Adenomatous polyposis coli promoter abnormalities in familial adenomatous polyposis and gastric adenocarcinoma and proximal polyposis of the stomach

Abstract

BACKGROUND

In some patients with familial adenomatous polyposis (FAP) who lack pathogenic variants in the coding region of the adenomatous polyposis coli (APC) gene, abnormalities in the promoter region have been identified. These promoter variants may also be associated with a spectrum of APC-associated polyposis syndromes, including gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS), which is characterized by gastric-restricted polyposis with minimal colorectal involvement. However, the clinical and molecular features of APC promoter-associated polyposis have not been comprehensively integrated.

AIM

To determine the genotype-phenotype correlations of APC promoter-region abnormalities.

METHODS

A systematic review of the PubMed database was conducted to identify and summarize evidence on promoter-associated FAP and GAPPS published between January 1993 and December 2025. The reference lists of eligible articles were screened to identify additional relevant studies. Altogether, 35 eligible studies, primarily comprising case reports and case series, were included. These studies were categorized according to the variant type and associated clinical phenotype.

RESULTS

The APC promoter-region abnormalities were classified into four categories. Class I consists of promoter 1B-limited deletions and is predominantly associated with an attenuated FAP phenotype, characterized by colorectal polyposis and cancer. Class II comprises variants affecting the Yin Yang 1 binding site and is subclassified into II-a (GAPPS phenotype), II-b (classical FAP), and II-c (mixed phenotypes), despite affecting a narrow genomic region. Class III includes point variants outside the Yin Yang 1 binding site, with one family exhibiting an attenuated FAP phenotype. Class IV comprises large structural alterations involving the entire promoter region and has been associated with the attenuated or classical FAP phenotype. The available evidence was limited and primarily derived from small case series.

CONCLUSION

Recognizing genotype-phenotype relationships improves understanding of APC promoter-associated diseases and may inform their clinical management. However, further studies are needed to refine these findings.

Key Words: Adenomatous polyposis coli; Promoter 1B; Familial adenomatous polyposis; Gastric adenocarcinoma and proximal polyposis of the stomach; Yin Yang 1; Deletion; Single-nucleotide variant

Core Tip: Abnormalities in the adenomatous polyposis coli promoter region are among the causes of familial adenomatous polyposis in patients without pathogenic variants in the coding region, as well as gastric adenocarcinoma and proximal polyposis of the stomach. In the present systematic review, the adenomatous polyposis coli promoter-region abnormalities were classified as follows: Deletions limited to promoter 1B; variants affecting the Yin Yang 1 (YY1) binding site; point variants outside the YY1 binding site; and large deletions or structural alterations encompassing the entire promoter-region. The variants affecting the YY1 binding site were further subclassified according to clinical phenotype.

INTRODUCTION

Familial adenomatous polyposis (FAP), an autosomal dominant hereditary polyposis syndrome, is primarily caused by pathogenic variants in the adenomatous polyposis coli (APC) gene. However, some patients with clinically diagnosed FAP lack detectable pathogenic variants in the APC coding sequence, suggesting alternative pathogenic mechanisms, including APC promoter-region abnormalities[1,2]. The APC gene has two alternative 5’ untranslated regions (UTRs), which are regulated by distinct promoters, i.e., 1A and 1B[1,3]. APC promoter usage is tissue dependent, with promoter 1B predominating in the gastric and colorectal mucosa and promoter 1A being more active in the brain[1,2]. In line with this tissue-specific regulation, promoter 1A is preferentially methylated in the normal gastric mucosa[4-6].

Gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS) is a rare but increasingly recognized hereditary gastric polyposis syndrome that is characterized by fundic gland polyposis and an increased risk of gastric adenocarcinoma. It typically presents with minimal colorectal or small intestinal involvement[7]. This condition arises from germline point mutations affecting the Yin Yang 1 (YY1) binding site within the APC promoter 1B, thereby placing GAPPS within the FAP spectrum[8].

