A 59-year-old man presents with fatigue and upper abdominal pain for the past week, which quickly becomes severe. A complete blood count (CBC) reveals moderate microcytic anemia. An upper endoscopy reveals diffuse mucosal erythema and nodularity, as well as a 2.5 cm ulcer with heaped up borders in the gastric fundus. The patient undergoes laparotomy with a partial gastrectomy. On gross inspection, there is a serosal defect adjacent to the mucosal ulcer. Representative sections are submitted for histologic and immunohistochemical evaluation. Results show the cells are CD45, CD20, MUM-1, BCL-2, and MYC diffusely positive while being CD3, CD5, CD10, CD21, CD23, keratin, S100, and c-KIT/CD117 negative.

Master List of Diagnoses

  • Burkitt lymphoma
  • Diffuse large B-cell lymphoma
  • Extranodal marginal zone/MALT lymphoma
  • Gastrointestinal stromal tumor
  • Mantle cell lymphoma
  • Metastatic melanoma
View slide image with DigitalScope

This case first appeared as Performance Improvement Program in Surgical Pathology (PIP) 2017, Case 26, and is diffuse large B-cell lymphoma of the stomach.

Criteria for Diagnosis and Comments

Histologic sections consist of stomach with a transmural infiltrate of pleomorphic dyshesive cells. For the most part, nuclear chromatin is open and nucleoli are prominent. Neoplastic cells are large (ie, nuclei larger than twice the size of nuclei in normal lymphocytes). Mitotic figures and apoptotic cells are abundant. In conjunction with the immunophenotype above, the morphologic features are diagnostic of diffuse large B-cell lymphoma (DLBCL).

The two most common lymphomas involving the stomach, extranodal marginal zone lymphoma/MALT lymphoma and DLBCL, have an approximately equal incidence. Gastric DLBCL occurs in two settings, as de novo neoplasia and as a transformation from a marginal zone B-cell lymphoma. However, like other marginal zone lymphomas, the incidence of transformation of gastric MALT lymphomas appears to be relatively low (approximately 5%). There appears to be a lower association with H. pylori infection when compared to MALT lymphoma. Whether arising de novo or as a transformation from a MALT lymphoma, DLBCL usually affects elderly patients. It usually presents in lower stages, though when compared to MALT lymphoma, more severe symptoms (ie, anemia, melena, etc) typically occur. Somewhat different from other lymphomas, de novo DLBCL in the stomach appears to have a worse prognosis than DLBCL transforming from MALT lymphoma. Higher stage and the presence of ascites are poor prognostic indicators. In the past, a distinction was made between the treatment strategies of MALT lymphoma and DLBCL. In many MALT lymphoma cases H. pylori eradication is an initial treatment choice. It was previously thought that eradication was not effective in DLBCL. However, it has recently been shown that some DLBCL cases can achieve complete remission after antibiotic therapy, though these cases have to be monitored closely for any signs of refractory behavior. The presence of t(11;18) – predictive of antibiotic resistance – does not seem to play a role in treatment algorithms for DLBCL since its presence is unusual, even in transformed cases.

Like other nodal and extranodal sites, gastric DLBCL can be categorized based on gene expression profiling studies. Generally speaking, these studies show two reproducible DLBCL molecular signatures, germinal center and activated B-cell types (a third 'unclassifiable' signature has also been reported). In general, DLBCL with a germinal center phenotype have better prognoses than those with an activated B-cell signature. Although gene expression profiling is not routinely used in clinical medicine, the Hans classification system uses immunohistochemical studies as a surrogate for molecular profiling. In general, the germinal center phenotype is CD10/BCL-6 positive and MUM-1 negative, and the activated B-cell phenotype is MUM-1 positive and CD10 negative. Approximately two-thirds of gastric DLBCL are of activated B-cell phenotype, and one-third are germinal center phenotype by the Hans classification system. Although the 2008 World Health Organization classification of hematopoietic tumors did not recommend the routine use of the Hans classifier for DLBCL, the current 2016 classification recommends it, when possible. Based on the immunophenotypic findings, this case would be best classified as activated B-cell type.

When compared to nodal cases in general, gastric DLBCLs more commonly have MYC translocations and BCL6 mutations; BCL2 mutations are less common. Chromosomal aneuploidies have been reported, though trisomy 3 (relatively common in MALT lymphoma) is unusual in DLBCL. However, other trisomies (12 and 18) have been reported in cases that have transformed from MALT lymphoma. As discussed previously, t(11;18) is very unusual in DLBCL.

Traditionally, morphologic and immunohistochemical characteristics were of prime importance in separating Burkitt lymphoma (BL) from DLBCL. Classically BL is a neoplasm of monomorphic medium-sized neoplastic lymphocytes with indistinct nucleoli. Likewise, immunohistochemical findings were thought to be relatively constant, with CD10 and BCL-6 being uniformly positive and BCL-2 negative (though this immunophenotype is not uncommon in DLBCL). In the past 10 years, gene expression profiling studies have shown that larger pleomorphic lymphomas, which would have been historically diagnosed as DLBCL, are molecularly similar to BL. In addition, BCL-2 positive BL cases can be seen, though they are typically not as diffusely positive as they were in this case. MYC translocation does not help differentiating between the two, as approximately 10% of DLBCL have MYC rearrangements. A new provisional entity in the 2016 classification, Burkitt-like lymphoma with 11q aberration has been reported to show more pleomorphism than classic BL. Although gastric BL is much less common than DLBCL, significant differences in treatment plans require differentiation between the two. In this case, the prominent pleomorphism and immunohistochemical findings rule out BL.

