Patients with ten or more tumor buds after surgical treatment for hepatocellular carcinoma had the poorest outcomes, with over fourfold higher overall and sevenfold higher disease-specific mortality than those without tumor budding, according to a recent study.
Researchers conducted a retrospective cohort study evaluating the prognostic value of tumor budding and tumor-stroma ratio (TSR) in hepatocellular carcinoma (HCC). The study, published in Human Pathology, investigated histopathologic markers that may improve survival prediction in patients with surgically and non-surgically treated HCC.
The analysis included 506 histologically confirmed cases of HCC diagnosed between 1986 and 2022. Among these, 101 patients underwent surgical treatment, while 405 received non-surgical or palliative management. Tumor budding and TSR were assessed on digitized hematoxylin- and eosin-stained slides using Aperio ImageScope. Tumor budding was characterized as isolated tumor cells or small clusters of up to four cells located at the tumor’s invasive margin. Patients were classified as bud-positive (≥1 bud) or bud-negative (0 buds) and further divided into three subgroups according to bud count: 1 to 4 buds, 5 to 9 buds, and at least 10 buds. TSR was quantified as the proportion of stroma within tumor tissue, with a 50 percent threshold used to differentiate between stroma-rich (≥50 percent) and stroma-poor (<50 percent) tumors.
In the surgical group, 44.6 percent of patients were bud-positive, while 55.4 percent were bud-negative. Tumor budding was found to be an independent risk factor for disease-specific mortality but was not associated with overall mortality. Tumor buds were consistently detected across multiple histological sections, appearing in 77 percent of slides from bud-positive tumors, confirming the reproducibility of the assessment.
In contrast, TSR showed no association with overall or disease-specific survival in either the surgical or non-surgical cohorts. Among surgically treated patients, a high TSR did not increase mortality risk. Correlation analyses demonstrated low sensitivity for detecting tumor budding from biopsy samples (22 percent) compared with resection specimens, and poor correlation between biopsy and resection samples for TSR.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
