How would you differentiate new onset inflammatory myositis from paraneoplastic phenomena in a patient with recurrent uterine cancer who is taking aromatase inhibitor?

Unfortunately, distinguishing whether the patient's dermatomyositis (DM) is "idiopathic" or paraneoplastic the patient with active/recent malignancy is incredibly difficult, if not impossible. While certain myositis-specific autoantibodies (MSA) such as anti-TIF-gamma and anti-NXP-2 are associated with markedly increased risk for cancer-associated dermatomyositis in adults, all subtypes of dermatomyositis can be paraneoplastic. Some MSA (such as anti-synthetase-associated MSA) have been associated with a lower risk for malignancy compared to other MSA, but reports of all DM subtypes as a paraneoplastic phenomenon are well-documented. Cohort-based assessments vary, but most studies suggest that ~5-15% of DM patients will be cancer-associated (and some cohorts suggest this risk may be ~30-40% in anti-TIF-gamma patients). This is typically estimated to be a ~6-7 times higher increase in malignancy risk than the expected background.

Just as importantly, even though underlying cancer is found within a year before/after the onset of DM in most paraneoplastic cases, the detection of cancer is heightened for at least 3 years prior/after DM diagnosis. (Most DM clinical trials which do not allow cancer-associated DM will exclude patient with malignancy within 3 years).

There are several reasons why cancer may be associated with DM: First and foremost, it is possible/likely that immune response to malignancy may trigger autoimmunity that subsequently leads to recognizable DM manifestations. Similarly, several authors have noted that dysregulation in pathways that are important for cell growth regulation and/or immune-system cancer surveillance (such as the TIF pathway, the SUMO pathway, and the PD-1/PD-L1 pathways) may play a key role in cancer-associated DM. Clinically, there is likely some contribution of immunosuppression to increased risk for malignancy (analogous to other autoimmune diseases). Lastly, patients who have active DM are likely to have closer monitoring and so may have higher rates of detecting cancer earlier than would be otherwise.

Cancer-associated DM may manifest with any malignancy, although most cohorts I have seen suggest that the risk is higher for adenocarcinomas or squamous cell carcinomas, and typically mirror the types of cancer typically seen in that age/general population (such as ovarian, breast, lung, pancreatic, GI cancer, and nasophayngeal cancer).

A key question is whether knowing a patient's dermatomyositis is cancer-associated would change management strategy compared to the hypothetical situation of a patient with a "idiopathic" DM and comorbid malignancy. In both scenarios, treatment of the cancer is typically necessary: Immune-stimulation from the cancer is often a driver for DM activity, and many patients will be difficult to control/refractory unless thier cancer is under good control (conversely, I have had several patients whose DM became much better after remission of their cancer). However, patients with DM typically need immunosuppression in addition to malignancy treatment, although I will sometimes alter/mitigate my therapy based on the oncologists treatment plan (i.e., I expect certain agents, like platinum-based regimens, to have an immunosuppressive effect in and of itself, and I may lower the intensity or hold other immunosuppressive agents while the patient is being treated with these). I also consider active or prior DM as a relative contraindication to immune-checkpoint inhibitor treatment, although this is a decision that needs to be made on an individual basis in conjunction with the patient and their oncologist based on the patient's goals of care and risk-tolerance. Finally, in both scenarios, patients will need to be routinely screened for recurrence of their cancer and/or secondary malignancy, especially if they experience a sudden/persistent worsening of their DM.

While there is still much debate about the optimal strategy for screening/monitoring for cancer in patients with DM, several authors have proposed strategies that make a lot of intuitive sense. I particularly like the "risk stratification strategy" proposed by Univ of Pittsburgh myositis group (Moghadam-Kia et al., PMID 32631604, Kardes et al., PMID 35970749).

Hope this answer helps!