About half of Western individuals will be diagnosed with at least one or more adenomatous polyps of the colon during their lives (van Hees, 2015). When ten or more adenomatous polyps are present, they suffer from what is termed polyposis. A UK-based study showed that ten or more adenomas were detected in about 1.1% of patients who underwent a colonoscopy as part of the population screening for CRC (Alexander, 2021).

In older guidelines (from before 2015), the recommendation was that people with ten or more polyps undergo genetic testing. Often, however, no (hereditary) predisposition is detected. The percentage of people in whom no (germline) pathogenic variant (PV) is found, so-called polyposis e.c.i. or colonic polyposis of unknown etiology (CPUE), even seems to increase (Terlouw, 2020), which seems partly due to more accurate endoscopic techniques. The Dutch guideline in 2015 introduced stricter criteria for referral to Clinical Genetics for genetic testing being >10 adenomas under the age of 60 years or >20 adenomas under the age of 70 years.

The question is, however, whether the criteria from the 2015 guideline are efficient. What are the findings in studies in patients with fewer than 10 adenomas, and is the cut-off at 70 years justified?

Description of the Guidelines

The 2015 Dutch guideline on hereditary colorectal cancer (CRC) recommended offering DNA testing to patients with 10 or more adenomas under 60 years of age and 20 or more adenomas under 70 years of age, or a personal history of desmoid tumors, hepatoblastoma, cribriform-morular papillary thyroid cancer, and CHRPE.

The European Hereditary Tumour Group (EHTG) guideline 2024 recommends a multigene germline panel in patients with >20 cumulative colorectal adenomas. The threshold may be reduced to 10 cumulative adenomas if diagnosed under the age of 60 years, or a family history of polyposis or CRC, or an extracolonic manifestation of FAP. This guideline also recommends that multi-gene germline testing (for CRC and polyposis syndromes) is performed in patients with gastrointestinal cancers under 50. Finally, they advise that somatic mutation testing for APC mosaic should be considered in patients with unexplained polyposis, who meet the criteria for testing mentioned above.

The British guideline from 2020 (Monahan, 2020) recommends gene panel testing in cases with a total of ≥10 adenomas <60 years of age, or patients ≥60 years of age with a ≥20 adenomas, or ≥10 with a family history of multiple adenomas or CRC.

The NCCN guideline 2024 (Hodan, 2024) advises testing with a personal history of ≥1 of the following criteria:

Recommend testing if ≥1 of the following criteria: ≥20 adenomas; known pathogenic variant in family; multifocal/bilateral congenital hypertrophy of retinal pigment epithelium (CHRPE); cribriform-morular variant of papillary thyroid cancer; family history of polyposis and family unwilling/unable to have testing.

Consider testing if personal history of ≥1 of the following criteria: between 10–19 cumulative adenomas, desmoid tumor, hepatoblastoma, unilateral CHRPE, or individual meets criteria for serrated polyposis (SPS) with at least some adenomas.

Description of studies

An overview of the outcome of genetic testing in adenomatous polyposis patients is given in Table 1.

Stanich (2019)

Stanich (2019) reported the outcome of constitutional DNA testing in 3789 patients with 10 or more colorectal polyps (75 with hamartomatous polyps). All patients had genetic testing of at least 14 colorectal cancer and polyposis genes (APC, BMPR1A, CDH1, CHEK2, EPCAM, MLH1, MSH2, MSH6, MUTYH, PMS2, PTEN, SMAD4, STK11, TP53). Three additional polyposis genes (GREM1, POLD1, and POLE) were evaluated in only a subset of patients, as they were added to the multigene panel assay during the study period. In addition, 1867 patients with between 20–99 colorectal polyps and 270 100+ colorectal polyps were tested. The adenomatous polyposis cohort was further stratified by age at testing. It was decided a priori to exclude any age groups with <10 patients within this cohort. Also, the low-risk variants APC p.I1307K and CHEK2 p.I1517T, as well as monoallelic MUTYH mutations, were not considered as mutation carriers in this analysis as they are not known to result in a polyposis phenotype.

