Pediatric Cancer Screening in Hereditary Gastrointestinal Cancer Risk Syndromes: An Update from the AACR Childhood Cancer Predisposition Working Group

Gastrointestinal (GI) polyposis in children requires careful evaluation based on polyp histology, family history, and other clinical features and should include a genetic evaluation. Patients require ongoing screening and surveillance for polyposis and may also require screening for both GI and non-GI cancers. We describe herein the guidelines for management of cancer risk in childhood GI polyposis syndromes developed by the American Association of Cancer Research (AACR) Pediatric Cancer Predisposition Workshop (1), as updated in 2023. Wherever possible, these recommendations are aligned with existing adult and pediatric gastroenterology guidelines (2–9). Gastrointestinal manifestations of hereditary replication repair deficiency syndromes (including mismatch repair-related disorders), PTEN Hamartoma Tumor Syndrome (PHTS), Li-Fraumeni Syndrome, and DICER1-related Tumor Predisposition Syndrome are included in other papers in this series (10–12). In this report, we focus on the major hereditary GI cancer syndromes with childhood manifestations, including familial adenomatous polyposis, Peutz-Jeghers syndrome (PJS), and juvenile polyposis syndrome (JPS). After each section, we summarize the updates from the 2017 recommendations. We also briefly review the age surveillance begins for adult-onset polyposis syndromes.

Familial adenomatous polyposis (FAP) is a hereditary cancer predisposition syndrome associated with GI polyposis and up to 100% risk of colorectal cancer. There is also risk of childhood malignancy including hepatoblastoma, thyroid cancer, desmoid tumor, and medulloblastoma (13). FAP is primarily caused by heterozygous germline pathogenic variants (PV) in APC which are de novo in 20% to 30% of diagnoses, with overall prevalence of between 1:1,100 and 1:3,700 (13). Children with FAP may also have associated physical features including osteomas, hypertrophy of the retinal pigmented epithelium, dental abnormalities, and epidermal cysts (13). There is a spectrum of colonic manifestations ranging from attenuated FAP (AFAP), with lower polyp burden and later onset colorectal cancer risk (typically patients with germline PV prior to codon 157 or after codon 1,595) to the most severe polyposis, with early onset of polyps and colorectal cancer risk (PV between codon 1,250 and 1,464), and classic FAP (PV in the remainder of the gene) (14). Screening is recommended for GI cancers and extracolonic malignancies, as outlined in Table 1. Prior guidelines have suggested a delay in genetic testing for familial APC variants until closer to age of first colonoscopy; however, given improved understanding of early childhood tumor spectrum, we continue to recommend testing in infancy or early childhood to initiate surveillance.

Surveillance with colonoscopy is a cornerstone of management, given the high lifetime risk of colorectal cancer. Onset of adenomatous polyps is most common in early to mid-adolescence; thus colonoscopy screening should start sometime between 10 and 15 years of age (3, 5, 15). Determination of exact timing should incorporate presence of symptoms, family history, and genotype–phenotype correlations (3, 16). Once initiated, surveillance should continue annually and include ongoing polypectomy, if possible. Onset of upper GI polyps and malignancy develops later than colonic manifestations, thus upper endoscopy is not recommended to begin until age 20 to 25 years. For individuals with AFAP, screening may start later and may also be spaced out if clinically appropriate based on phenotype.

Colectomy is recommended to prevent colon cancer in all patients with FAP apart from those with significantly attenuated disease, who may be managed with colonoscopy surveillance. It is important that colectomy is discussed in an age-appropriate manner with adolescent and young adult patients. Decisions about the timing of colectomy are complex and based on multiple factors, including polyp burden and histology, desmoid tumor risk, patient and family understanding of risks and benefits, quality-of-life, access to care, and socio-economic factors including insurance coverage or national health practices (16, 17). These decisions should therefore be discussed with patient, parent (for pediatric age patients), surgeon, and gastroenterologist, with involvement of psychosocial providers. Surgery at a center with experience in FAP colectomy is crucial, and in patients at higher risk for desmoid tumor, a single-stage laparoscopic surgery is preferred to limit risk of abdominal desmoid tumor (18, 19). Additionally, ongoing surveillance of rectum and/or pouch is required post-colectomy, as there is still risk of cancer in residual rectal tissue (8).