With increasing recognition of GAPPS, understanding APC promoter-associated polyposis is becoming clinically important. However, the clinical and molecular features of APC promoter-associated polyposis have not been comprehensively integrated. In the present systematic review, we summarize the germline abnormalities affecting the APC promoter region, including APC promoter-associated FAP and GAPPS, and propose a genotype-phenotype-oriented classification framework to facilitate the clinical interpretation and management of these conditions.

MATERIALS AND METHODS

We conducted a comprehensive literature search of the PubMed database to identify peer-reviewed articles published between January 1993 and December 2025, without restrictions on study design. Additionally, the reference lists of all eligible articles were manually screened to identify further relevant studies. The literature search was performed using the following keywords: “APC” AND “promoter” AND (“FAP” OR “GAPPS”). We applied PubMed’s automatic term mapping to capture terminology and indexing variations. In the present study, observational studies (case reports, case series, and cohort-based studies) involving human participants or tissues were included. Although animal and in vitro studies were reviewed to provide mechanistic or contextual insights, they were not included in the primary synthesis of clinical genotype-phenotype associations. We excluded the following studies: Articles written in languages other than English; review articles, editorials, and commentaries without original data; studies not primarily focused on the APC promoter-region abnormalities in either FAP or GAPPS; and studies mainly focusing on extracolonic or nongastrointestinal manifestations (e.g., thyroid or lung cancer) without detailed characterization of the gastrointestinal phenotypes.

The present systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Given that all of the included studies were case reports or case series, the study quality was assessed qualitatively. To enhance transparency and minimize reporting bias, the study protocol was prospectively registered in the International Prospective Register of Systematic Reviews (registration number: CRD420251267200). In the main text, the genetic variants were annotated according to the reference sequence NM_001127511.3 and the GRCh38/hg38 genome build. However, the variant nomenclature presented in the Supplementary Table 1 has been retained as originally reported in the respective publications.

RESULTS

Search results

Figure 1 presents a flow diagram detailing the processes of study identification, screening, eligibility assessment, and exclusion. The initial literature search yielded 65 articles. Among these articles, two were excluded because they were written in languages other than English; 10 were review articles without original data; 18 did not primarily focus on the APC promoter-region abnormalities in either FAP or GAPPS; and six primarily focused on extracolonic or nongastrointestinal manifestations. After applying these exclusion criteria, 29 articles were selected from the initial PubMed search. After manual review of the reference lists, six additional articles were identified. Altogether, 35 articles were finally included in the present systematic review.

Figure 1 Flow diagram summarizing the processes of identification, screening, eligibility, and exclusion of the literature search. APC: Adenomatous polyposis coli.

Classification of APC promoter-region abnormalities

To the best of our knowledge, no report has demonstrated that deletions confined to the APC promoter 1A region alone, without promoter 1B involvement, cause FAP. Based on a comprehensive literature review, we propose a novel classification system for APC promoter-region abnormalities. Table 1 provides a structured summary of each class, including their representative variants and associated clinical features. These abnormalities are categorized into the following four classes: Deletions limited to promoter 1B (class I); variants affecting the YY1 binding site (class II); point variants outside the YY1 binding site (class III); and large deletions or structural alterations encompassing the entire promoter region (class IV). Figure 2 shows a schematic overview of the APC promoter region, variant classes, and their potential phenotypic implications. Class II variants are further subclassified into classes II-a, II-b, and II-c according to their associated clinical phenotypes (GAPPS, FAP, or mixed type). A detailed list of the reported variants and associated phenotypes from each included study is provided in Supplementary Table 1.

