| |||||||||||||||||||||||||||||||||
Whether patients with rheumatic diseases are at an increased risk for lymphoma is unclear. Some authors report an increased risk that is independent of immunosuppressive therapy and could be associated with an altered immune status due to the disease process4,5,6. Other epidemiologic studies, however, have failed to demonstrate an increased risk of lymphoma in these patients regardless of whether immunosuppressive therapy is given7. Some studies have found that an increased risk of lymphoma in patients with rheumatic diseases may be related at least in part to immunosuppressive therapy with azathioprine and cyclophosphamide8. Although several reports have described methotrexate therapy as relatively safe for rheumatic diseases,9,10,11,12 recent studies suggest that disorders of immune suppression may arise as a complication of such therapy13,14.
We describe two patients with rheumatic diseases in whom malignant lymphomas (large-cell lymphoma and Hodgkin's disease) developed during treatment with methotrexate. The neoplastic cells in both patients contained EBV latent membrane protein detected by immunohistochemical studies and viral transcripts demonstrated by in situ hybridization. In each patient, the tumor regressed after the withdrawal of methotrexate. These case reports demonstrate that lymphomas whose clinical behavior is similar to that of many lymphomas occurring in patients after transplantation can develop in patients with rheumatoid arthritis or dermatomyositis who are treated with methotrexate. The immune status of patients with rheumatic diseases, the effect of cytotoxic therapy in such patients, and the relation between the immune system and the development of lymphoproliferative disorders in this setting merit further investigation.
Case Reports
Patient 1
An 86-year-old woman with an 18-year history of severe, seropositive, erosive rheumatoid arthritis was seen on October 7, 1991, with a 1-month history of a rapidly expanding mass of the left thenar eminence. She had been treated with several medications, including aspirin, other nonsteroidal antiinflammatory agents, gold sodium thiomalate, and sulindac (Clinoril). For approximately three years before presentation, she had received methotrexate, initially at a weekly dose of 10 mg given intramuscularly and then at a weekly dose of 7.5 mg orally, together with a daily dose of 5 mg of prednisone. She had had no joint-related pain for approximately 12 months while receiving this regimen.
The lump in the thenar eminence was erythematous and minimally tender. The patient had also noticed smaller lesions on the first metacarpal-phalangeal joints of both hands. She gave no history of fever, night sweats, or a substantial change in weight. Physical examination revealed a soft-tissue mass in the left thenar eminence measuring approximately 3 to 4 cm in diameter and two separate soft-tissue masses, each measuring less than 1 cm in diameter, involving the first metacarpal-phalangeal joints of the left and right hands. There was no evidence of peripheral adenopathy or hepatosplenomegaly. She had no known risk factors for infection with the human immunodeficiency virus (HIV). Laboratory studies demonstrated antinuclear antibodies at a titer of 1:160 and the absence of a monoclonal gammopathy on serum immunoelectrophoresis. Additional laboratory studies revealed a lactate dehydrogenase level of 213 U per liter, a white-cell count of 6400 per cubic millimeter, a hematocrit of 0.39, and a platelet count of 224,000 per cubic millimeter. Radiographic studies confirmed the presence of a soft-tissue mass with a definite matrical structure between the left thumb and second metacarpal joint. A biopsy of the left thenar mass was performed on November 1, 1991, and treatment with methotrexate was discontinued at that time. Histologic examination of the mass revealed a large-cell lymphoma. Shortly thereafter, without additional therapy, the soft-tissue masses were found to have regressed, and by November 18, 1991, all three soft-tissue masses showed nearly complete resolution. Sixteen months after the diagnosis, the patient had no evidence of disease.
Patient 2
A 65-year-old woman with a history of breast cancer presented with a rash involving her entire body. The results of a skin biopsy performed in May 1990 confirmed the diagnosis of dermatomyositis. The patient underwent a workup for cancer that included chest films and abdominal and pelvic computed tomographic scans. These studies demonstrated only mild fullness of the head of the pancreas without other abnormalities.
