Endoscopic Diagnosis and Clinical Course of Isolated Upper Gastrointestinal Amyloidosis: A Retrospective Observational Study

Article information

Korean J Helicobacter Up Gastrointest Res. 2025;25(4):363-370

Publication date (electronic) : 2025 December 4

doi : https://doi.org/10.7704/kjhugr.2025.0076

Je Seong Kim¹, Young Eun Seo¹, Chae June Lim¹, Chan Muk Im¹, Hyung Hoon Oh¹, Ki Hyun Kim¹, Sung Bum Cho¹, Joo Yeon Koo², Young Eun Joo¹, Wan Sik Lee^,1

¹Departments of Internal Medicine, Chonnam National University Medical School, Hwasun, Korea

²Department of Pathology, Chonnam National University Hwasun Hospital, Chonnam National University Medical School, Hwasun, Korea

Corresponding author Wan Sik Lee, MD, PhD Department of Internal Medicine, Chonnam National University Medical School, 264 Seoyang-ro, Hwasun-eup, Hwasun 58128, Korea E-mail: jadelook@hanmail.net

Received 2025 October 31; Revised 2025 November 12; Accepted 2025 November 19.

Abstract

INTRODUCTION

An amyloid is a fibrillar protein characterized by a β-sheet secondary structure (known as cross-β) that readily forms aggregates [1]. Amyloidosis is a disease characterized by the extracellular deposition of amyloid proteins [2]. To date, 36 proteins have been identified as amyloidogenic. Of these, 14 are exclusively associated with systemic amyloidosis, reflecting their potential to affect multiple organs, whereas 19 are primarily implicated in localized diseases, where deposition is restricted to a single organ or tissue [3].

Amyloid light-chain (AL) amyloidosis is the most common form of this disease, with an estimated incidence of 3–13 cases per million people per year. In contrast, amyloid A (AA) amyloidosis occurs at a rate of approximately 2 cases per million per year. The typical age of onset for both types is between 55 and 60 years [4]. Amyloidosis may be either hereditary or acquired, and late-onset disease does not preclude a hereditary origin [5]. Approximately one-third of cases are hereditary, while acquired forms represent about half of all cases. The latter typically occurs at an older age and likely reflects an age-related decline in protein homeostasis mechanisms [6].

Gastrointestinal involvement is relatively common in systemic amyloidosis, whereas isolated gastrointestinal amyloidosis is rare [7]. According to a recent systematic review, the median age of patients with isolated gastrointestinal amyloidosis was 64.4 years, with a male-to-female ratio of 2:1. The stomach is the most frequently involved organ, followed by the small intestine and the colon [8]. Gastrointestinal amyloidosis may present with a wide spectrum of clinical manifestations, including bowel dysmotility, weight loss, gastrointestinal bleeding, ulceration or perforation, and malabsorption [2].

The endoscopic features of gastrointestinal amyloidosis are diverse and may include depressed lesions, submucosal tumorlike masses, nodular changes, and thickened folds [1,7]. Active ulcers or ulcer scars can also be observed as manifestations of amyloid deposition [9,10]. Vascular dilatation is another common finding, and magnified observation using narrow-band or blue laser imaging often reveals distortion of the microvasculature [1]. Because these lesions are frequently misinterpreted as other conditions, it is crucial to recognize endoscopic findings suggestive of amyloidosis. Therefore, this study aimed to characterize the endoscopic features and the clinical course of patients with isolated upper gastrointestinal (UGI) amyloidosis.

METHODS

RESULTS

Baseline characteristics

The cohort included 7 males (63.6%) and 4 females (36.4%) with a median age of 72 years (Table 1). Two patients were diagnosed with involvement of both the stomach and duodenum, while four patients had isolated gastric amyloidosis and five had isolated duodenal disease.

Table 1.

Clinical and endoscopic features of patients with isolated UGI amyloidosis

Endoscopic findings

Regarding endoscopic features, yellowish patches were the most common finding in the duodenum, whereas flat-depressed lesions, flat-elevated vascular lesions, and ulcerative lesions were observed with similar frequencies in the stomach, followed by yellowish patches. Ulcerative lesions were observed in two patients (18.2%), one of whom presented with melena as the initial symptom due to ulcer bleeding. Flat-elevated lesions were uncommon in our series, being noted in only two patients (18.2%).