Of the 1342 patients with between 10–19 adenomatous colorectal polyps, a total of 31 of these patients (2,1%) had a PV in one of the adenomatous polyposis genes, (10 in APC (1%), 10 MUTYH biallelic (1%),1 in POLE (0,1%) and 1 patient  had CMMRD (0,1%)), 13 (1%) had a PV in a hamartomatous polyposis gene (2 in SMAD4 (0,1%), 5 in BMPR1A, 3 in PTEN (0,3%), 1 in STK11, 2 in GREM1) and 41 had a PV in one of the MMR genes (3,3%), 17 in CHEK2 (1,3%) and 1 in CDH1.

In the 1637 patients with between 20-100 adenomatous polyps a PV in adenomatous polyp gene was found in 139 patients (8.5%), a PV in a hamartomatous polyposis gene in 40 (2.4%) and 19 in an MMR gene (1.1%), 3 in TP53 (0.2%), 22 in CHEK2 (1.3%) and 5 in CDH1 (0.3%),

In 219 patients with more than 100 adenomatous polyps, 90 patients (41%) had PV in adenomatous polyp genes, 9 (4.1%) in a hamartomatous gene, and only 2 PVs were found in an MMR gene, 1 in TP53 and 4 in CHEK2.

Authors conclude that the diagnostic yield of pathogenic variants remained above 5% in all ages and cohorts, despite a decrease with age. In the multiple adenoma cohort, the yield was higher in those patients with a personal or family history of cancer. In patients with 10–19 adenomas, though only 2.1% had a PV in the “traditional” polyposis predisposition gene, another 1,1% had a PV in a hamartomatous polyposis gene, 3.3% had a PV in one of the MMR genes, and 1.3% had a PV in the CHEK2-gene. The question is whether or not these MMR gene variants would also have been found through CRC tumor testing. Interestingly, in the supplemental files, the percentages of patients with polyps and no history of CRC are mentioned, in the group with 10-19 adenomas, the prevalence of PV in the MMR genes is almost two times lower, around 1.7%, (12/685), but still considerable high and therefore convincing evidence to include the MMR genes also in the polyposis panel. Although in most panels these genes are already included because of also the possibility of biallelic MMR mutation, causing a severe tumor phenotype with adenomatous polyposis, CMMRD.

In this study CHEK2 was also included in the analysis, explaining about 20% of all the PVs found. Still, the prevalence of around 1-1.5% in most groups is slightly more often than in the general (western) populations; in gnomAD (4.1.0 nonUKB), 0.23% is a carrier (https://gnomad.broadinstitute.org/variant/22-28695868-AG-A?dataset=gnomad_r4), in the Netherlands this is estimated to be > 0.6 % (personal note from molecular geneticist). Also, the clinical value of the findings of a CHEK2 variant is questionable since there are no clinical consequences if there is not a family history of breast cancer, according to the Dutch guidelines for breast cancer.

Unfortunately, the table showing the prevalence in the Adenoma Cohorts by Age does not specify which nonpolyposis colorectal cancer gene variant was found per age group, so although the almost 8% PV found in the age group above 80 could concern variant in moderate penetrant risk genes such as CHEK2 or it could highly selected cases based on family history or other cancer history. Notable, the younger age groups have a lower percentage of found variants.

Grover (2012)

A study in 2012 describes the rate of APC and biallelic MUTYH pathogenic variants (PVs) found in 7225 individuals with colorectal polyposis. PVs in APC or common European founder biallelic PVs in MUTYH were identified in 87/970 (9%) individuals with 10–19 adenomas and 559/3253 (17%) individuals with 20–99 adenomas. (Grover 2012) There was an incremental increase in the odds of a PV with an increasing number of adenomas and earlier age at adenoma diagnosis, particularly under the age of 50 years.

Terlouw (2020)

Terlouw (2020) studied the outcomes of DNA testing in 2082 polyposis patients tested between 1992 and 2017. Contrary to the other studies, Terlouw et al, also included individuals without adenomas (n = 336) but with extracolonic manifestations, CRC with KRAS codon 12 G>T transversion, a first-degree relative >10 adenomas and/or multiple primary CRCs. In 4% APC and 1% biallelic MUTYH variants were detected in these 0-adenoma patients.

