Abstract
OBJECTIVE: There is a well-recognized association between cancer and myositis, so cancer screening at diagnosis is recommended. We aim to report the results of our cancer screening strategy and to ascertain the reliability of using PET/CT to identify cancer-associated myositis (CAM) in a large cohort of patients with myositis from a single center over 10 years. METHODS: This retrospective observational study included all patients diagnosed with any type of myositis except for inclusion body myositis. Cancer screening strategy was individualized according to clinical and serological data, including PET/CT as the main test to detect occult cancer (OC). Procedures derived from a positive PET/CT were registered. Qualitative data expressed as percentages, and quantitative data as the median with the interquartile range were analyzed. A ROC curve was used to estimate the reliability of PET/CT for CAM diagnosis. RESULTS: Seventy-seven out of 131 patients underwent a PET/CT for OC screening. The performance of the PET/CT in patients with myositis at disease onset yielded an area under the curve ROC of 0.87 (0.73-0.97) for CAM diagnosis. Invasive procedures in 7 (9%) patients without a final diagnosis of cancer did not cause derived complications. Patients not evaluated for OC did not develop cancer after a median follow-up of 3.3 years (1.7-6.7). CONCLUSION: Cancer screening strategy should be individualized. PET/CT at myositis onset seems to be an efficient approach to rule out CAM. This practice does not seem to significantly increase harm to patients related to the additional tests needed to clarify inconclusive results.