| OBJECTIVEFirstly, we postulate that conventional imaging features may provide reliable information for HCC prognosis and provide better guidance with respect to treatment strategies. To test our hypothesis, we conducted this study on patients with a single HCC lesion above BCLC stage A (>5 cm) with no extrahepatic metastases.Secondly, we explored texture analysis as a prognostic and patient stratification approach in the determination of the appropriate therapeutic option, LR or TACE, for patients with single large HCCs. Herein, two questions were raised:(1) are the textural parameters of the primary tumor, calculated from baseline CT, related to prognosis? (2) Does texture analysis have the potential to provide an additional view for treatment modification between LR and TACE?SIGNIFICANCESingle large HCC include a broad round of patients with prognosis heterogeneity. In our clinical practice, not all patients with BCLC stages above A exhibited a greater survival benefit with LR. Moreover, LR may increase perioperative complications and prolong postoperative recovery compared to TACE. Therefore, to assess whether a patient scheduled to undergo LR would be better suited for TACE and vice versa, reliable prognostic markers for patient stratification are needed.Conventional CT image might by largely overlooked and conveyed information on HCC prognosis. If feasibility of detailed grouping by conventional image features and/or textural features in patient stratification are confirmed, conventional image phenotype and texture analysis could potentially be used to inform the LR vs. TACE decision-making strategy.BACKGROUNDHCC and BCLC staging systemLiver cancer is the fifth and seventh most frequently diagnosed cancer, ranking second and sixth in the causes of cancer-related deaths, for men and women, respectively. Hepatocellular carcinoma (HCC) represents the majority of liver cancer types.The Barcelona Clinic Liver Cancer (BCLC) staging system, which links therapeutic strategies to treatment outcomes, is commonly used to decide treatment modalities. BCLC is a well-recognized staging system for HCC and has been adopted by the European Association for the Study of the Liver and the European Organization for Research and Treatment of Cancer, as well as by the American Association for the Study of Liver Diseases.Controversies on classification of single large HCCHowever, controversies have recently arisen regarding treatment strategies in certain situations, such as single tumors>5 cm without vascular invasion or extrahepatic metastasis should be classified as stage A or B. Considering the uncertainty, we adopted the solution of adding a separate subgroup of stage AB proposed by one recent study, in which stage AB was defended as a single HCC larger than 5 cm without vascular invasion or extrahepatic metastasis, Child Pugh A-B, and PST 0 or 1.Controversies on initial treatment of single large HCCAccording to the BCLC staging system, LR is recommended for HCC patients up to stage A disease, while TACE is the best choice for stage B. This controversies may alter therapy-based decision-making in cases of single HCCs> 5 cm, particularly for potential LR candidates.Basing on BCLC staging system, the diameter of a single HCC may not be a contraindication for liver resection (LR). In addition, other studies also pointed out that asymptomatic patients with a solitary HCC without vascular invasion or extrahepatic spread and with well-preserved liver function could also be considered for transcatheter arterial chemoembolization (TACE). One study showed that, regardless of BCLC stage, LR resulted in a greater survival benefit than locoregional therapy such as TACE and ablation. Another study showed that, for HCC patients with resectable multiple lesions with a BCLC stage higher than A, LR resulted in better survival than conventional TACE. However, most Asian patients with HCC have diseased liver parenchyma, such as hepatitis B virus infection and/or hepatitis B virus-related cirrhosis, and LR in this population is therefore associated with a high risk of complications.Feasibility of a subclassification systemProposals for a subclassification system for BCLC stage B tumors have emerged in recent years. One study proposed a stratification system aimed toward tailoring therapeutic interventions based on both the evidence available to date and expert opinions. Another study suggested taking the Child-Pugh score and extended criteria for transplantation into account. In clinical practice, the decision to treat with LR or TACE is made using a combination of clinical symptoms, laboratory test results, and pathological biomarkers, whereas the CT images routinely acquired during treatment and follow-up are largely overlooked. Conventional assessment of tumor size and enhancement of cross-sectional images are far from satisfactory in the determination of an appropriate therapeutic strategy, due to insufficient imaging of the inherent properties of the tumor and interobserver variability in image interpretation.Conventional CT image featuresFor years, prognostic studies mainly focused on clinical or biological markers. However, routinely acquired computed tomography (CT) and magnetic resonance (MR) images were largely undervalued. Endorsed by major clinical practice guidelines. CT and MR imaging played emerging crucial roles in the diagnosis, staging and characterization of