Also, a relatively large cohort of patients with 1–9 adenomas was included (n = 328; APC tested n = 217 and MUTYH tested n = 309), one APC and six biallelic MUTYH variants were identified (2% variant detection rate). In this group, the APC PV carrier had already developed adenomas by age 20 and had an FDR with >100 polyps. Of the MAP patients, four were affected with CRC between the ages of 39 and 53. Information on KRAS status in tumor DNA was available for one patient, showing a somatic KRAS c.34G > T transversion.

In the group with 10–19 adenomas (n = 406; APC tested n = 324 and MUTYH tested n = 401), three FAP (1%) and six MAP patients (1,5%) were diagnosed who all developed adenomas aged under 60. Approximately 9% APC and 10% MUTYH biallelic PV mutations were found in patients with between 20-50 adenomas; this number is 12% and 18%, respectively, for 50-99 polyps and more than 70% and 7% in case someone has more than 100 adenomas.

No MUTYH or APC variants were found in patients with fewer than 20 adenomas aged over 70 (n = 82). One MAP patient was found in the patients with over 20 adenomas aged over 70 (1/90, 1.1%).

One weakness of this study was that not all patients with low adenoma counts were tested for both APC and MUTYH. Moreover, variants in other genes were not taken into account. Many of the patients were tested for POLE/D, MSH3, and NTHL1 on a research basis, the proven variant carriers were excluded in this study. Possible variants in other genes, such as SMAD4, BMPR1A, and PTEN, might be present, albeit likely only in a small percentage.

Based on this study, in contrast with the outcome of Stanich 2019, testing patients with adenomas above the age of 70 should be taken cautiously since the variant detection rate was 1%. The higher percentage found in the study of Stanich in older age groups (>70 and >80 yrs) is likely explained by also including MMR genes in the panel, which unexpectedly showed germline PVs up to 6-7% likely (exact percentage not given, it was grouped together with CHEK2, CDH1 en TP53 germline variants). More research is needed to see whether this finding of MMR PVs in older aged patients is because of selection bias (polyposis patients with also CRC who did not undergo IHC MMR testing before, for example). Notably, the same group published in 2022 a study (Jain et al.) on adenoma counts in known Lynch syndrome patients, showing that 6% of patients with Lynch syndrome had 10 or more cumulative adenomatous polyps. Authors note that this is similar to rates of multiple adenomas seen in the general population in a recent study by Sullivan et 2022.

For now, it seems too early to propose MMR testing in patients with >10 adenomas, but in case of a history of advanced colonic neoplasia or Lynch related tumor or an affected family history of Lynch related tumors this should be considered.

Also, in general, other more specific circumstances might warrant panel testing for polyposis genes, such as polyps below age 20 and numerous primary CRCs (≥2) and between 5-10 polyps depending on age and in consultation with clinical geneticist.

The 2% heterozygote MUTYH carriers detected in this study is higher than expected based on the 1% prevalence reported in the Exome Aggregation Consortium database but similar to what Grover (2012) found in patients with 20 adenomas [Grover et al]. Although this could mean that MUTYH heterozygotes might possibly have a higher risk for developing adenomas, this has no clinical consequences because the total number is still low, and CRC risk is not significantly increased for MUTYH (Nielsen, 2012).

In patients with <10 adenomas and CRC younger than 50, the detection rate of MAP may also be above 5% (see also submodule Verwijscriteria voor klinisch genetisch onderzoek bij CRC en genpanelanalyse), specifically when a hot spot KRAS G>T is present. Referral and DNA analysis in patients with CRC under the age of 50 are covered in submodule Verwijscriteria voor klinisch genetisch onderzoek bij CRC en genpanelanalyse and will not be discussed separately in this submodule.

