|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2005.04.6227 on March 6 2006 © 2006 American Society of Clinical Oncology. Prospective Study of Survival Outcomes in Non-Hodgkin's Lymphoma Patients With Rheumatoid Arthritis
From the Departments of Medicine, Preventive and Societal Health, and Pathology, University of Nebraska Medical Center, and Omaha Veterans' Affairs Medical Center; Department of Medicine, Mayo Clinic, Rochester, MN; Department of Hematology, University of College Hospital Galway, Galway, Ireland. Address reprint requests to Ted R. Mikuls, MD, MSPH, Assistant Professor, Section of Rheumatology and Immunology, Department of Medicine, 983025 Nebraska Medical Center, Omaha, NE 68198-3025; e-mail: tmikuls{at}unmc.edu
PURPOSE: Although preliminary studies suggest that non-Hodgkin's lymphoma (NHL) complicating rheumatoid arthritis (RA) may be a clinically distinct entity compared with that occurring in the general population, studies examining the impact of antecedent RA on survival are limited. In this prospective study, we examined the association of RA with survival in patients with NHL. PATIENTS AND METHODS: Using two large lymphoma registries, we identified patients with evidence of RA preceding NHL. Survival in RA patients was compared with that of controls using proportional hazards regression, adjusting for the effects of age, sex, lymphoma diagnosis-to-treatment lag time, calendar year, International Prognostic Index score, and NHL grade. RESULTS: The frequency of NHL subtypes was similar in RA patients (n = 65) and controls (n = 1,530). Compared with controls, RA patients with NHL had similar overall survival (hazard ratio [HR] = 0.95; 95% CI, 0.70 to 1.30) but were at lower risk of lymphoma progression or relapse (HR = 0.41; 95% CI, 0.25 to 0.68) or death related to lymphoma or its treatment (HR = 0.60; 95% CI, 0.37 to 0.98), but were more than twice as likely to die from causes unrelated to lymphoma (HR = 2.16; 95% CI, 1.33 to 3.50). CONCLUSION: RA is associated with improved NHL-related outcomes, including a 40% reduced risk of death occurring as a result of lymphoma or its treatment and approximately a 60% lower risk of lymphoma relapse or progression compared with non-RA controls. However, the survival advantage gained in RA from the acquisition of lymphomas with favorable prognoses is negated through an increased mortality from other comorbid conditions.
The risk of lymphoma is increased in patients with rheumatoid arthritis (RA) and other systemic inflammatory disorders. Confirmed by several investigations,1-6 Isomaki et al7 were the first to report this association, observing a 2.5-fold increased risk of lymphoma in RA patients compared with the general population. Although the reason for this association is incompletely understood, several possible explanations have emerged including the role of immunosuppression,8-12 Epstein-Barr virus infection,13-17 and unchecked systemic inflammation.18-20 While its etiology has not been adequately defined, it is clear that the lymphomas arising from RA are a heterogeneous group of malignancies with diverse clinical features. In a small cohort of RA patients who all had diffuse large B-cell lymphoma, median survival from the time of lymphoma diagnosis was only 6 months,19 suggesting that lymphomas complicating RA may portend a poor prognosis. Alternatively, there have been reports of spontaneous lymphoma regression in RA following the withdrawal of immunosuppression,16,21 implying that RA-associated lymphomas may follow a more benign course. To more closely examine whether lymphomas in RA are clinically distinct, we examined the association of antecedent RA with survival outcomes in patients with incident non-Hodgkin's lymphoma (NHL).
