Hereditary Factors in Esophageal Adenocarcinoma

During the last decades there has been a dramatic increase in the incidence of esophageal adenocarcinoma (EAC) in Western countries [1,2,3]. Despite improvements in multimodality therapy, the prognosis of patients with EAC remains poor [4]. Barrett's esophagus (BE) is the predominant risk factor for EAC in addition to age, male gender and Caucasian ethnicity [5]. BE is a premalignant condition in which the normal squamous epithelial lining in the lower esophagus is replaced by columnar intestinal cells [6]. BE is considered a long-term complication of severe chronic gastroesophageal reflux disease (GERD). The susceptibility for GERD may in turn be influenced by factors such as obesity, alcohol consumption and nicotine abuse [7].

The vast majority of BE and EAC cases are sporadic and caused by somatic mutations [8], i.e. mutations that may occur in any cell of the body except for germ cells. However, over the last decades several families have been identified with clustering of EAC [9,10,11,12,13,14]. This observation suggests that one or more inherited factors might play a role in the initiation of EAC in these families. Familial clustering of cancer is important thanks to the success of implementing genetic testing and screening methods for cancer syndromes [15,16,17]. However, for cancers that are not covered in familial risk management guidelines, such as EAC, awareness of the true familial risk is needed to provide rational advice [17]. In terms of clinical genetics, for a true familial risk, the number of affected family members needs to be higher than is to be expected by chance alone [18].

The purpose of the present study was to review the literature on the occurrence and characteristics of familial EAC. Previous studies introduced and persevered the definition familial BE, i.e. two or more first- or second-degree family members diagnosed with BE, EAC or gastroesophageal junction adenocarcinoma (GEJAC) [19]. These studies considered familial BE and familial EAC to be part of the same genetic trait, because EAC appears to arise from BE and both conditions share the same epidemiologic risk factors. We hypothesize that familial EAC can be distinct from most familial BE. Since BE is much more prevalent among the common population, familial BE does not necessarily have to be the underlying condition of familial EAC. If two or more first-degree family members are diagnosed with EAC, it is unlikely that this can be explained by chance alone, based on the absolute risk of 0.12-0.5% for malignant transition of BE into EAC [20,21]. Therefore, familial EAC might be the result of accelerated malignant progression from familial BE, or familial EAC might arise without familial BE as the premalignant condition. In both scenarios involvement of specific germline mutations driving familial EAC can be envisaged.

In the literature familial EAC has been grouped with familial BE and has been termed familial BE, which is defined as two or more family members diagnosed with BE, EAC or GEJAC [19]. Two studies estimated the prevalence of familial BE by reporting the proportion of patients diagnosed with BE, EAC or GEJAC that had at least one other family member diagnosed with BE, EAC or GEJAC. Chak et al. [22] reported that about 7% fulfill the criteria of familial BE, which is in line with the 6% reported by Ash et al. [23]. Including GEJAC in familial BE can be criticized, since a tumor present on the gastroesophageal junction can originate either from the esophagus or the gastric cardia. In the last instance the tumor probably does not arise from BE.

BE is generally accepted as the premalignant lesion for EAC. BE is a complication of chronic GERD. The prevalence of BE in the common population is estimated at 2% [24], while the prevalence of BE among patients with GERD is approximately 10% [25]. The annual risk of developing EAC from BE is estimated to be between 0.12 and 0.5% [20,21]. Two case-control studies about the prevalence of GERD among relatives of BE patients suggested a familial predisposition for GERD in these families [26,27]. Among 27 and 47 relatives of patients diagnosed with BE or EAC, the prevalence of BE was reported to be 18% (n = 5) and 28% (n = 13), respectively [28,29]. More importantly, the prevalence of EAC described among 20 families with a strong familial expression of GERD, BE and EAC was estimated at 31% [12]. This finding suggests a true familial risk of EAC in these families. In contrast, a Swedish population-based, nationwide case-control study did not find an association between a positive history of esophageal cancer among first-degree relatives and the risk of EAC [30]. In addition, an Italian case-control study revealed no difference in the prevalence of cancer in general between patients diagnosed with BE and patients with reflux esophagitis plus healthy controls. However, relatives of patients diagnosed with esophageal or gastric cancer had an increased risk of BE, particularly if the affected relative was younger than 50 years at the time of diagnosis [31]. It should be noted that the prevalence of familial BE is relatively low, hence in a randomly taken cohort of patients with BE only a few patients will be part of family with clustering of BE and/or EAC. Although the previously mentioned studies are based on relatively small samples sizes, the prevalence of EAC in these families with clustering of EAC appears to be distinctly higher than in the common population, and this also accounts for the prevalence of BE.

Known risk factors for the malignant transition of BE into EAC are increasing age, male gender and Caucasian ethnicity [7]. No differences in gender, ethnicity and in addition in nicotine abuse and alcohol consumption were reported between patients with familial BE and/or EAC and sporadic cases [22,23,28,32]. Contradictory results were observed regarding the prevalence of obesity, with some studies reporting no difference in prevalence [28,32], while others reported a lower body mass index for patients with familial BE and/or EAC compared to sporadic cases [22,33].

