Genetic differences in familial adenomatous polyposis syndrome in a Hungarian population: A prospective single center study

Abstract

BACKGROUND

Familial adenomatous polyposis (FAP) is a disorder of autosomal dominant inheritance that is responsible for around 1% of colorectal cancer (CRC) cases.

AIM

To determine the mutation profile of FAP-specific to the Hungarian population.

METHODS

This prospective single-center study enrolled patients with clinically suspected FAP or attenuated FAP (aFAP). Whole-exome next-generation sequencing was performed to detect variants of 50 FAP priority genes and 173 CRC predisposing genes or other CRC disease-associated genes. To identify larger deletions and insertions, a multiplex amplifiable probe hybridization technique was used. The identified genes were then classified according to the American College of Medical Genetics and Genomics guidelines.

RESULTS

A total of 26 index patients with clinically suspected FAP (n = 21) and aFAP (n = 5) were enrolled. APC gene alterations were confirmed in 92.31% of the cases (region 1B deletion, n = 2; whole-gene deletion, n = 4; frameshift mutation, n = 2; nonsense mutation, n = 5, and splice mutation, n = 1), with the remaining two cases having CHEK2 and MSH3 gene alterations. According to pathogenicity, 21 cases had pathogenic mutations, 6 cases had likely pathogenic mutations, and 16 cases had variants of unknown significance (VUS). The most frequent of the latter were the POLE (n = 5) and PIEZO1 (n = 4) gene variants.

CONCLUSION

Germline mutations in the APC gene were confirmed in more than 90% of Hungarian patients with clinically suspected FAP. Although the role of VUS genes is unclear, they are highly likely to play a role in the development of CRC.

Key Words: Polyposis syndrome; Genomics; Familial adenomatous polyposis; Genetic testing; APC; Germline mutation; Colorectal cancer

Core Tip: This prospective single-center study assessing the mutation profile of 26 Hungarian patients with clinically suspected familial adenomatous polyposis (FAP) or attenuated FAP confirmed APC gene mutations in over 90% of cases, confirming its critical role in FAP development. Variants of uncertain significance, particularly in POLE and PIEZO1, were also frequently detected and may contribute to colorectal cancer risk, though their clinical relevance remains unclear, highlighting the need for further research and genotype-phenotype correlation.

INTRODUCTION

Genetic factors play an important role in the development of colorectal cancer (CRC), which is supported by the fact that around 20%-30% of cases have a family history that includes at least one first- or second-degree relative with CRC. Although polygenic inheritance has been most frequently observed, single-gene mutations with a high probability of inheritance, which account for nearly 5% of cases, are also of particular importance given that such tumors usually appear at a young age as part of multiple tumor syndromes[1-3]. Hereditary tumor syndromes, the pathomechanism of which is characterized by the presence of germline genetic mutations, represent a phenotypically diverse group of diseases dependent on genetic alterations[1,4-6]. Familial adenomatous polyposis (FAP), which is responsible for approximately 1% of CRC cases, is one of the most well-known and common types of single gene inherited cancer syndrome. The birth incidence of FAP is 1/8300-10000, occurring equally in both sexes, while the prevalence is 1/11300-37600[7]. This condition is characterized by a large number of adenomatous colonic polyps (> 1000) starting at a young age, often in childhood, with a 100% probability of malignant transformation throughout one’s lifetime. FAP shows autosomal dominant inheritance, which is caused by a mutation in the APC gene on the long arm of chromosome 5 (5q 21-22)[8]. The attenuated FAP (aFAP) is considered a phenotypic variant of the disorder wherein only variable reduction in the function of APC gene is detectable, delaying disease onset by around 40 years or later. Furthermore, it is characterized by fewer polyps (10-100) and a lower risk of malignant transformation[8]. Recent rapid advances in genetics and sequencing systems have facilitated the identification of certain disease-specific genes and their mutations responsible for the development of tumor syndromes[2,4,9-11]. Over the past decade, a growing number of studies have shown that appropriate screening protocols can significantly reduce mortality from hereditary tumor syndromes, which has prompted the establishment of international guidelines for both genetic testing and endoscopic screening[10,12-14]. Several countries, including Hungary, have limited access to genetic testing technology for suspected polyposis syndromes and lack a registry of patients and their genetic abnormalities. Unfortunately, Hungary ranks among the countries with the highest CRC incidence and mortality rates in Europe. Over the past three decades, CRC-related mortality in Hungary has steadily increased compared to the European Union average[15,16]. In an effort to improve these epidemiological outcomes, Hungary launched a nationwide, population-based, two-stage CRC screening program in 2018, targeting individuals aged 50 to 70 at average risk. Despite these public health efforts, the identification of hereditary CRC cases, including polyposis syndromes, remains limited in clinical practice. As a result, reliable data on the prevalence of these hereditary conditions in the Hungarian population are still lacking. Therefore, the current study aimed to identify germline genetic variants underlying FAP syndrome through whole-exome sequencing and compare the identified genetic variants with databases and international publications to determine mutations specific to the Hungarian population.