Patient Population and Case Ascertainment Records from the Nebraska Lymphoma Study Group (NLSG; 1984-2002) and the Mayo Clinic (Rochester, MN; 1988-1998) lymphoma registries were reviewed. Both databases include sociodemographics, baseline clinical and laboratory data, and comprehensive longitudinal data pertinent to lymphoma diagnoses. Working Formulation diagnoses were reclassified according to the World Health Organization by a board-certified hematopathologist after review of available morphologic and immunophenotypic data. Cases with insufficient data for definitive classification were left unclassified. Considering the small number of patients with Hodgkin's disease, we limited our analyses to patients with NHL. For both databases, cause of death was ascertained primarily through medical record review and next-of-kin interview. When necessary, these records were supplemented through review of death certificates and the use of data from the National Death Index. Possible RA cases were identified by different means based on the database of origin and on the calendar year. Possible RA patients from the Mayo Clinic registry were identified by cross-linking administrative diagnostic codes for NHL and RA. NLSG case subjects who were enrolled from 1990 to 1998 were asked in a baseline questionnaire whether they had RA. For NLSG patients enrolled outside of this period (1984-1989 and 1999 to 2002), all available records were reviewed. RA status was then classified as definite or probable based on the quality of evidence in available medical records. RA cases were excluded if the date of NHL diagnosis preceded RA onset. Definite cases included those who had a documented RA diagnosis from a board-certified rheumatologist or documentation of at least four of seven American College of Rheumatology (ACR) classification criteria.22 Probable RA cases included those with a diagnosis from a nonrheumatologist plus documentation of at least one of the following: (1) radiographic erosions characteristic of RA, (2) RA-associated deformities on examination (ie, swan-neck deformity), or (3) use of a disease-modifying antirheumatic drug. For all definite and probable RA cases, records were reviewed to define rheumatoid factor status, date of RA onset, and exposure to disease-modifying therapies. Based on its inclusion of a large sample of well-characterized and representative lymphoma cases uniformly staged and treated, the NLSG served as the source of controls. With an anticipated disease prevalence of less than 1%,23 records of control subjects were not reviewed. To assure that the comparator group had similar characteristics as RA cases, controls included all other NLSG subjects with NHL meeting the following criteria: (1) subjects with the same NHL subtypes observed among RA cases using the Working Formulation (uniformly available); (2) those subjects between the ages of 40 and 90 years (the age range of RA subjects); (3) those subjects with an initial lymphoma diagnosis-to-treatment lag time of less than 90 days; and (4) a calendar year of initial lymphoma treatment between January 1, 1984, and December 31, 2002. Initial lymphoma treatments were categorized as potentially curative (chemotherapy, surgical resection, radiation, or a combination of modalities) versus conservative (palliative/observation). Lymphoma grade was categorized as follows: aggressive lymphomas included diffuse large B-cell, follicular grade 3, Burkitt and Burkitt-like, composite lymphomas with a diffuse large B-cell component, and B-cell immunoblastic subtypes; indolent lymphomas included follicular grades 1 and 2, small lymphocytic, and mucosal-associated lymphoid tumor (MALT) subtypes. Because their natural course is difficult to predict, indeterminate grade NHLs were excluded. Excluded cases included RA patients with mantle cell (n = 1), peripheral T-cell (n = 1), and unclassifiable (n = 5) subtypes, leaving 65 cases eligible for the analysis (44 from Mayo Clinic and 21 from the NLSG). Eligible controls were also limited to those with either aggressive or indolent subtypes.
Analyses We created cumulative incidence curves to compare the risk of cause-specific death between the RA cases and controls in the presence of competing risks of death from other causes. The specific causes of death examined included lymphoma and treatment-related deaths and deaths not related to lymphoma or lymphoma treatment. For the analysis of each type of death, other causes of death were considered competing events. Cumulative incidence curves were also created to examine the risk of lymphoma relapse or progression between the RA cases and controls with death as a competing risk. Ninety-five percent CIs for cumulative incidence at 60 months were calculated using the log (-log) transformation. Cox regression analysis, adjusted for competing risks, was used to explore the association of RA status with specific cause of death in addition to risk of lymphoma relapse or progression.26,27 Covariates in the analyses included those described in the previous paragraph for the overall and event-free survival analyses. To explore the effect of lymphoma case mix further, we repeated the regression modeling after limiting the data set to RA cases and controls with diffuse large B-cell NHL, which was the most prevalent subtype. We also examined the effect of RA case classification by repeating the analyses and limiting the data set to definite RA cases and controls. Analyses were performed using either SAS System for Windows, version 9.1 (SAS Institute, Cary, NC) or S-plus (Version 3.3, Statistical Sciences, Seattle, WA) version 6.2 for Windows (Insightful Corporation, Seattle, WA).
We identified 65 NHL cases with either definite (n = 54) or probable (n = 11) antecedent RA. During this follow-up interval, there were 1,530 non-RA controls eligible for the analyses. The median follow-up time was 77.4 months for RA cases and 60.8 months for controls. Characteristics of RA cases and controls are listed in Table 1. In contrast with the controls, patients with RA were more often women (62% v 49%; P = .06).
