Background/Purpose: To inform guidance for cancer surveillance in patients with idiopathic inflammatory myositis (IIM), we conducted a retrospective cohort study in a single tertiary referral center to evaluate the use and yield of computed tomography (CT) for cancer surveillance within distinct myositis-specific autoantibody strata from 2007 through 2020.

Methods: In 2020, we reviewed the electronic medical records of all patients who previously enrolled in our Myositis Research Registry to identify those who met the following criteria: (i) Probable or definite DM/PM by Peter and Bohan, (ii) IMNM by the 2003 ENMC Criteria, (iii) Classic DM rash (Gottron’s/heliotrope) and consistent histopathology on skin biopsy. Myositis specific and associated autoantibodies were assayed using Euroimmun line blot (16 Ag IgG), ELISA (anti-Mi2 and anti-TIF1ƴ [MBL], anti-HMGCR [INOVA Diagnostics]), and in-house immunoprecipitation (anti-NXP2). We also reviewed their charts and outside medical records for any CT chest and abdomen/pelvis studies ordered for cancer surveillance/screening purposes. Our myositis center’s strategy for malignancy screening/surveillance includes at least one CT chest, abdomen, and pelvis upon IIM diagnosis. Data on CT imaging was collected with researchers blinded to both myositis-specific autoantibody and cancer status. Two time periods were analyzed: Cancer screening tests performed (i) after cohort enrollment, and (ii) after IIM-symptom onset, both through 12/31/2020.

Results: Among 1174 patients, 1054 patients had at least one myositis specific or associated autoantibody. From cohort enrollment onwards, the number of cancers that occurred within 0-3 yrs, 3-5 yrs, and >5 yrs after IIM symptom onset was 35, 15, and 34, respectively. The number/% of CT chest scans that were positive for cancer were 6/1.34% (0-3 yrs), 2/1.18% (3-5 yrs), and 3/1.38% ( >5 yrs). The number/% of CT a/p positive for cancer was 9/2.90% (0-3 yrs), 0/0% (3-5 yrs), and 3/1.91% ( >5 yrs). Of the 35 cancers diagnosed within 0-3 yrs from IIM symptom onset, more than half were diagnosed by methods other than CT chest, abdomen and pelvis imaging. Amongst 273 anti-TIF1g-positive patients, 181 CT chest or a/p scans were performed, 5 (2.7%) of which were positive within 0-3 yrs. In contrast, amongst 248 patients with Jo1, PL12, PL7, OJ, or EJ autoantibodies, 178 CT chest or a/p scans were performed, 1 (< 1%) of which was positive.

From IIM symptom onset onwards, the number of IIM patients diagnosed with cancer that occurred within 0-3 yrs, 3-5 yrs, and >5 yrs after IIM symptom onset was 70, 16, and 36. The percent of screening/surveillance CT chest and CT a/p scans that were positive for cancer was comparable to the time window from cohort enrollment onwards (Table 2).

Conclusion: In a tertiary referral center population, the number of patients needed to be screened by CT imaging to detect one cancer within the first 3 yrs of IIM was approximately 70 for CT chest and 40 for CT a/p. CT imaging is of higher yield in patients with anti-TIF1ƴ antibodies compared to patients with antisynthetase autoantibodies. Whether CT imaging leads to improvement in cancer or IIM outcomes remains to be determined. Funded in part by NIAMS K23AR075898.

« Back to ACR Convergence 2021

ACR Meeting Abstracts - https://acrabstracts.org/abstract/use-and-yield-of-computed-tomography-as-a-cancer-surveillance-method-in-idiopathic-inflammatory-myositis/