MATERIALS AND METHODS

Involvement of patients

This prospective single-center study was conducted at University of Szeged, Hungary between 2018 and 2021 in close collaboration between Department of Internal Medicine and Department of Medical Genetics. The research was limited to genetic testing of patients with clinically suspected polyposis syndrome caused by FAP or aFAP. The inclusion criteria were: (1) The presence of more than 10 colonoscopy-confirmed colonic polyps, indicating the possibility of polyposis syndrome; (2) With morphological features characteristic of FAP or aFAP; and (3) Histopathological examination confirming the adenomatous nature of the polyps. Patients with the following characteristics were excluded from the study: (1) Histopathology confirmed non-adenomatous polyp, indicating the possibility of hamartomatous polyposis syndrome (e.g., Juvenile polyposis syndrome and Peutz-Jeghers syndrome) or serrated polyposis syndrome; (2) Lack of endoscopic and/or genetic screening findings supporting the presence of FAP or aFAP among first-degree relatives of index patients; and (3) Patients under the age of 18 regardless of the clinical probability of FAP or aFAP. Patients were recruited prospectively after receiving detailed information regarding the study and signing the informed consent forms. All included patients were offered genetic consultation and the possibility of genetic and endoscopic screening of their first-degree relatives, following the recommendations of the current genetic and gastroenterological guidelines[10,17]. Before testing the enrolled FAP/aFAP patients, we also searched for FAP-specific variants in a control group of 123 patients with hereditary ophthalmological, dysmorphological, or neurological diseases who provided prior consent to the storage of their blood samples in a biobank of University of Szeged Department of Medical Genetics and their potential future scientific use, with rigorous protection of their personal data. This study was approved by the National Public Health Centre Institutional Committee of Science and Research Ethics (No. 5503-5/2018/EÜIG). All included patients have signed an informed consent form for the scientific use of their medical data. This study was performed in accordance with the ethical principles stated in the Declaration of Helsinki.

Methods of genetic testing

Whole-exome next-generation sequencing (NGS) was performed using peripheral blood samples according to the manufacturer’s protocol (QAGEN GmbH, Hilden, Germany). The SureSelectQXT reagent kit (Agilent Technologies, Santa Clara, CA, United States) was used for library construction. NGS was performed using an Illumina NextSeq 550 using the 300 cycles mid output kit version 2.5 (Illumina, Inc., San Diego, CA, United States)[18]. The reads were aligned to the hg19 human reference genome using the Burrows-Wheeler Aligner. The identified variants were filtered based on three gene sets, namely FAP priority genes (50 genes), FAP predisposing genes, and other CRC disease-associated genes (173 genes), using panel based bioinformatic evaluation[10,19]. The APC gene is of particular importance in the study of FAP and CRC gene mutations, as mutations in this gene are a major cause of the disease. However, other genes are also involved in the developmental process. Different genes may also be responsible for the development of extraintestinal manifestations during the progression of the disease[20]. The selection of genes was guided by the literature and previous studies, and therefore we focused on genes with a proven role in the development of FAP and CRC. However, genes that have been found in several disease-associated studies but have not been shown to play a role were also included in the panel.

For variant annotation, we used the Genome Analysis Toolkit, which achieved an average on-target coverage of 290.9%, with 97.8% of the targets being covered at least 20 ×. For variant annotation, we used ANNOVAR software (version 2.0.2, July 17, 2017)[21]. The potentially pathogenic variants identified were validated through conventional capillary sequencing (Sanger sequencing method). Analysis of sequencing data and interpretation of the differences found [pathogenic variants, presumptively pathogenic variants, or variants of unknown significance (VUS)] were performed according to the current American College of Medical Genetics and Genomics (ACMG) professional guidelines using Franklin bioinformatics sites (https://franklin.genoox.com accessed January 13, 2023), as well as Varsome and SpliceAl[19,22]. Given that the technique is not suitable for the detection of larger deletions and insertions, a multiplex amplifiable probe hybridization (MLPA) (SALSA-MLPA probemix P043-E1 APC, IVD certified, MRC Holland) technique was used for the identification of these genetic alterations. The analytical sensitivity and specificity of the test was approximately 98%-99%[18,23].

Data collection

The medical records of the patients were collected using the MedSolution medical recorder. The occurrence of gastrointestinal and extraintestinal cancers in the index patient’s family was assessed, and a family tree was created during gastroenterological and/or genetic consultation.