RA patients had a mean disease duration of 14 years (standard deviation, 10 years) before NHL diagnosis, most of the patients were rheumatoid factor positive (69%), and almost all had prior exposure to disease-modifying therapies (95%). The most common previously used disease-modifying drugs included hydroxychloroquine and gold salts, followed by methotrexate and sulfasalazine (Table 1). The most frequently observed NHL diagnoses in RA patients included diffuse large B-cell (n = 28, 43%) and follicular subtypes (n = 17, 26%; Table 2). The distribution of grade (aggressive and indolent) was not significantly different between controls and RA cases (P = .07), though low-grade lymphomas were more common in patients with RA than in controls (34% v 23%). Compared with controls, RA patients were more likely to receive a conservative therapeutic approach (palliative/observation) as their initial lymphoma treatment (15% v 2%; P < .001). There were no differences between RA cases from the two databases based on age, sex, International Prognostic Index score, or lymphoma grade.
Over the entire follow-up period (median, 5.1 years; range, 0.1 to 19.6 years), there were a total of 941 deaths, including 44 patients with RA. Deaths unrelated to lymphoma or its treatment through the entire follow-up period were most commonly due to coronary artery disease and stroke, altogether accounting for 13.6% of all deaths (n = 6) among the RA cases and 7.7% (n = 69) among the controls. After 60 months of follow-up, there were 25 deaths among RA cases and 675 deaths from the control group (Table 3). Among RA patients, the cumulative incidence of deaths unrelated to the lymphoma or lymphoma treatment was higher than the incidence of death among non-RA controls (15% in RA cases v 7% in controls at 60 months; Fig 1A). In contrast with RA patients with lymphoma, the cumulative incidence of lymphoma and treatment-related death at 60 months was higher among controls (34% in controls v 21% in RA cases; Fig 1B). At 60 months, the cumulative incidence for lymphoma progression or relapse was 21% for RA cases compared with 42% for non-RA controls (Fig 1C). Based on cumulative incidence curves, the risk of specific causes of death, lymphoma progression, or relapse was similar for RA cases from the two centers.
After adjusting for multiple confounders, RA was not associated with overall survival (HR = 0.95; 95% CI, 0.70 to 1.30; Table 4). However, when compared with controls with lymphoma, RA patients had a 30% lower risk for the combined outcome of death, lymphoma progression, or relapse (HR = 0.70; 95% CI, 0.52 to 0.95). In multivariable analyses, adjusting for competing events, RA was associated with a 59% reduced risk for progression or relapse (HR = 0.41; 95% CI, 0.25 to 0.68) and a 40% reduced risk of death from lymphoma or its treatment (HR = 0.60; 95% CI, 0.37 to 0.98) compared with non-RA controls. In contrast, lymphoma patients with RA were more than twice as likely (HR = 2.16; 95% CI, 1.33 to 3.50) as controls to die as a result of causes unrelated to the lymphoma or its treatment.
When the analysis was limited to patients with diffuse large B-cell NHL, the results were similar. Among patients with diffuse large B-cell lymphoma, there was no association of RA with overall survival (HR = 1.02; 95% CI, 0.66 to 1.57), though there were nonsignificant trends favoring improved survival in RA patients compared with control participants in terms of event-free survival (HR = 0.81; 95% CI, 0.53 to 1.26) and death from lymphoma or lymphoma treatment (HR = 0.63; 95% CI, 0.29 to 1.36). The risk of NHL relapse or progression remained significantly lower in RA patients (HR = 0.42; 95% CI, 0.20 to 0.91), whereas deaths unrelated to lymphoma or its treatment were again more frequent in patients with RA (HR = 1.64; 95% CI, 0.85 to 3.15) compared with controls. Limiting the analysis to cases with definite RA did not alter these results. Compared with control participants, patients with definite RA (including all NHL subtypes) had similar overall survival (HR = 0.90; 95% CI, 0.64 to 1.26) and a lower likelihood of progression or relapse (HR = 0.40; 95% CI, 0.23 to 0.70), but were at increased risk of death unrelated to lymphoma or its treatment (HR = 2.07; 95% CI, 1.24 to 3.46). Likewise, the inclusion of RA cases and non-RA controls with indeterminate-grade NHL did not alter our findings.