The patient was initially treated with prednisone, which resulted in some improvement of her rash; however, the rash worsened when the dose of prednisone was tapered. On February 20, 1991, treatment with methotrexate was begun. A 5-mg test dose was followed by a weekly dose of 37.5 mg given intramuscularly. Recurring fevers soon developed, beginning approximately 36 hours after each weekly dose of methotrexate and lasting approximately 48 hours. In early March 1991, the patient's lactate dehydrogenase level was found to be 818 U per liter. A skin infection subsequently developed, at which time methotrexate was withheld and antibiotics were administered. Treatment was then switched to low-dose methotrexate consisting of three 2.5-mg tablets weekly. In May 1991, a mass developed in the right inguinal region. Treatment with methotrexate was continued, and the patient was monitored. She had no risk factors for HIV infection.
On August 15, 1991, the patient was reevaluated. She had lost 9 kg (20 lb) during the six-month period since the previous evaluation and her fevers had progressed. Physical examination revealed right axillary and bilateral inguinal lymphadenopathy and changes characteristic of dermatomyositis throughout the skin. A chest film revealed multiple nodular opacities, measuring approximately 1.5 cm in diameter, involving both lower lung fields that were considered highly likely to represent metastatic disease. A computed tomographic scan showed retroperitoneal and bilateral inguinal lymphadenopathy and hypodense lesions involving both lobes of the liver. Methotrexate therapy was discontinued.
On September 9, 1991, a biopsy of the right inguinal lymph node was performed, which showed features diagnostic of nodular sclerosing Hodgkin's disease. A clinical evaluation approximately four weeks after the discontinuation of methotrexate, during which time the patient received no additional therapy, revealed regression of the lymphadenopathy. The patient was seen again on November 11, 1991, at which time there was complete resolution of inguinal adenopathy and of the pulmonary nodules on the basis of chest films. Liver function was normal. A follow-up computed tomographic scan showed resolution of retroperitoneal and inguinal adenopathy. Twelve months after the follow-up scan, there was still no clinical evidence of adenopathy.
Methods
Histologic and Immunohistochemical Studies
Histologic sections were stained with hematoxylin and eosin. Fresh-frozen tissue was not available for study. Immunohistochemical studies of paraffin sections were performed with a biotin-streptavidin method described previously15. Incubation with primary antibody (Table 1) was followed by incubation first with a biotinylated goat antimouse second-stage antibody (Jackson Immunoresearch Laboratories, West Grove, Pa.) and then with peroxidase-conjugated streptavidin (Jackson Immunoresearch Laboratories). An IgM-specific second-stage antibody (Jackson Immunoresearch Laboratories) was used for Leu-M1, an IgM isotype. Diaminobenzidine was used as a chromogen.
|
In situ hybridization studies of EBV RNA were performed with a 30-base oligonucleotide probe complementary to a portion of the EBER-1 gene, as described previously16. Paraffin sections 5 microm thick were deparaffinized, rehydrated, predigested with pronase, prehybridized, and then hybridized overnight at a concentration of 0.25 ng of biotinylated probe per microl. The probe was detected by exposing the samples to avidin-alkaline phosphatase conjugate and then eliciting the signal with McGadey's substrate. The appearance of a blue-black color within the nucleus darker than that present in background levels was considered a positive reaction. For each assay, a known EBV-positive neoplasm served as a positive control and an EBV-negative lymphoid tissue served as a negative control.
Results
Patient 1
Histologic sections of the soft-tissue mass in the thenar eminence showed a diffuse large-cell lymphoma associated with extensive areas of necrosis, similar to those seen in lymphomas occurring in patients after transplantation3. The malignant lymphoid cells had varied cytologic features; some were large and noncleaved, some were immunoblastic, and others were plasmacytic (Figure 1A). Mixed with the malignant cells was a minor population of reactive lymphoid cells. The lymphoma infiltrated skeletal muscle, connective tissue, and nerves. Immunohistochemical studies performed on paraffin-embedded sections demonstrated that the malignant-cell population had a CD20+ (B-lineage cells), CD30+ immunophenotype (Table 1). The neoplastic cells showed strong reactivity for EBV latent membrane protein (Figure 1B). In situ hybridization studies demonstrated the presence of EBV RNA in the nuclei of the neoplastic cells (Figure 1C).