Figure 2 Schematic representation of the adenomatous polyposis coli promoter 1B-1A region and proposed mechanisms underlying the phenotypic divergence of promoter-associated variants. A: The relative positions of promoter 1B, exon 1B, promoter 1A, and exon 2 are shown with approximate genomic distances. Class I-IV alterations are mapped according to their reported extents. Promoters 1A and 1B give rise to distinct adenomatous polyposis coli (APC) transcripts with alternative non-coding first exons (5′ untranslated regions). The representative RefSeq transcripts are NM_000038.6 (promoter 1A-derived) and NM_001127511.3 (promoter 1B-derived); B: Expanded view of the APC promoter 1B Yin Yang 1 (YY1)-binding region. The closely adjacent variants within this region have been associated with distinct clinical phenotypes, including phenotypes that are predominantly gastric (gastric adenocarcinoma and proximal polyposis of the stomach), predominantly colorectal (familial adenomatous polyposis), and mixed. The YY1-binding site is indicated within the promoter 1B region. The schematic illustrates the potential mechanisms linking these variants to phenotypic diversity, including an altered YY1-dependent transcription, tissue-specific regulation (e.g., promoter 1A methylation), and residual APC activity differences. These mechanisms are presented as a conceptual model, as the precise molecular basis of phenotypic divergence remains unclear. APC: Adenomatous polyposis coli; YY1: Yin Yang 1; GAPPS: Gastric adenocarcinoma and proximal polyposis of the stomach; FAP: Familial adenomatous polyposis.

Table 1 Classification of familial adenomatous polyposis and gastric adenocarcinoma and proximal polyposis of the stomach associated with adenomatous polyposis coli promoter-region alterations.

Class	Genetic alteration	Typical variants	Predominant phenotype	Gastric involvement	Colorectal involvement	Cancer risk
I	Promoter 1B-limited deletion	Large deletions in promoter 1B	Classical FAP	Absent-mild	Severe	High CRC risk
II-a	YY1 binding site SNV	c.-191T>C, c.-191T>G, c.-195A>C	GAPPS	Severe	Absent-mild	High GC risk
II-b	YY1 binding site SNV	c.-192A>T, c.-190G>A	Classical FAP	Absent-mild	Severe	Moderate CRC risk
II-c	YY1 binding site indel	c.-192_-191delATinsTAGCAAGGG	Mixed GAPPS and AFAP	Severe	Moderate	High GC and moderate CRC risks
III	Non-YY1 binding site SNV	c.-40G>A	AFAP	Absent	Moderate	Moderate CRC risk
IV	Large structural alteration	Chromothripsis and large upstream CNVs	Classical FAP or AFAP	Not available	Severe	Moderate CRC risk

FAP: Familial adenomatous polyposis; CRC: Colorectal cancer; YY1: Yin Yang 1; SNV: Single-nucleotide variant; GAPPS: Gastric adenocarcinoma and proximal polyposis of the stomach; GC: Gastric cancer; AFAP: Attenuated familial adenomatous polyposis; CNV: Copy number variation.

Class I: Deletions limited to promoter 1B

Class I is characterized by deletions confined to the APC promoter 1B region, first reported by Rohlin et al[1] in 2011 in Swedish families with clinically diagnosed FAP who lack pathogenic APC coding variants. Promoter 1B deletions were identified through copy number variation (CNV) analyses, including multiplex ligation-dependent probe amplification (MLPA), and supported by expression studies demonstrating reduced promoter 1B-driven APC transcription through partial gene silencing, consistent with a classical FAP phenotype[1]. Such deletions have since been reported in multiple independent families[1,2,9-18].

The associated phenotype is predominantly classical FAP, with high incidences of colorectal cancer and duodenal neoplasia. Gastric polyposis has been reported in a limited number of families[2,10], and in some cases, earlier disease onset in successive generations has been noted[9,16]. Transcript-specific analyses have demonstrated that promoter 1B deletions selectively reduce the APC-1B transcripts without substantially affecting APC-1A expression, although the degree of compensation varies across tissues and individuals[1,2,9]. This variability may reflect the deletion size differences, particularly when the upstream regulatory elements of promoter 1A are involved[8,9].

Class II: Variants affecting the YY1 binding site

Class II comprises sequence variants affecting the YY1 binding site within the APC promoter 1B region. To date, the reported variants are predominantly single-nucleotide variants, with relatively few small indels described. These variants cluster in the c.-190 to c.-195 region corresponding to the YY1 binding site, which has been functionally validated through binding and reporter assays[8].

