Gastric diffuse large B-cell lymphoma (DLBCL) comprises most of all gastrointestinal lymphomas, accounting for approximately 55%–65% of all cases. This condition predominantly affects male patients and those older than 50 years and is usually aggressive. The clinical presentations are nonspecific and can mimic other diseases. In this report, we present a case of a 66-year-old male patient with a history of hematemesis and melena, with significant weight loss for 3 months. Esophagogastroduodenoscopy (EGD) revealed multiple stomach ulcers. A biopsy taken during EGD, which was followed by immunohistochemical staining, confirmed the diagnosis of non-Hodgkin lymphoma, specifically germinal center B-cell (GCB)–subtype DLBCL. A positron emission tomography (PET) scan was performed, revealing the involvement of the mesenteric lymph nodes and an infiltrative lesion that extended from the stomach to the pancreas. Based on the Ann Arbor staging system, the diagnosis was identified to be stage IV gastric DLBCL. Following diagnosis, we administered rituximab, cyclophosphamide, hydroxydaunorubicin, vincristine, and prednisone (R-CHOP) to the patient.

B-cell lymphoma, endoscopic findings, gastrointestinal lymphoma, PET, R-CHOP

Gastric diffuse large B-cell lymphoma (DLBCL) is the most prevalent type of gastrointestinal lymphoma, accounting for approximately 55%–65% of cases.^[1] This malignancy typically affects males over 50 and is characterized by its aggressive nature.^[1,2] However, the clinical manifestations of gastric DLBCL are often nonspecific, including symptoms such as nausea, vomiting, gastrointestinal bleeding, and weight loss, which can easily be mistaken for other conditions. Diagnostic tools such as esophagogastroduodenoscopy (EGD) with biopsy and immunohistochemical analysis are crucial for accurate diagnosis. Imaging modalities like positron emission tomography (PET) scans provide critical insights into disease staging and therapeutic responses. This report discusses the case of a 66-year-old male presenting with nonspecific gastrointestinal symptoms, ultimately diagnosed with stage IV gastric DLBCL.

A 66-year-old male patient presented with a history of hematemesis, melena, nausea, vomiting, and significant weight loss over the past 3 months. He had no significant medical or surgical history and denied taking any medications. He did not smoke cigarettes, drink alcohol, or use illicit drugs. Physical examination revealed normal vital signs and systemic examination results; however, a soft tissue tumor was found in the upper left back region. Multiple gastric ulcers were found during EGD (Fig. 1). Biopsy samples were collected during EGD, and histopathological results indicated a suspicion for gastrointestinal stromal tumor with a differential diagnosis of lymphoma and adenocarcinoma. Immunohistochemical staining was conducted to confirm the histopathological diagnosis. Table 1 presents the immunophenotypic profile of the tumors. As displayed in Fig. 2 , the tumor cells showed immunopositivity for CD20, negativity for CD3, and a high Ki-67 index (60%–70% of tumor cells). The final diagnosis was non-Hodgkin lymphoma, specifically GCB-subtype DLBCL, based on the Hans algorithm.

Figure 1.

EGD showing multiple gastric ulcers.

Figure 2.

A. Hematoxylin and eosin staining showing large tumor cells arranged in a diffuse pattern (400×); B. Immunohistochemistry staining showing a positive result for CD20 (400×); C. CD3 staining showing a negative result (400×); D. Ki-67 staining indicating a high proliferation index (400×).

A further workup for tumor staging included a PET scan, which revealed thickening and focal hypermetabolic lesions in the gastric and pyloric walls as well as in the cauda of the pancreas. An infiltrative lesion, which extended from the stomach to the pancreas, was noted. This is consistent with malignancy. Moreover, a mass with fluorodeoxyglucose uptake was observed in the left supraspinatus, which led to the destruction of the left scapula. Based on the Ann Arbor staging system, the final diagnosis was stage IV gastric DLBCL. The patient’s age, Ann Arbor stage IV, elevated LDH levels, >1 extranodal site, and good performance status resulted in a total International Prognostic Index score of 4.

