarre mitotic figures (eFig. 456.6). In focal anaplasia, these changes must be confined to restricted foci within the primary tumor, without marked nuclear or mitotic atypia in the remainder of the lesion. Diffuse anaplasia is defined as the finding of localized anaplasia with severe nuclear unrest elsewhere, any anaplasia outside the tumor capsule or in metastases, or involvement of multiple areas within one slide.27

Staging

Different staging systems have been employed by NWTS (National Wilms Tumor Study) and SIOP (International Society of Pediatric Oncology), as the staging relates to timing of primary surgery. NWTS staging has been based on stage prior to chemotherapy.28 SIOP employs a staging system based on examination of the tumor after preoperative chemotherapy. COG has continued the use of the NWTS-5 staging system, with a few modifications based on the results of that trial. Patients are now defined as Stage III if any intraoperative spill occurs or if any form of preoperative biopsy is performed. CT-only pulmonary lesions will be classified as Stage IV disease. Tables 456-1 and 456-2 describe the staging systems currently in use by COG and SIOP.

Treatment

Surgery, chemotherapy, and radiation therapy are all components of the multimodality therapy that has led to much improved cure rates for Wilms tumor (WT), as well as other renal tumors.

A subset of patients deemed very low risk (Stage I FHWT, age < 2 years, tumor weight < 550 gms) may have a good overall survival with surgery alone. All other patients with WT also receive adjuvant or neoadjuvant chemotherapy, with radiation therapy reserved for higher-risk patients.

Surgery

Surgical removal of primary tumor is the basis of cure in renal tumors. In WT, two schools of thought exist on the timing of primary nephrectomy. In North America, upfront nephrectomy is almost universally practiced, unless tumor is deemed inoperable at presentation. Advantages of this approach, advocated by NWTS and COG, include immediate removal of primary tumor burden and opportunity for complete initial pathologic and surgical staging. SIOP trials are based on preoperative chemotherapy, with nephrectomy done after 6 to 12 weeks of chemotherapy. Advantages include easier surgical resection if tumor has decreased in size with chemotherapy and assessment of response to chemotherapy. Disadvantages include the risk of initiation of therapy without a confirmed pathologic diagnosis (and an accepted misdiagnosis rate of up to 5%)29,30 and loss of upfront biologic studies on tumor tissue.

Chemotherapy

A number of chemotherapy drugs are known to be active in renal tumors. eTable 456.4 lists the most standardly used drugs, as well as their mechanism of action and common toxicities. The NWTS and SIOP studies (as well as other cooperative studies) have refined regimens of combination chemotherapy appropriate to stage and histology. Reasonably, tThose with lower-stage disease and more benign histology receive fewer agents for a shorter duration of time. Current COG studies are also using the biologic information of the presence of absence of LOH of 1p and 16q to assign therapy. Table 456-3 outlines current chemotherapy regimens for FH WT being studied in the first COG renal trials. As the overall cure rate of anaplastic Wilms tumors, rhabdoid tumors, and clear cell sarcoma are still unacceptably low, intensification of therapy with higher dosages and addition of different chemotherapy agents are being studied in those tumors.

Radiation Therapy

Radiation therapy has been shown to be very effective against Wilms tumor (WT). This efficacy is balanced against the known risks of the acute and long-term toxicities of radiation therapy. Acute toxicities include hepatopathy, pneumonitis, and cardiac dysfunction, and late effects include risk of second tumors, decrease in function of radiated renal tissue, and late cardiac and lung toxicity.31 Currently, Radiation therapy is reserved for higher-stage FH WT or anaplastic WT. The role of radiotherapy for patients with FH WT with pulmonary metastases is being investigated in the first COG trial for Higher Risk FH WT Patients. Patients with complete resolution of their pulmonary lesions after 6 weeks of chemotherapy will not receive radiation therapy but slow responders will. Current SIOP trials avoid the use of radiotherapy in FH WT patients with pulmonary disease with acceptable event-free survival (EFS), using doxorubicin-containing chemotherapy regimens.

Special Considerations

A special subset of patients are those who present with bilateral tumors (approximately 7% of all WT patients) or whom are at known risk of developing bilateral tumors (those with certain associated syndromes, see eTable 456.2). In these patients, upfront surgery is avoided and partial nephrectomies are often performed in an effort to balance therapy with preservation of renal function. For this reason also, radiation therapy is avoided when possible in patients with bilateral disease.

Complications and Late Effects

Almost all patients with a history of Wilms tumor (WT) will have a single kidney, or, in the case of patients with a history of bilateral WT, significant reduction in functioning renal tissue bilaterally. Therefore, precautions and follow-up appropriate to any child with a single kidney should be put in place. Consideration of restriction of activities with a high likelihood of damage to the remaining kidney is recommended. Blood pressure should be monitored at least yearly, and careful attention should be given to any evidence of hypertension.

The therapies that offer cure for renal tumors carry risks of long-term toxicity. Patients who receive radiotherapy are at greater risk for second tumors. Patients who receive adriamycin have a slight risk of subsequent cardiac toxicity. Etoposide and cyclophosphamide both confer a risk of secondary leukemia or myelodysplasia. Infertility is associated with high cumulative doses of cyclophosphamide. Ifosfamide and carboplatin may contribute to long-term kidney dysfunction.

Prognosis

Prognosis is dependent on histology and stage of tumor. Although overall survival of patients with pediatric renal tumors approaches 90%,32 patients with low-stage and FH WT are most likely to do well. The best prognosis is seen in patients with FH WT < 2 years of age with tumors < 550 grams. Anaplastic histology confers a worse prognosis than FH WT, lymph node metastatic, renal capsular invasion, or renal vessel invasion; loss of heterozygosity of 1p and 16q all confer worse prognosis. Prognosis of survival after relapse is better for patients who did not receive either doxorubicin or radiation therapy for their initial tumor, relapsed distantly, and relapsed greater than 1 year after completion of therapy.33-35

Prevention and Control

Even in the absence of any physical findings, patients who have been identified as having genetic predisposition syndromes such as WAGR or BWS should undergo routine imaging because of their known risk of developing WT. A schedule of ultrasound every 3 to 4 months from birth to age 5 has been suggested for patients with WAGR, and up to age 7 or 8 in patients with associated overgrowth syndromes.36,37 Prophylactic nephrectomy in patients with Denys-Drasch has been suggested but is not universally recommended.

Future Directions

Although much has been learned about the genetics of Wilms tumor (WT), the underlying gene defect in the majority of WT has not been found. Active research in gene expression and expression profiling is underway, with the hope of using these techniques to further stratify patients. Such techniques may be useful in either identifying the patients who may do well with even less therapy than currently considered standard, and identifying those patients who, despite meeting all criteria to classify as standard or low risk, do not respond to current standard therapy.

The cure rates for anaplastic WT, rhabdoid tumor of the kidney, clear cell sarcoma, and some renal cell carcinoma remain unacceptably low. The study of both intensified chemotherapy regimens, and use of novel agents, including biologics, is underway for these patients.

Although reasonable rates of salvage exist for patients with initial low risk disease (treated with 2-drug therapy and no radiation therapy), the overall cure of the remainder of patients with recurrent WT is low. Study of intensified chemotherapy regimens or autologous stem cell transplant may be appropriate for those patients.

As in many pediatric cancers, the motto “cure is not enough” applies to WT. It is hoped that identifying underlying genetics of WT will lead to insight into eventual prevention of this tumor.

I would like to acknowledge the efforts and contributions of Dr. Kirsten Ecklund for the radiology images, Dr. Harry Kozakewich for the pathology images, and Dr. Bradley Linden for the presurgical image.