Introduction
Hepatocellular carcinoma (HCC) is the most common liver malignancy and the third leading cause of death among various cancers [
1]. Liver transplantation, resection, and ablation are the curative therapies for patients with early HCC [
2]. Unfortunately, the majority of HCC patients are not suitable for curative treatment at the time of diagnosis because of poor liver function, multifocal disease, vascular involvement, and extrahepatic spread [
3]. Transarterial chemoembolization (TACE) is widely used as a bridge to liver transplantation, or as the standard treatment for patients with intermediate HCC [
4]. Nevertheless, the therapeutic efficacy of TACE varies greatly due to the high heterogeneity of HCC [
5]. Several studies have evidenced that the overall response rates following TACE range from 15 to 85% and the cumulative rates of local tumor progression at 1, 3, and 5 years are 33%, 52%, and 73%, respectively [
6,
7]. Therefore, it is crucial to preoperatively estimate tumor response to TACE treatment which may aid in guiding subsequent therapeutic strategies.
Magnetic resonance imaging (MRI)-based evaluations that are noninvasive and repeatable can be used to preoperatively assess treatment response. Several scholars have reported that larger lesion diameter, irregular margin, arterial peritumoral enhancement, satellite nodule, and apparent diffusion coefficient (ADC) value are associated with therapeutic efficacy of TACE treatment [
8‐
11]. Although these imaging characteristics are encouraging, they are not sufficient for the individual evaluation of tumor response to TACE, and the ability to predict TACE efficacy in HCC is limited when a high degree of tumor heterogeneity.
Radiomics, an emerging and non-invasive approach, can extract high-throughput quantitative data from multi-modality imaging and characterize tumor heterogeneity, which may potentially guide individual medicine [
12,
13]. Numerous studies have demonstrated that radiomics-based models effectively identify the diagnosis and pathological characteristics or predict therapeutic efficacy and prognosis of cancer patients for clinical decision-making [
14‐
17]. Recently, there has been increasing interest in evaluating radiomics patterns of the region surrounding the visible tumor [
15,
17‐
19]. Recurrence or metastasis of HCC is mainly intrahepatic, indicating that the peritumoral liver tissue may be a favorable soil for the spreading hepatoma cells [
20]. Several scholars have reported that HCC patients with microvascular invasion (MVI), epithelial cell adhesion molecule (EpCAM), programmed death ligand 1 (PD-L1) expression, and higher CD68 + cell density in peritumoral tissues have a significantly higher risk of recurrence or metastasis and cancer-related death [
21‐
24]; thus, peritumoral tissues might have valuable predictive information of HCC prognosis. Several recent studies have reported that CT or MRI-based radiomics on intratumoral and peritumoral regions can effectively predict MVI, vessels encapsulating tumor clusters (VETC), anti-PD-1 treatment efficacy, and prognosis of resection or TACE in patients with HCC, which may achieve an enhanced prediction of the individualized risk estimation [
15,
17‐
19,
25‐
27]. However, the value of intratumoral and peritumoral radiomics based on MRI in predicting treatment response of HCC after TACE remains unknown.
Therefore, the present study aimed to determine whether radiomics assessment of HCC peritumoral regions based on contrast-enhanced MR (CE-MR) images could provide valuable information about TACE response and enhance the ability of intratumoral radiomics for the prediction of treatment efficacy of TACE in patients with HCC.
Discussion
In the study, we constructed various radiomics models of intratumoral, peritumoral, and intratumoral combined peritumoral derived from CE-MR images for preoperatively predicting treatment response of TACE in patients with HCC. Our study confirmed that intratumoral combined peritumoral radiomics models performed better than the intratumoral model. Furthermore, a combined nomogram incorporating clinical-radiological risk factors and the optimal T-PTR (3 mm) rad-score was developed and validated, and demonstrated satisfactory performance, calibration, and clinical utility. The proposed radiomics approach successfully predicted TACE treatment efficacy and may facilitate individualized treatment decision-making for patients with HCC.