Several studies reported a lower age at diagnosis for patients with familial BE and/or EAC compared with sporadic cases [9,13,23,33,34]. Other studies could not confirm these observations [22,28,32]. In a study on 20 families with a strong familial expression of GERD, BE and EAC, the age at diagnosis of EAC appeared to be 5-10 years younger when compared to sporadic cases [12]. This finding is consistent with the concept of the presence of a germline mutation, which generally results in a lower age at disease onset when compared with sporadic cases. This concept is based on the ‘Knudson's two hit hypothesis' for the complete (bi-allelic) inactivation of tumor suppressor genes, i.e. individuals born with already one inactivating germline mutation are likely to develop the second somatic inactivating hit earlier in life than individuals without an inherited predisposition, who have to develop both somatic hits during life in the same cell before tumorigenesis occurs. The second hit may be influenced by environmental and/or by other genetic factors [35].

Since 1978 there have been several case reports on families with clustering of BE and EAC [9,10,11,12,13,14,34]. All studies suggest a pattern of inheritance consistent with an autosomal dominant genetic trait [9,10,11,12,13], which likely reflects genetic predisposition to the disease. Nevertheless, it can be anticipated that familial EAC can also be caused by common environmental exposures in family members or by a combination of both. However, the segregation analysis (i.e. an analysis to determine whether a certain gene is involved in the distribution of a phenotypic trait) of Sun et al. [36] provided the first epidemiologic evidence in support of a genetic etiology for familial BE, EAC or GEJAC. The pattern of inheritance was found to be consistent with a rare autosomal dominant genetic trait.

Hereditary tumors are generally caused by the presence of germline mutations, which are present in the reproductive cells of one or both of the parents of the affected offspring. Germline mutations are therefore present in all the cells of the affected offspring and can be transmitted from one generation to the next. Colorectal cancer has well-defined familial syndromes due to specific gene mutations. For example, patients with Lynch syndrome have germline mutations in mismatch repair (MMR) genes, and attenuated familial adenomatous polyposis, familial adenomatous polyposis (FAP) and Gardner's syndromes are all caused by germline mutations in the adenomatous polyposis coli (APC) gene [37].

Esophageal lesions are rarely described in these hereditary colorectal cancer syndromes. However, several studies reported the occurrence of BE and/or EAC in patients with (attenuated) FAP or Gardner's syndrome, which are both caused by mutations in the APC gene. Gupta et al. [38] described a family with a father who tested positive for exon 4 deletion in APC. All three of his sons were diagnosed with attenuated familial adenomatous polyposis based on the presence of multiple polyps (<100) throughout the entire colon. The middle son (41 years old) was additionally diagnosed with EAC on the background of BE. A biopsy of the father revealed BE with low-grade dysplasia. In the youngest son BE was also confirmed in histopathological biopsies. The oldest son had endoscopic findings compatible with esophagitis without metaplastic changes. Although the simultaneous occurrence of these two potentially inherited disorders in a single family may be due to chance alone, it is also possible that the disorders are linked. This suggests that deletions of the APC gene could play a role in the pathogenesis of familial BE and subsequently familial EAC. In another study with 36 FAP patients, 6 patients were additionally diagnosed with BE (16.7%); interestingly, the average age at diagnosis of BE was 20 years younger than that in the non-FAP patients with BE [39]. In (attenuated) FAP, APC gene mutations lead to increased nuclear β-catenin levels and activation of the Wnt pathway. In EAC, although through a different mechanism of Wnt activation, nuclear β-catenin is also frequently increased [40]. It can be anticipated that the mutation in APC in patients with (attenuated) FAP might increase the chance of developing EAC.

EAC has not been associated with Lynch syndrome. However, 3-5% of EAC appears to have MMR deficiency [41]. It is unknown, however, whether these tumors arise as part of Lynch syndrome or whether they are due to somatic MMR gene mutations.

Orloff et al. [42] sought to identify genes associated with BE and EAC predisposition. Germline mutations in three candidate genes were identified in approximately 11% of patients diagnosed with BE and EAC, the most commonly affected gene being macrophage scavenger receptor 1 (MSR1) [7%], followed by activating signal cointegrator 1 complex subunit 1 (ASCC1) and collagen triple helix repeat containing 1 (CTHRC1). However, further studies are needed to determine the role of these genes in the development of familial EAC.

Clustering of EAC in a family is rare but may be caused by an autosomal dominant genetic trait. In view of the poor prognosis of EAC, efforts should be made to identify a genetic predisposition for EAC in these families. Finding a genetic predisposition in familial EAC would facilitate identifying non-affected family members with the predisposition for this type of cancer. Endoscopic surveillance of relatives with the predisposition may lead to the detection of preneoplastic or early neoplastic stages of EAC. Since patients diagnosed with earlier stages of EAC have a better chance for definite cure, this strategy could probably improve outcome [4].

We would like to thank Wirchor Bramer, biomedical information specialist, Medical Library, for coordinating the literature search.

The authors disclose no conflicts of interest.