Update on surveillance for Wilms tumor and hepatoblastoma in Beckwith-Wiedemann Syndrome and other predisposition syndromes

Abstract

Wilms tumors are commonly associated with predisposition syndromes many, but not all, of which include overgrowth. Several of these syndromes also include a risk of other embryonal malignancies – particularly hepatoblastoma. Guidelines for surveillance in this population were published in 2017 and recently members of the AACR Pediatric Cancer Working Group met to update those guidelines with a review of more recently published evidence and risk estimates. This perspective serves to update pediatric oncologists, geneticists, radiologists, counselors and other healthcare professionals on revised diagnostic criteria, review previously published surveillance guidelines and harmonize updated surveillance recommendations in the North American and Australian context for patients with overgrowth syndromes and other syndromes associated with Wilms tumor predisposition.

Introduction

Embryonal tumors such as Wilms tumor (WT) and hepatoblastoma (HB) are commonly associated with several predisposition syndromes. Although Beckwith-Wiedemann syndrome/spectrum (BWS/BWSp) and WT1-related disorders are the most common, there are many other syndromes with reported increased risk of WT. In some syndromes, there is also an increased risk of HB. This paper will review the evidence of risks of WT and, where relevant, HB in each syndrome and provide recommendations for surveillance.

This is an update of the 2017 consensus statement from the AACR predisposition syndrome and surveillance guideline working group with several key changes to the structure of the manuscript. Not all WT predisposition syndromes include somatic overgrowth as a feature. Therefore, this paper focuses more broadly on all syndromes that have clinical features other than WT that would be detected early in life allowing for surveillance. Since the last consensus statement, several new genes have been associated with WT predisposition (TRIM28, DIS3L2, REST, CTR9, FBXW7, NYNRIN, KDM3B). Variants in these genes either do not have an associated syndrome or that syndrome is not yet defined, and these are characterized in an upcoming publication by the AACR Pediatric Cancer Working Group. Recommendations for surveillance for patients with these variants are discussed in the companion paper. Additional data suggests that in BWS, HB risk ends earlier than previously noted.

While there are existing guidelines for WT surveillance from the SIOP-Europe Genome Working Group(1), and for BWS from a European focused international consensus(2), the guidelines presented here are written primarily from the North American perspective. An important distinction between these two approaches is the degree of risk considered to be sufficient for tumor surveillance. In these guidelines, 1% risk of tumor is the threshold to recommend surveillance (Table 1).

Table 1:

Summary of tumor risks and surveillance recommendations

Syndrome	Molecular variant	Estimated WT Risk	Other neoplasms	WT surveillance?	Other surveillance?	Follows general surveillance protocol?
Beckwith-Wiedemann Syndrome / Spectrum *	IC1 GOM	20%	HB ACC Rhabdo Others	Yes	HB	Yes
	pUPD 11p15.5	8%	HB ACC Rhabdo	Yes	HB	Yes
	IC2 LOM	~1%	HB	Yes	HB	Yes
	CDKN1C	<<1%	Neuroblastoma	No	Neuroblastoma	No
	Others	4% - 10%	HB ACC Rhabdo	Yes	HB	Yes
WAGR	11p13 del	45% - 60%	Germ cell tumor	Yes	No	Yes
WT1-related Disorders	WT1	38% - 43%	Germ cell tumor	Yes	No	Yes
Simpson Golabi Behmel Syndrome	GPC3	5.1	HB	Yes	HB	Yes
Perlman Syndrome	DIS3L2	70%	No	Yes	No	Yes
Bohring Opitz Syndrome	ASXL1	>1%	HB	Yes	HB	Yes
Trisomy 18	T18	>1%	HB	Shared decision making	HB	Consideration of WT surveillance until 12 years of age
Osteopathic Striata with Cranial Sclerosis	AMER1	>5%	No	Shared decision making	No	Consideration of surveillance until 12 years of age
Mulibrey Nanism	TRIM37	8%	RCC Thyroid Cancer Thecomas	Yes	No	Yes
PIK3CA-related Overgrowth Syndrome	PIK3CA	>1% in CLOVES, <1% in others	No	Shared decision making in CLOVES	No	Yes
2p24 dup/2q37 del	2p24/2q37	Undefined	Neuroblastoma	Shared decision making	Neuroblastoma in discussion with family	Yes
Sotos Syndrome	NSD1	<1%	Leukemia Embryonal tumors Pineoblastoma	No	No	No
Weaver Syndrome	EZH2	<1%	Neuroblastoma	No	Neuroblastoma	No

^*Acceptable risk is 1% in North America and 5% in most of Europe. This leads to different risk thresholds for tumor screening guidelines. These guidelines are based on the North American risk threshold of 1%.

