Targeting cancer screenings based on idiopathic inflammatory myositis (IIM) subtype, autoantibodies, and age may help to maximize cancer detection while limiting false positives.
In a retrospective, single-center study conducted at Johns Hopkins University in Baltimore, researchers found that when screening patients with IIM for cancer via CT imaging, the diagnostic yield (number of cancers detected/tests performed) was highest in patients with dermatomyositis and the autoantibody anti–TIF1-gamma. Screening patients below age 40 years was associated with lower diagnostic yields and higher false positives, regardless of subtype.
Because of the well-known association between IIM and contemporaneous cancer, newly diagnosed patients with IIM often undergo screening. Yet, there is little research on the most efficient assessment approaches, Christopher Mecoli, MD, an assistant professor of medicine at John Hopkins University School of Medicine and lead author of the study, told this news organization. “There has been a lot written about how these patients should be evaluated for cancer. Unfortunately, the majority of literature is based on eminence,” he said. This study is “one of the first pieces of real data to inform that conversation,” he added.
The research was published online in Arthritis Care & Research.
In the study, Dr. Mecoli and colleagues looked at 1,086 patients enrolled in the center’s Myositis Research Registry from 2003 through 2020. The analysis included patients with a diagnosis of dermatomyositis, polymyositis, immune-mediated necrotizing myopathy (IMNM), and antisynthetase syndrome (ASyS). The researchers also looked at myositis-specific autoantibodies, including anti–TIF1-gamma, –Jo1, and –HMGCR. Patients were excluded from the analysis if they had a cancer diagnosis prior to their IIM onset.
Among patients included in the analysis, the average age of IIM onset was 49 years, and median follow-up duration was 5.3 years. Most patients were female (71%), 68% were white, 21% were Black, 3.6% were Asian, and 7.4% had a listed race of other or unknown. About 66% of all patients received a chest CT scan within 3 years of IIM onset, and 51% received an abdomen/pelvis CT in that same time frame. False positives were defined as the percentage of scans that led to a noncancerous biopsy.
During the study period, 62 patients had a cancer diagnosis within the first 3 years of IIM onset, with the most common cancers being breast (19%), melanoma (13%), and cervical/uterine (10%). Of 1,011 chest scans performed, 9 led to a cancer diagnosis (0.9%), compared with 12 of the 657 abdomen/pelvis (a/p) CT scans (1.8%). Patients with the dermatomyositis-specific autoantibody anti–TIF1-gamma had the highest diagnostic yield (2.9% in chest CT and 2.4% in a/p CT). Regardless of autoantibodies, dermatomyositis patients above 40 years of age had a diagnostic yield of 1.4% in chest CT and 2.7% in a/p CT. For patients under the age of 40 with polymyositis, IMNM, and ASyS, the diagnostic yield for all CT scans was 0.0%. The diagnostic yield in patients under 40 with dermatomyositis was also low (0.0% in chest CT, 0.8% in a/p CT).
The false-positive rate for all chest CT scans was 2.8%, with patients with IMNM and ASyS having the highest frequency of false positivity (both 4.4%). “Based on our data, CT chest imaging in ASyS and IMNM patients are associated with the most harm from a cancer screening perspective,” the authors write. In a/p CT, patients with dermatomyositis under 40 and patients with ASyS had the highest false-positive rates (4.9% and 3.8%, respectively).
“Age was a really big deal in terms of predicting diagnostic yield and false-positivity rate,” Dr. Mecoli said, particularly in patients with dermatomyositis. “This subgroup has historically been thought to have the biggest dissociation with cancer,” he said, but in patients under 40, “it doesn’t look like CT scans were that helpful. They were not picking up a lot of cancers, and they were leading to a lot of false-positive results.”
Still, Rohit Aggarwal, MD, of the division of rheumatology and clinical immunology at the University of Pittsburgh, Pennsylvania, noted that the diagnostic yields of 1%-2% and even 2%-4% in higher-risk populations were high. By comparison, lung cancer screening trials had a diagnostic yield of about 1%, and trials examining CT screening for colorectal cancers had diagnostic yields of 0.5%, the authors write.
“The key message for me is that we should definitely perform CT scans of the chest, abdomen, and pelvis within 3 years of diagnosis – typically at presentation – if the patient has any risk factor for increased risk of cancer, which include dermatomyositis and age above 40,” Dr. Aggarwal toldthis news organization. He was not involved with the research. There are also other clinical factors to consider that were not included in the study, he added, such as severe dysphagia, patients with refractory treatment, and male sex.
Both Dr. Aggarwal and Dr. Mecoli agreed that there are limitations to this single-center, retrospective study that make it difficult to generalize the results. Similar studies should be conducted at other institutions to see if these associations hold true, Dr. Mecoli said. A prospective study could also help control for factors such as selection bias, Dr. Aggarwal added. “I don’t think these are definitive data, but I think these data were needed at retrospective levels” to plan future research, he said.
The study was supported in part by grants from the National Institutes of Health, the Jerome L. Greene Foundation, the Donald B. and Dorothy L. Stabler Foundation, the Huayi and Siuling Zhang Discovery Fund, and Dr. Peter Buck. Dr. Mecoli and Dr. Aggarwal have disclosed no relevant financial relationships.
A version of this article first appeared on Medscape.com.