H. pylori testing was performed in eight patients, and only one (12.5%) tested positive, whose infection was successfully eradicated with standard therapy. Multiple lesions were more frequent than solitary lesions. Three patients (27.3%) were identified as having AA amyloidosis and showed positive immunoreactivity for AA, whereas the remaining cases were classified as non-specific.

Two patients underwent endoscopic ultrasound (EUS) (Fig. 1). In Case 1, a large flat-depressed lesion initially suspected to be a granular cell tumor showed submucosal hypoechoic thickening with preservation of the muscularis propria and serosal layers. In Case 4, the stomach showed diffusely yellowish mucosa, and EUS demonstrated linear isoechoic to hypoechoic lesions along the muscularis mucosa and submucosal layers.

Fig. 1.

Representative EUS findings in patients with gastric amyloidosis. A: EUS image of Case 1 showing submucosal wall thickening with hypoechoic features (red arrow). B: EUS image of Case 4 showing submucosal wall thickening with a preserved mucosal layer (red arrow). EUS, endoscopic ultrasound.

Clinical manifestations

Six patients (54.5%) were asymptomatic at the time of diagnosis (Table 2). Abdominal pain, including epigastric discomfort, was reported in three patients (27.3%), while gastrointestinal bleeding occurred in two patients (18.2%). Most patients did not receive specific treatment, except for one who showed disease progression during follow-up. The mean follow-up duration was 1.7 years, with a maximum duration of 4.8 years. Among the 11 patients, only one died, and the cause of death was unrelated to gastrointestinal amyloidosis.

Table 2.

Clinical course and outcomes of patients with isolated UGI amyloidosis

Evaluation to exclude systemic amyloidosis

Additional examinations were performed to exclude systemic amyloidosis and confirm isolated UGI involvement (Table 3). Colonoscopy was conducted in four patients (36.4%), bone marrow examination in seven (63.6%), and 2D echocardiography in four (36.4%). Patients who did not undergo these evaluations were clinically judged to have no systemic involvement on the basis of the absence of relevant symptoms. Seven patients (63.6%) had underlying diseases that were either previously treated or were currently managed.

Table 3.

Examinations performed to exclude systemic amyloidosis

Comparison of clinical and endoscopic features between isolated and systemic amyloidosis

For comparative analysis, five patients who were initially excluded because of systemic involvement or multiple myeloma were also included (Table 4). The isolated amyloidosis group (n=11) tended to include older patients than the non-isolated group (n=5) (68.6±12.0 years vs. 55.8±24.1 years, p=0.311). There were no significant differences between the two groups in terms of sex distribution (male: 63.6% vs. 40.0%; p=0.596), disease progression (9.1% vs. 20.0%; p>0.999), or mortality (9.1% vs. 40.0%; p=0.214).

Table 4.

Comparison of clinical and endoscopic features between isolated and systemic amyloidosis

Yellowish patch-like lesions were more frequent in the isolated group (45.5% vs. 20.0%), whereas ulcerative lesions were more common in the non-isolated group (18.2% vs. 60.0%); however, these differences were not statistically significant (p=0.488). Similarly, nonspecific amyloidosis was predominant in the isolated group (72.7% vs. 20.0%), whereas AA type amyloidosis was more common in the non-isolated group (27.3% vs. 60.0%; p=0.087). All p-values were greater than 0.05, indicating no statistically significant differences between the groups.

