The European League Against Rheumatism Sjögren’s Syndrome Disease Activity Index measured at the time of diagnosis predicted the development of lymphoma and death in Spanish patients with severe primary Sjögren’s syndrome in a large, multicenter registry.
"We identified a specific hematological and immunological profile (cytopenias, hypocomplementemia, monoclonal band, and cryoglobulinemia) as laboratory predictors of hematological neoplasia in these patients," said lead study author Dr. Pilar Brito Zerón. "If you have an SS [Sjögren’s syndrome] patient with these features, you have to be very careful because this patient has a higher probability of developing a lymphoma."
"Physicians have had an activity index tool for other diseases for a long time, but there was nothing for SS until recently," when the EULAR Sjögren’s Syndrome Disease Activity Index (ESSDAI) was published in 2010, Dr. Brito Zerón said. "In Spain, we have one of the largest cohorts of SS patients in the world," so it was a good opportunity to test the ESSDAI.
Dr. Brito Zerón of Hospital Clinic in Barcelona and her colleagues studied patient records from the GEAS-SS multicenter registry, a cohort of 921 patients with SS from 20 medical centers in Spain, and retrospectively calculated their 2010 ESSDAI. During a mean follow-up period of 75 months, 25 (3%) of 904 patients developed lymphoproliferative disease; 17 were excluded because they had lymphoma before their primary SS diagnosis. Two-thirds were MALT (mucosa-associated lymphoid tissue) lymphomas, 80% of which were located in the parotid glands.
The investigators found that the following baseline features at diagnosis were most associated with lymphoma development: male gender (hazard ratio [HR], 5.78; 95% confidence interval [CI], 2.14-15.63); cryoglobulins (HR, 4.44; 95% CI, 1.86-10.58); monoclonal serum band (HR, 4.23; 95% CI, 1.38-13.02); C3 values less than 0.82 g/L (HR, 3.75; 95% CI, 1.38-10.19); C4 values less than 0.07 g/L (HR, 3.22, 95% CI, 1.08-9.61); and older age (HR, 1.04; 95% CI, 1.00-1.07). Gender, low C3, monoclonal band, and cryoglobulins were significant independent variables related to lymphoma, Dr. Brito Zerón reported at the annual European Congress of Rheumatology.
An ESSDAI score of one or greater in the constitutional (HR, 4.06; 95% CI, 1.54-10.70) and hematologic (HR, 2.59; 95% CI 1.16-5.78) domains was associated with the development of lymphoma, with hematologic activity being independently associated. In the constitutional domain, patients with the highest degree of activity – including fever greater than 38.5° C, night sweats, and/or involuntary weight loss of at least 10% – showed the highest risk of developing lymphoma (HR, 9.11; 95% CI, 2.51-33.12).
At the time of diagnosis with the 2002 primary SS classification criteria, patients had a mean baseline ESSDAI of 5.81. During follow-up, the patients accumulated another mean 3.34 points for a cumulative ESSADI of 9.15. A large majority of patients were women (94%) and had a mean age of nearly 54 years at the time of diagnosis. Most of the 921 patients in the registry had xerostomia (96%), xerophthalmia (95%), positive ocular tests (93%; 805 of 863), grade 3-4 parotid scintigraphy (88%; 598 of 676), and positive salivary gland biopsy (88%; 424 of 482). Cytopenias occurred in 34% overall, including anemia (17%), leucopenia (20%), and thrombocytopenia (9%). Immunologic disease characteristics of the patients included positive autoantibody tests for antinuclear antibodies (90%), anti-Ro (73%), rheumatoid factor (57%), and anti-La (46%). Others had low C4 (12%) or C3 (9%) levels and low cryoglobulins (12%) or monoclonal gammopathy (9%).
The investigators also correlated the baseline ESSDAI score with mortality. After an average follow-up of 75 months, 83 (9%) patients died. Deaths were attributed to causes related to SS (27 patients), cardiovascular disease (20 patients), infections (17 patients), and other causes (11 patients). The cause of death was unknown in eight patients.
The active ESSDAI domains that were associated with death were the constitutional (HR, 2.66; 95% CI, 1.38-5.11), pulmonary (HR, 2.13; 95% CI, 1.09-4.16), and biologic (HR, 3.01; 95% CI, 1.91-4.76), with the pulmonary and biologic domains being independently associated with death.
Further analysis revealed that a score of one or greater in the constitutional, lymphadenopathy, hematologic, and biologic domains was predictive of death related to SS (HRs ranging from 2.59 to 7.88), while activity at the constitutional, cutaneous, pulmonary, renal, neurologic, and hematologic domains predicted mortality related to infection (HRs ranging from 3.7 to 9.29). The investigators found no associations between activity in specific ESSDAI domains and death from cardiovascular disease or other causes.
"Activity of constitutional and lymphadenopathy domains, closely related to lymphoma, correlated with death caused by SS itself, while activity in the main extraglandular sites of involvement (in which high doses of corticosteroids and immunosuppressive agents are used) correlated principally with death caused by infection," Dr. Brito Zerón said. "ESSDAI is a useful tool to score systemic activity in patients with primary SS not only in prospective studies, but also in clinical trials that evaluate the efficacy of a specific drug."