|
The examination of sections from the right inguinal lymph node revealed features diagnostic of nodular sclerosing Hodgkin's disease. Nodules rich in lymphocytes, with scattered eosinophils, neutrophils, and plasma cells, were separated by bands of fibrous tissue. A few Reed-Sternberg cells and their variants were present (Figure 2A and Figure 2B). The immunophenotypic profile was typical of that seen in nodular sclerosing Hodgkin's disease, with expression of CD15 (Leu-M1) and CD30 (Ber-H2) and the lack of expression by Reed-Sternberg cells of antigens associated with B and T cells (Table 1). In situ hybridization studies showed evidence of EBV RNA in virtually all the Reed-Sternberg cells and their variants, whereas reactive lymphocytes were generally negative for EBV RNA (Figure 2C). Reed-Sternberg cells and their variants also expressed EBV latent membrane protein on immunohistochemical analysis (Figure 2D).
|
We describe the occurrence of hematolymphoid tumors during treatment with methotrexate in two patients with connective-tissue diseases. One patient had large-cell lymphoma, and the other presented with nodular sclerosing Hodgkin's disease. Both lymphomas were associated with EBV. In both patients, the tumor regressed after treatment with methotrexate was stopped. The clinical behavior of these two cases of lymphoma is similar to that of many lymphomas that develop in patients who receive immunosuppressive therapy after organ transplantation.
It is well known that patients who receive immunosuppressive therapy after organ transplantation are at increased risk for lymphoproliferative disorders. The morphology of these lesions is varied. Some have a polymorphous appearance and consist of B lymphocytes at various stages of activation and differentiation, whereas others have a monomorphous appearance, most typically that of diffuse large-cell lymphoma3. These tumors usually express EBV antigens and viral transcripts and may regress on withdrawal of the immunosuppressive agents1. A generally accepted hypothesis for the development of these tumors is that the lack of immune surveillance in these patients allows the proliferation of EBV-transformed B lymphocytes. Initially, this process appears to be polyclonal; with progression, the most proliferative clone predominates, so that monoclonal tumors can be identified17. Interestingly, studies of patients with rheumatoid arthritis have shown the inability of T cells to prevent the proliferation of EBV-infected B cells in vitro18,19.
Although Hodgkin's disease has been reported after organ transplantation,20,21 it is not a neoplasm that has been typically associated with immunosuppression, and regression without therapy has not to our knowledge been reported in patients with Hodgkin's disease. The histologic features of the lymph-node-biopsy specimen from Patient 2 were considered diagnostic, and the immunophenotypic findings were typical of those described in Hodgkin's disease22. Previous studies have demonstrated that 19 to 48 percent of the cases of Hodgkin's disease that occur in the general population are associated with EBV23,24,25. A recent study found that Hodgkin's disease in HIV-positive patients is significantly more likely to be associated with EBV than is Hodgkin's disease in HIV-negative patients26. The role of EBV in the pathogenesis of Hodgkin's disease requires additional investigation.
The unavailability of fresh-frozen tissue from our patients prohibited molecular analysis for immunoglobulin gene rearrangements or for the presence of clonal EBV genome. Although such molecular studies provide additional information about the biology of these lymphoid proliferations, they do not necessarily provide information about the clinical behavior of these tumors. As described by Nalesnik and coworkers in an extensive review of post-transplantation lymphoproliferative disorders, some patients may die as a result of their polyclonal lymphoid proliferations, whereas other patients with monoclonal neoplasms may be cured by reducing the level of immunosuppression27.
The association between immunosuppressive therapy and lymphoma in patients with connective-tissue disorders is controversial. An increased risk of hematolymphoid neoplasms in patients with rheumatoid arthritis who were treated with immunosuppressive therapy was reported in one study8. The drugs implicated in this study were azathioprine and cyclophosphamide. Other reports have described an increased risk of lymphoma in patients with rheumatoid arthritis; however, this risk was not associated with therapy but instead appeared to be related to the disease process itself4,5,6. On the other hand, some studies have failed to identify any increased risk of lymphoma in patients with rheumatoid arthritis, regardless of therapy7. In a recent report, an EBV-positive lymphoma developed in a patient with rheumatoid arthritis during treatment with cyclosporine; however, no mention was made of the effect of the withdrawal of cyclosporine therapy28.