Beckwith–Wiedemann Syndrome/Spectrum

Beckwith–Wiedemann syndrome (BWS) is an overgrowth and cancer predisposition syndrome caused by epigenetic and/or genetic alterations(2). It is classically characterized by pre- and postnatal overgrowth manifested through features included organomegaly, macroglossia, abdominal wall defects, lateralized overgrowth, and embryonal tumors(2). BWS has been redefined as a spectrum, or the Beckwith-Wiedemann spectrum (BWSp), which ranges in severity from the classically described phenotype as above, atypical BWSp with fewer and milder clinical features, and isolated lateralized overgrowth (ILO)(2). Many cases of BWSp are mosaic, and clinical features vary between patients. ILO, also referred to as isolated hemihyperplasia (IHH), is considered a more subtle presentation of BWSp leading to a spectrum of features associated with a variety of structural, genetic, or epigenetic abnormalities localized to chromosome 11(3). The most common tumor types associated with BWS are WT and HB; however, additional tumors have been reported, including neuroblastoma (NBL), rhabdomyosarcoma, pheochromocytoma, and adrenocortical carcinoma(2-4).

Disruption of imprinted growth regulatory genes that are normally expressed in a parent-of-origin–specific manner including H19, IGF2, and CDKN1C on chromosome 11p15 cause BWSp(5). Most cases of BWSp (85%) are not inherited but are instead due to epigenetic changes on chromosome 11p15. The most common of these changes is loss of methylation at imprinting control region 2 (IC2) (KCNQ1OT1:TSS-DMR) or gain of methylation at imprinting control region 1(IC1) (H19/ IGF2:IG-DMR)(5). Paternal uniparental isodisomy for part or all of chromosome 11 (pUPD11) can also cause BWSp. In pUPD11 both copies of this region of chromosome 11 are derived from the father. In some cases, the UPD extends to the whole genome(6). These children also need an evaluation for other imprinting syndromes including Angelman syndrome. Also, rare causes of hereditary BWSp include pathogenic variants (PVs) on the maternally derived copy of CDKN1C, maternally or paternally inherited deletions or duplications of the 11p15 region, and chromosomal rearrangements.

Tumor risk by molecular subgroup in BWSp

For BWSp, the overall tumor risk typically falls within the range of 8-12%(2,7). Among reported tumors, 47% are WT and 25% are HB. Neuroblastomas occur less frequently and are primarily seen in patients with CDKN1C PVs. The remaining tumor types are less common and do not meet the risk threshold for cancer surveillance. Individuals with IC1 gain of methylation have the highest overall cancer risk at up to 28%, most commonly WT(2,7). Patients with pUPD11 have a risk range of 16-30%, most commonly for WT and HB(2,7). In patients with genome-wide paternal uniparental isodisomy, cancer risk is likely higher than pUPD11although these cases are too rare to define actual risk(2,6); these tumors also appear to occur later in life than in pUPD11. Individuals with IC2 loss of methylation have an overall risk of 2-3% and most tumors reported are HB(2,7).

Differentiated screening by epigenetic or genetic subtype in blood has been suggested by some groups(2), while others, including the previous AACR guidelines(4), suggest a standardized screening protocol. Recent data have shown that the genetic or epigenetic subtype in blood is not always the same as that found in the tumor or the normal organ tissue from which the tumor arose(7,8). Furthermore, the level of epigenotype mosaicism in the blood and tissue epigenotypes can vary within one individual(7,8), which makes relying on blood epigenotype unreliable for BWSp. We therefore recommend that all patients with BWSp undergo screening for HB and WT except for patients with CDKN1C PVs.

