Real-world safety profile of PD-1 inhibitors in Epstein–Barr Virus–Associated lymphoproliferative disorders: A retrospective analysis from west China hospital

Abstract Background PD-1 inhibitors are increasingly utilized beyond oncology, including in Epstein-Barr virus–associated hemophagocytic lymphohistiocytosis (EBV-HLH) and chronic active Epstein-Barr virus disease (CAEBV). However, data on adverse events (AEs) in these inflammatory conditions remain limited.Methods We retrospectively analyzed 97 patients treated with PD-1 inhibitor at West China Hospital between 2020 and 2024. Of these, 74 had EBV-HLH and 23 had CAEBV. Laboratory values, imaging, and clinical symptoms were reviewed before and after PD-1 inhibitor therapy. Adverse events were graded per CTCAE v5.0. HLH reactivation was assessed using HLH-2004 criteria. Comparative analyses were conducted using paired t-tests for continuous variables and chi-square or Fisher's exact tests for categorical data. Additionally, adverse event data were compared between EBV-HLH, CAEBV from our center and solid tumor cohorts from published literatures. Notably, patients in our cohort received PD-1 inhibitor, sindilizumab, at a dose of 100 mg every month, while standard regimens in solid tumor trials typically employed 200 mg every three weeks.Results HLH reactivation occurred in 9.5% (7/74) of EBV-HLH patients and 4.3% (1/23) of CAEBV patients, with a median onset of 5–7 days following PD-1 inhibitor initiation; no HLH episodes were reported in solid tumor cohorts (p < 0.01 vs both EBV-HLH and CAEBV groups). Fever was reported in 47.3% of EBV-HLH and 39.1% of CAEBV patients, typically within 24 hours of PD-1 inhibitor administration-significantly higher than the <15% reported in solid tumor trials (p < 0.01). Anemia was observed in 89.2% and 82.6% of EBV-HLH and CAEBV patients, respectively, with Grade ≥3 anemia in 59.5% and 34.8%, compared to 10–15% overall and 2–4% Grade ≥3 anemia in solid tumor cohorts (p < 0.001).Thrombocytopenia occurred in 78.4% and 56.5%, with Grade ≥3 events in 36.5% and 21.7%, respectively, compared to 5–10% and <2% Grade ≥3 in solid tumor patients (p < 0.001).Neutropenia was observed in 43.2% and 21.7%, with Grade ≥3 neutropenia in 24.3% and 13.0%, respectively, significantly exceeding the 6–8% and 1–3% Grade ≥3 rates seen in solid tumor trials (p < 0.001). Renal toxicity, defined by elevated serum creatinine, was observed in 21.6% of EBV-HLH and 17.4% of CAEBV patients, compared to <5% in solid tumor studies (p = 0.0037). Hepatic toxicity, defined by elevation of liver enzymes or bilirubin, occurred in 41.9% of EBV-HLH and 34.8% of CAEBV patients, versus 5-7% in solid tumor cohorts (p < 0.01). Supporting lab data showed a significant increase in total bilirubin (33.9 µmol/L to 64.3 µmol/L; p = 0.022), a non-significant rise in direct bilirubin (25.1 µmol/L to 49.9 µmol/L; p = 0.1673), and elevated ALP levels (p = 0.0181).Cardiac marker elevation (troponin, BNP, or myoglobin) occurred in 14.9% of EBV-HLH and 13.0% of CAEBV patients, exceeding the <5% reported in solid tumor cohorts (p < 0.05). Inflammatory markers (CRP, PCT, IL-6) increased in over 70% of patients within 1-2 days, consistent with PD-1 inhibitor induced hyperinflammation. Additionally, one hypothyroidism developed in CAEBV patient, acute hypoxia occurred in one CAEBV patient with pulmonary EBV involvement before PD-1 inhibitor administration, and two neurological events in two EBV-HLH patients with one mood change and the other dysgraphia. Dermatologic events were limited to three mild skin rashes: two in patients with EBV-HLH and one in a patient with CAEBV. Conclusions PD-1 inhibitors in EBV-HLH and CAEBV patients are associated with a distinct and intensified immune toxicity profile compared to solid tumor cohorts. Statistically significant increases in hematologic, renal, hepatic, and inflammatory toxicities underscore the need for vigilant monitoring and individualized risk assessment. These findings highlight the importance of adapting PD-1 inhibitor based regimens with caution when repurposing them for EBV-driven non-malignant disorders.