Idiopathic multicentric castleman disease unresponsive to one IL-6 inhibitor may respond to another IL-6 inhibitor

Abstract Background Idiopathic multicentric Castleman disease (iMCD) is a morbid hematologic disorder defined by lymphadenopathy with a distinct histology and systemic inflammation that can lead to multiorgan failure. The only FDA-approved therapy is siltuximab, an antagonist targeting interleukin 6 (IL-6), but second-line treatment options remain limited for the roughly half of patients with refractory disease. Tocilizumab, an inhibitor targeting the IL-6 receptor, is often considered interchangeable with siltuximab due to their shared pathway inhibition, though its efficacy has not been studied in siltuximab-refractory patients. We identified a cohort of patients whose disease is refractory to one IL-6 inhibitor but interestingly responds to another, suggesting a potential second-line treatment option for iMCD patients. We sought to define this cohort of responders to a second IL-6 inhibitor by comparing them to patients refractory to both siltuximab and tocilizumab. Methods We utilized two registries: ACCELERATE, the largest international registry of iMCD patients, that leverages an expert panel to confirm every patient diagnosis, and Peking Union Medical College Hospital, the largest single-center registry of Chinese iMCD patients. We identified patients treated with siltuximab and tocilizumab at different stages of their treatment (n=23). We categorized patients into two groups: refractory/response (REF/RES; N=9) or refractory/refractory (REF/REF; n=14). We excluded patients from the REF/REF group if they received less than 3 doses of IL-6 inhibition (n=6) or had insufficient records to assess a response (n=2), resulting in a final REF/REF cohort (N=6). Results Among the 15 patients with an assessable response to both IL-6 inhibitors, 9 (60%) responded to the second IL-6 inhibitor (REF/RES). Four of 9 patients (44%) received siltuximab as the first IL-6 inhibitor and tocilizumab as the second. Two of 9 patients had severe disease prior to their first IL-6 inhibitor and one of 9 prior to the second. Five of 9 patients received an IL-6 inhibitor as first-line therapy. Six of 9 patients received corticosteroids with the first IL-6 inhibitor, and 3 received other medications (hydroxychloroquine, rituximab, sirolimus, and chemotherapy). Five of 9 patients had progressive disease to the first IL-6 inhibitor, and 4 had stable disease. Five of 9 patients received the second IL-6 inhibitor immediately after nonresponse to the first (median time between inhibitors: 14.4 months [1.6-25.3 months]). Three of 9 patients received corticosteroids with the second IL-6 inhibitor, and 2 received other medications (hydroxychloroquine and sirolimus). Four of 9 patients achieved a complete response on the second IL-6 inhibitor, and the remaining 5 achieved a partial response. One patient discontinued the second IL-6 inhibitor entirely. No patients relapsed on the second IL-6 inhibitor, with a median duration of follow-up of 14.7 months [8.8-56.8 months]. Of the 9 patients in the REF/RES cohort, 5 were female, and the median age at diagnosis was 34 years. Of the 6 patients in the REF/REF cohort, 2 were female, and the median age at diagnosis was 47 years. The idiopathic plasmacytic lymphadenopathy (IPL) clinical subtype was most common in the REF/RES cohort (67%), whereas the REF/REF cohort was predominantly the severe subtype TAFRO (thrombocytopenia, anasarca, fever, renal dysfunction/reticulin fibrosis, organomegaly; 83%) (p<0.05). Plasmacytic histopathology was most common in the REF/RES cohort (56%), and hypervascular histopathology was seen in all REF/REF patients (100%) (p<0.05). Constitutional symptoms were present in both groups (REF/RES, 89%; REF/REF, 67%), with comparable disease severity scores, and inflammatory marker abnormalities by C-reactive protein, erythrocyte sedimentation rate, hemoglobin, albumin, and creatinine. Conclusions We describe a cohort of patients whose disease was refractory to one IL-6 inhibitor but notably responded to another, despite the perception that these therapies are interchangeable. Responses to the second IL-6 inhibitor were encouragingly durable, with patients showing sustained disease control based on available follow-up. Despite limited numbers, we observed that responders to a second IL-6 inhibitor were more likely to have the IPL subtype. In a disease with few alternatives, this cohort suggests an unexpected therapeutic benefit to trialing a second IL-6 inhibitor following an initial nonresponse.