Recent data have shown that there are no reported cases of WT or HB in patients with CDKN1C PVs(9). Out of the 126 patients reported in this group, four of whom had NBL (3.2%)(9). Other tumor types were observed in <1% of cases. We therefore recommend that the subset of patients with BWSp due to CDKN1C PVs be screened for NBL (these screening guidelines are covered in an upcoming publication by the AACR Pediatric Cancer Working Group) and not for HB and WT (which means abdominal ultrasounds for NBL but no AFPs).

WT1-Related Disorders

The WT1-related tumor predisposition syndromes are defined as risk of tumors associated with germline PVs in the WT1 gene and include syndromes also known as Denys-Drash Syndrome,(10) Frasier Syndrome,(11) Meacham syndrome, or Nephrotic Syndrome type 4.(12)^,(13) Although these disorders are now all considered part of a WT1-related tumor predisposition spectrum, eponyms are still commonly used. A number of patients with WT1-related syndromes have differences of sexual development (DSD) and are (46,XY) phenotypic females.(14) Recommendations for management of individuals with these syndromes should include input from professionals experienced in DSD care. Degenerative renal disease, without evidence of initial response to steroids, is a hallmark of the WT1-related tumor predisposition syndromes and may include proteinuria and mesangial or focal glomerulosclerosis. When these syndromes are initially recognized, often through nephrology evaluation, surveillance should be initiated with abdominal ultrasound every three months through age seven as for other predisposition syndromes with risk of WT. If renal disease progresses to end stage renal disease and the child undergoes renal transplant, consideration can be given for nephrectomy of the non-functioning kidney as a means of primary prevention.

Risk of gonadal tumors including gonadoblastoma or dysgerminoma is characteristically observed in patients with Frasier syndrome with splice variants most commonly in exons 9 or 10, whereas those with missense variants in exons 8 or 9 are less likely to develop gonadal tumors but are at risk of WT. The absolute risk of developing gonadoblastoma is difficult to ascertain as for many patients the constitutional exon 9/10 splice site variants are found after a gonadal tumor diagnosis. Patients with streak or undescended gonads are at risk for gonadoblastoma – this risk is particularly high in 46XY females(15). Prophylactic removal of streak ovaries to prevent gonadal tumors could be a consideration in consultation with endocrinologists, urologists, gynecologists, or DSD specialists but a discussion of this procedure is beyond the scope of these guidelines.

In addition to the WT1-related syndromes associated with severe renal disease, WT1 PVs, usually within the zinc finger regions, are observed in a small percentage of apparently sporadic WT(16). These patients may have minor genitourinary abnormalities including cryptorchidism or hypospadias.

The WT1-related tumor predisposition syndromes, however, are distinct from WAGR syndrome which involves a large germline deletion spanning a larger section of the chromosomal region and includes both WT1 and PAX6, the latter causing pan-ocular disease including aniridia. The risk for WT in WAGR syndrome is high (45% - 60%) but occurs in the same timeframe as in other WT predisposition syndromes and therefore tumor surveillance recommendations are the same as for other WT tumor predispositions. Additional screening to evaluate kidney health should extend into adolescence and beyond as these patients are at risk for chronic kidney disease as observed in up to 51% of patients with WAGR syndrome(17).

Simpson-Golabi-Behmel syndrome

Simpson-Golabi-Behmel syndrome (SGBS) is an overgrowth syndrome characterized by hypertelorism, frontal bossing, macroglossia, macrosomia, overgrowth, kidney abnormalities, thick lips, supernumerary nipples, prognathism, developmental delay, organomegaly, and embryonal tumors(18). SGBS is caused by PVs or deletions of GPC3, which maps to the X-chromosome. Most reported cases are males; however, females with skewed X-inactivation have been reported(19).

To date, 189 cases of SGBS have been reported(20). The most frequently observed tumor types were HB (5.1%) and WT (5.1%)(20). Other tumor types were observed in <1% of cases(20). The mean age at which WT were diagnosed was 28 months, ranging from 0 to 84 months. The mean age at which HB was diagnosed was 17 months, ranging from 9 months to 36 months. We recommend screening patients with SGBS with the same approach as BWSp. There are not currently enough data for a formal screening guideline for females with SGBS, however screening should be considered if there are features of SGBS and based on discussion with the family.

