Posted: Thursday, October 6, 2022
Patients with advanced hepatocellular carcinoma have an increased risk of early death. Yewei Zhang, MS, MD, of The Second Affiliated Hospital of Nanjing Medical University, Nanjing, China, and colleagues identified risk factors of early death in this patient population and established predictive nomograms. Their findings, published in BMC Gastroenterology, suggested the nomograms were accurate and may benefit clinicians in predicting the likelihood of early death based on patient risk factors.
“The nomograms may help oncologists and clinicians identify risk factors and probability of early death more quickly and accurately, so as to allow for more precise individualized treatment plans to be formulated, thereby improving the patients’ survival probability and quality of life,” the investigators stated.
Using the Surveillance, Epidemiology, and End Results (SEER) database, the researchers retrospectively evaluated 6,603 patients diagnosed with stage IV hepatocellular carcinoma between 2010 and 2015. Overall, 21% had stage IVA disease, and 79% had stage IVB disease. Early death was defined as death that occurred within 3 months of initial diagnosis. The researchers constructed predictive nomograms, performed an internal validation, and used decision curve analysis to verify the clinical application value of the models.
For patients with stage IVA disease, the multivariable analyses indicated the risk factors for early death included age, tumor size, histologic grade, alpha-fetoprotein level, fibrosis score, tumor state, surgery, radiotherapy, and chemotherapy. The risk factors for stage IVB disease were age, histologic grade, alpha-fetoprotein level, tumor stage, node stage, bone metastasis, lung metastasis, surgery, radiotherapy, and chemotherapy. For stage IVA and IVB disease, the concordance indices were 0.820 and 0.785, respectively. The nomogram calibration plots also indicated that the predicted early death was consistent with the actual value, and the results of the decision curve analysis demonstrated that the nomograms may have clinical application.
Disclosure: The study authors reported no conflicts of interest.