In this study of lymphoma patients, RA was associated with better NHL-related outcomes when compared with non-RA controls, including a 40% reduced risk of deaths occurring from lymphoma or its treatment and an approximate 60% lower risk of lymphoma relapse or progression. However, our results also show that any survival advantage gained in RA patients from the acquisition of more indolent lymphomas seems to be negated through an increased mortality from other comorbid conditions. Consistent with many studies showing increased cardiovascular mortality in RA,28-30 our results suggest that at least part of the excess mortality in lymphoma can be attributed to coronary artery disease and stroke, which together were two times more frequent in RA patients with lymphoma than non-RA controls. There have been reports of spontaneous lymphoma regression in RA, particularly with the withdrawal of immunosuppressive therapy.16,21 Salloum et al16 and Usman and Yunus21 have independently reported on NHL response rates in RA with methotrexate withdrawal used as a primary intervention. Of a combined group of 34 patients in these series,16,21 nine patients (26%) experienced complete remission, a rate that exceeds the rate expected for de novo NHL.31,32 In addition to suggesting that immunosuppressive therapy may play a causal role in some lymphomas, these reports also indicate that lymphomas developing in RA patients may be clinically distinct and may portend favorable long-term outcomes. The lower rate of lymphoma relapse, progression, and tumor-related deaths in RA patients parallels observations of NHL arising in other populations, including those who have been treated with immunosuppressive agents and in patients with acquired immunodeficiency.33 Although lymphomas developing post-transplantation are usually of aggressive large-cell subtypes, they may regress completely when immunosuppressive treatments are decreased or discontinued, even in the absence of chemotherapy or radiation treatments.34 Lymphomas developing in this setting have been attributed to the level of immunosuppression and resulting infection caused by the Epstein-Barr virus.35 However, reports examining the role of immunosuppression and the Epstein-Barr virus in RA-related lymphomas have been inconsistent with case series16,17 showing higher rates of viral infection and population-based studies19,36 showing lower rates of viral infection. Because in situ hybridization data were not routinely available for most patients in this study, we were unable to examine the impact of Epstein-Barr virus infection on survival. Our results differ from those of a large Swedish cohort study that examined lymphoma subtypes in patients with RA.19 In that study, investigators identified 33 cases of NHL in a large population of RA patients observed over a 20-year period. They found that diffuse large-B cell subtypes were overrepresented, accounting for two thirds of all NHL tumors in RA patients. For the 22 RA patients with diffuse large-B cell lymphoma, the median overall survival was only 6 months (range, 2 weeks to 13 years). In contrast, we observed a similar frequency of diffuse large B-cell lymphomas in RA patients compared with non-RA controls, and among our RA patients with this subtype we found an overall median survival approaching 5 years (95% CI, 9 to 93 months). There are important differences between the Swedish study and our report that may explain this disparity. The Swedish cohort predates ours, with follow-up occurring between 1964 and 1984, which is notable because calendar year was a significant determinant of survival in our study. This may relate, at least in part, to the substantial advances in lymphoma treatment over this period. Previous exposure to disease-modifying antirheumatic drugs was also far more frequent among cases in our study compared with the Swedish study (95% v 49%), which is noteworthy because unchecked systemic inflammation has been cited as a potential risk factor for lymphoma development.18,20 Although controls from one of the participating centers were not available for this study, it is unlikely that systematic differences in patients from the two registries explain our findings. There were no differences in RA patients from the two databases in terms of age, sex, lymphoma subtypes, cause-specific deaths, or distribution of International Prognostic Index scores. Due to the heterogeneity of previous RA treatments received, we were not able to examine the impact of specific disease-modifying therapies on survival. The lack of comprehensive data regarding the discontinuation of immunosuppressive therapies in RA subjects may be important, since discontinuation of these agents could paradoxically lead to both a good tumor response and increased comorbidity related to unchecked inflammation. Because of limited exposure and a follow-up interval that predated their availability, we could not assess the association of previous exposures to RA biologic therapies with survival. Additionally, our analysis was limited to NHL patients from registries with predominantly white enrollees and therefore it may not be generalizable to other RA populations. Another limitation of our study is that we did not have the opportunity to clinically examine each RA case. However, our case validation methods, including the requirement that definite cases meet American College of Rheumatology criteria based on medical record review and/or confirmation by a rheumatologist, make it unlikely that we incorrectly classified patients as having definite RA. This method is consistent with validation studies of diagnostic criteria that use reports from subspecialists as the gold standard.37 Although we included probable RA cases in the analysis, our results were not changed when we limited the analysis to those participants with definite disease. It is possible that individuals with milder cases of RA or those with RA who did not seek medical attention may have gone unrecognized. However, the large size of the comparison group makes false negatives a less important concern because inclusion of RA patients in the control group would only serve to bias our results toward the null. These results have important implications. When the natural course of RA is complicated by lymphoma, it is essential to recognize that other comorbidities (eg, coronary artery disease and stroke) may persist as determinants of survival and should be addressed as part of a comprehensive management approach. Additional studies are warranted to more closely examine the biologic characteristics of lymphoma developing in the context of RA and to assess whether the distinct clinical phenotype observed in our study can be predicted at a molecular level. In addition to furthering our understanding of lymphoma pathogenesis, such information would represent an important advance in lymphoma management and would ultimately allow for a more informed risk assessment of new and evolving RA therapies implicated in playing causal roles in lymphoproliferative disease.