Statistical analysis

Statistical analysis was performed using Microsoft Office 365 Excel software version 2021 (Microsoft Corporation, Redmond, WA, United States). Descriptive statistics was used to describe the clinical and demographic data of the included patients and determine the prevalence of genetic alterations in the study population. Categorical variables were presented as event rates and relative frequencies, whereas continuous variables were presented as means with standard deviations and medians with ranges.

RESULTS

Characteristics of index patients

Prior to the inclusion of index patients, the control group was tested for FAP-specific gene variants, with no gene differences having been found. This study enrolled 26 index patients (7 men, 26.9% and 19 women, 73.1%) with clinically suspected FAP (n = 21) or aFAP (n = 5). The average age of the patients at clinical diagnosis of FAP or aFAP was 30.6 ± 12.1 years (range: 14-61 years; median: 26.5 years), whereas study inclusion and genetic testing were on average 12.7 ± 9.8 years afterward (range: 3-41 years; median: 9 years). Regarding the number of polyps, only 9 (34.6%) patients had less than 100 polyps during the first colonoscopy, whereas the remaining 17 (65.4%) patients all had over 100 polyps, with most having an uncountably high number of polyps. CRC developed in 7 (26.9%) patients, among whom 3 already had malignancies at the first colonoscopy upon FAP diagnosis, whereas another 3 were found to have tumors during screening examinations 31, 12, and 3 years after clinical aFAP diagnosis. In the remaining one patient, early polyp carcinoma was found in the bowel resectate only after prophylactic surgery (restorative proctocolectomy with ileal pouch anal anastomosis). FAP-related extraintestinal premalignant (duodenal polyps n = 4, gastric polyps n = 4, ovarian cyst n = 2) and malignant lesions (desmoid tumor, n = 1; breast cancer, n = 1; ovarian cancer, n = 1; and testicular cancer, n = 1) were reported in 9 (34.6%) patients (Table 1).

Table 1 Clinical characteristics of patients, n (%).

Characteristics of index patients (n = 26)	Number
Gender
Male	7 (26.9)
Female	19 (73.1)
Age at diagnosis (median age)	30.6 (14-61)
Number of polyps
< 100	9 (34.6)
> 100	17 (65.4)
CRC in medical history	7 (26.9)
Extracolonic manifestations
Desmoid cancer	1 (3.8)
Breast cancer	1 (3.8)
Ovarian cancer	1 (3.8)
Testicular cancer	1 (3.8)
Duodenum polyp	4 (15.4)
Gastric polyp	4 (15.4)
Ovarian cyst	2 (7.7)
Indications for surgery
Prophylactic RP-IPAA	12 (46.1)
CRC	6 (23.1)
No surgery	8 (30.8)
CRC in family history	13 (50.0)
Polyposis in family history	20 (76.9)

CRC: Colorectal cancer; RP-IPAA: Restorative proctocolectomy with ileal pouch anal anastomosis.

Detected FAP-specific gene variants

Of the 26 patients analyzed, 19 were confirmed to have a FAP-specific gene mutation following whole-exome sequencing and MLPA techniques, 16 (61.5%) and 3 (11.5%) of whom had FAP phenotype and aFAP phenotype genetic alterations, respectively. Although clinically characteristic features of FAP were present, in 8 cases (30.8%) no clear result was obtained even after performing the MLPA technique, at least no abnormality was found in the APC gene. The negative APC mutation also shows that there is still an unknown area in FAP research and raises the possible role of other predisposing genes. However, in these cases VUS genes were identified, suggesting that these genes may play a role in the development of FAP.

A mutation in the FAP-specific APC gene was confirmed in 16 patients, whereas a mutation in the CHEK2 gene and MSH3 gene, which is also a confirmed FAP pathogenic gene variant, was detected in 2 patients. Several cases were found to have a mutation in exon 16. We also examined one family (IDs 9, 10, 11, and 12) in which all members had whole-gene deletion, with 3 cases having a FAP phenotype and 1 case having an aFAP phenotype (Table 2).

Table 2 Detected APC germline mutations and other mutations: Adenomatous polyposis coli and other familial adenomatous polyposis-specific gene mutations in this study.