In the present study, most of HCC patients receiving TACE had BCLC A or B stage, which was consistent with previous studies [
18,
31]. The BCLC system recommends TACE as the standard therapy for intermediate HCC. TACE could be a candidate treatment option for early-stage patients who are unsuitable for resection or ablation due to old age, hepatic dysfunction, severe comorbidities, and tumor location [
4]. This treatment stage-migration strategy is well established and recommended by international guidelines [
28,
32]. Therefore, TACE clearly has a critical role in the treatment of HCC at early stage, and HCC patients included in our study reflect the real conditions in clinical setting.
Patients with HCC receiving TACE have various treatment efficacy and clinical outcomes [
6,
7]. Objective response after first TACE course has been proved to be an independent and robust prognostic predictor for clinical outcomes, which may aid in guiding individual therapeutic strategies in HCC patients [
33]. Several recent studies [
34‐
37] have constructed radiomics models based on preoperative single MRI sequence or multiparametric MRI (MP-MRI) to predict tumor response of HCC patients receiving TACE, and the AUCs ranged from 0.692 to 0.866 in the validation cohort; however, their studies only focused on intratumoral radiomics features. Pathologically, peritumoral parenchyma is representative of cancerous heterogeneity, and the crucial information can be indicated by changes in the area surrounding tumors, such as biological aggressiveness, microinvasion, and micrometastasis [
15,
18,
38]; thus, accurate evaluation of the neighboring tissue around tumors may also be useful in predicting treatment response and prognosis of TACE in patients with HCC.
Previous studies have reported that radiomics analysis of intratumoral combined 3 mm, 5 mm, and 10 mm peritumoral regions can provide valuable information for prognosis prediction in HCC [
18,
19,
30,
39,
40]. In addition, according to the practice guidelines for the pathological diagnosis of primary liver cancer (2015 update) [
41], the liver tissues within 10 mm surrounding the tumor are defined as the adjacent areas around the cancer, where the probability of MVI is high. Therefore, our study selected the most stable and predictive radiomics features from intratumoral, 3 mm, 5 mm, and 10 mm peritumoral regions for radiomics model construction, which can quantitatively assess the heterogeneity and invasiveness of intratumoral and peritumoral tissues in a non-invasive way. In the present study, the PTR (3 mm), PTR (5 mm), and PTR (10 mm) radiomics models showed comparable performance compared with the TR model, which indicated that peritumoral tissues were possess of a clinical value in assessing treatment efficacy. Yang et al. [
42] reached a similar result that peritumoral radiomics model obtained equivalent performance compared with the intratumoral model in predicting MVI in HCC patients with the AUCs of 0.714 and 0.728, respectively. The radiomics features contributed to peritumoral model construction in the study were most derived from AP images. This finding was in agreement with previous studies, in which the presence of peritumoral enhancement in AP images indicated more aggressive biological behavior [
18,
31].
We further combined intratumoral and peritumoral rad-scores to establish T-PTR (3 mm), T-PTR (5 mm), and T-PTR (10 mm) radiomics models for predicting TACE response. In our study, T-PTR radiomics models demonstrated better predictive performance compared with the TR radiomics model, which indicated that peritumoral radiomics might potentially enhance the ability of intratumoral radiomics for TACE response prediction. This might be interpreted that arterial peritumoral enhancement and irregular margin presented in the peritumoral area are independent predictors of prognosis in HCC patients [
18,
19]. Chen et al. [
30] found that intratumoral and peritumoral (5 mm, 10 mm, and 20 mm) radiomics models based on contrast-enhanced CT images performed better than the intratumoral radiomics model in predicting the first TACE response with the AUCs of 0.790, 0.810, 0.750, and 0.720, respectively, which was consistent with our study. Additionally, our study demonstrated that the T-PTR (3 mm) radiomics model achieved the best-performing performance among the seven radiomics models. A similar study reported by Liu et al. [
40] found that intratumoral and peritumoral (3 mm) radiomics model showed better performance compared with radiomics models on intratumoral, peritumoral (3 mm), peritumoral (5 mm), and intratumoral and peritumoral (5 mm) for predicting 1-year survival of HCC after hepatectomy. Only one of the published studies, conducted MRI-based radiomics on intratumoral and peritumoral regions for TACE prognosis prediction [
18]. In their study, radiomics models based on the entire tumor volumetric of AP (AP
ETV), PVP
ETV, and the border extensions of 1 mm, 3 mm, and 5 mm on the PVP (PVP
B1, PVP
B3, and PVP
B5) were constructed to predict recurrence-free survival (RFS) of HCC patients after TACE. The best C-index results of PVP
ETV and PVP
B3 radiomics models were 0.727 and 0.714 in the validation dataset, respectively. However, the above research only performed radiomics analysis of whole areas including intratumoral and peritumoral regions, and did not explore the individual contribution of the area around the tumor to predictive model construction; thus, it was unable to determine the significance of the separate peritumoral region in predicting recurrence or prognosis. Compared with the previous study [
18], our study may have the following advantages: first, radiomics features derived from three-phase enhanced MR images might more fully reflect tumor heterogeneity and vascularization patterns, which is helpful for efficacy estimation; second, the individual peritumoral (3 mm, 5 mm, and 10 mm) radiomics models were constructed, and the valuable peritumoral distance was determined; third, intratumoral combined peritumoral radiomics analysis may contain more prognostic information, and potentially provide a more accurate and effective approach of individualized efficacy prediction for HCC patients.