Patients with FAP are at increased risk of papillary thyroid cancer (PTC), particularly the cribriform-morular variant (cmvPTC) histology which is considered pathognomonic for FAP diagnosis. cmvPTC is associated with a low risk of invasive behavior, although multifocality is more common in FAP when compared to sporadic cases (20, 21). The estimated incidence of PTC in patients with FAP falls between 2% and 9%, with a strong female predominance (22–25). These diagnoses are primarily made in early adulthood, although thyroid cancer has been reported in late adolescence (22).

Based on these data, we recommend initiating thyroid cancer screening with ultrasound for all patients with FAP starting at age 16 years and repeating every 2 years. Physical exam is less sensitive for PTC screening, thus was not preferred by this committee. Because patients with FAP have a risk of benign thyroid nodules (estimated to be between 9% and 44%), fine-needle aspiration biopsy should be considered prior to surgery (21, 22, 24, 26). Consultation with a pediatric endocrinologist with experience in the evaluation of thyroid nodules and tumor predisposition syndromes is recommended.

The incidence of hepatoblastoma in FAP may be as high as 2.5% based on retrospective studies and further prospective evaluation of hepatoblastoma risk in FAP is needed (27). Of patients with FAP and hepatoblastoma, 96% presented prior to age of 7 years, and the median age of onset was 20 months (compared to non-FAP median age of onset of 13 months) (27). Given the low morbidity associated with screening, and the significant survival benefit of early identification of hepatoblastoma as well as ability to limit late effects from intensive chemotherapy, we recommend screening for hepatoblastoma in patients with FAP (28). This includes abdominal ultrasound and serum AFP every 3 months from time of FAP diagnosis to 7 years of age. More research is needed on the benefit and burden of hepatoblastoma screening in FAP, and clinicians should work with families to make informed decisions about early in life screening.

Desmoid tumors can cause significant morbidity in patients with FAP, as desmoid tumors can be both locally aggressive and treatment refractory (29). Desmoids tend to be associated with variants between codon 543 to 713 and 1,310 to 2,011 (19). Desmoids are more common in females, during adolescence and early adulthood, during pregnancy, and with family history of FAP-associated desmoid tumors (14, 19, 30, 31). Screening with abdominal/pelvic MRI can be considered 1 to 3 years after colectomy in a patient with a family history of desmoid tumors or a genotype in the at-risk region, to continue every 5 to 10 years.

Medulloblastoma patients now routinely undergo molecular characterization, and FAP is associated with WNT-activated, CTNNB1 mutation negative medulloblastoma which have improved outcomes and tailored therapy (32). The risk for medulloblastoma in individuals with FAP is estimated to be <1% in childhood with some female predominance (33). As such, routine neuroimaging is not recommended. Rather, recommendations include careful annual neurologic exam, educating parents about signs and symptoms of brain tumors (such as severe or worsening headaches, headaches with emesis, and focal neurologic symptoms), and low threshold for rapid evaluation with neuroimaging.

Specific updates to the 2017 AACR FAP recommendations include, most significantly, a stronger recommendation for hepatoblastoma screening. Other updates include variant specific recommendations for desmoid tumor screening, commentary regarding timing of colectomy and FAP-associated medulloblastoma evaluation, the recommendation for thyroid ultrasound for thyroid cancer screening and including endocrinology expertise.

Gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS) is a rare condition caused by PV in the APC promoter 1B and characterized by a propensity toward gastric, rather than colon, cancer not described in 2017 paper (34). Screening with upper endoscopy is recommended to start at age 15 years, with consideration of prophylactic gastrectomy by age 30 to 35 years factoring in presence of fundic gland polyps, dysplasia, and family history of cancer (35). GAPPS does not include a high risk of colorectal cancer but a baseline colonoscopy should be performed once at age 15 years (or at time of diagnosis if later; ref. 5).

