Background High risk of cancers in patients with myositis is partly explained by common expression of myositis-associated autoantigens in cancer tissues and in regenerating muscles. Malignancies found in myositis patients would be either cancers directly involved in myositis development or cancers that are not different from those found in general population.
Objectives To compare the standardized incidence ratio (SIR) of cancers related and unrelated to myositis activity in Korean patients with myositis
Methods Medical records of 283 patients with polymyositis (PM) or dermatomyositis (DM) were reviewed to identify 52 cancer cases. SIR was calculated using cancer incidence of period-, age- and sex-matched Korean population. Under the hypothesis that cancer tissues should temporally overlap with active myositis in patients whose cancers are involved in myositis development, cancer cases were divided according to the presence (group A, cancers related to myositis activity, n=32) or absence (group B, cancers unrelated to myositis activity, n=20) of either newly diagnosed, remnant, or recurred cancers at any time during active phase of myositis. SIRs of groups A and B were compared.
Results The mean (±standard deviation) age at myositis diagnosis was 47.7±15.1 years in all patients and 77.7% were female. The overall SIR was 3.0 [2.2-4.0] (4.4 [2.6-6.8] in men; 2.51 [1.7-3.6] in women). Although a majority of cancers were found from 50's to 70's, the highest SIR, 22.8 [4.7-66.5], was observed in 20's. When stratified by intervals between myositis and cancer diagnoses, no elevated SIR was noted except for 10.3 [6.8-15.1] within 1 year of myositis diagnosis, and for 3.2 [1.2-6.9] during the next 2 years. The most frequent cancer was non-Hodgkin's lymphoma (NHL, 23.0 [9.3-47.4]). SIRs of esophageal cancer (23.6 [4.9-68.8]), multiple myeloma (MM, 22.9 [2.8-82.7]), lung cancer (6.4 [2.9-12.1]), and adenocarcinoma of unknown primary (ACUP, 25.7 [7.0-65.7]) were also higher than general population.
When compared groups A and B, group A tended to be more male-predominant (14/30 vs 5/22, p=0.077) and had an older age at myositis diagnosis than group B (60.5±11.1 vs 49.3±16.6 years, p=0.022). There was no difference in PM and DM distribution. The SIR in group A was 1.8 [1.2-2.6] (3.0 [1.6-5.1] in men; 1.3 [0.8-2.2] in women) while in group B, it was 1.2 [0.8-1.9] (1.4 [0.5-3.0] in men; 1.2 [0.6-2.0] in women). Elevated SIRs were observed in 60's in group A (2.6 [1.4-4.5]) and in 20's in group B (22.8 [4.7-66.5]). The SIR within 1 year of myositis diagnosis was 9.9 [6.4-14.7] in group A while in group B, there was no temporal relationship between myositis and cancer. Elevated SIRs were observed for esophageal cancer (23.6 [4.9-68.8]), NHL (16.4 [5.3-38.3]), and ACUP (25.6 [7.0-65.7]) in group A while lung cancer (3.6 [1.2-8.3]) in group B.
Conclusions Cancer risk in Korean patients with myositis was approximately 3 times higher than general population. SIRs of NHL (23.0 [9.3-47.4]), esophageal cancer (23.6 [4.9-68.8]), MM (22.9 [2.8-82.7]), lung cancer (6.4 [2.9-12.1]), and ACUP (25.7 [7.0-65.7]) were significantly higher than general population. Two distinctive groups of cancers were noted according to the relationship with myositis activity in terms of incidence, temporal relationship with myositis, age distribution, and histologic types of cancers.
Disclosure of Interest None declared