APC gene mutations	Patient ID	Transcript number	Exon	Nucleotide changes	Zygosity	Type of mutations	PhyoP100 way score	Phenotypes	Detection method
	1	NM 000038.6		1B region deletion	Heterozygous			FAP	WES
	2	NM 000038.6		1B region deletion	Heterozygous			FAP	WES
	8	NM 000038.6	5	c.505 508del:p.I169fs	Heterozygous	Frameshift		FAP	WES
	9	NM 000038.6			Heterozygous	Genomic deletion		FAP	WES
	10	NM 000038.6			Heterozygous	Genomic deletion		FAP	WES
	11	NM 000038.6			Heterozygous	Genomic deletion		aFAP	WES
	12	NM 000038.6			Heterozygous	Genomic deletion		FAP	WES
	13	NM 000038.6	16	c.3180 3184del:p.Q1062	Heterozygous	Frameshift	0	FAP	WES
	14	NM 000038.6	16	c.C2805A:p.Y935X	Heterozygous	Nonsense	1.441	aFAP	WES
	15	NM 000038.6	10	c.C1248G:p.Y416X	Heterozygous	Nonsense	5.029	FAP	WES
	16	NM 000038.6	16	c.C3340A:p.R1114X	Heterozygous	Nonsense	3.923	FAP	WES
	17	NM 000038.6	16	c.G4057T:p.E1353X	Heterozygous	Nonsense	6.23	FAP	WES
	18	NM 000038.6	14	c.C1660T:p.R554X	Heterozygous	Nonsense	0.753	FAP	WES
	22	NM 000038.6	7	C646-2A>T	Heterozygous	Splice	6.462	FAP	WES
	23	NM 000038.6	5	c.505 508del:pI169fs	Heterozygous	Frameshift	6.5e-06	FAP	WES
	24	NM 000038.6	16	c.C2805A:p.Y935X	Heterozygous	Nonsense	1.441	FAP	WES
Other gene mutation	Patient ID	Transcript number	Exon	Nucleotide changes	Zigosity	Type of mutations	PhyoP100 way score	Phenotypes	Detection method
CHEK2[39]	6	NM 007194.4	4	c.T407C:p.I157T	Heterozygous	Missense	6.65	FAP	WES
MSH3[44,45]	21	NM 002439.5	5	c.998delT:p.Y334Ifs11	Heterozygous	Frameshift	0	aFAP	WES
		NM 002439.5	20	c.G2756A:p.G919D	Heterozygous	Missense	1.000		WES

FAP: Familial adenomatous polyposis; WES: Whole-exon sequencing; aFAP: Attenuated familial adenomatous polyposis.

Our results were evaluated using prediction and clinical databases (ClinVar and Varsome). Patients with confirmed APC gene mutations had the following types of mutations: Region 1B deletion (2 cases), whole-gene deletion (4 cases), frameshift mutation (2 cases), nonsense mutation (5 cases), and splice mutation (1 case). The CHEK2 gene alteration was a missense mutation, whereas the patient with the MSH3 gene alteration had both frameshift (1 case) and missense (1 case) mutations.

Classification of detected gene mutations

During the whole-exome sequencing and MLPA technique, mutations in other genes were also found in our patients. The analysis of sequencing data and interpretation of differences found (pathogenic variant, likely pathogenic variant and VUS) were performed according to the current ACMG guidelines[10]. Patients were classified as follows according to the pathogenicity of the mutations in the genes found: 21 with a pathogenic variant, 6 with a likely pathogenic variant, and 16 with a VUS mutations. According to the study, VUS genes were relevant given their uncial role in the development of the disease. Among these genes, POLE (n = 5) and PIEZO1 (n = 4) were the most frequently detected (Table 3).

Table 3 Classification of the detected gene mutations: Classification of patients based on pathogenicity of mutations in genes found.

Classification	Identified gene mutations	Transcript number	Zigosity	Frequency of gene mutations (%)	Detection method
	NTHL1[25]	NM 002528.7	Heterozygous	2 (7.69)	WES
Pathogenic	APC	NM 000038.7	Heterozygous	19 (73.08)	WES
	MSH3[44,45]	NM 002439.5	Heterozygous	3 (11.54)	WES
Likely pathogenic	CHEK2[39]	NM 007194.4	Heterozygous	6 (23.08)	WES
	POLE[24,25]	NM 006231.4	Heterozygous	5 (19.23)	WES
	FLCN[47]	NM 144997.7	Heterozygous	1 (3.85)	WES
	PIEZIO1[29,30,44,46]	NM 001142864.4	Heterozygous	4 (15.38)	WES
	ATM[46,48,49]	NM 000051.4	Heterozygous	1 (3.85)	WES
VUS	PMS2[55]	NM 000535.7	Heterozygous	1 (3.85)	WES
	PIK3CA[50,51]	NM 006218.4	Heterozygous	1 (3.85)	WES
	RNF43[52,53]	NM 017763.6	Heterozygous	1 (3.85)	WES
	PDGRA[54]	NM 006206.6	Heterozygous	1 (3.85)	WES
	c.3949G>C.pGlu1317Gln	NM 000038.6	Heterozygous	1 (3.85)	WES

VUS: Variants of unknown significance; WES: Whole-exon sequencing.