DISCUSSION

Proteins can adopt a wide range of conformational states within living organisms. Under certain conditions, they may undergo structural alterations that lead to the formation of nonfunctional and potentially pathogenic protein aggregates [11]. Ultimately, highly ordered assemblies known as amyloid fibrils form and are closely associated with numerous human diseases [6]. Several types of amyloidosis are clearly hereditary, with genetic factors playing a pivotal role in their pathogenesis [12]. Point mutations or deletions in precursor proteins can promote structural rearrangements that predispose to fibril formation [13]. In addition, systemic conditions such as multiple myeloma, lymphoma, or chronic dialysis can trigger the development of amyloidosis [14]. Histopathological examination of amyloidosis reveals a characteristic green–yellow–orange birefringence under cross-polarized light following Congo red staining [15]. Immunohistochemical analysis is performed using antibodies against AA amyloid, amyloid P components, and κ or λ light chains [16]. Our patients demonstrated consistent findings (Fig. 2).

Fig. 2.

Histopathologic findings in patients with upper gastrointestinal amyloidosis. A: Amyloid deposits stained with Congo red (Congo red stain, 40×). B: Apple-green birefringence under polarized light (Congo red stain under polarized light, 100×). C: Positive immunostaining for the amyloid P component (Amyloid P stain, 100×). D: Positive immunostaining for κ and λ light chains (kappa and lambda light chain stain, 200×).

Amyloidosis has the potential to affect virtually every organ system [17]. When the gastrointestinal tract is involved, the amyloid deposits exhibit type-specific demographic characteristics. For instance, patients with AA amyloidosis typically have a median age of 64 years, which is younger than that of patients with other types [16]. In our study, the AA amyloidosis group had a slightly older median age of 74 years (range, 63–82 years) compared with patients with nonspecific types, a finding inconsistent with previous reports. The remaining patients were not classified as having a specific amyloid type. These findings may be explained by the fact that our study included only patients with isolated gastrointestinal amyloidosis, whereas other forms such as AL and amyloid Transthyretin (ATTR) account for most cases of systemic amyloidosis [18].

The endoscopic features of gastrointestinal amyloidosis are highly variable and often atypical [19]. Erosions, ulcers, and even lesions that mimic malignancy can be observed, and in some cases, endoscopically normal-appearing mucosa have been histologically confirmed as amyloidosis through biopsy [20,21]. Several attempts have been made to classify these endoscopic findings. Said et al. [20] categorized lesions into groups such as normal appearance, erythema, and erosion, whereas Niu et al. [22] proposed a classification based on mucosal morphology, including normal, elevated, and thickened mucosa, emphasizing the mucosal status. A recent Korean study classified endoscopic findings into five types: protruding, granular, hemorrhagic, ulcerative, and nonspecific [23].

To simplify and standardize endoscopic characterization, we propose a classification system comprising four categories for gastric amyloidosis and three categories for duodenal amyloidosis (Fig. 3). Diffuse, flat, and yellowish lesions were defined as the yellowish patch-like type, representing the characteristic endoscopic appearance of duodenal amyloidosis in our cohort, whereas they were less frequently observed in the stomach. This type typically exhibits vascular dilatation, with central areas lacking visible vasculature, resulting in a yellowish mucosal appearance. The flat-depressed and flat-elevated types were characterized by a complete loss of mucosal vascularity and a homogeneous surface texture, which distinctly demarcated the lesion margins. These two types were differentiated based on their morphology, appearing either depressed or elevated, and the involved mucosa showed a whitish discoloration due to a lack of visible vascularity. The ulcerative-type showed heterogeneous endoscopic findings. In one patient, a large deep ulcer with easy-touch bleeding mimicked advanced gastric cancer, whereas in another patient, shallow ulcers with hematin deposits suggested benign ulcerative changes.

Fig. 3.

Proposed classification of upper gastrointestinal amyloidosis.

The proposed classification has the advantage of being simple, intuitive, and easy to apply in clinical practice. Regular endoscopic follow-up is warranted for isolated UGI amyloidosis, and a straightforward and easily recalled classification system is particularly useful during serial examinations. This system enables endoscopists to accurately recognize previously diagnosed lesions and compare them consistently with prior findings, thereby improving the continuity and quality of patient management.