Perlman Syndrome

Perlman syndrome is a rare autosomal recessive overgrowth disorder caused by PVs in the DIS3L2 gene that encodes a protein involved in mRNA and microRNA processing(21). Aside from overgrowth, children with Perlman syndrome have multiple significant syndromic features including characteristic facies, fetal hydrops, ascites, hydronephrosis, hypotonia, and developmental delay(22). Based on collections of case series and reports, ~70% of survivors of the neonatal period will develop WT(22,23). Despite prevalent comorbidities, survival after appropriate multimodal therapy has been reported in the literature in children with Perlman syndrome whose tumors have been detected on surveillance(24,25). Therefore, surveillance is recommended for children with this condition to facilitate earlier diagnosis and less invasive management.

Bohring-Optiz Syndrome

Bohring-Opitz Syndrome (BOS) is a condition defined by PVs in ASXL1 and characterized by a distinctive clinical presentation. Patients with BOS have a combination of atypical facial features, microcephaly, growth issues, developmental delays, upper extremity deformities and seizures(26,27). Despite the small number of patients with BOS identified to date, there is concern about increased risk for malignancies in patients with BOS, as ASXL1 has been identified as a tumor suppressor gene(28). Three patients with BOS have been diagnosed with WT: two bilateral WT diagnoses (ages 1.5 and 5 years), and one unilateral WT diagnosis (age 4 years)(29,30). One additional patient was diagnosed with nephroblastomatosis at age 5 months on autopsy(31). Based on available data, the risk for WT is estimated to be >1%(30)^,29. Recently, the number of reported cases of hepatoblastoma reported in patients with BOS have increased to three suggesting an overall risk >1% as the overall number of reported patients with BOS is <300. The cases were diagnosed at ages 15 months, 18 months, and 2 years(32). Given this estimated increased risk, the recommendation is to pursue WT and HB screening for patients with BOS. No other tumor screening is currently recommended based on available literature.

Trisomy 18

Trisomy 18, also known Edwards Syndrome, is a constitutional chromosomal abnormality characterized by an additional copy of chromosome 18. Most often it is characterized by a full extra copy of chromosome 18 present in all cells, referred to as complete trisomy 18. Less commonly, mosaicism for trisomy 18 or constitutional presence of an additional partial segment of chromosome 18q can occur.

Complete trisomy 18 is characterized by distinctive craniofacial/skeletal findings and a constellation of multi-system conditions, disorders, and defects, including cardiovascular, pulmonary, gastrointestinal, genitourinary, and neurologic/central nervous system involvement. The spectrum of features present in affected children confers a poor prognosis with demise in neonates/infants a common occurrence. However, an increasing proportion of children have prolonged survival. Children with mosaicism for trisomy 18 or partial trisomy of chromosome 18q have a variable, often milder, phenotype.

The overall risk of malignancy in children with trisomy 18 is presumed to be at least 1%, with HB and WT predominating(4). However, cancer risk estimation is complicated by the high mortality rate in infancy. To date, there have been about 50 cases of HB and 21 cases of WT reported in children with trisomy 18(33,34). Other benign and malignant tumors have rarely been reported(34,35).

Based on these cancer risks, WT and HB surveillance may be warranted for some affected children. However, the decision whether to screen should be considered within the context of a shared decision-making paradigm incorporating the parental goals of care and the parental and medical provider estimation of potential risks of tumor-directed therapies and interventions(36-38). Depending on the outcome of these longitudinal discussions, surveillance may be initiated, deferred, or discontinued after initiation.

The lack of normal standards for AFP levels in trisomy 18 makes interpretation of levels challenging and so both the absolute level and the short-term trend should be assessed(34). Due to multiple published cases of WT occurring in older children with trisomy 18, predominantly female, renal ultrasounds should be performed every 3 months to 12 years of age.

