| |||||||||||||||||||||||||||||||
Case Report
The patient was a 79-year-old woman who presented with a dry mouth. A lip biopsy revealed the histologic features of Sjogren's syndrome. Eleven years before, she had undergone a partial gastrectomy, with a histologic diagnosis of benign peptic ulcer. Two years after the lip biopsy, she reported weight loss, and a biopsy of the gastric remnant showed a low-grade B-cell MALT lymphoma. Staging procedures revealed bone marrow involvement, but no other detectable sites of disease.
Methods
We reviewed sections of formalin-fixed, paraffin-embedded tissue from the original gastrectomy specimen and subsequent biopsies of the lip, stomach, and bone marrow that were stained with hematoxylin and eosin.
Immunohistochemical Analysis
Paraffin sections from all specimens except the bone marrow were immunostained by the streptavidin-biotin-immunoperoxidase technique, including previous digestion with trypsin when required, with antibodies to kappa and lambda immunoglobulin light chains. Sections of formalin-fixed, paraffin-embedded hyperplastic tonsils were used as positive controls, and omission of the primary antiserum served as a negative control.
PCR Analysis
We prepared paraffin-embedded tissue for PCR, using the method of Wright and Manos,5 and amplified rearranged heavy-chain genes using the seminested PCR method of Wan et al.,4 in which primers are directed to the joining region (first-round primer, 5'TGAGGAGACGGTGACC3'; nested second-round primer, 5'GTGACCAGGGTNCCTTGGCCCCAG3') and to the conserved framework-three segment of the variable region (5'ACACGGC[C/T][G/C]TGTATTACTGT3'). Test samples were run in parallel with a polyclonal control (reactive tonsil), a monoclonal control (Raji B-cell line), and a negative control (no template DNA). Extractions from each specimen were undertaken at different times and performed in duplicate to minimize the risk of cross-contamination. Products were analyzed on 10 percent polyacrylamide gels, stained with ethidium bromide, and viewed under ultraviolet radiation.
Cloning and Sequencing
PCR products derived from the gastric resection, the subsequent endoscopic gastric biopsy, and the lip biopsy were electrophoresed on a 10 percent polyacrylamide gel, and in each case a sample from the larger of the two bands was reamplified to enhance for the sequence. The products were run on a 4 percent low-melting-temperature agarose gel, and the appropriate fragment was cut from the gel, purified with the Mermaid Kit (Bio 101, La Jolla, Calif.), and ligated into Bluescript-SKII phagemid (Stratagene, La Jolla, Calif.) before transfection into SURE bacteria (Stratagene). Colonies were screened by PCR, as described above, and six recombinants from each site that yielded a fragment of an appropriate size were selected for sequencing by the Sanger method, with the Sequenase kit (United States Biochemical, Cleveland). Computer analyses of the resulting DNA sequences were performed on the server of the National Center for Biotechnology Information at the National Library of Medicine6 and the sequence-analysis software (release 5) from the Genetics Computer Group7. The D and J regions were determined from published sequences8,9.
Results
Sections from the original gastrectomy specimen showed an ulcer overlying a heavy lymphocytic infiltrate. Both this specimen and the subsequent gastric-biopsy specimen revealed the histologic features of low-grade gastric B-cell lymphoma of MALT type (Figure 1 and Figure 2). Reactive B-cell follicles were surrounded by a diffuse infiltrate of centrocyte-like cells that invaded gastric glands, forming lymphoepithelial lesions; these were most obvious in the gastric-biopsy specimen. The lip-biopsy specimen (Figure 3) contained several discrete minor salivary glands, in each of which there was a chronic inflammatory infiltrate composed of small lymphocytes and scattered larger cells and plasma cells. There was some infiltration of duct epithelium, but the appearance in no way suggested lymphoma. The infiltrate was strikingly confined to the glands and did not spread into the surrounding tissues. The bone marrow specimen revealed paratrabecular deposits of low-grade B-cell lymphoma.
|
|
|
The results of PCR analysis are shown in Figure 4. Two PCR products, representing rearrangement of the immunoglobulin-heavy-chain gene in both alleles, were seen in samples of test DNA from each of the biopsy specimens. Repeat reactions with digested material from further sections produced identical results, and control specimens produced appropriate patterns of amplification.