HCC. Although the use of hepatobiliary agents for MR imaging can provide additional diagnostic information (especially for small HCC), studies of imaging features as they pertainal to HCC prognosis and patient selection are limited.Corona enhancement and mosaic architecture are two imaging features favoring the diagnosis of large HCC. However, neither their prognostic values nor their impacts on the selection of liver resection (LR) vs. transcatheter arterial chemoembolization (TACE) as treatment modalities have been established. Corona is pointing to enhancement of the venous drainage area in the peritumoral parenchyma. Studies shown that corona enhancement might convey information on microvascular invasion and metastatic satellites. HCCs with corona enhancement findings tend to be diagnosed as progressed, hypervascular HCC. Mosaic architecture is the presence of randomly distributed internal nodules or compartments within a mass that differ in shape on enhancement. Mosaic patterns on characterization and differential diagnosis were previously described in Yersinia colitis. In HCC, mosaic architecture is characteristic of tumor heterogeneity with histologic and cytologic variations more common in large HCC, manifesting by confounding factors such as fibrous separations, necrosis, hemorrhage, copper deposition, and fatty infiltration.Interestingly, corona enhancement can be confused with a peritumor capsule in its appearance; the distinction between the two patterns is that the corona pattern fades in the hepatic venous phase while a peritumor capsule manifests as a progressively enhancing rim at delayed phase. Similarly, mosaic architecture may tend to be confused with necrosis, which manifested as hyperintensity on T2 weighed images. Proper recognition of these two patterns is crucial in image interpretation. In our study, a third radiologist was consulted to obtain consensus evaluation and reduce the probability of misinterpretation.Texture analysisRadiomics is an emerging research field that aims to utilize the full potential of medical imaging. This includes texture analysis, which is assumed to reflect tissue heterogeneity. Heterogeneity is a well-recognized feature reflecting alterations in tissue patterns, likely occurring due to cell infiltration, abnormal angiogenesis, microvasculature and necrosis. Two studies reported an association between image traits and tumor glucose metabolism and stage, hypoxia and angiogenesis, respectively. One study suggested an association between image traits, including textural features, and underlying gene expressions in HCC. Another stated that the radiomic signature could be transferred from lung to head-and-neck cancer, suggesting that this signature identifies a general prognostic tumor phenotype. In fact, texture analysis has shown feasibility in the differential diagnosis of liver cancers, hepatic fibrosis detection/staging, and prediction of postoperative hepatic insufficiency. Texture analysis for oncology has shown promising results in the field of survival analysis and treatment response evaluation in non-small cell lung cancer, colorectal cancer, locally advanced squamous cell carcinoma of the head and neck, and so on. Inspired by these studies, extetion to HCC prgnosis and treatment evaluation is not unexpected.An explanation for relationship between texture analysis and HCC prognosis might be attributed to angiogenesis. One study pointed out that a combination of 28 imaging traits including textural features could reconstruct 78% of the global gene expression profiles in HCC, including vascular endothelial growth factor (VEGF) and so on, revealing cell proliferation, liver synthetic function, and patient prognosis. Additionally, computer simulations and clinical studies directly suggested that hepatic texture observed on CT might reflect liver vascularity. In HCC, angiogenesis regulated by VEGF and other factors could directly affect HCC prognosis.Interestingly, we noted that, not all the features showed significant correlation with survival in RS. Consistent with many studies, not all features were related to the endpoint. One study also pointed out that features should be robustly reflect the complexity of the individual volumes, but cannot be overly complex or redundant. We hypothesized that subtle difference in the formula of specific features might lead to different relationships with HCC. Notably, although patient characteristics such as tumor size, AFP level, and Child-Pugh class were taken into account, statistical difference was not observed in cox regressions. Suggesting texture features might have additional prognostic potential; in fact, there is no unequivocal validation of these features in HCC prognosis.Texture analysis is associated with challenges in image acquisition. In a previous phantom study, texture parameters were demonstrated to be relatively sensitive to tube voltage, but to be independent of the tube current. Additionally, one study showed that hepatic texture features were less sensitive to changes in CT acquisition parameters. Slice thickness is another major determinant of textural parameters, with one study revealing that a slice thickness of< 3 mm was optimal for feature grading. Thus, we carefully excluded images outside this criterion in the present study, which might have partly reduced the influence of textural parameter reproducibility in prognostic evaluation.METHODWe