Osteopathia Striata with Cranial Sclerosis

In 2009, germline loss of function variants in AMER1 were identified as the cause of the very rare condition X-linked dominant osteopathia striata with cranial sclerosis (OSCS) and were thought to not have an increased risk for cancer (39). PV in AMER1 have complete penetrance with variable expressivity for OSCS. Radiographic findings of sclerosis of the skull and long bones are the hallmark characteristics of OSCS(40,41). Common features include macrocephaly, cleft palate, conductive hearing loss, and longitudinal striations of the long bones. Some individuals have mild learning difficulties and extra skeletal anomalies. Males with somatic mosaic PV in AMER1 have similar features as females.

There have been multiple case reports and a series of individuals with germline AMER1 PV and cancer, most commonly WT. Thus far, 6 females are known to have had WT with age at diagnosis ranging from 7 months to 12 years (42-45), 2 of whom had bilateral tumors, and one additional male died neonatally with bilateral multifocal nephroblastomatosis on autopsy(46). Other single cancers have been reported(47)^,(48). As this is a rare disease with <100 cases, the 7 reported cases imply a WT risk significantly higher than 1% and so surveillance is recommended.

Mulibrey-Nanism

Mulibrey (muscle, liver, brain, and eye) Nanism (MN) is caused by biallelic inheritance of PVs in TRIM37, and the most common variant is a Finnish founder variant, c.493-2A>G.(49) Associated features include growth deficiency, constrictive pericarditis, characteristic facial features, insulin resistance, vascular liver lesions, infertility (male and female), and a risk for benign and malignant tumors. To date the largest study regarding cancer risk comes from a Finnish cohort which reported 210 tumors in 66/89 (74%) affected individuals(49). Since that first publication (which formed the basis of recommendations in the previous guidelines), additional information about renal tumors has been reported. An update of data from the Finnish cohort found that 8/101 (8%) patients developed WT(50). A systemic review reported on fourteen cases of WT in individuals with MN. The mean age of diagnosis is 1.6 years (0.6-3.7) and all cases achieved remission(51). The frequency of WT in this population meets criteria for screening. Other cancers have predominately occurred in adulthood, and other than WT, each individual cancer type has been reported in ≤2% of affected individuals(49).

PIK3CA-related disorders/PROS

PIK3CA is one of the most commonly mutated genes in solid cancers(52). Unlike other PI3K-signaling alterations, PIK3CA-related overgrowth syndromes (PROS) are generally benign overgrowth disorders caused by postzygotic activating mutations in PIK3CA typically acquired during day 20 to day 56 postgastrulation(53). They are generally found in mesodermal and neuroectodermal derivative tissues and are not typically germline(53). The PROS phenotype is quite variable and includes several disorders previously described as unique and now unified by PVs in a common gene. These include Klippel-Trenaunay Syndrome (KTS), Megalencephaly-capillary malformation (MCAP) syndrome, and Congenital lipomatous overgrowth, vascular malformations, epidermal nevi, scoliosis (CLOVES) syndrome amongst others. Several studies have specifically reviewed the risk of the development of cancer. Only a few cases of cancer have been reported in patients with PROS including WT(54-56). Out of 483 patients with PROS and a documented PIK3CA PV reviewed in a meta-analysis, there were 6 cases of WT and 2 cases of nephroblastomatosis (1.6%). However, this number may be subject to ascertainment bias as PROS phenotypes can be subtle and detection of a pathogenic PIK3CA variant requires careful selection of tissue for biopsy; additionally, not all of these patients with renal findings are molecularly confirmed suggesting that the denominator could include more unconfirmed cases. It should be noted that of these 8 patients, at least 7 of them had the CLOVES phenotype. A separate review of patients with MCAP syndrome written before the causative PIK3CA variant was described found 2/112 patients had developed a WT(57). Four children with PROS were reported to have other tumor types(58). While the risk of WT in PROS could be higher than the 1% surveillance threshold depending on how this literature is interpreted, the risk may be specifically in those with CLOVES syndrome. We therefore recommend a conversation about surveillance in patients with PROS and a CLOVES phenotype with decisions on surveillance for patients made through shared decision-making with the family.

