O30. The Risk of a Subsequent Cancer Diagnosis after PMR: A Primary Care Database Study

Muller, Sara; Mallen, Christian D.; Belcher, John; Helliwell, Toby; Hider, Samantha L.

Abstract

Background: PMR is the commonest inflammatory rheumatological disorder managed in primary care. Other inflammatory conditions such as RA and psoriasis have been associated with an increased incidence of cancer, specifically haematological malignancies and lymphoma. Small PMR studies have not shown consistent associations [1,2]. The aim of this study was to investigate whether there is an association between PMR and cancer in a primary care population.

Methods: All individuals in the General Practice Research Database (GPRD) aged ≥50 years and with a Read-coded diagnosis of PMR who were treated with corticosteroids between 01/01/1987 and 31/12/1999 were matched by age, gender, practice and consultation year to up to five patients without PMR. Those with a prior history of vascular disease or cancer were excluded.

The outcome of interest was a diagnosis of cancer following PMR diagnosis. This was defined as having received a Read code in Chapter B (neoplasms), excluding subchapter B7 (non-malignant neoplasms). Individuals were followed-up until first cancer diagnosis, death or May 2011, whichever was earliest. Event times were compared using a Cox regression model with robust standard errors to allow for matching. The model was adjusted for age group and having ever smoked. An interaction term was fitted to investigate changes in the association between PMR and cancer with time from diagnosis. The types of cancer occurring in the first 6 months were compared in those with and without PMR.

Results: 2877 PMR patients were matched with 9942 non-PMR patients. 9329 (73%) were female and mean age was 72 (S.D.: 9) years. Median (inter-quartile range) time under observation was 7.8 (3.43, 12.31) years. 23.2% of PMR patients (n = 667) and 19.5% of non-PMR patients (n = 1938) received a diagnosis of cancer in the study period. After adjusting for age group and ever smoking, PMR was significantly associated with a cancer diagnosis within 6 months [HR 1.96 (1.18, 2.42)], but after this time, there was no association (Table 1). No statistically significant difference was seen in the types of cancer between the two groups.

Conclusions: Analysis from a large cohort of general practice consulters suggests that there is no long-term association between PMR and malignancy. Whilst cancer diagnosis was twice as likely in the first 6 months after original PMR diagnosis, this may reflect an incorrect original PMR diagnosis, rather than a true association with malignancy. This highlights that clinicians should strongly consider malignancies as part of the differential diagnosis for PMR.

Citation

Muller, S., Mallen, C. D., Belcher, J., Helliwell, T., & Hider, S. L. (2013, April). O30. The Risk of a Subsequent Cancer Diagnosis after PMR: A Primary Care Database Study. Presented at Annual Meeting of the British Society for Rheumatology and British Health Professionals in Rheumatology, Birmingham, ENGLAND

• Presentation Conference Type: Lecture
• Conference Name: Annual Meeting of the British Society for Rheumatology and British Health Professionals in Rheumatology
• Conference Location: Birmingham, ENGLAND
• Start Date: Apr 23, 2013
• End Date: Apr 25, 2013
• Deposit Date: Aug 29, 2023
• Publisher URL: https://academic.oup.com/rheumatology/article/52/suppl_1/i33/1929284
• Related Public URLs: https://academic.oup.com/rheumatology/issue/52/1