The patient was treated with the R-CHOP chemotherapy regimen, which was administered every 3 weeks for a total of six cycles. The patient tolerated R-CHOP well, except for neutropenia, which was effectively managed with secondary growth factor support. No infectious complications occurred during treatment. After 6 cycles of R-CHOP, the treatment continued with rituximab alone for up to eight cycles. A PET-CT scan performed at the end of therapy showed complete remission. Two years after the final dose of R-CHOP, a PET scan evaluation was performed, revealing no evidence of disease relapse (Fig. 3 ).

Figure 3.

A comparison of PET/CT scans after six cycles of R-CHOP showed favorable results based on the Deauville score of 1. Scan A was taken on June 29, 2022, and scan B was taken on March 18, 2023; C. A PET scan performed two years after the completion of chemotherapy demonstrates no evidence of disease relapse, indicating sustained remission. Scan C was taken on July 29, 2024.

Gastric lymphoma is a rare malignancy, but gastric DLBCL is the most common extra-nodal site of lymphoma.^[1,3] The clinical symptoms of gastric DLBCL are nonspecific and can mimic those of other diseases, such as epigastric discomfort, unexplained weight loss, anorexia, vomiting, and occult bleeding.^[4] Hematemesis and melena are uncommon but can occur, with or without other gastric symptoms.^[5] This makes DLBCL of the stomach challenging to diagnose, which can lead to misdiagnosis and delayed treatment.^[3,6] During physical examination, palpable mass and lymphadenopathy are often observed.^[3]

Gastric lymphoma can present as ulcers, heterogeneous masses, or a combination of both in multiple locations at various stages.^[6] In this patient, multiple gastric ulcers were identified via EGD, which led to hematemesis and melena, indicative of gastric bleeding.^[7]

Initial EGD-biopsy of this case suggested a gastrointestinal stromal tumor (GIST) or gastric adenocarcinoma. GIST is recognized as the most common mesenchymal neoplasm of the gastrointestinal tract, while gastric adenocarcinoma is an epithelial malignancy. Although these neoplasms arise from different cellular lineages, both GIST and gastric adenocarcinoma—particularly those displaying invasive growth—can closely mimic lymphomas in terms of cytological appearance. Shared features include increased cellularity, submucosal infiltration and extension, ulceration of the overlying mucosa, areas of necrosis, and elevated mitotic activity.^[8,9] These overlapping morphological features, combined with the frequent issue of limited or non-representative biopsy specimens—particularly in image-guided or endoscopic procedures—present a significant diagnostic challenge.^[10] Therefore, immunohistochemical staining is essential to achieve an accurate diagnosis and guide appropriate clinical management.

Diagnosis of DLBCL requires the expression of one or more B-cell markers by the tumor cells, specifically CD19, CD20, CD79a, or PAX5. In this case, the tumor was immuno-negative for CD117 (KIT), DOG1, CDX2, and CK7, effectively excluding neoplasms of mesenchymal and epithelial origin. Conversely, strong immunopositivity for CD20 and CD45, along with a high proliferative index, supported the diagnosis of gastric DLBCL. Subclassification of DLBCL into two distinct subtypes based on the cell of origin is crucial for prognostic prediction and the selection of potential targeted therapies. Gene expression profiling is the gold standard method, yet it is not widely accessible. The Hans algorithm is the most utilized IHC-based classifier, incorporating staining for CD10, BCL6, and MUM1. This present gastric DLBCL case is classified as GCB-subtype according to Hans’s algorithm, which is characterized by positive CD10 staining. Prior studies have identified a varied proportion of the GCB subtype in different gastrointestinal DLBCL sites. The GCB subtype generally exhibits a more favorable prognosis than the non-GCB subtype and is less frequently reported in gastric DLBCL (27%–42%).^[11-14]

PET scans were conducted to stage the disease and guide appropriate treatment. PET scans demonstrated thickening and focal hypermetabolic lesions in the stomach, pylorus wall, and cauda of the pancreas. A PET scan also revealed a mass in the left supraspinatus region causing destruction of the left scapula, indicating extranodal involvement beyond the primary gastric site. Extranodal spread involving the supraspinatus muscle and scapula is an exceedingly rare presentation. A thorough literature review reveals limited documented cases of gastric DLBCL presenting concurrently with masses in the supraspinatus region, causing scapular destruction. A similar case of a 45-year-old male with gastric DLBCL and multifocal bone involvement, including the left femur, tibia, fibula, and scapula, who achieved partial remission following R-CHOP chemotherapy and autologous peripheral blood stem cell transplantation has been reported.^[15] However, the case did not report involvement of the supraspinatus muscle. This highlights the uniqueness of such presentations and underscores the importance of advanced imaging and histopathological confirmation in diagnosis.