The POLE gene is responsible for encoding the catalytic subunit of the DNA polymerase epsilon. The enzyme plays a role in DNA repair and chromosomal DNA replication. Mutations in the POLE gene have also been associated with the development of colon cancer[24-26]. Bellido et al[27] investigated the role of POLE gene variants in CRC and FAP patients. Patients carrying POLE exonuclease mutations had colonic adenomas, duodenal adenomas or CRC. The mean age of onset of CRC was 40 years. In another study, brain tumors and ovarian tumors were also observed in the presence of POLE gene variants. In this study, the POLE P436S gene variant, which is pathogenic according to the ACMG classification was tested[28]. The protein encoded by the PIEZO1 gene is a mechanically activated ion channel that links mechanical forces to biological signals. The encoded protein contains 36 transmembrane domains and functions in a homotetrameric form. Notably, studies have shown that the PIEZO1 gene defects were not only associated with dehydrated hereditary stomatocytosis but also related to colon cancer[29,30]. PIEZO1 is expressed in cells and tissues, including tumor cells. The role of PIEZO1 has been investigated in the migration of tumor cells in CRC metastasis and gastric cancer. High expression of PIEZO1 has been observed in CRC tissues, which may serve as a prognostic factor[31]. Furthermore, a role for PIEZO1 has been observed in the mechanism of breast cancer development[32].

Correlation between gene mutations and disease phenotype

Among the 26 patients analyzed, 7 were diagnosed with CRC (patient IDs: 1, 3, 8, 10, 21, 22, and 24), the development of which was attributed to mutations in pathogenic or likely pathogenic genes in the majority of cases. One patient with rectal cancer (ID10) and another with synchronous cancers of the rectum and the hepatic flexure (ID24) had a mutation only in the APC gene. In patient ID22, the APC gene variants were associated with a likely pathogen variant CHEK2 gene mutation, whereas in patient ID8, two VUS gene mutations (FLCN and POLE) were confirmed aside from the APC gene alteration in the background of the sigmoid cancer. In patient ID3, one pathogenic variant, one likely pathogenic variant, and one gene VUS were simultaneously identified in the background of a cecal and ascending colonic cancer, whereas in patient ID1 and ID2, only one gene VUS mutation (PIEZO1) and one APC gene view deletion were detected. Every second patient was younger than 40 upon CRC diagnosis, with an average age of 42.6 years (Table 4).

Table 4 Correlation between gene mutations and tumor phenotype.

Patient ID	Gene mutations	Polyp number	Colectomy	CRC (location)	Age at the diagnosis of CRC	Extracolonic manifestation
1	APC, PIEZO1	> 100	Prophylactic RP-IPAA	Yes (coecum)	24.5	Duodenal and gastric polyp
2	APC, PIEZO1	> 100	Prophylactic RP-IPAA	No		Duodenal and gastric polyp
3	POLE, NTHL1, CHEK2	10-100	Prophylactic RP-IPAA	Yes (coecum, ascending colon)	31	No
4	POLE, NTHL1, PIK3CA, CHEK2	10-100	No	No		No
5	PDGFRA, CHEK2	10-100	No	No		No
6	APC	> 100	Prophylactic RP-IPAA	No		Duodenal and gastric polyp
7	POLE	> 100	Prophylactic RP-IPAA	No		Ovarian tumor
8	APC, FLCN, POLE	> 100	Therapeutic RP-IPAA	Yes (sigmoid colon)	50.5	No
9	APC	> 100	Prophylactic RP-IPAA	No		No
10	APC	> 100	Therapeutic RP-IPAA	Yes (rectum)	49.5	No
11	APC	10-100	No	No		No
12	APC	> 100	Prophylactic RP-IPAA	No		No
13	APC, ATM MSH3	> 100	Prophylactic RP-IPAA	No		No
14	APC	10-100	No	No		No
15	APC	> 100	Prophylactic RP-IPAA	No		No
16	APC, CHEK2, PIEZO1	> 100	Prophylactic RP-IPAA	No		Breast cancer
17	APC	> 100	Prophylactic RP-IPAA	No		No
18	APC, PMS2, CHEK2	> 100	Prophylactic RP-IPAA	No		Duodenal polyp
19	MSH3	10-100	No	No		No
20	APC	10-100	No	No		No
21	MSH3 (pathogenic), MSH3	10-100	Therapeutic RP-IPAA	Yes (rectum)	64	No
22	APC, CHEK2	> 100	Therapeutic RP-IPAA	Yes (sigmoid colon)	31	Testis cancer
23	APC, PIEZO1, POLE, RNF43	> 100	No	No		No
24	APC	> 100	Therapeutic RP-IPAA	Yes (rectum, hepatic flexure)	48	Gastric polyp
25	APC	> 100	Prophylactic RP-IPAA	No		Desmoid tumor
26	APC	10-100	No	No		No

CRC: Colorectal cancer; RP-IPAA: Restorative proctocolectomy with ileal pouch anal anastomosis.