2p24 duplication and 2q37 deletion

Germline 2p24.3 duplication including the MYCN and DDX1 genes has also been reported in some patients with WT, including bilateral and familial cases(59-61). In one case, the duplication was inherited from the mother who had no history of cancer(59). There are also reports of NBL in patients with partial trisomy 2p (including the 2p24 region), most of which include the ALK gene in addition to MYCN(62-68), with one exception(69). Additionally, there have been reports of WT in children with 2q37 microdeletions, particularly with breakpoints at or proximal to 2q37.1(70,71). Due to the rarity of these chromosomal abnormalities, the tumor risk and spectrum is not well-established, and there are no established surveillance guidelines. Surveillance may be considered for the patient and/or close relatives depending on several factors, including the specific genetic abnormality, tumor type, laterality, and age of diagnosis, as well as the family history of tumors and presence or absence of the duplication in family members. If surveillance is agreed upon by the physician and the family, the surveillance recommendations for WT and/or NBL can be used as a guide.

Sotos Syndrome

Sotos syndrome is a disorder of pre- and post-natal somatic overgrowth in which malignancies in childhood have been reported. The most common underlying genetic cause of Sotos syndrome is a PV in NSD1, a gene involved in developmental epigenetic regulation(72) although recently other genes including NFIX and APC2 have been implicated with overlapping phenotypes(73).

Physical characteristics of Sotos syndrome include characteristic facies and pre- and post-natal growth with height and weight often exceeding two standard deviations above the mean(74). Other clinical findings include variable intellectual disabilities including autism.

Leukemias and multiple types of embryonal and other pediatric solid tumors have been reported in patients with Sotos syndrome, however these reports are mostly as single case studies or small series. The broad spectrum of tumors observed in Sotos syndrome is described in the previous AACR series(75) to which more recently a pineoblastoma observed in a single patient has been reported(76) as has a complex hamartoma of skin(77). However, despite the broad spectrum of childhood cancer types reported, the overall cancer risk is low, estimated to not exceed 3-5%. The risk of any one type of cancer including WT is likely 1% or less. Because of the relatively low risk, as well as the diverse types of cancer reported in low numbers, no routine surveillance is recommended. However, the healthcare professional caring for a child with Sotos syndrome should be aware of the reports of cancer with this diagnosis and should promptly evaluate any sign or symptom to suggest a developing cancer diagnosis.

Weaver Syndrome

Weaver syndrome is an overgrowth syndrome initially identified by a clinical presentation of tall stature, variable intellectual disability and distinct facial features(78), that is associated with underlying pathogenic EZH2 variants(79). Somatic mutations in EZH2 have been noted in numerous types of malignancies(80), leading to concern that patients with Weaver syndrome may be at increased risk for tumor development. The most common malignancy reported is NBL, which has been noted in a total of 7 patients(78,81-86). The largest review notes the risk for development of NBL to be ~9% (5/56 patients included in cohort). In contrast to other overgrowth conditions, there are no reports of WT or HB. Screening for NBL is warranted in Weaver syndrome patients, which is addressed in the article by Kamihara et al as part of this series (87). No additional tumor screening recommended based on current data.

Overall approach to cancer screening/surveillance protocols

The AACR workshop recommendations are based on North American healthcare culture and may differ from practices in other regions. In North America, screening is advised at a 1% risk threshold, while Europe typically uses a 5% threshold. These recommendations aim for a standardized, minimally invasive, universally applicable screening protocol to improve early tumor detection and reduce morbidity and mortality.

It is essential to have comprehensive discussions and informed counseling within the context of a patient's specific syndrome and tumor risk, preferably led by knowledgeable healthcare professionals. One should also acknowledge that these guidelines may result in more frequent and extended screenings for some individuals. However, when surveyed, parents of patients with BWSp indicated that screening provided comfort as did their knowledge of their child's actual risk(88).

Age brackets for tumor risk may differ between syndromic and sporadic cases. To simplify screening recommendations, our AACR workshop committee suggests a uniform approach for the established cancer predisposition syndromes with a WT risk above 1% with exceptions noted above. Additional screening for HB via monitoring serum AFP measurement levels is advised for BWSp, trisomy 18, SGBS, and BOS. To aid in interpretation of AFP results, AFP norms for BWSp have been developed(89). These recommendations aim to detect tumors early and cover the age range when 95% of tumors typically appear(90,91). The suggested screening interval is every three months to reduce the risk of interval tumor development. As more data accumulate and genetic testing becomes more accessible and understood, future screening recommendations may become more nuanced, considering genetic determinants and syndromic origins of WT and HB.

