Search

Aims Immunoglobulin (Ig)G4-related disease (IgG4-RD) is an increasingly recognized fibroinflammatory condition that commonly exhibits multisystem involvement, with localized (e.g. inflammatory pseudotumours that can mimic malignancy) or diffuse (leading to organ dysfunction) patterns of tissue involvement. The 2012 Boston criteria have standardized the histopathological approach to the diagnosis of IgG4-RD and require one or more of the cardinal morphological features with prominence of IgG4+ plasma cells and an IgG4+/IgG+ plasma cell ratio of at least 40%. The relative prevalence of the morphological criteria varies between anatomical sites, but granulomas are rarely found and, indeed, their presence would usually deter a pathologist from making this diagnosis. The aim was to characterize two cases of IgG4-RD in which granulomas were present and to highlight this as an unusual feature of the condition. Methods and results We describe two cases in which the features of IgG4-RD were present within lymph nodes, together with granulomas. This is a recognized but rare morphological pattern of IgG4-RD. Conclusions While an unusual finding, the presence of granulomas should not preclude a diagnosis of IgG4-RD in the appropriate clinicopathological context.

We have compared light chain immunohistochemistry in reactive lymphoid tissue and a series of paraffin-embedded B-cell lymphomas using standard trypsin digestion with a heat mediated epitope retrieval method. Fifty-seven B-cell lymphomas (18 high grade, 29 low grade and 10 cases of nodular lymphocyte predominance Hodgkin's disease), two reactive lymph nodes and eight tonsils fixed for known times between 12 h and 2 years were studied. Paraffin-embedded tissue was stained with polyclonal anti-kappa and anti-lambda antibodies. For each antibody staining was performed on two sections, one treated with trypsin digestion and one with microwave heating. Sections were scored from 0 to + + + with 0 representing poor staining and + + + excellent staining. A score of ++ was considered satisfactory. Light chain restriction was recorded if present. Satisfactory staining was obtained in 34/59 cases using trypsin digestion and 56/59 cases using heat retrieval. Light chain restriction was demonstrated in 32/57 (56%) B-cell lymphomas using trypsin digestion and 52/57 (91%) using heat retrieval. Satisfactory staining was obtained in tonsils fixed for up to 48 h using trypsin digestion and up to 2 years using heat retrieval. We have shown that for light chain immunostaining a heat mediated epitope retrieval method produces more consistent and satisfactory results and is effective over a greater range of fixation times than traditional trypsin digestion.

Bcl-2 protein expression was studied in a series of 58 MALT lymphomas using a monoclonal antibody which recognises this protein in routinely processed paraffin embedded tissue. Thirty-three of 58 cases showed heterogeneity for bcl-2 expression, 18 of 58 cases were bcl-2 positive and 7 of 58 were bcl-2 negative. High grade and low grade MALT lymphomas showed different patterns of staining. All 21 low grade tumours were positive for bcl-2, though in seven cases only a proportion of the neoplastic cells expressed this protein. In the 37 high grade tumours the majority of the neoplastic cells were negative with seven cases showing no reactivity at all. These findings give further support to the theory that MALT lymphomas differ in pathogenesis to nodal lymphomas and suggest that the good prognosis of MALT lymphomas may partly be explained by the fact that they maintain a normal pattern of bcl-2 expression.

Aims: To determine the immunophenotype of gastric and intestinal diffuse large B-cell lymphomas and investigate the clinical significance of patterns of antigen expression. Methods and results: Immunohistochemistry was performed for detection of CD10, Bcl-6, Bcl-2, MUM1 and p53 in paraffin-embedded tissue from 29 patients with primary diffuse large B-cell lymphoma of the stomach and intestine. Statistical analysis was performed using χ2 and Fisher’s exact tests. Survival data were analysed by the Kaplan–Meier method and compared using a log rank test. Thirteen of the 29 cases were of germinal centre phenotype (CD10+ or CD10–, Bcl-6+ and MUM1–). Sixteen were of activated B-cell phenotype (all CD10– and either Bcl-6–, or Bcl-6+ and MUM1+). Sixteen cases showed Bcl-2 expression. There was a statistically significant difference (P < 0.05) in immunophenotype of the neoplastic cells relating to tumour site. Of 15 gastric lymphomas, 11 were of activated B-cell phenotype and 9/14 intestinal tumours were of germinal centre phenotype. No significant survival difference was found between groups with regard to expression of any of the antigens investigated. Conclusions: Primary gastric and intestinal large cell lymphomas appear to show different patterns of antigen expression. This suggests that these tumours arise by different mechanisms and/or from different causes. Gastric diffuse large B-cell lymphomas are usually of activated B-cell phenotype that may reflect a relationship with low-grade gastric lymphomas of marginal zone type. Intestinal diffuse large B-cell lymphomas usually show a germinal centre phenotype, suggesting an origin from germinal centre B cells. In this study the tumour immunophenotype was not associated with any difference in survival.

We have compared the value of immunohistochemical and polymerase chain reaction (PCR) techniques in distinguishing follicular lymphoma from follicular hyperplasia in formalin-fixed paraffin-embedded tissues of 41 follicular lymphomas, 15 reactive lymph nodes and 5 reactive tonsils. Immunohistochemistry demonstrated bcl-2 protein in the follicle centre cells of 97% of follicular lymphomas whereas monoclonal immunoglobulin light chain was detected in 83% of cases. Assessing the lowest proliferating follicle of each case by MIB-1 immunostaining, proliferation fractions in the lymphomas varied from 0.5% to 59% (mean 15.6%). Over 80% of lymphomas had proliferation fractions of less than 25%. PCR detected gene rearrangement either at the bcl-2 locus, or at the IgH locus, or at both loci in 32%, 44% and 61% of lymphomas, respectively. The follicle centre cells of the reactive lymph nodes and tonsils were all bcl-2 protein negative and polytypic for kappa and lambda light chains. Proliferation fractions of the lowest proliferating follicle in each reactive case ranged from 30.5% to 86.8% (mean 64.9%). Rearrangements of the bcl-2 or IgH loci were not detected in any reactive case. This study demonstrates that bcl-2 and light chain immunostaining are the most consistently helpful aids to diagnosing follicular lymphoma. A low proliferation fraction also indicates lymphoma but a high proliferation fraction does not exclude the diagnosis. Immunostaining with a combination of anti bcl-2 and MIB 1 antibodies is a sensitive and specific method for identifying follicular lymphoma, is technically simple to perform and easy to interpret. In occasional cases, where immunostaining gives equivocal results, PCR analysis can confirm lymphoma, but a negative result does not exclude the diagnosis.