Hepatocellular carcinoma (HCC), the predominant pathological type of primary liver cancer, is the fifth most prevalent cancer and the third leading cause of cancer-related death worldwide. According to the World Health Organization (WHO), HCC accounts for over 700,000 deaths annually, making it a significant public health concern worldwide [1]. Hepatitis B virus (HBV) and hepatitis C virus (HCV) infections are common etiological risk factors for the initiation and progression of HCC [2]. Patients with hepatitis B-positive hepatocellular carcinoma (HCC) exhibit a more rapid degree of tumor progression and a worse prognosis. Despite advancements in treatment over the past decades, the mortality of patients with HCC continues to increase, emphasizing the need for more effective therapies [3]. According to the latest Barcelona Clinic Liver Cancer (BCLC) clinical staging system [4] for treatment recommendation, Transcatheter Arterial Chemoembolization (TACE) has been suggested as a treatment commonly used in clinical for patients with early and intermediate HCC. While surgical resection or ablation are generally recommended as first-line treatment in BCLC A patients where possible, many patients are not suitable for resection or ablation due to portal hypertension, unsuitable anatomy, severity of liver disease, or non-liver comorbidities [5]. TACE is therefore often used in patients with single solitary tumor who are not suitable for resection or ablation, or otherwise require upfront transplantation [6].

The TACE response varies with each patient because HCC comprises a highly heterogeneous patient population [7]. The objective TACE response rate remains approximately 32–57 % in HCC patients [8,9], however, initial complete response (CR) at first TACE is still the most robust predictor for favorable outcome in HCC [9]. Lao et al. found that hepatitis B virus reactivated in 14.5 % of HCC patients after TACE, stabilized in 66.7 %, and decreased in 18.8 % [10]. Therefore, early assessment of response to TACE, particularly complete response, is essential to make personalized treatment strategies and improve prognosis. Clinical evidence has indicated that some clinical factors, e.g., abnormal alpha-fetoprotein (AFP) level and pretreatment platelet count (PLT) can be independent markers for predicting prognosis of TACE-treated HCCs [11,12]. TACE has been reported to cause HBV reactivation and secondary hepatitis, and HBV reactivation may more commonly present with more severe clinical symptoms [13]. There are no reports of HBV as a clinical factor to predict the TACE response.

Magnetic resonance imaging (MRI) is regarded as a noninvasive and widely used tool for diagnostic imaging, detailed assessment of the liver and tumor, and angiography of the blood vessels [5]. MRI with some functional techniques, such as diffusion-weighted imaging (DWI) and intravoxel incoherent motion (IVIM), plays an important role in assessing early outcomes after TACE treatment [[14], [15], [16], [17]]. However, in clinical practice, owing to the lack of specific biomarkers in preoperative MRI, therapeutic efficiency of TACE for HCCs can only be subjectively characterized and interpreted by experienced physicians [18].

Radiomics is an emerging image post-processing technology that converts traditional medical image information into high-throughput quantitative features [19], and can be used to quantitatively assess tumor biological heterogeneity and prognosis [20,21]. MRI-based radiomics has shown promising results in HCC, with studies demonstrating its potential to predict response to TACE，improve patient stratification and guide personalized treatment strategies [[21], [22], [23], [24]]. Kuang et al. [22] and Cannella et al. [23] predicted TACE response of HCC based on T2W+ arterial phase images from DCE-MRI and gadoxetate disodium-enhanced MRI, respectively. Kong et al. constructed a nomogram model by combining multi-sequence MRI features and clinical factors for predicting TACE response [24]. Besides the internal tumor area, the information from the peritumoral area could also be valuable to predict the biological behavior of HCC [[25], [26], [27], [28]]. A report has shown that irregular margins for the liver tumor and arterial peritumoral enhancement are highly associated with therapeutic efficiency of TACE in HCC [11]. The peritumoral histopathologic changes within a 1 cm distance of the tumor could be a predictor of recurrence in patients with chronic viral hepatitis-associated HCC [25]. Another group also found that radiomics features extracted from the peritumoral area with a radial distance of 2 mm outside the primary lesion in the MRI can be used for the diagnosis of MVI [26]. Song et al. found that the combined model based on clinical-radiological and 3-mm peritumoral radiomics features was more valuable than the clinical-radiological model or radiomics model alone for evaluating the recurrence-free survival of HCC patients after c-TACE [27]. Zhao et al. conducted a small, single-center study based on contrast-enhanced MRI, which found MRI-based radiomics model from the intratumoral and peritumoral region (3 mm) demonstrated superior performance in the estimation of treatment response for HCC patients treated with TACE [28].To our knowledge, there has been no report explored the potential value of peritumoral plus tumoral area of HCC based on multiparameter MRI for predicting complete response to initial TACE in early and intermediate HCC.

This study aims to improve the accuracy and clinical utility of early prediction of initial TACE efficacy by establishing a clinical-radiomics nomogram integrating multi-sequence MR features from both intra- and peritumoral areas with clinical features. This study will provide important guidance for the development of more personalized and effective treatment strategies for HCC.