Table 1 a-c: outcomes of DNA testing in polyposis patients

Table 1a. Prevalence gene variants in patients with 1-10 adenomas

1b. Prevalence gene variants in patients with 10-20 adenomas

Table 1c. Prevalence gene variant in patients with 20-100 adenomas

 Van Puijenbroek (2012)

In a Dutch study the KRAS c.34G > T mutation was detected in 10 of 210 colon tumors, in one patent biallelic MUTYH PVS were detected (10%). This was in a 61-year-old patient with a cecum carcinoma and three adenomas. MUTYH hotspot mutation screening in 182 patients without the somatic c.34G>T KRAS mutation led to the detection of three monoallelic germline and no biallelic MUTYH mutation carriers.

Aimé (2015)

A French study compared 86 adenomas and 19 CRCs of 30 MAP patients to 135 adenomas and five CRCs of 47 familial adenomatous polyposis (FAP) patients. Secondly, the germline MUTYH gene sequence was analyzed in patients carrying the KRAS c.34G>T in CRCs diagnosed between 2008 and 2012. The c.34G>T was present in 39.7% of MAP adenomas versus 1.6% of FAP adenomas (P < 0.01). Sensitivity and specificity for detecting MAP were 39.7% and 98%, respectively. Sensitivity increased with the number of adenomas tested (P = 0.039). KRAS exon 2 analysis was performed on 2239 CRC and 2.2% harbored the c.34G>T transversion. Among 28 carriers of the KRAS c.34G>T mutation, biallelic MUTYH mutations were detected in seven patients (25%). One patient did not have any polyp or family history and did not fulfill criteria for MUTYH testing.

Disel (2024)

A recent study analyzed a large pan-cancer comprehensive genomic dataset from patients profiled (tissue next-generation sequencing) during routine clinical care. Pathogenic MUTYH alterations were identified in over 2% of 229 120 solid tumors in 30 tumor types, including germ cell (3.6%), adrenal gland (3.1%), colorectal tumors (2.7%), and small intestinal tumors (2.6%). Across tumor types, a majority of the MUTYH alterations were predicted to be monoallelic. A minority was biallelic, but this also included germline monoallelic cases with loss of heterozygosity. No difference was made in this study for biallelic germline PV carriers. So, it was not possible to calculate sensitivity or specificity for KRAS testing in this study to find MAP.

Terlouw (2025)

The rate of APC mosaicism was 9.4% in 541 patients with a broad spectrum of polyposis phenotypes. In patients who did not meet the scope of Dutch guidelines, the detection rate was 2.3% (5/219). In patients with ≥20 adenomas before the age of 60 or ≥30 adenomas before the age of 70, the detection rate was ≥10%. Of 34 mosaic patients who underwent an esophagogastroduodenoscopy, 26% were diagnosed with gastroduodenal polyps. In one patient, the mosaic variant was detected in semen, but none of the children tested in this cohort inherited the mosaic variant.

Figuur 1 (submodule Verwijscriteria voor klinisch genetisch onderzoek bij adenomateuze polyposis): Detection rates of APC mosaicism subdivided in adenoma count groups and stratified for age at first adenoma.

400 polyposis patients tested somatically for APC mosaicism, a recurrent splice variant in APC c.835-8A>G was detected in a significant proportion of colorectal lesions. This variant fits a mutation signature caused by the genotoxin colibactin. Further analysis of all APC mutations showed that at least one APC variant was detected in approximately 30% of patients that matched a colibactin-associated mutation signature.

Georgeson 2024 (preprint)

Another recent study of a large cohort of CRCs also describes that occurrence of the signature SBS88 is strongly associated with the presence of the APC splice variant c.835-8A>G and thus further confirmed the association found by Terlouw, in 7.5% (398/5,292) of the intestinal CRCs examined.

No systematic review of the literature was performed. To answer the following question, “What are the endoscopic and histologic referral criteria to perform genetic testing in patients with adenomatous polyposis?”, recent international guidelines were used; the EHTG, British and NCCN guideline (Zaffaroni, 2024; Monahan, 2020; Hodan, 2024). The EHTG guideline is based on a recent structured literature search (Zaffaroni, 2024). Articles of interest that were published after the publication of this guidelines or were otherwise relevant were also included. These include Terlouw (2025), Georgeson (2025), Disel (2024), Terlouw (2020), Stanich (2019), Aimé (2015), Grover (2012), and van Puijenbroek (2012).