Castleman disease: 2026 so far

Multiple series and cases pointed to long-lasting local control.

In the months covered, the clearest pattern was that unicentric Castleman disease (UCD, one enlarged lymph node region) was usually treated as a localized problem. Complete surgical removal was commonly followed by imaging showing no local recurrence. This came from several retrospective series (including retroperitoneal, head/neck, and intrapulmonary UCD) and many case reports in unusual locations where imaging first suggested other tumors. Reports also repeatedly noted that the diagnosis often became definite only after pathology (microscope testing of tissue) from surgery or biopsy. They also described that how difficult surgery was could vary by location and vascularity (how many blood vessels were involved). In these reports, outcomes looked most favorable when the mass could be fully removed, while uncertainty before surgery and procedure risk depended on the anatomy.

Research covered in: February 2026, March 2026, April 2026, May 2026, June 2026, July 2026, August 2026, September 2026

Real-world cohorts and registry data helped set expectations for response.

In the months covered, several real-world reports described improvement in iMCD (idiopathic multicentric Castleman disease, disease in multiple lymph node areas without a known cause) after IL-6-blocking medicines (drugs that block an inflammation signal), including siltuximab or tocilizumab. These included routine-care cohorts and a registry analysis comparing children and adults. Early in 2026, a small retrospective series reported disease control with first-line siltuximab in adults. Later reports added a single-center cohort showing partial responses across iMCD subtypes, and an ACCELERATE registry analysis suggesting similar response rates in pediatric and adult iMCD (with or without corticosteroids). Several single-patient reports also described benefit in harder-to-recognize situations (for example, lung involvement, IgA vasculitis, or TAFRO features), while noting that lab improvement may come before full symptom relief in some cases. Taken together, these real-world sources repeatedly connected IL-6 blockade with improvement in symptoms and/or lab markers, while also showing that response and recovery timing differed from person to person.

Research covered in: March 2026, April 2026, May 2026, June 2026, August 2026, September 2026

Stopping treatment was reported, with relapse clustered early.

One management update in the months covered described what happened after stopping therapy following deep lab normalization in iMCD-TAFRO (a subtype with thrombocytopenia, anasarca, fever, reticulin fibrosis, and organomegaly). A multicenter retrospective report said most people stayed in biochemical remission (normalization of disease-related lab tests) after stopping therapy once labs fully normalized. It also described a smaller group who worsened within roughly the first year, and then improved again after retreatment using similar approaches. A separate retrospective report described sustained remission for more than five years after stopping a thalidomide-cyclophosphamide-prednisone regimen, but key details (including how many patients) were not provided in the summary. In these reports, some patients were able to remain in remission off therapy after a deep response, but relapse needing retreatment was also seen, especially within the first year after stopping in the iMCD-TAFRO series.

Research covered in: January 2026, April 2026

Retrospective studies explained why IL-6 results can be confusing.

Two retrospective studies in the months covered highlighted that IL-6 and CRP can move in opposite directions depending on the treatment being used. With IL-6-blocking therapy, CRP (C-reactive protein, a blood marker of inflammation) tended to drop even when measured IL-6 rose sharply. This made IL-6 harder to interpret during IL-6 blockade. By contrast, IL-6 trends were described as more informative in patients on non-IL-6-blocking regimens, where rising IL-6 could come before biochemical worsening. One small study also noted that rituximab was associated with smaller IL-6 and CRP decreases that did not reach statistical significance. These findings mean lab trends during treatment may need to be read alongside the specific regimen, because the same marker can behave very differently across therapies.

Research covered in: June 2026

Small studies and single cases tried JAK, mTOR, and IL-1 approaches.

In the months covered, several small studies and single-patient reports described what was tried when IL-6 blockade did not bring iMCD under control, but the evidence stayed early and mixed. These included a very small exploratory trial of sirolimus with only two enrolled patients, case reports of ruxolitinib in relapsed/refractory TAFRO, and one case where adding IL-1 blockade (anakinra, a drug that blocks another inflammation signal) to ongoing IL-6 blockade was followed by complete durable remission. One single-patient translational analysis (lab-based testing to connect biology with treatment response) also described broad decreases in immune mediators over 16 weeks on sirolimus after tocilizumab non-response. June also included one case report of autologous CD19 CAR T-cell therapy (a personalized immune-cell treatment made from a patient's own T cells) in HHV-8-negative iMCD, with complete clinical and biochemical remission lasting more than 12 months, mild cytokine release syndrome (an inflammatory reaction after immune therapy), and low antibody levels requiring IVIG replacement. Overall, these reports show that clinicians used several non-IL-6 strategies in severe or refractory situations, but the chance of benefit and the safety tradeoffs across patients remain uncertain because most evidence came from very small numbers.

Research covered in: January 2026, June 2026, August 2026