Castleman disease as a spectrum: Characterization of novel subtypes using data from the spanish GELTAMO registry

Abstract Introduction Castleman disease (CD) unrelated to HHV-8 or POEMS syndrome comprises a heterogenous group of disorders characterized by lymphoid proliferation. Based on the number of lymphadenopathies (LADs), the presence of constitutional symptoms and laboratory abnormalities, these CD cases have traditionally been classified into unicentric (UCD) and idiopathic multicentric (iMCD) types, the latter being further subclassified into iMCD-TAFRO (thrombocytopenia, anasarca, fever, reticulin fibrosis/renal dysfunction, organomegaly), iMCD-IPL (idiopathic plasmacytic lymphadenopathy) and iMCD-NOS (not otherwise specified). Oligocentric CD (oligoCD) and asymptomatic MCD (aMCD) have recently been identified as novel subtypes of CD (Pierson, et al. BloodAdv 2025; Zhang, et al. BloodAdv 2024). From a national physician-managed registry, we aimed to evaluate the clinical-biological features and prognosis of novel CD subtypes and compare them with UCD and iMCD-NOS. Methods We performed a multicenter retrospective analysis including patients enrolled in the physician-managed Spanish registry GELTAMO-CAST-2020-018 with confirmed diagnosis of CD, negative for HHV8 and not classified as MCD-POEMS. The project was approved by the Ethical Committee of Germans Trias i Pujol Hospital (PI-20-103). After centralized revision and exclusion of alternative causes of LADs, cases were classified as UCD, iMCD, oligoCD and aMCD. Patients with 2-4 contiguous LADs but not fulfilling iMCD criteria were defined as oligoCD. Patients with >4 disseminated LADs and no clinical criteria of iMCD were categorized as aMCD. Survival and clinical-biological features of oligoCD and aMCD subtypes were compared to UCD and iMCD-NOS. Results Out of 176 enrolled patients, 91 CD cases diagnosed from September 1998 to March 2024 were included in this study. Median age was 42 years (range 5-83) with 55% of patients identified as male, 41% as female, and 4% as other. At initial classification, 57 (63%) cases were categorized as UCD and 34 (37%) as iMCD (iMCD-NOS = 31; iMCD-TAFRO = 3). After revision, 13 cases (14%) were reclassified as oligoCD, including 5 and 8 cases previously diagnosed as UCD and iMCD-NOS, respectively. Additionally, 4 cases (4%) initially classified as iMCD-NOS were reclassified as aMCD. No statistically significant differences in age or sex distribution were observed between the different subgroups. Hypervascular histopathology was predominant in oligoCD (8/13) and aMCD (3/4), as it was in UCD (44/52), while plasmacytic histology prevailed in iMCD-NOS (11/19). Hemoglobin and albumin levels were significantly higher in patients with oligoCD and aMCD, whereas CRP and ESR values were significantly lower, compared to patients iMCD-NOS (p < 0.05). In contrast, no significant differences in these laboratory values were observed when comparing either subtype to UCD (p>0.05). 13/17 (76%) patients were treated at diagnosis (oligoCD: surgery 4/11, steroids 2/11, anti-CD20 5/11; aMCD: surgery 1/2, anti-CD20 1/2). The overall response rate was 91% in patients with oligoCD, including complete responses in 8/11 cases and partial responses in 2/11. In the aMCD group, the overall response rate was 100%, with complete responses observed in both treated cases. Altogether, these findings suggested that oligoCD and aMCD exhibited clinical and laboratory characteristics resembling those of UCD. After a median follow-up of 2.85 years (range: 6 months to 24 years), no statistically significant differences were observed between UCD and oligoCD in terms of either overall survival (OS, p=0.40) or progression-free survival (PFS, p=0.45). In contrast, a significant difference in PFS was found between oligoCD and iMCD-NOS (p=0.037), with the median PFS not reached in the oligoCD group and 2.4 years in the iMCD-NOS group. A trend toward a difference in OS was also noted (p=0.078), with the median OS not reached in oligoCD and 9.25 years in iMCD-NOS. Cases of aMCD were excluded from the survival analysis due to the limited sample size. However, it is worth noting that, at data cut-off, no deaths or relapses had occurred in this group. Conclusions Both oligoCD and aMCD exhibit clinical-biological features and outcomes that are comparable to those observed in UCD. Our findings align with recent reports regarding the novel CD subtypes, supporting the need to approach unrelated HHV-8 and POEMS CD as a continuum of clinicopathological entities rather than a binary classification.