PJS is a hereditary cancer predisposition syndrome associated with intestinal polyposis in early childhood, with estimates of prevalence ranging from 1:25,000 to 1:280,000 (36). Clinically, diagnosis results from a variety of presentations, including family history of PJS, GI bleeding or intussusception from PJS polyps, development of distinct pigmented oral freckling (also present in perianal region and dorsum of hands and feet), or rarely with precocious puberty related to ovarian and testicular tumors (4, 37). PJS polyps have pathognomonic broad bands of mucosal smooth muscle. PJS is caused by PV (including intragenic or whole gene deletions) involving STK11 and testing is recommended at time of clinical presentation or, for familial disease, in early childhood to ensure start of recommended screening (38). Screening recommendations for PJS are included in Table 2.

The incidence of intussusception in PJS is more than 20% by age 10 years and more than 50% by age 20 years (39). For this reason, GI screening is crucial to identify early polyps and is recommended to start at 8 years and to include endoscopy, colonoscopy, and small bowel study [either video capsule endoscopy (VCE) or magnetic resonance enterography (MRE; ref. 40)]. Additionally, families should be carefully educated to recognize the signs and symptoms of intussusception and promptly seek urgent care. It is also important that providers recognize the polyp lead-point of an intussusception, as radiographic reduction is difficult if ileocolic and not recommended if ileoileal. As the malignant potential of PJS polyps is unclear, an interval of 2 to 3 years between GI evaluation is recommended if polyps are identified (4, 41). If no polyps are identified on the initial studies in a child who is 8 years of age or older, it is considered safe to conduct next GI surveillance at 18 years of age provided no symptoms develop in the intervening years.

Children with PJS are also at risk of developing sex-cord tumors of the ovaries and testes (41). In male patients, there is approximately a 9% incidence of large-cell calcifying Sertoli cell tumors of the testes (LCCSCT) which have a low risk for malignant transformation, thus do not necessarily require intervention. However, LCCSCT may be associated with increased estrogen secretion with resultant gynecomastia, macro-orchidism, and growth acceleration (41). For this reason, patients should be screened with annual physical exam with attention to linear growth, breast tissue and testicular volume. Aromatase inhibitors may decrease advancing gynecomastia and linear growth (42). There is no known utility of screening testicular ultrasound, as testicular microlithiasis associated with LCCSCT does not have a clear correlation with testicular cancer risk in pediatric patients (43).

In females, there is risk of sex cord tumors with annular tubules (SCTAT), with incidence between 10% and 20%. SCTAT are benign and treated with unilateral salpingo-oophorectomy, though they can also present with growth acceleration and precocious puberty (44). In female patients with PJS, in the setting of precocious puberty, ultrasound of the pelvis is indicated to identify any masses, and referral to endocrinology should be made (42). In adulthood, patients are also at risk for breast, gynecologic, pancreatic, and lung cancers with screening for these malignancies starting at or more than 25 years (45).

Specific updates to the 2017 AACR PJS recommendations include a clear recommendation regarding intussusception education, the use of MRE if VCE is not available for small bowel screening and clarified recommendations regarding ovarian and testicular screening in childhood.

JPS is a hereditary polyposis syndrome in which individuals develop hamartomatous, juvenile-type polyps (JP) in the upper and lower GI tract, with a prevalence of between 1:16,000 and 1:100,000 (46, 47). Patients may meet clinical criteria for the syndrome, and/or they may have a germline PV in BMPR1A or SMAD4. Clinical criteria include at least one of the following: (i) at least five pathologically defined JP in the colon; (ii) JP in both the upper and lower GI tract; (iii) any number of JP and a family history of JPS (48). Cancer incidence in JPS is estimated to be between 10% and 50% for colorectal cancer and up to 20% for gastric cancer, although as described below, cancer risks vary with genetic etiology (1, 49–51). For patients with familial disease, genetic testing is recommended prior to 15 years or earlier if onset of symptoms.

Screening via endoscopy and colonoscopy is necessary to identify polyps prior to progression to malignancy; additionally, given the inflammatory nature of these polyps, removal prevents bleeding and subsequent anemia. Screening should start at onset of symptoms such as rectal bleeding or anemia, or if asymptomatic starting at age 12 to 15 years for colonoscopy and 15 years for upper endoscopy (Table 3). Subsequent upper and lower endoscopy should be continued every year, with option to space to every 2 to 3 years if polyps are not identified as determined by patient’s gastroenterologist. This is aligned with other pediatric professional guidelines, although some suggest starting endoscopy closer to 18 years of age (2, 5, 52).