Class II-a includes variants associated with the GAPPS phenotype. The reported variants include isolated point mutations, such as c.-191T>C, c.-191T>G, c.-192A>G, and c.-195A>C, as well as compound variants, including c.-195A>C in combination with c.-125delA[8,19-33]. Although c.-125delA lies outside the core YY1 binding site, it segregates with GAPPS in multiple families.

Although GAPPS is predominantly characterized by gastric manifestations[34,35], colonic polyps have also been observed more frequently in affected individuals. Those colonic polyps exhibit nuclear accumulation of β-catenin on immunohistochemistry, similar to that observed in gastric lesions, suggesting that the GAPPS-associated variants may not confer a strictly gastric-limited phenotype[36]. Accordingly, colorectal evaluation at diagnosis, along with regular surveillance, is recommended[37].

Class II-b includes variants associated with the FAP phenotype, including c.-190G>A and c.-192A>T[38-40]. Although gastric lesions, including fundic gland polyposis, have occasionally been noted, they lack the gastric-predominant phenotype characteristic of GAPPS and instead present with colorectal polyposis and cancer[8,38]. Functional studies have shown that the c.-190G>A variant disrupts YY1-dependent transcriptional regulation of the APC promoter 1B based on the dual-luciferase reporter assay results. However, the mechanisms underlying the phenotypic divergence among the closely adjacent variants remain unclear.

Class II-c includes variants associated with phenotypes exhibiting features of both FAP and GAPPS within the same family. To date, only a single family carrying the APC promoter 1B indel variant c.-192_-191delATinsTAGCAAGGG has been reported[41]. This variant was identified through germline testing with a segregation analysis and supported by detailed clinical and pathological evaluations. Affected individuals presented with 30-100 adenomas, consistent with an attenuated FAP (AFAP) phenotype. This indel variant may more profoundly disrupt the APC promoter 1B activity, reducing the APC expression below the threshold required to suppress the development of colorectal polyps.

Class III: Point variants outside the YY1 binding site

Class III comprises the non-YY1 binding site point variants, with only c.-40G>A reported so far[42]. This variant was identified through sequencing of the APC promoter 1B and 5′ UTR and was found to segregate with an AFAP phenotype without gastric polyposis. Functional characterization using luciferase reporter assays incorporating the APC 5′ UTR demonstrated that c.-40G>A creates an out-of-frame upstream AUG start codon, leading to translational repression of canonical APC expression. Although preserved promoter 1A activity has been proposed to explain the attenuated phenotype, direct comparisons of APC expression and tissue-specific regulation are lacking.

Class IV: Large deletions or structural alterations encompassing the entire promoter region

Class IV comprises large deletions or structural alterations involving the entire APC promoter region, reported only in a few studies[16,43]. Large deletions involving the APC promoter region have been identified by CNV analyses, including MLPA, and inferred to encompass promoters 1A and 1B[16]. Another deletion involving the APC promoter region was identified through genetic analyses and shown, based on expression studies, to cause transcriptional silencing of the APC allele. However, its precise boundaries were not defined, and promoter 1B involvement could not be conclusively determined[43]. The clinical phenotype is typically classical FAP with colorectal cancer, although gastric involvement has not been described.

A single case of large-scale promoter-region rearrangement due to chromothripsis has been reported, which was identified by germline whole-genome sequencing combined with optical genome mapping[44]. This event led to the physical separation of promoter 1B from the APC coding region and reduced the APC expression, but it was associated with an AFAP phenotype. Disruption of additional genes in the Wnt signaling pathway may contribute to this phenotypic variability[44]. These findings suggest that extensive structural alterations may lead to heterogeneous clinical presentations, depending on the extent of genomic disruption.