In this study, during the construction of the clinical-radiological model, ALP, tumor size, and satellite nodule were independent predictors associated with treatment response of HCC after TACE. Previous researches on TACE clarified that a higher ALP value was an independent risk factor for unfavourable overall survival (OS) [
43,
44]. Our study showed that abnormal ALP value was a significant predictor for poor response of HCC. Additionally, ALP has already been included in the Chinese University Prognostic Index, a HCC staging system that assigns a score of 3 when ALP is > 200 IU/L, indicating the potential role of ALP in predicting the prognosis of HCC patients [
45]. Tumor size has been broadly recognized as a major predictive factor of treatment response for TACE [
9,
33]. Larger tumors usually have more satellite lesions or daughter nodules making it difficult for TACE to achieve CR [
46]. In our study, maximal tumor size > 5 cm was a significant predictive factor for NR, a result similar to the study by Jeong et al. [
47]. Several studies reported that satellite nodule surrounding the main tumor was closely related to tumor grade, MVI, and early recurrence (ER) after resection, and TACE treatment efficacy and prognosis [
10,
17,
37,
48]. Our study demonstrated that the presence of satellite nodule was inclined to show NR to TACE treatment. This may be interpreted that the development of satellite nodule favors vascular invasion and also tumor recurrence [
48].
We ultimately developed a combined nomogram integrating the T-PTR (3 mm) rad-score with clinical-radiological risk indicators for treatment response prediction. The combined nomogram achieved good calibration and the strongest predictive performance based on AUCs in the training (nomogram vs. radiomics model vs. clinical-radiological model, 0.910 vs. 0.884 vs. 0.789) and validation (nomogram vs. radiomics model vs. clinical-radiological model, 0.918 vs. 0.911 vs. 0.782) cohorts. The novel combined nomogram was evaluated by a decision curve to clarify the clinical usefulness, which may offer insight into clinical outcomes on the basis of threshold probability, from which the net benefit could be derived [
36]. Our results clearly demonstrated that the combined nomogram could obtain more net benefit than either the treat-all-patients or the treat-none-patients strategies across a wide range of threshold probabilities. Therefore, our novel nomogram may provide a reliable and efficient tool that enables visualized and personalized decision-making for the treatment management of patients with HCC.
This study has several limitations. Firstly, this was a retrospective study at a single center, which may introduce selection bias. The sample size was relatively small, especially for the independent testing cohort. A larger cohort population from multi-center is further needed to externally validate the robustness and reproducibility of the predictive models and reinforce the conclusions of our study. Secondly, the ROIs were manually delineated by radiologists, and thus is time-consuming and prone to error and user variability. It’s essential to develop an automatic and reliable liver tumor segmentation tool. Thirdly, it should be noted that MP-MRI data are not included in this study. In the future, we will attempt to develop a radiomics approach based on MP-MRI for response evaluation after TACE. Fourthly, for patients with multifocal HCCs, our study chose the largest lesion for radiomics analysis. A further direction will be considered to perform the per-lesion level study, as well as to explore how to comprehensively analyze radiomics features of multifocal lesions for treatment efficacy prediction. Finally, our study provided a promising tool for the precise prediction of treatment response after the initial TACE. In the future, we will try to explore MRI-based intratumoral and peritumoral radiomics features associated with RFS of HCC patients treated with TACE.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.