Methotrexate therapy for patients with connective-tissue disease has not generally been associated with the development of cancer10,11,12; however, two recent reports record the development of lymphoma in patients treated with methotrexate for rheumatoid arthritis13,14. The first report describes a patient with rheumatoid arthritis who was taking low-dose methotrexate (5 to 10 mg per week) for 33 months and in whom a follicular large-cell lymphoma developed. No mention was made of the effect of the withdrawal of methotrexate therapy, and the authors did not state whether the EBV viral genome was present or absent in the tumor13. Therefore, as the authors suggest, this case could represent a coincidental occurrence of lymphoma and rheumatoid arthritis. The second report describes complications of therapy with low-dose methotrexate in four patients with rheumatic diseases. These complications included disseminated herpes zoster in two patients, Pneumocystis carinii pneumonia in one patient, and diffuse adenopathy and an abdominal-wall mass in a fourth patient with rheumatoid arthritis, who had been treated for seven years with methotrexate (10 to 15 mg per week) and prednisone (5 mg per day). A biopsy of the abdominal-wall mass was performed and revealed a lymphoproliferative disorder interpreted as "consistent with diffuse small cleaved cell lymphoma." Because of a localized herpes zoster infection, methotrexate was withdrawn (treatment with prednisone was continued), and within four weeks the patient began to improve without additional therapy. Over the next six months, all evidence of adenopathy and the abdominal mass resolved. The results of any EBV studies of the lymphoproliferative disorder were not reported14.
The case reports we describe clearly show that methotrexate therapy can have a role in the development of EBV-associated lymphoma in patients with rheumatoid arthritis or dermatomyositis. The disappearance of tumor after the withdrawal of methotrexate suggests that the therapy was a major factor in the generation of these tumors. The absence of an association between the development of lymphoma and methotrexate treatment in patients with psoriasis9 raises the possibility that other factors, such as compromise of the immune status by the disease process, may also have an important role. This may particularly be true for rheumatoid arthritis, a disorder in which patients have an abnormally elevated frequency of EBV-infected B lymphocytes and in which there has been shown to be a major T-cell defect in EBV-specific suppression18,29.
Supported in part by grants (34233, 33119, and 50341) from the National Cancer Institute, National Institutes of Health.
We are indebted to Dr. Mickey J. Myhre for sending the biopsy specimens from Patient 2 to us and to Eva Pfendt for technical assistance with immunohistochemical studies.
Source Information
From the Department of Pathology, Stanford University Medical Center, Stanford, Calif. (O.W.K., M.R., R.A.W., R.F.D.); the Department of Pathology, City of Hope National Medical Center, Duarte, Calif. (L.M.W.); the Department of Molecular and Experimental Medicine and Division of Hematology/Medical Oncology (G.J.D.Z.), the Division of Rheumatology (P.K.H.), and the Department of Pathology (B.A.R.), Scripps Clinic and Research Foundation, La Jolla, Calif.; and the Division of Hematology/Oncology, Mountain States Tumor Institute, Boise, Idaho (P.G.M.).
Address reprint requests to Dr. Kamel at the Department of Pathology, Stanford University Medical Center, 300 Pasteur Dr., Stanford, CA 94305.
References
| |||||||||||||||||||||||||||||||||
Related Letters:
Reversible Lymphomas
Shiroky J. B., Newkirk M. M., Kamel O. W., van de Rijn M., Warnke R. A., Dorfman R. F.
Extract |
Full Text
N Engl J Med 1993;
329:1657-1658, Nov 25, 1993.
Correspondence
Cyclosporine and Methotrexate for Severe Rheumatoid Arthritis
Schlesinger N., Huppert A., Hoch S., Malleson P., Steinberg A. D., Rosh J. R., Birnbaum A. H., van de Rijn M., Kamel O. W., Storb R., Thomas E. D., Tugwell P., Yocum D., Pincus T., Wells G.
Extract |
Full Text
N Engl J Med 1995;
333:1567-1569, Dec 7, 1995.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | 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
