Pediatric idiopathic multicentric castleman disease is often severe and responds to siltuximab

Abstract

Abstract Background: Idiopathic multicentric Castleman disease (iMCD) is a potentially fatal hematologic disorder marked by widespread lymphadenopathy and inflammation. Siltuximab, an interleukin-6 (IL-6) inhibitor, is the only FDA-approved treatment. Limited data exist on pediatric iMCD or its responsiveness to siltuximab. To fill this gap and guide pediatric treatment, we provide a systematic characterization of pediatric iMCD and treatment efficacy. Methods: We queried ACCELERATE, the largest iMCD registry with expert-confirmed diagnoses. We analyzed two panel-confirmed iMCD cohorts: children (under 19 years old at symptom onset) and adults. We assessed disease severity with a modified CHA score [C-reactive protein (CRP), hemoglobin, albumin] within 90 days of diagnosis. Regimens were defined as treatments started within two weeks of each other and categorized into the following six groups: siltuximab±corticosteroids (CS), tocilizumab±CS, immunomodulator(s)±CS, anti-IL-6+immunomodulator±CS, chemotherapy±other treatments, or CS monotherapy. Immunomodulators included sirolimus, rituximab, ruxolitinib, anakinra, lenalidomide, eculizumab, plasmapheresis, and others. We defined clinical response as a ≥50% decrease in the proportion of abnormal symptoms and laboratory measures after regimen initiation. A clinical response was durable if no new regimens were started within one year of the index regimen. We excluded regimens with unclear responses and grouped adverse events per the Medical Dictionary for Regulatory Activities. Results: We analyzed two cohorts: children (n=30) and adults (n=96) with panel-confirmed iMCD diagnoses. Symptom onset occurred at a median age of 15 years in children and 40 years in adults, ranging from 1 to 74 years. Hypervascular lymph node histology was more common in children compared to adults (83% vs 52%, p<0.05). The severe TAFRO subtype of iMCD (thrombocytopenia, anasarca, fever, renal dysfunction/reticulin fibrosis, organomegaly) was more prevalent in children (77% vs 55%, p<0.001). Children had higher CHA scores at diagnosis (1.67 vs 1.31, p<0.05), indicating greater disease severity, due to worse anemia (hgb 9.3 vs 10.2 g/dL, p<0.05) and higher CRP (138.3 vs 89.1 mg/L, p<0.05). Two children died, 1 and 9 months following diagnosis, compared to eight adults after a median of 38 months after diagnosis. 30 children received 68 regimens and 96 adults received 317 regimens. The most administered regimen, siltuximab±CS, was given to 14 (47%) children and 55 (57%) adults. Twelve of the 14 children (86%) and 35 of the 48 adults (73%) with evaluable regimens achieved a clinical response to siltuximab±CS (p=0.48). We observed durable responses in 7/11 (64%) of siltuximab-treated children and 21/43 (49%) of adults (p=0.50). Clinical response rates were similar between children and adults across other treatments, with no significant differences: tocilizumab±CS (67% [4/6] vs 53% [8/15]), immunomodulators±CS (43% [3/7] vs 53% [28/53]), anti-IL6+immunomodulators±CS (67% [6/9] vs 59% [10/17]), chemotherapy (67% [6/9] vs 88% [21/24]), and CS monotherapy (75% [3/4] vs 23% [9/40]). Durable response rates were comparable between children and adults across treatments, with no significant differences: tocilizumab±CS (40% [2/5] vs 50% [7/14]), immunomodulators±CS (29% [2/7] vs 25% [13/52]), anti-IL6+immunomodulators±CS (38% [3/8] vs 53% [8/15]), chemotherapy (38% [3/8] vs 52% [11/21]), and CS monotherapy (25% [1/4] vs 0% [0/40]). Nineteen (63%) pediatric patients had 71 medication-associated adverse events (AEs), mainly steroid-induced hyperglycemia and hypertension. Similarly, 67 (70%) adult patients had 275 medication-associated AEs, mostly rash and nausea. Of the siltuximab-treated children, 36% (5/14) had AEs like bilirubinemia, tinnitus, and fatigue. Among adults, 64% (30/55) experienced mostly rash and fatigue. Conclusions: In the largest pediatric iMCD analysis to date, we found more severe disease with higher CHA scores and more TAFRO cases than in adults. We observed high response rates to siltuximab in children, which were similar to those in adults. Since the phase 2 siltuximab trial excluded children and had few TAFRO cases, these results are reassuring with no new safety concerns. Despite being the largest, best-validated pediatric iMCD cohort to date, findings should be interpreted cautiously due to the registry's observational, non-randomized, and non-controlled design and further research is needed.