WT screening

WT screening is influenced by age-related factors. A comprehensive meta-analysis indicates that the mean age for diagnosing WT in individuals with BWSp is around 24 months, with the majority of cases occurring before the age of 3 and 95% presenting prior to their 7^th birthday(91). Similar age-related incidence has been reported for patients with WT1-related disorders. Compared to patients who did not receive tumor screening, these patients were diagnosed earlier, demonstrating that screening was effective in WT(91).

For WT screening, we propose initiating screening at birth (or upon diagnosis of the specific syndrome) with renal ultrasound every 3 months up to the child's 7^th birthday. In cases where HB is also a concern, full/complete abdominal ultrasounds are recommended every 3 months until the child's 3^rd birthday. Afterward, these patients can transition to renal ultrasound screenings every 3 months until their 7^th birthday. Given the slightly increased risk for adrenal tumors in BWSp, especially in patients with pUPD11 and genome-wide paternal uniparental isodisomy, the adrenal glands should be imaged as part of these ultrasounds.

Additionally, we advocate for biannual physical examinations conducted by a specialist, such as a geneticist or pediatric oncologist. These evaluations should encompass ongoing education about tumor manifestations, reinforcement of the rationale for screening, and adherence to the screening regimen. They should also address other syndrome-specific manifestations.

HB screening

HB screening is strongly advised in the context of syndromes described here with an increased HB risk. The relative risk is substantially elevated in these children, estimated at 2,280 times that of the general population in BWSp (the most well described syndrome)(92). The cumulative incidence of HB in patients with BWSp in the literature showed that most patients presented with HB prior to 30 months of age(90). Based on these results, we recommend screening for HB until the 3^rd birthday (36 months). We recommend HB screening to include alpha-fetoprotein (AFP) measurements and complete abdominal ultrasounds every three months until the 3^rd birthday. This represents a shorter HB screening duration than what was recommended in the 2017 guidelines, with screening decreased from 48 months to 36 months.

Serum AFP screening is highly sensitive for detecting HB and its elevation often precedes HB detection via ultrasound. This screening is recommended for patients with BWSp/IHH, trisomy 18, and SGBS. Screening for HB in familial adenomatous polyposis is included in the upcoming gastrointestinal polyposis paper by the AACR Pediatric Cancer Working Group.

The key for interpretation of AFP measurements is to track the AFP results over time and monitor the normal downward trend, considered in the context of age and clinical status. In addition, AFP norms for BWSp have been developed and can be used to interpret AFP results, as AFP values in children with BWSp tend to be elevated during the early years of life compared to standard pediatric values(89). Such interpretation should ideally be carried out by or in consultation with physicians experienced in AFP monitoring in these syndromes, particularly geneticists and oncologists affiliated with cancer predisposition programs. Small, incremental rises within reference ranges should not trigger additional testing, as these can be attributed to various factors such as intercurrent illnesses or developmental milestones like teething. This underscores the importance of gathering a comprehensive medical history when discussing AFP results.

Substantial increases in AFP values (exceeding 50-100 ng/mL) warrant further investigation, including reevaluation of the most recent ultrasound imaging and typically repeating AFP measurements after 6 weeks. Although different intervals for repeat testing have been recommended in isolated cases, it remains uncertain whether repeating the AFP measurement more frequently than every 6 weeks significantly impacts clinical outcomes. If two successive increases occur, further imaging via MRI or contrast enhanced US is recommended. In cases of markedly larger increases (exceeding 1,000 ng/mL), repeat testing for validation is advised, and if confirmed, expedited additional imaging becomes imperative.

Radiologic considerations

Standard ultrasound is recommended for the regular tumor screening with complete abdominal or renal ultrasounds. Fasting is not typically required as imaging the gallbladder is not the focus of these studies. When a concerning finding is noted on regular ultrasound, in some cases, a contrast ultrasound can be used to clarify these findings, in particular, to differentiate hemangiomas from HB in the liver. The advantages of contrast ultrasounds in young children are that they do not require sedation for imaging. If contrast-enhanced ultrasound is not available, MRI using a hepatobiliary contrast agent is the test of choice for evaluation of liver masses. If MRI is not available, contrast enhanced CT can also be used.