Based on the Ann Arbor staging system, the patient was diagnosed with stage IV gastric DLBCL, IPI score 4. Although various staging systems are available for gastric DLBCL, the Ann Arbor staging system remains the most widely utilized in clinical practice.^[16] Other staging systems—the Lugano, TNM, and Paris—are also used, though their accuracy remains debated. A standardized staging system is essential for the effective management of gastric DLBCL.

The etiology of gastric DLBCL is not yet clear. It may arise primarily de novo or secondarily from low-grade mucosa-associated lymphoid tissue lymphoma (MALT), with coinfection by H. pylori serving as a predisposing factor. Furthermore, chronic gastritis can predispose patients to gastric DLBCL.^[1,2] Compared with low-grade MALT lymphoma, high-grade gastric lymphoma is associated with a lower complete remission rate and shorter survival.^[1] We believe that the etiology of gastric DLBCL in this patient primarily originated de novo rather than as a secondary transformation from low-grade MALT lymphoma.

Various treatment modalities, including surgery, radiotherapy, and chemotherapy, have been utilized to treat gastric DLBCL. Nevertheless, the current treatment approach has shifted away from surgery toward chemotherapy. Referring to recent studies, for patients with an IPI score of 3-5, immunochemotherapy consisting of polatuzumab vedotin, rituximab, cyclophosphamide, doxorubicin, and prednisone (Pola-R-CHP) is preferred.^[17,18] Based on the phase 3 POLARIX study, progression-free survival was significantly higher in patients treated with Pola-R-CHP than in those treated with R-CHOP after a median follow-up of 28.2 months (76.7% [95% CI, 72.7–80.8] vs. 70.2% [95% CI, 65.8–74.6] at 2 years), although no significant difference was observed in overall survival (88.7% [95% CI, 85.7–91.6] in the Pola-R-CHP group vs. 88.6% [95% CI, 85.6–91.6] in the R-CHOP group).^[18] Since polatuzumab vedotin was not covered by the patient’s insurance, R-CHOP, the previous gold standard therapy for DLBCL, was administered as a substitute for Pola-R-CHP. It is typically administered once every 21 days for an average of six to eight cycles, depending on the patient’s disease and health status.^[1,2,19,20] In this case, we administered R-CHOP for six cycles and decided to continue with rituximab, completing a total of eight cycles. A PET/CT scan conducted at the end of therapy demonstrated complete remission, with no evidence of disease progression after two years.

This case underscores the importance of early recognition and accurate diagnosis of gastric DLBCL, a malignancy often presenting with nonspecific symptoms. Advanced diagnostic tools and timely treatment, such as R-CHOP, are crucial for achieving remission and improving outcomes in this aggressive yet treatable disease.

Ethical statement

• This study was conducted following ethical standards. As per institutional guidelines, formal ethics approval was not required for this case report.
• The authors declared that no clinical trials were used in the present study.
• The authors declared that no experiments on humans or human tissues were performed for the present study.
• Written informed consent was obtained from the patient for publication of this case and any accompanying images.
• The authors declared that no experiments on animals were performed for the present study.
• The authors declared that no commercially available immortalized human and animal cell lines were used in the present study.

Conflict of interest

The authors have declared that no competing interests exist.

Artificial Intelligence (AI) use

The authors accept full responsibility for the content of the manuscript, including the disclosure of any use of AI. No AI tools were used in the preparation of this manuscript.

Funding

No funding was reported.

Author contributions

All authors have contributed equally.

Author ORCIDs

Wulyo Rajabto https://orcid.org/0000-0002-8231-418X

Lugyanti Sukrisman https://orcid.org/0000-0003-4498-7735

Agnes Stephanie Harahap https://orcid.org/0000-0001-8920-7873

Maria Pyrhadistya https://orcid.org/0000-0003-0736-1709

Ainun Safitri https://orcid.org/0009-0002-2665-9957

Data availability

All of the data that support the findings of this study are available in the main text.