The authors indicated no potential conflicts of interest.
Supported in part by training grants from the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases (Grant No. K23 AR0500004-01A1) and the Arthritis Foundation (T.R.M.). This study was approved by the institutional review boards at both the University of Nebraska Medical Center (Omaha, NE) and Mayo Clinic (Rochester, MN). Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Prior P, Symmons D, Hawkins C, et al: Cancer morbidity in rheumatoid arthritis. Ann Rheum Dis 43:128-131, 1984 2. Gridley G, McLaughlin J, Ekbom A, et al: Incidence of cancer among patients with rheumatoid arthritis. J Natl Cancer Inst 85:307-311, 1993 3. Mellemkjaer L, Linet M, Gridley G, et al: Rheumatoid arthritis and cancer risk. Eur J Cancer 32A:1753-1757, 1996[CrossRef] 4. Hakulinen T, Isomaki H, Knekt P: Rheumatoid arthritis and cancer studies based on linking nationwide registries in Finland. Am J Med 78:29-32, 1985[Medline] 5. Tennis P, Andrews E, Bombardier C, et al: Record linkage to conduct an epidemiologic study on the association of rheumatoid arthritis and lymphoma in the province of Saskatchewan, Canada. J Clin Epidemiol 46:685-695, 1993[CrossRef][Medline] 6. Thomas E, Brewster D, Black R, et al: Risk of malignancy among patients with rheumatic conditions. Int J Cancer 88:497-502, 2000[CrossRef][Medline] 7. Isomaki H, Hakulinen T, Joutsenlahti U: Excess risk of lymphoma, leukemia and myeloma in patients with rheumatoid arthritis. J Chronic Dis 31:691-696, 1978[CrossRef][Medline] 8. Kinlen L: Incidence of cancer in rheumatoid arthritis and other disorders after immunosuppressive treatment. Am J Med 78:44-49, 1985[Medline] 9. Silman A, Petrie J, Hazleman B, et al: Lymphoproliferative cancer and other malignancy in patients with rheumatoid arthritis treated with azathioprine: A 20 year follow up study. Ann Rheum Dis 47:988-992, 1988 10. Matteson E, Hickey A, Maguire L, et al: Occurrence of neoplasia in patients with rheumatoid arthritis enrolled in a DMARD registry. J Rheumatol 18:809-814, 1991[Medline] 11. Jones M, Symmons D, Finn J, et al: Does exposure to immunosuppressive therapy increase the 10 year malignancy and mortality risks in rheumatoid arthritis? A matched cohort study. Br J Rheumatol 35:738-745, 1996 12. Asten P, Barrett J, Symmons D: Risk of developing certain malignancies is related to duration of immunosuppressive drug exposure in patients with rheumatic diseases. J Rheumatol 26:1705-1714, 1999[Medline] 13. Kamel O, van de Rijn M, LeBrun D, et al: Lymphoid neoplasms in patients with rheumatoid arthritis and dermatomyositis: Frequency of Epstein-Barr virus and other features associated with immunosuppression. Hum Pathol 25:638-643, 1994[CrossRef][Medline] 14. Liote F, Pertuiset E, Cochand-Priollet B, et al: Methotrexate related B lymphomproliferative disease in a patient with rheumatoid arthritis: Role of Epstein-Barr virus infection. J Rheumatol 22:1174-1178, 1995[Medline] 15. van de Rijn M, Cleary M, Variakojis D, et al: Epstein-Barr virus clonality in lymphomas occurring in patients with rheumatoid arthritis. Arthritis Rheum 39:638-642, 1996[Medline] 16. Salloum E, Cooper D, Howe G, et al: Spontaneous regression of lymphoproliferative disorders in patients treated with methotrexate for rheumatoid arthritis and other rheumatic diseases. J Clin Oncol 14:1943-1949, 1996 17. Dawson T, Starkbaum G, Wood B, et al: Epstein-Barr virus, methotrexate, and lymphoma in patients with rheumatoid arthritis and primary Sjogren's