Patients with clinical JPS without a germline PV in SMAD4 or BMPR1A (termed mutation-negative JPS) have a different clinical phenotype including a lower polyp burden in adulthood, earlier age of diagnosis, and less likely family history of disease—there is also a trend toward lower risk of cancer in these patients (53). For this reason, some guidelines for adults recommend spacing out screening in patients for whom subsequent polyps have not been identified after those found in childhood. For those JPS patients with mutation-negative disease, more than 5% of individuals have a family history of JPS (53), and a baseline colonoscopy for parents and siblings >15 years should be considered. The decision to do sibling screening, given lower risk of sibling disease, should be a shared decision with the patient’s family.

Finally, patients with SMAD4 variants have unique features including increased likelihood of gastric polyps and are at risk of hereditary hemorrhagic telangiectasia (HHT). Thus, patients with SMAD4 mutations should be screened per guidelines developed for individuals with HHT not included in the table (54).

In patients with early-onset JPS, or JPS of infancy (JPI), patients may have extremely early onset of symptoms including rectal bleeding and hypoalbuminemia. The molecular basis is chromosomal deletions in 10p23 (which includes both PTEN and BMPR1A genes; ref. 55). Thus, patients require screening with endoscopy and colonoscopy for JPS (as above), as well as screening for PHTS (11). While patients usually present with symptoms, if genetically diagnosed presymptomatically, monitoring with fecal occult blood testing could be considered starting at time of diagnosis.

Specific updates to the 2017 AACR JPS recommendations include the clinical distinction between mutation-positive and mutation-negative JPS and the addition of recommendations for JPI. Additionally, the recommendation for small bowel screening for JPS was removed.

The focus of these recommendations is the management of children with pediatric-onset polyposis syndromes. Adult-onset syndromes are increasingly identified in pediatric patients via broader sequencing methods, like high-risk colorectal cancer panels, or secondary findings from exome or genome analysis. For this reason, we have included Table 4, which has the most updated age-of-onset for cancer screening. In the setting of diagnosis of an adult-onset GI cancer risk syndrome, we recommend reference to adult guidelines, including the National Comprehensive Cancer Network, which are regularly updated (5). Cascade genetic testing for adult-onset genes is not recommended for children, as described in detail in the paper in this series on identification of adult-onset cancer predisposition syndromes in children (56). Genetic counseling at any age may be appropriate and/or beneficial.

Pediatric patients with GI polyps, or genetic predisposition for developing GI polyps, require multidisciplinary management to include close coordination of care with pediatric gastroenterology. Further research is needed to determine the pediatric impact, if any, of adult-onset GI cancer predisposition syndromes. Additionally, as screening is highly dependent upon endoscopic procedures and imaging, future study of novel surveillance methods is greatly needed to improve quality of life in these patients.

K.W. Schneider reports non-financial support from AACR during the conduct of the study as well as personal fees from University of Colorado, University of Pennsylvania, and ACMG and other support from Journal of Genetic Counseling outside the submitted work. H. Lesmana reports personal fees from Pharming outside the submitted work. H. Scott reports other support from Medicare Benefits Schedule; grants and other support from Omico; and grants from Cancer Council SA, NHMRC Australia, Medical Research Future Fund (MRFF), Cancer Australia, RAH Research Foundation, Leukaemia Foundation of Australia, The RUNX1 Research Program, and Tour de Cure during the conduct of the study. A. Bauer reports personal fees from Egetis Pharm and IBSA Pharm outside the submitted work. S.E. Plon reports membership on a scientific advisory panel for Baylor Genetics. No conflicts of interest were disclosed by the other authors.

S.P. MacFarland, K.E. Nichols, C.C. Porter, and S.E. Plon are supported in part by the St. Baldrick’s Foundation. The work of SAS is supported by the intramural research program of the Division of Cancer Epidemiology and Genetics, National Cancer Institute. K.A. Schultz receives funding from National Institutes of Health (R37CA244940), Rein in Sarcoma, and Pine Tree Apple Classic Fund.