APC gene variants were predominantly detected in patients with over 100 polyps (n = 15, 88.2%). Moreover, 9 of 17 patients with over 100 polyps had only APC gene alteration, whereas the remaining 6 cases had an APC gene that was associated with another gene variant (CHEK2, n = 3; PIEZO1, n = 4; POLE, n = 2; FLCN, n = 1; RNF43, n = 1; PMS2 n = 1). Only two patients had a single-gene mutation in POLE or PIEZO1. All three patients with the aFAP specific gene mutation (patient IDs: 11, 14 and 26) had a polyp count below 100, and this phenotype was present in only one patient with the FAP-specific gene mutation (patient ID20). All patients with confirmed pathogenic mutations in the MSH3 and NTHL1 genes fewer than 100 polyps at the index colonoscopy. Upper gastrointestinal polyps were found in five cases, among whom three had both duodenal and gastric polyps with the following gene variants: A mutation of the APC gene apart from PIEZO1 in patient ID1 and ID2 and only an APC gene variant in patient ID6. In patient ID18 with duodenal polyps, the APC gene mutation was accompanied by other pathogenic variants, such as PMS2 and CHEK2, whereas in patient ID24 with gastric polyps, only the APC gene variant was detected. Upper gastrointestinal polyps were diagnosed in 21.1% and 50% of the patients with APC gene and PIEZO1 variants, respectively. APC and PIEZO1 gene mutations were also the most frequent in the extraintestinal manifestations of the tumor syndrome.

VUS genes in families

A total of 9 VUS gene variants were identified in 12 of the 26 patients enrolled herein. Only 1 patient had a VUS gene alteration as a single mutation, whereas the remaining 11 patients had a combination of pathogenic (n = 8), likely pathogenic (n = 1), or both gene alterations (n = 2). Among the 12 patients with VUS gene variants, 6 had at least one first-degree relative involved in the stud, which indicated that three families in the study cohort were affected by VUS gene alterations.

Aside from the APC gene alteration, two variants of the PIEZO1 gene (c.G2546A:p.R849H and c.T2132A:p.M711K) were equally detected in the ID1 and ID2 siblings who showed the same disease phenotype (number of colonic polyps > 100, duodenal and gastric polyposis). Their mother was clinically diagnosed with both polyposis syndrome and CRC but did not undergo genetic testing either within or independent of our study. In contrast to the ID1 and ID2 siblings, the other two patients with the PIEZO1 gene variant had no upper gastrointestinal polyposis based on their clinical records.

In the other family, the mother (ID3) and her two daughters (ID4 and ID5) had different gene variants. In the mother, the POLE (c.C1306T:p.P436S) VUS was described in combination with the pathogenic NTHL1 and the likely pathogenic CHEK2 (c.T407C:p.I157T) gene variant. Among these, the CHEK2 gene alteration was present in both daughters, whereas the NTHL1 and POLE gene alterations were found only in the older daughter (ID4). Simultaneously, both daughters also had a novel gene VUS, the PIK3CA (c.A1544G:p.N515S) gene variant in ID4, and the PDGFRA (c.A2897G:p.H966R) gene variant in ID5. The disease phenotype was similar in all three patients who had 10-100 polyps in the colon and no extraintestinal manifestations. The mother was diagnosed with a duplex right colon tumor (cecum and ascending colon) at the age of 31. Her daughters were under 30 years of age, had underwent regular surveillance colonoscopies and polypectomies, and had no cancer. However, the older daughter (ID4) had a higher clinical cancer risk given that she had several polyps with high grade dysplasia removed, unlike her sister. This may be explained by the fact that the older daughter had a similar mutation profile as her mother, which, unlike her sister, also carries the pathogenic NTHL1 gene variant and a POLE gene VUS. In patient ID7, only the POLE P436S variant was identified and an ovarian tumor developed as an extracolonic manifestation.

In the third family, a father-daughter relationship was found, but the VUS gene (c.G2756A:p.G919D) was identified only in the father (ID21) and was not inherited by the daughter (ID19). The disease phenotype was similar in both the father and daughter, both of whom had 10-100 polyps in the colon. The father was diagnosed with rectal cancer at the age of 64; however, it remains undetermined how the VUS gene influenced the development of this cancer (Table 5).

Table 5 Gene variants of unknown significance.

