Objectives: Rheumatic disorders associated with cancer include a variety of conditions, most of which have no features distinguishing them from idiopathic rheumatic disorders. It is generally held that an extensive search for occult malignancy in most rheumatic syndromes is not recommended unless accompanied by specific findings suggestive of malignancy. The objective of this review are to identify rheumatic syndromes associated with cancer, to call attention to features that may suggest the presence of a hidden cancer, and to examine the role to additional clinical and laboratory data as clues to the possible neoplastic cause of those syndromes.
Methods: A MEDLINE search of the literature dealing with cancer-associated rheumatic syndromes was conducted.
Results: Review of the literature identified significant progress in this area. First, the association of malignancy with certain rheumatic syndromes was convincingly established, such as asymmetric polyarthritis presenting in the elderly with an explosive onset, rheumatoid arthritis with monoclonal gammopathy, Sjögren's syndrome with monoclonality, hypertrophic osteoarthropathy, dermatomyositis, polymyalgia rheumatica with atypical features, Lambert-Eaton myasthenic syndrome, palmar fasciitis and arthritis, eosinophilic fasciitis poorly responsive to corticosteroid therapy, erythema nodosum lasting more than 6 months, and onset of Raynaud's phenomenon or cutaneous leukocytoclastic vasculitis after age 50 years. Second, the list of cancer-associated rheumatic syndromes was extended by including additional entities such as benign edematous polysynovitis, sacroiliitis, adult-onset Still's disease, dermatomyositis sine myositis, systemic sclerosis, Sweet's syndrome, osteomalacia, skeletal hyperostosis, antiphospholipid syndrome, and essential mixed cryoglobulinemia. Third, evidence was provided substantiating that certain long-standing rheumatic syndromes, in particular rheumatoid arthritis, Felty's syndrome, Sjögren's syndrome, dermatomyositis, systemic sclerosis, systemic lupus erythematosus, and temporal arteritis behave like "premalignant conditions." Fourth, it was shown that the recognized tumor markers alpha-fetoprotein, prostate-specific antigen, CA-125, CA 19-9, and CA-3 have low sensitivity and specificity in screening for occult cancer in a population of rheumatic patients, whereas the presence of a monoclonal gammopathy in rheumatoid arthritis and the monoclonal antibody 17-109 in Sjögren's syndrome are reliable signs of malignant transformation.
Conclusions: The presence of specific rheumatic syndromes and certain clinical and laboratory findings may justify a workup for hidden cancer. Studies of the epidemiology of the cancer-associated rheumatic syndromes and evaluation of the validity of aforementioned clues in prospective studies are goals for future investigations.