BIRMINGHAM, ENGLAND – In the first 6 months after the diagnosis of polymyalgia rheumatica, the risk of cancer was almost doubled, with a hazard ratio of 1.96, in a primary care–based matched cohort study of more than 12,000 individuals.
After 6 months, the cancer risk subsided, with a hazard ratio of 1.03 at 6-12 months after diagnosis, 1.04 at 1-2 years, 1.05 at 2-5 years, 1.1 at 5-10 years, and 1.00 after 10 years, reported Sara Muller, Ph.D., of the Research Institute for Primary Care & Health Sciences, Keele University, England.
The drop-off in cancer diagnosis may reflect "a differential diagnosis issue. Maybe there are early cancer symptoms that are being diagnosed as PMR [polymyalgia rheumatica]," she added at the British Society for Rheumatology annual conference.
Dr. Sara Muller
Monitor PMR patients closely for the first 6 months after diagnosis, Dr. Muller advised. Future research needs to try to tease out how to identify PMR patients who might actually have cancer from those whose condition is limited to joint problems.
PMR is the most common inflammatory rheumatologic condition in older adults. The study was performed to see if there was any link between PMR and cancer, as has been seen for rheumatoid arthritis and lymphoma.
Case reports indicate PMR has been misdiagnosed as renal, testicular, gastric, or hematologic (lymphoma) cancer. The results of a Swedish study (Rheumatology 2010;49:1158-63) suggest that cancer risk is slightly increased in patients diagnosed with PMR or giant cell arteritis. These were all secondary care studies, however, so Dr. Muller and her associates decided to look at a primary care population, where most cases of PMR are diagnosed and treated.
Using the U.K. General Practice Research Database (GPRD), now known as the Clinical Practice Research Datalink, the research team identified 2,877 cases of PMR in individuals aged 50 years or older who were diagnosed between 1987 and 1999. Each case was then matched to five individuals without PMR as controls (n = 9,942). The development of cancer was assessed from 1987 to 2011. Patients with a prior history of cancer or vascular disease were excluded from the study. The cohort was 73% female, and the mean age of the study population was 72 years.
The median observation time was 7.8 years, with some patients followed for more than 20 years. During this time, 667 (23.2%) cases of cancer were diagnosed in patients with PMR and 1,938 (19.5%) in those without, giving respective cancer diagnosis rates of 27.7 and 24.4 per 1,000 person-years.
"In those people with PMR, there were more genitourinary cancers, which were mainly prostate cancers, than in those without PMR," observed Dr. Muller. She added that there were also more cancers affecting the lymphatic system and hematopoietic tissue, and unspecified cancers categorized as "other" by the GPRD coding system.
Conversely, PMR patients were less likely than those without PMR to have cancers of the bone, connective tissue, skin and breast, digestive system and peritoneum, and the respiratory tract and intrathoracic organs.
However, these were only trends and not statistically significant. "We could not really look at the statistical significance of these differences in types of cancer because, despite this being possibly one of the largest datasets where you would find this kind of information, we still didn’t really have enough numbers to make any formal statistical analysis," Dr. Muller said.
The Royal College of General Practitioners Scientific Foundation Board supported the research. Dr. Muller had no conflicts of interest.