Patient ID	VUS genes	Transcript number	Exon	Nucleotide changes	Zigosity	PhyoP100 way score	Detection method
1	PIEZO1	NM 001142864	19	c.G2546A:p.R849H	Heterozygous	1.591	WES
	PIEZO1	NM 001142864	16	c.T2132A:p.M711K	Heterozygous	3.907	WES
2	PIEZO1	NM 001142864	19	c.G2546A:p.R849H	Heterozygous	1.591	WES
	PIEZO1	NM 001142864	16	c.T2132A:p.M711K	Heterozygous	3.907	WES
3	POLE	NM 006231	13	c.C1306T:p.P436S	Heterozygous	9.953	WES
4	POLE	NM 006231	13	c.C1306T:p.P436S	Heterozygous	9.953	WES
	PIK3CA	NM 006218	10	c.A1544G:p.N515S	Heterozygous	4.744	WES
5	PDGFRA	NM 006206.6	22	c.A2897G:p.H966R	Heterozygous	6.868	WES
7	POLE	NM 006231	13	c.C1306T:p.P436S	Heterozygous	9.953	WES
8	FLCN	NM 144997.7	4	c.C115A:p.P39T	Heterozygous	1.645	WES
	POLE	NM 006231	2	c.C154T:p.R52W	Heterozygous	1.914	WES
13	ATM	NM 000051	22	c.G3257A:p.R1086H	Heterozygous	6.258	WES
	MSH3	NM 002439	14	c.C2041T:p.P681S	Heterozygous	5.178	WES
16	PIEZO1	NM 001142864	18	c.G2423A:p.R808Q	Heterozygous	7.747	WES
	PIEZO1	NM 001142864	18	c.G2344A:p.G782S	Heterozygous	3.99	WES
18	PMS2	NM 000535.7	6	c.G682A:p.G228S	Heterozygous	5.643	WES
21	MSH3	NM 002439	20	c.G2756A:p.G919D	Heterozygous	1.000	WES
23	POLE	NM 006231	2	c.C154T:p.R52W	Heterozygous	1.914	WES
	PIEZO1	NM 001142864	31	c.T4252C:p.Y1418H	Heterozygous	4.794	WES
	RNF43	NM 017763	3	c.G319A:p.V107I	Heterozygous	6.609	WES

VUS: Variants of unknown significance; WES: Whole-exon sequencing.

DISCUSSION

This has been the first comprehensive genotype study of patients with clinically suspected FAP and aFAP syndrome in Hungary. Despite the small number of patients included, our cohort can be considered representative of the Hungarian population, with our results already reflecting the distribution observed therein. One of the major advantages of the current study is that unlike other similar studies, we not only examined APC and MUTYH-associated gene variants via whole-exome NGS but also extended it to other FAP priority and CRC-associated gene variants. In addition, no data specific to the Hungarian population have so far been published regarding this matter, highlighting the novelty of our study and the opportunity, its provides for comparing the detected genetic mutations with described variants in other countries. Moreover, the extension of our study could serve as a good basis for the establishment of a Hungarian polyposis register.

After confirming the clinically proposed diagnoses of FAP and aFAP, the presented results support our expectations and were also consistent with those presented in the literature. Among the 26 patients studied, 18 (69.23%) had been clearly genetically confirmed to have FAP or aFAP based on the detection of APC gene alterations, with 6 cases having deletion (23.08%). This roughly corresponds to the proportions reported in other studies that performed genetic testing for suspected adenomatous polyposis syndrome[33,34]. In contrast, the study carried out by Kerr et al[35] at the Mayo Clinic, which involves the largest number of consecutive patients so far (n = 1591), found an APC gene mutation rate of only 25%. This could obviously not be attributed to the method of genetic testing but rather to its indication. International guidelines recommend screening for polyposis syndrome in patients with over 10 adenomatous polyps below the age of 60. Therefore, physicians affiliated with the Mayo Clinic frequently submit samples for the APC testing of patients with relatively few polyps to exclude the possibility of aFAP despite understanding that the pretest probability of finding a germline APC alteration was quite low[35]. However, a clear confirmation of an APC gene mutation can confirm the presence of FAP, which can promote the development of a substantial number of colonic polyps and increase the risk of CRC. Accordingly, we failed to find FAP specific gene variants in the control group, which included a relatively large number of patients with hereditary ophthalmological, dysmorphological, or neurological diseases without colonic polyps, similar to those included a Czech study[36].