Genetic counseling considerations

The genetic testing and counseling article in this AACR CPWG series addresses the many nuances and updates to testing children for cancer predisposition(93). As treatment of the cancer and future management may be directly impacted by identifying a cancer predisposition syndrome, we support germline genetic testing for all children diagnosed with WT or HB. Specifically, we support recommendations stated in GeneReviews that highlight certain populations for germline testing but also allows for further genetic evaluation all children with WT in conversation with the family and in accordance with institutional standards(94). Recommended testing may include 11p15 methylation analysis for BWSp, SNP array for BWSp and chromosomal deletions or duplications such as 2p24 duplication, as well as single gene or multi-gene panel testing. However, the recognition of features associated with syndromes in patients and families can help guide the differential diagnosis. Clinical suspicion for a syndrome that is often or always due to mosaic genetic changes, such as BWSp or PROS, may necessitate genetic testing on multiple tissues.

Identifying a cancer predisposition syndrome in a child with cancer also allows for cascade testing of parents and siblings(93). Siblings who test positive for cancer predisposition syndrome but do not have cancer should follow the surveillance recommendations put forth in this manuscript. Of note, BWSp is often diagnosed clinically in the setting of negative genetic testing. BWSp occurs sporadically for the most part, but siblings should undergo clinical evaluation if they have any BWSp features. Genetic evaluation should only be performed for siblings when a heritable cause of BWSp is identified in the proband. Many of the other syndromes included in this review, such as BOS and WAGR syndrome, have a high de novo rate, but cascade testing should still be considered.

Final considerations

This article presents an overview of multiple syndromes which include WT as a common risk, with some syndromes known to include risks of other tumors. For simplicity, we recommend a standard surveillance strategy using ultrasounds every three months in early childhood until the 7^th birthday, to cover the reported age range in which 95% of WT develop. These ultrasounds should be renal if the syndrome includes only WT or complete abdominal ultrasounds until the 3^rd birthday followed by renal ultrasounds until the 7^th birthday if the syndrome also includes HB. It is acknowledged that as more is learned about these various syndromes and the consequence of predisposition gene variants on velocity of tumor development, the age-related windows of tumor susceptibility, degree of observed overgrowth, and possible interactions between the predisposition genes with other gene variants or other factors, that more nuanced screening recommendations will be developed. For the rarer syndromes, there may be additional cancer risks beyond WT that have not been reported such that the clinician caring for these children should remain alert for signs or symptoms of other types of malignancy which would prompt evaluation. Lastly, shared decision-making and other personal and familial factors may influence the degree of surveillance to be undertaken in these and other cancer predisposition syndromes. Levels of parental anxiety versus levels of reassurance associated with cancer screening may be considered. Shared decision making should be considered in all surveillance plans but has been specifically emphasized in this review for syndromes with a high very early mortality or in syndromes with very low or non-specific types of tumor risks.

Abbreviations

BOS

Bohring-Opitz Syndrome

BWS

Beckwith-Wiedemann Syndrome

BWSp

Beckwith-Wiedemann Spectrum

CLOVES

Congenital lipomatous overgrowth, vascular malformations, epidermal nevi, scoliosis

DSD

Differences of sexual development

HB

Hepatoblastoma

HH

Isolated hemihyperplasia

IC1

Imprinting control region 1

IC2

Imprinting control region 2

ILO

Isolated lateralized overgrowth

KTS

Klippel-Trenaunay Syndrome

OSCS

Osteopathia striata with cranial sclerosis

MCAP

Megalencephaly-capillary malformation

MN

Mulibrey Nanism

NBL

Neuroblastoma

PROS

PIK3CA-related overgrowth syndromes

pUPD11

Paternal uniparental isodisomy of chromosome 11

PV

Pathogenic variant

SGBS

Simpson-Golabi-Behmel syndrome

WT

Wilms tumor