Clinicopathologic Features and Survival Determinants of Hepatosplenic T‐Cell Lymphoma: A Pooled Patient‐Level Dataset Analysis

ABSTRACT

Hepatosplenic T‐cell lymphoma (HSTCL) is an ultra‐rare, aggressive T‐cell neoplasm for which evidence regarding prognosis and optimal therapy remains limited. We conducted a retrospective pooled patient‐level analysis of 529 patients from 198 sources. The median age was 32 years, with 71.7% male. Hepatosplenomegaly and bone marrow involvement were present in 99.7% and 91.8% of evaluable patients, respectively. Among 436 patients with evaluable overall survival (OS), the median OS was 12.0 months, and the median progression‐free survival was 5.0 months. Multivariable analysis demonstrated that hemoglobin less than 8 g/dL (HR 3.36, 95% CI 1.61–7.00; p  = 0.0012) and underlying immunosuppression (HR 3.00, 95% CI 1.77–5.06; p  < 0.0001) independently predicted inferior OS, while time‐dependent stem cell transplantation (SCT) was associated with improved survival (HR 0.31, 95% CI 0.15–0.65; p  = 0.0021). A provisional two‐point score incorporating severe anemia and immunosuppression stratified patients into low‐, intermediate‐, and high‐risk groups, with median OS of 26, 5, and 3 months, respectively (log‐rank p  < 0.0001; C‐index 0.706). Platinum‐containing first‐line regimens were associated with lower mortality compared to CHOP‐like therapy (HR 0.43, 95% CI 0.25–0.75; p  = 0.003). The SCT survival association was consistent across complementary analyses. Nonresponsive, stable, or progressive disease at transplantation predicted inferior post‐transplant survival compared to complete response (HR 2.84, 95% CI 1.32–6.10; p  = 0.007). These findings identify severe anemia and immunosuppression as adverse prognostic features, support non‐CHOP platinum‐based induction and early SCT evaluation after disease control. However, because this is a retrospective pooled design, the treatment associations and the provisional prognostic score still require external validation.