AUTHORS
Kamilia Moalem, MD, Claudia Anna Maria Fulgenzi2, Kai Hosmann3, Jose Mercado-Matos, MD, PhD4, Peter Kelleher, PhD5, Hartmut Stocker6, Francesca Duca, MD7, Robert Yarchoan, MD8, Mark Bower, MD, PhD9, Kathryn Lurain, MD, MPH10, Pascal Migaud, MD11, Alessia Dalla Pria, MD, PhD12, Ramya Ramaswami, MBBS, MPH13
AFFILIATIONS
1 National Institutes of Health, Bethesda, Maryland, United States
2 Imperial College London, London, United Kingdom, National Centre for HIV Malignancies Chelsea and Westminster, London, United Kingdom
3 St-Joseph Hospital, Berlin-Tempelhof, Berlin, Germany
4 National Institutes of Health, Bethesda, Maryland, United States
5 Imperial College London, London, United Kingdom, Imperial College London, London, United Kingdom
6 Charite Universitatsmedizin Berlin, Berlin, Germany, St-Joseph Hospital, Berlin-Tempelhof, Berlin, Germany
7 Imperial College London, London, United Kingdom, University of Milano-Bicocca, Milan, Italy
8 National Institutes of Health, Bethesda, Maryland, United States
9 Imperial College London, London, United Kingdom, Imperial College London, London, United Kingdom, National Centre for HIV Malignancies Chelsea and Westminster, London, United Kingdom
10 National Institutes of Health, Bethesda, Maryland, United States
11 Charite Universitatsmedizin Berlin, Berlin, Germany, St-Joseph Hospital, Berlin-Tempelhof, Berlin, Germany
12 Imperial College London, London, United Kingdom, Imperial College London, London, United Kingdom
13 National Institutes of Health, Bethesda, Maryland, United States
ABSTRACT
Background: Kaposi sarcoma herpesvirus (KSHV)-associated multicentric Castleman disease (MCD), which occurs predominantly in people with HIV (PWH) shares clinical and laboratory features with hemophagocytic lymphohistiocytosis (HLH). Secondary HLH related to hematologic malignancies carries a poor prognosis. MCD is described across CD4+ T-cell counts. The factors that drive secondary HLH and its clinical significance in MCD are poorly defined. Furthermore, it is unclear how HIV-specific factors or the presence of other KSHV-associated conditions, such as Kaposi sarcoma (KS), relate to the onset of secondary HLH in MCD. Rituximab is an essential component of MCD treatment and is administered with chemotherapy in stage-stratified approach in cases of severe illness or concurrent KS. We evaluated secondary HLH in MCD and the HIV-related features that are associated with it, in the largest international cohort reported to date.
Methods: We conducted a retrospective analysis of prospectively collected data on adults with HIV and pathologically confirmation of KSHV-associated MCD across 3 international centers (United Kingdom, Germany and the United States) from 2008 to 2026. HLH was defined by HScore >168 and/or >5 of 8 HLH-2004 criteria at MCD presentation, prior to MCD-directed therapy. Continuous variables were compared by Mann-Whitney test and binary variables by chi-squared test. Associations with HLH were examined using univariate and multivariable logistic regression, considering binary variables of recent HIV diagnosis (<1 year prior to MCD), HIV viral load (>1000 copies/ml), CD4+ T-cell count (<250 cells/µl), and concurrent KS. KSHV viral load levels were not included as they were not uniformly measured between the participating sites. The effect of secondary HLH diagnosis on overall survival (OS), evaluated from the time of MCD diagnosis to death from any cause was compared by Kaplan-Meier method and log-rank test.
Results: Among 176 evaluable patients with MCD and HIV (median CD4+ T-cell count 232 cells/µl and median HIV VL <20 copies/ml), 91% were male and 48 (27%) met HLH criteria at MCD flare. At the time of MCD presentation, 64% had concurrent KS, 53% of all patients had been on antiretroviral therapy for more than 6 months. Those with MCD and HLH phenotype presented with a more severe clinical picture: 60% had an ECOG performance status of >2, as compared to 23% of those without HLH (P<0.0001). Patients with HLH had lower CD4+ T-cell counts (median 139 vs 275 cells/µL, p=0.005) and higher HIV viral loads (median 562 vs 50 copies/mL, p=0.001) than patients without HLH.
In the univariate analyses, the HLH phenotype was associated with a recent HIV diagnosis (<1 year), uncontrolled HIV viremia (>1000 copies/ml), low CD4+ T-cell counts (<250 cells/µl) and concurrent KS (all P<0.05). In the multivariable model, the HLH phenotype was independently characterized by recent HIV diagnosis (adjusted odds ratio (OR) 3.5, 95% CI 1.6–7.5, p=0.002) and KS (adjusted OR 2.5, 95% CI 1.1–6.0, p=0.032), with a non-significant trend toward association for uncontrolled HIV viremia (adjusted OR 2.0, 95% CI 0.96–4.3, p=0.064). The univariate association with low CD4+ T-cell counts did not persist after adjustment (adjusted OR 1.6, 95% CI 0.7–3.5, p=0.28) within this model. Among those with MCD and secondary HLH, 11 received rituximab monotherapy (23%) and the remaining were administered rituximab with chemotherapy, including etoposide (40%), liposomal doxorubicin (29%) or other treatments including anthracycline-based combination chemotherapy (EPOCH, 8%). Despite the more severe presentation, HLH phenotype was not associated with reduced overall survival among those with MCD and HIV (5-year OS 91% vs 79%; log-rank P=0.08).
Conclusions: In HIV-associated MCD, the HLH phenotype characterizes a severe presentation, marked by poor performance status independent of Hscore, which occurs in the context of recent or uncontrolled HIV infection, rather than a distinct process. Despite this severity, secondary HLH did not confer a significant survival disadvantage when patients received timely, rituximab-based risk-stratified therapy at experienced centers.
DISCLOSURES
Kamilia Moalem, MD: No relevant conflicts of interest to declare.