|
|
The finding of identical PCR products, representative of the rearrangement of both immunoglobulin-heavy-chain alleles, in biopsy material from the stomach, lip, and bone marrow implied that the same neoplastic clone was present in each site. Sequencing of one of the PCR products from the gastric-resection, gastric-biopsy, and lip-biopsy specimens confirmed that rearranged immunoglobulin-heavy-chain CDRIII regions were amplified in each case; furthermore, the close similarity in DNA sequence showed that the dominant B-cell population at each site was derived from a common clone. The sequences amplified from each site differed by at least one nucleotide, so it is unlikely that the PCR products resulted from cross-contamination. It is possible, however, that some of the differences observed may have arisen during the amplification reactions, since the enzyme we used (Taq polymerase) lacks a 3'-5' proofreading function10.
Molecular analysis of the sequential biopsy specimens in this case enabled us to follow the neoplastic clone from its first manifestation as a gastric MALT lymphoma, which was probably incompletely resected,11 through its involvement in the development of Sjogren's syndrome 11 years later, to its further manifestation as recurrent gastric lymphoma with dissemination to the bone marrow. These studies confirm the remarkable indolence of low-grade gastric B-cell lymphomas of MALT type and provide evidence that eventual dissemination is not necessarily associated with genetic alteration at the immunoglobulin-heavy-chain locus.
The immunoglobulin variable region that was sequenced showed no notable accumulation of somatic mutations during the progression of the disease. This is also true in myeloma,12 B-lineage chronic lymphocytic leukemia,13 and precursor B-lineage acute lymphoblastic leukemia,14 but not in follicular lymphoma15 and hairy-cell leukemia,16 in which hypermutation in the CDR regions has been observed during clonal evolution. Other changes at the immunoglobulin locus have been observed during disease progression, among them the secondary rearrangements seen in B-lineage acute lymphoblastic leukemia14 and the class switching observed in myeloma17 and hairy-cell leukemia16. There may be changes in the immunoglobulin locus in MALT lymphomas during the course of the disease, but they have yet to be identified. Although our data suggest that the immunoglobulin gene is stable in MALT lymphoma, we have only sequenced one of the rearranged alleles in a single patient.
Malt lymphomas of other sites, including the salivary gland, lung, and thyroid, exhibit similar clinical behavior, and our results suggest that monoclonal B-cell proliferations at these sites, in an appropriate histologic setting, are good evidence of lymphoma18 rather than a prelymphomatous lesion, as others have suggested19.
The involvement of the neoplastic B-cell clone in the lesion responsible for Sjogren's syndrome in this patient is intriguing. It is conceivable that this syndrome was an unusual manifestation of a secondary salivary-gland lymphoma, thus providing a good illustration of the tendency of MALT lymphomas to disseminate to other mucosal sites. The histologic appearance of the salivary-gland lesion argued against this, however, and was more in keeping with the participation of the clone in a chronic inflammatory response. This would imply that the neoplastic cells were still functionally capable of recognizing immunologic signals, a property for which there is evidence in MALT lymphomas in the form of orderly plasmacytic differentiation,1 follicular colonization,20 and the response of at least one variety (immunoproliferative small-intestine disease) to antibiotics21. The presence of the neoplastic clone in the salivary glands might also argue in favor of a common immune system involving lymphocytes committed to mucosal sites22.
In tracing the course of this disease over a 13-year period, we found PCR to be a powerful tool with which to identify lymphoma populations in sequential biopsy specimens without the need for an amplifiable molecular genetic abnormality, such as a bcl-2 gene rearrangement. In this way, with sequential biopsy specimens, the behavior and mode of dissemination of B-cell lymphomas can be accurately investigated without fresh tissue.
Supported by the Cancer Research Campaign of Great Britain.
Source Information
From the Department of Histopathology, University College London Medical School (T.C.D., H.P., A.C.W., L.P., P.G.I.), and the Department of Oral Pathology, Institute of Dental Surgery, Eastman Dental Hospital (P.M.S.), both in London.
Address reprint requests to Professor Isaacson at the Department of Histopathology, University College London Medical School, University St., London WC1E 6JJ, United Kingdom.
References
| |||||||||||||||||||||||||||||||
Related Letters:
Mucosa-Associated Lymphoid-Tissue Lymphoma
Zelenetz A. D., Diss T.C., Pan L.X., Isaacson P.G.
Extract |
Full Text
N Engl J Med 1993;
329:1657, Nov 25, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
