Application of Machine Learning Techniques to Assess Alpha-Fetoprotein at Diagnosis of Hepatocellular Carcinoma

Abstract

Hepatocellular carcinoma (HCC) is the most common primary liver tumor and is associated with high mortality rates. Approximately 80% of cases occur in cirrhotic livers, posing a significant challenge for appropriate therapeutic management. Adequate screening programs in high-risk groups are essential for early-stage detection. The extent of extrahepatic tumor spread and hepatic functional reserve are recognized as two of the most influential prognostic factors. In this retrospective multicenter study, we utilized machine learning (ML) methods to analyze predictors of mortality at the time of diagnosis in a total of 208 patients. The eXtreme gradient boosting (XGB) method achieved the highest values in identifying key prognostic factors for HCC at diagnosis. The etiology of HCC was found to be the variable most strongly associated with a poorer prognosis. The widely used Barcelona Clinic Liver Cancer (BCLC) classification in our setting demonstrated superiority over the TNM classification. Although alpha-fetoprotein (AFP) remains the most commonly used biological marker, elevated levels did not correlate with reduced survival. Our findings suggest the need to explore new prognostic biomarkers for individualized management of these patients.

Prognostic Impact of Metabolic Syndrome and Steatotic Liver Disease in Hepatocellular Carcinoma Using Machine Learning Techniques

“…Additionally, these patients are at risk of developing HCC in the absence of liver cirrhosis, which would necessitate a biopsy to establish the diagnosis [ 5 , 70 ]. It is anticipated that in the coming years, MASLD will surpass the main causes historically associated with the development of these tumors, owing to the rise in obesity, the implementation of HBV vaccination strategies in newborns, and the introduction of new treatments for HCV [ 8 , 19 ]. Although only around 1% of MASLD patients will specifically die from liver-related causes, its high prevalence should prompt us to consider this disease.…”

“…At present, there is no molecule or biomarker available capable of predicting the course of the disease or facilitating its diagnosis in early stages for inclusion in the screening program. Although the determination of alpha-fetoprotein (AFP) is widely used due to its easy availability, it does not provide benefits for screening, while its true utility in the prognosis of these patients is still debated [ 19 ]. Therefore, it is crucial to focus on researching other factors and signaling pathways related to the pathogenesis of these tumors, which could impact patient survival by favoring their detection in early stages.…”

“…The eXtreme Gradient Boosting (XGB) algorithm was chosen as the method for developing the learning model, and it was compared with other systems commonly used in the scientific literature [ 24 ]. XGB was selected for its fast execution, its ability to adapt to large volumes of data, and its accuracy in results obtained to date in various areas of clinical practice [ 19 , 25 ].…”

Personalized Assessment of Mortality Risk and Hospital Stay Duration in Hospitalized Patients with COVID-19 Treated with Remdesivir: A Machine Learning Approach

“…In this study, the extreme gradient boost (XGB) method was used as the reference algorithm, thanks to its notable features, such as fast execution, scalability, and high processing capability through parallel computing [21,22]. XGB consistently outperforms other algorithms in accurately solving various data science problems [22][23][24]. Additionally, a comparative analysis was conducted with other supervised ML systems.…”