syndrome: Case series. J Rheumatol 28:47-53, 2001 18. Baecklund E, Ekbom A, Sparen P, et al: Disease activity and risk of lymphoma in patients with rheumatoid arthritis: Nested case-control study. BMJ 317:180-181, 1998 19. Baecklund E, Sundstrom C, Ekbom A, et al: Lymphoma subtypes in patients with rheumatoid arthritis: Increased proportion of diffuse large B cell lymphoma. Arthritis Rheum 48:1543-1550, 2003[CrossRef][Medline] 20. Wolfe F: Inflammatory activity, but not methotrexate or prednisone use predicts non-Hodgkins lymphoma in rheumatoid arthritis: A 25-year study of 1,767 RA patients. Arthritis Rheum 41:S188, 1998 (suppl 9; abstr) 21. Usman A, Yunus M: Non-Hodgkin's lymphoma in patients with rheumatoid arthritis treated with low dose methotrexate. J Rheumatol 23:1095-1097, 1996[Medline] 22. Arnett F, Edworthy S, Bloch D, et al: The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31:315-324, 1988[Medline] 23. Silman A: Rheumatoid arthritis, in, Silman A, Hochberg M (eds): Epidemiology of the rheumatic diseases, Vol 1 (ed 2). New York, NY, Oxford University Press, 2001, pp 31-71 24. Cox D: Regression models and life tables (with discussion). J R Stat Soc B 34:187-220, 1972 25. The International Non-Hodgkin's Lymphoma Prognostic Factors Project: A predictive model for aggressive non-Hodgkin's lymphoma. N Engl J Med 329:987-994, 1993 26. Gooley T, Leisenring W, Crowley J, et al: Estimation of failure probabilities in the presence of competing risks: New representations of old estimators. Stat Med 18:695-706, 1999[CrossRef][Medline] 27. Lunn M, McNeil D: Applying Cox regression to competing risks. Biometrics 51:524-532, 1995[CrossRef][Medline] 28. Wolfe F, Mitchell D, Sibley J, et al: The mortality of rheumatoid arthritis. Arthritis Rheum 37:481-494, 1994[Medline] 29. Gabriel S, Crowson C, O'Fallon M: Comorbidity in arthritis. J Rheumatol 26:2475-2479, 1999[Medline] 30. Symmons P, Jones M, Scott D, et al: Long-term mortality outcome in patients with rheumatoid arthritis: Early presenters continue to do well. J Rheumatol 25:1072-1077, 1998 31. Gattiker H, Wiltshaw E, Galton D: Spontaneous regression in non-Hodgkin's lymphoma. Cancer 45:2627-2632, 1980[CrossRef][Medline] 32. Drobyski W, Qazi R: Spontaneous regression in non-Hodgkin's lymphoma: Clinical and pathogenetic considerations. Am J Hematol 31:138-141, 1989[CrossRef][Medline] 33. Knowles D, Chamulak G, Subar M, et al: Lymphoid neoplasia associated with the acquired immunodeficiency syndrome (AIDS): The New York University Medical Center experience with 105 patients (1981-1986). Ann Intern Med 108:744-753, 1988 34. Starzl T, Nalesnik M, Porter K, et al: Reversibility of lymphomas and lymphoproliferative lesions developing under cyclosporin-steroid therapy. Lancet 1:583-587, 1984[Medline] 35. Nalesnik M, Jaffe R, Starzl T, et al: The pathology of posttransplant lymphoproliferative disorders occurring in the setting of cyclosporine A-prednisone immunosuppression. Am J Pathol 133:173-192, 1988[Abstract] 36. Mariette X, Cazals-Hatem D, Warszawki J, et al: Lymphomas in rheumatoid arthritis patients treated with methotrexate: A 3-year prospective study in France. Blood 99:3909-3915, 2002 37. MacGregor A, Bamber S, Silman A: A comparison of the performance of different methods of disease classification for rheumatoid arthritis: Results of an analysis from a nationwide twin study. J Rheumatol 21:1420-1426, 1994[Medline] Submitted October 14, 2005; accepted January 20, 2006.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