Consistent with the literature, our study population had an average age of 30 years at FAP/aFAP diagnosis, indicating that the disease started at a young age and that the number of polyps ranged from 100 to 1000. Moreover, several cases of polyposis had been found in the families of the patients involved, confirming that the inheritance pattern of the disease. CRC developed in 7 (26.9%) of our patients, among whom 4 already had tumors upon clinical diagnosis of FAP, whereas the remaining 3 developed tumors in the non-prophylactically removed colon. This finding supports the need for surgical intervention among those with confirmed hereditary tumor syndrome[37]. Furthermore, in the presence of FAP, extracolonic manifestations, such as desmoid tumors, osteomas, or upper gastrointestinal polyps and tumors, should be expected in approximately 70% of cases. Among the patients included herein those who developed extracolonic tumors, such as ovarian tumors and breast cancer, possessed the following gene variants: Ovarian tumor (POLE) and breast cancer (APC, PMS2, and CHEK2). The literature also revealed that the POLE and CHEK2 gene variants may play a role in the development of not only FAP and CRC but also ovarian tumors and breast cancer[38,39]. The most common of these is congenital hypertrophy of the retinal pigment epithelium, which is a benign and asymptomatic retinal lesion affecting approximately 70%-75% of patients[40]. This fact is supported by other research that has investigated the role of POLE in polyposis and CRC, as well as our identification of the POLE P436S gene variant. The POLE P436S variant has been associated not only with CRC and polyposis, but also with ovarian tumorigenesis. Upper gastrointestinal tract polyps (stomach and duodenum) have been reported in half of patients, whereas desmoid tumors have been observed in nearly 10% of patients based on literature data[41]. PIEZO1 expression is observed not only in healthy cells but also in tumor cells which may serve as a prognostic factor. The PIEZO1 R849H and PIEZO1 M711K gene variants identified in our study were not tested in the context of polyposis or CRC. In our study, duodenal and gastric polyps were observed in patients who had both variants simultaneously. It is difficult to draw conclusions due to the small number of elements in our study, but further research may support a role for PIEZO1 in CRC or polyposis syndrome, which could lead to new findings in this research. Similarly to our study, a Turkish study found that 4% of patients had desmoid tumors[42]. Moreover, our study showed that 3.8% of patients had a desmoid tumor and that 15.5% and 15.4% of the cases had some type of gastric or duodenal polyp, respectively[43]. Although these extracolonic manifestations are typical of FAP, the low incidence rates may be attributed to the small number of patients[42,43].

The current study not only sought to identify for APC gene variants but also performed extended analysis to search for FAP priority genes, FAP predisposing genes, and other CRC disease-associated genes. In most of our cases, we could confirm the pathogenic mutations of APC genes. However, unlike the Turkish study mentioned earlier, we were also able to detect other gene variants, such as MSH3, NTHL1 and CHEK2 gene alterations[44,45]. The results for whole-exome sequencing, Sanger sequencing, the MLPA technique was unclear in 8 patients, although such patients had a clinical pathology consistent with FAP. However, these patients were found to have gene VUS, which also occurred in several cases. Some of these gene alterations have been associated to gastrointestinal diseases and CRC. Therefore, the investigation of mutations in these gene VUS may be important given their yet unclear role in disease development, despite being present in patients[27,31,46-55]. Identifying individuals at high risk for CRC, accurate risk assessment, and appropriate selection of screening protocols can significantly reduce the likelihood of developing CRC. By increasing the number of patients included and extending the study nationwide, our study may help define the mutation profile within the Hungarian population and evaluate CRC risk, thereby enhancing the significance of the identified gene variants. Our study also has significant clinical relevance considering that genetic testing of high risk patients and their family members is currently not available in routine patient care.

The major limitation of prospective enrollment in our study was the relatively low incidence of FAP and aFAP, which was further hampered by the fact that genetic testing was only performed in cases at high risk for a hereditary tumor syndrome. Given the relatively small number of patients included herein, we can draw limited conclusions about the overall characteristics of the Hungarian population and associations between the genotype and phenotype. Furthermore, the time-consuming and expensive molecular technique (whole-exon sequencing) and evaluations performed in the current study, as well as occasional unclear results provided by the MLPA technique, prevented us from drawing any conclusions.

CONCLUSION

In conclusion, germline mutations in the APC gene had been confirmed in over 90% of Hungarian patients with clinically suspected FAP. Although the majority of the detected abnormalities were pathogenic or likely pathogenic for FAP and aFAP syndrome, the substantial proportion of gene VUS, of which the POLE and PIEZO1 mutations were the most frequent, needs to be highlighted. The identification of these genes is also particularly important from a scientific standpoint considering their yet unclear role in the development of the polyposis syndrome and CRC. Hence, further studies on phenotype-genotype correlations are required to determine their malignancy risk. Patients with FAP who had previously undergone surgery frequently developed extraintestinal tumors, therefore, the ongoing surveillance, with follow-up every two years, including genetic counseling and clinical evaluation. Additionally, for family members undergoing screening, it is essential to expand the genetic analysis beyond the APC gene to include VUS genes found. Since the clinical significance of the VUS genes identified in our study and their potential impact on CRC risk remains unclear, they do not currently warrant changes to the existing surveillance protocols. A reliable assessment of genotype-phenotype correlations would require a larger patient cohort and long-term follow-up, which could be significantly supported by the creation of a national registry. Extending our study nationwide would not only facilitate the determination of the mutational profile of the Hungarian population but also allow for the evaluation of CRC risk, thereby enhancing the significance of the identified gene variants.