DISCUSSION

The present systematic review synthesizes current evidence on germline APC promoter region abnormalities and proposes a genotype-phenotype-oriented classification framework, highlighting distinct biological subgroups of APC promoter-associated polyposis centered on promoter 1B alterations. YY1 is a context-dependent transcription factor that functions as an activator and repressor, exhibiting oncogenic or tumor-suppressive roles depending on the cellular context[45-49]. Within the APC regulatory landscape, YY1-binding site variants reduce the APC expression and drive gastric fundic gland polyposis in GAPPS, whereas YY1 upregulation suppresses the APC expression and promotes tumorigenesis in endometrial cancer cells[50,51]. Consistent with this context-dependent role, results of dual-luciferase reporter assays have demonstrated that YY1 overexpression can repress the APC promoter 1B activity, even in the absence of promoter variants[38].

Nuclear β-catenin accumulation in the colonic polyps of patients with GAPPS supports its inclusion within the FAP disease spectrum[36]. However, the marked heterogeneity of clinical phenotypes indicates that the APC promoter-region abnormalities cannot be interpreted as uniform pathogenic events. A recent study using patient-derived tissues, human gastric organoids, and inducible APC-mutant mice demonstrated region-specific Wnt responses to APC dysfunction, which is consistent with the “just-right” Wnt signaling model[52]. The Wnt responses to APC dysfunction differ between the gastric corpus and antrum, contributing to region-specific fundic gland polyp formation[52]. Heterozygous APC loss increased epithelial proliferation in the corpus but not in the antrum, whereas homozygous loss was not sustained in the corpus but induced hyperproliferation in the antrum. In addition to the tissue-specific differences in promoter usage and epigenetic regulation, these findings suggest that phenotypic variation may partly reflect the differences in the tissue-specific Wnt signaling thresholds.

Current guidelines recommend germline multigene panel testing, including APC and other polyposis-associated genes (e.g., MUTYH, POLE, and POLD1) in patients with adenomatous polyposis[37]. However, some patients remain genetically unresolved, in which case, APC promoter-region abnormalities, intronic variants, and APC mosaicism should be considered. In this context, our classification framework may help guide subsequent genetic evaluation of APC promoter-region abnormalities.

Specifically, CNV analysis (e.g., MLPA) is useful for detecting promoter region deletions (classes I and IV), whereas targeted sequencing of the APC promoter 1B region is required to identify point variants (classes II and III). In patients with predominantly gastric phenotypes suggestive of GAPPS, prioritization of promoter 1B sequencing may be reasonable.

Although evidence remains limited, the proposed classification framework may have implications for surveillance strategies. Class I alterations and class II-b variants associated with a classical FAP phenotype may require careful upper and lower gastrointestinal surveillance according to the established FAP management principles. Contrarily, class II-a variants (GAPPS) may warrant focused gastric surveillance and consideration of prophylactic total gastrectomy, alongside colorectal evaluation. Class III variants may be managed using the AFAP strategies with individualized colorectal surveillance. Class II-c and IV alterations should be managed according to the clinical phenotype, given the heterogeneity of the underlying variants.

The present review has several limitations. First, due to the low incidence of APC promoter-related FAP, most of the included studies had a relatively small sample size, thereby limiting the ability to draw definitive conclusions on penetrance and phenotypic variability across populations. Second, the available evidence was derived from retrospective observational studies, consisting predominantly of case reports and case series. Third, the clinical descriptions of gastrointestinal phenotypes were highly heterogeneous across studies, accompanied by substantial variability in surveillance strategies and reporting, particularly regarding gastric polyp burden. Fourth, functional evidence relies largely on in vitro analyses, which may not fully capture the complexity of the in vivo regulation of APC expression across different tissue contexts. Finally, the possibility of publication bias cannot be excluded, as families with distinctive phenotypes may be preferentially reported.

CONCLUSION

Despite its limitations, the present systematic review delineates the genotype-phenotype relationships across the APC promoter-region abnormalities and highlights their clinical heterogeneity. Organizing these alterations within a unified framework provides a conceptual basis for interpreting organ-specific disease expression and may inform individualized surveillance strategies. Further research is needed to elucidate the molecular mechanisms linking promoter abnormalities to tissue- and region-specific dysregulations of APC expression, as well as to accumulate additional clinical cases.