It remains unclear whether the endoscopic features and clinical manifestations of isolated gastrointestinal amyloidosis differ from those of systemic amyloidosis involving the gastrointestinal tract. In the present study, the isolated amyloidosis group tended to include older patients, possibly because patients with systemic amyloidosis might have died before being diagnosed with gastrointestinal involvement. In patients with the isolated form, yellowish patch-like lesions were the most common finding in the duodenum, consistent with previous reports [17,22]. In contrast, the stomach exhibited more diverse morphological patterns without a dominant trend. These findings suggest that endoscopic manifestations are broadly similar irrespective of systemic involvement [17]. In our study, among patients with systemic involvement, the ulcerative type was the most common, which may reflect deeper submucosal invasion. Regarding the amyloid subtypes, AA-type amyloidosis was more frequent in the systemic group, although this difference was not statistically significant. This finding is consistent with previous observations that AA amyloidosis is typically associated with systemic diseases, as the AA protein tends to deposit more readily in the gastrointestinal tract [19].

However, the diagnostic role of EUS in gastrointestinal amyloidosis has not yet been fully clarified. Amyloid deposition may involve either the mucosal or submucosal layer, with duodenal lesions more frequently demonstrating submucosal infiltration than gastric lesions [22]. Previous studies have also reported that AL and ATTR amyloid proteins tend to accumulate predominantly within the submucosa, whereas AA proteins are more commonly confined to the mucosal layer [24,25]. On EUS, amyloid deposits typically appear as hypo- to isoechoic areas [1]. In our series, EUS revealed linear hypoechoic lesions with submucosal invasion, which were consistent with these observations.

The mainstay of the management of isolated gastrointestinal amyloidosis is careful observation and regular endoscopic follow-up. As reported in previous studies, most cases of isolated disease remain stable without progressing to systemic amyloidosis and generally exhibit a favorable prognosis [26,27]. However, if disease progression occurs, there is a potential risk of systemic involvement, and systemic chemotherapy, which is commonly used to treat systemic amyloidosis, may be warranted. In systemic AL amyloidosis, a combination regimen of daratumumab, cyclophosphamide, bortezomib, and dexamethasone (the Dara-CyBorD regimen) is regarded as the treatment of choice [28]. In our series, only one patient (Case 1) received chemotherapy because of disease progression from the duodenum to the stomach, accompanied by abdominal pain and diarrhea. The patient was treated with the CyBorD regimen, resulting in marked symptomatic improvement. Apart from this case, no other patient showed disease progression, suggesting that isolated gastrointestinal amyloidosis generally follows a benign clinical course.

Gastrointestinal hemorrhage occurs in approximately 25%– 45% of patients with gastrointestinal amyloidosis, and the main bleeding sources include ulceration, ischemia, and mucosal friability [29]. Gastrointestinal amyloidosis can occasionally lead to life-threatening massive bleeding, in which case surgical resection should be considered [17,30]. In our cohort, Case 3 presented with melena, and duodenoscopy revealed large, deep gastric ulcers. Despite three sessions of successful endoscopic hemostasis, bleeding recurred, and surgical resection was recommended; however, the patient declined. After two weeks of supportive care alone, the patient died of a spontaneous intracranial hemorrhage, which was not directly related to amyloidosis but may have been influenced by uncontrolled hypertension.

This study had several limitations. Owing to the rarity of this condition, the sample size was small, which precluded a meaningful statistical analysis. In addition, some patients were lost to follow-up 12 months after diagnosis, limiting the assessment of the long-term disease course. Moreover, a few patients declined systemic evaluations such as 2D echocardiography or bone marrow biopsy, raising the possibility of underdiagnosed systemic amyloidosis. Future studies should aim to delineate the differences between isolated and systemic gastrointestinal amyloidosis and determine whether isolated UGI amyloidosis represents a distinct clinical entity.

In conclusion, isolated UGI amyloidosis appears to be a distinct clinical entity characterized by diverse endoscopic features. The awareness of amyloidosis as a potential underlying cause is crucial when interpreting duodenoscopic findings. Although the overall prognosis is generally favorable, careful long-term follow-up is warranted because of the potential for disease progression and systemic involvement. Familiarity with characteristic endoscopic appearances and the natural disease course is essential for accurate diagnosis and optimal management.

Supplementary Materials

Article information Continued