Given the location, MALT lymphoma is a diagnostic consideration. Like other MALT lymphomas, gastric MALT lymphoma can have a varying minority of dispersed large lymphoid cells, which, if found in small clusters, can cause confusion with DLBCL. In addition, the lack of a nodular or follicular architecture in many cases can resemble a diffuse growth pattern, which can be mistaken for DLBCL. A CD21 immunostain can help in problematic cases, as MALT lymphoma will be expected to show some residual follicular dendritic meshworks (DLBCL lacks this finding). In this case, the sheet-like growth pattern of large cells, diffuse architecture, and lack of residual follicular dendritic meshworks rules out MALT lymphoma.

Gastric mantle cell lymphoma is rare, though it is seen in cases with generalized gastrointestinal involvement. Unusual histologic variants such as large cell and pleomorphic mantle cell lymphomas can be confused with DLBCL, especially when an immunohistochemical panel reveals aberrant CD5 coexpression in the neoplastic B-cells. In addition, although highly sensitive for mantle cell lymphoma, cyclin D1 positivity can also be seen in other lymphoid malignancies. In questionable cases, SOX11 immunohistochemistry and fluorescence in situ hybridization (FISH) studies for the t(11;14) CCND1 translocation will resolve the diagnosis. In this case, both the negative CD5 and cyclin D1 rules out mantle cell lymphoma.

Gastrointestinal stromal tumors (GISTs), particularly epithelioid types, can show a similar pattern of spread and morphologic findings. Although epithelioid GISTs are typically relatively monomorphic, they can occasionally be pleomorphic and show high-grade cytology. Such cases would be impossible to differentiate between DLBCL without immunohistochemical studies. A simple algorithm of CD45-positivity and CD117-negativity will rule out GIST.

Melanoma is one of the more common metastatic malignancies of the stomach. In the absence of melanin production, cytologic features can be very similar to those seen in DLBCL. However, immunohistochemical results will readily distinguish between the two.

  1. Concerning gastric diffuse large B-cell lymphoma (DLBCL), which of the following statements is most characteristic?

    1. De novo cases of gastric DLBCL tend to have a better prognosis than transformed cases of marginal zone lymphoma.
    2. H. pylori eradication has no therapeutic role.
    3. Most cases are of the “activated B-cell” type.
    4. Most cases arise as a transformation from MALT lymphoma.
    5. t(11;18) is a relatively common cytogenetic abnormality.
  2. In differentiating between Burkitt lymphoma (BL) and DLBCL, which of the following statements is most accurate?

    1. BCL2 positivity rules out BL.
    2. Gastric BL is more common than gastric DLBCL.
    3. Histologic, immunophenotypic, and cytogenetic studies are all necessary.
    4. MYC translocation rules out DLBCL.
    5. Nuclear pleomorphism rules out BL.
  3. A gastric mucosal biopsy reveals a diffuse destructive population of enlarged lymphocytes. Immunohistochemical studies are positive for CD20 and CD5, and negative for SOX11. Subsequent FISH studies are negative for both MYC and CCND1 translocations. Which of the following statements is most accurate?

    1. BL is a diagnostic consideration.
    2. DLCBL is ruled out.
    3. MALT lymphoma is a diagnostic consideration.
    4. Mantle cell lymphoma is ruled out.
    5. This immunophenotype characterizes a common gastric lymphoma.


  1. Conconi A, Franceschetti S, Aprile von Hohenstaufen K, et al. Histologic transformation in marginal zone lymphomas. Ann Oncol. 2015;26(11):2329-2335.
  2. Ferreri AJ, Govi S, Ponzoni M. The role of Helicobacter pylori eradication in the treatment of diffuse large B-cell and marginal zone lymphomas of the stomach. Curr Opin Oncol. 2013;25(5):470-479.
  3. Odze, RD, Goldblum JR. Surgical Pathology of the GI Tract, Liver, Biliary Tract and Pancreas. 3rd ed. Philadelphia, PA: Elsevier-Saunders; 2015:854-856.
  4. O’Malley DP, Goldstein NS, Banks PM. The recognition and classification of lymphoproliferative disorders of the gut. Hum Pathol. 2014;45(5):899-915.
  5. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2016;127(20):2375-2390.
  6. Zullo A, Hassan C, Andriani A, et al. Primary low-grade and high-grade gastric MALT-lymphoma presentation. J Clin Gastroenterol. 2010;44(5):340-344.


Brad B. Bryan, MD, FCAP
Surgical Pathology Committee
Central Oregon Pathology Consultants
Bend, OR

Answer Key

  1. Most cases are of the “activated B-cell” type. (c)
  2. Histologic, immunophenotypic, and cytogenetic studies are all necessary. (c)
  3. Mantle cell lymphoma is ruled out. (d)