retrospectively analyzed 275 patients with a single HCC lesion>5 cm without extrahepatic metastasis treated with LR or TACE. Typical imaging features of corona (corona-or corona+) and mosaic (mosaic-or mosaic+) patterns were consensus-classified by two radiologists with 5 and 4 years of experience in abdominal CT interpretation, respectively. In LR patients, the overall survival (OS) and time to progression (TTP) were compared between corona enhancement negative (corona-) vs. positive (corona+) and mosaic architecture negative (mosaic-) vs. positive (mosaic+) patients. Furthermore, by the combination of corona and mosaic, LR patients were divided into negative for both corona and mosaic patterns (LR-/-), positive for only one feature (LR+/-), and positive for both (LR+/+); their OS and TTP were compared to those of the TACE group. Cox regression was performed to identify independent factors for OS.To explore the prediction power of textural features comparing with radiographic features, six typical subjective imaging features were involved and consensus-classified by two radiologists, including the shape, capsule, corona, mosaic, node-in-node, and enhanced region relative to the entire tumor.130 patients with a single large HCC (> 5cm) initially treated by LR or TACE were enrolled. For BCLC stage C, only patients with branch vascular invasion were included.130 patients were classified as stage AB (without vascular invasion) and stage C (with vascular invasion).For each pre-treatment examination,1.25-mm axial images obtained at the portal venous phase through the largest cross-sectional area of the tumor were selected and transferred to two personal computers for texture analysis. The process of texture analysis comprised three steps:(1) image filtration, (2) wavelet analysis and (3) feature extraction. The first two steps were performed using MATLAB software. Laplacian of Gaussian (LoG) spatial band-pass filters were used to reduce the sensitivity to noise. Three a values (0,1.0, and 1.5) and a single filter-width of σ*5 pixels were used. Each image is decomposed into 1,2, or 3 levels and reconstructed in three directions (diagonal, horizontal and vertical). Two radiologists independently performed textural feature extraction and quantification using ImageJ software. For each reader, a user-defined irregular ROI was drawn manually around the largest cross-sectional tumor outline and copied to the nine derived texture feature maps. Subsequently, the values of the texture features were measured and saved for further analysis.Inter-observer agreement on textual features was evaluated using intraclass correlation coefficients (ICCs). Cox regression were used for feature screening. Kaplan-Meier analysis and Cox regression were used for survival analysis. One way-ANOVA or Kruskal-Wallis H was used to compare the identified textural parameters among the subgroups. Post hoc multiple comparisons were performed using Bonferroni’s correction or Dunnett’s T3 test.RESULTSConventional image features:Corona and mosaicIn the survival plots for LR, corona-had better OS and TTP than corona+, and mosaic-had better OS than mosaic+. There was no significant difference in TTP between the subgroups. On Cox regression analysis, corona enhancement, but not mosaic architecture, was a significant factor for OS, whereas neither were a significant factor for TTP. In TACE patients, neither corona nor mosaic patterns had significant correlations with OS or TTP.Since corona and mosaic patterns had limited prognostic value in the TACE group but were significant prognostic factors in LR patients, subgrouping by corona and/or mosaic pattern was not performed for the TACE group (which was used as a reference in this analysis), whereas LR patients were subgrouped into LR-/-(negative for both corona and mosaic patterns), LR+/-(either a positive corona or mosaic pattern), and LR+/+(positive for both corona and mosaic patterns).In subgroup analysis, our results showed that the patients in the LR-/-achieved better prognosis than LR+/+patients. Moreover, LR-/-patients had a significant survival benefit (for both for OS and TTP), as did LR+/-patients. Meanwhile, LR+/+ had limited survival compared to LR-/-and LR+/-, and showed no significant difference compared to TACE patients. The consequences of these findings may be as follows:1) for patients negative for both corona and mosaic patterns or positive for only one of them, LR should be recommended if possible; and 2) for patients positive for both patterns, LR does not provide a survival benefit compared to TACE. Therefore, considering the possibly of increased perioperative risks and prolonged postoperative recovery, TACE might be recommended as the first-line therapy for patients positive for both patterns. Multivariate regression showed that PEI, BCLC stage, and subgroup were significant predictors of OS.Conventional image features vs. textural features There were no significant differences in the patient baseline demographics and characteristics. All texture features, calculated from two sets of regions of interest (ROIs), showed excellent agreement (ICC value,0.799-0.999).In this study, none of the candidate clinical and imaging variables showed significant differences. Multivariate Cox models showed that only Wavelet-2-H (filter 1.0) in LR and wavelet-2-V (filter 0 and 1.0) in TACE were related to survival. Without detailed subgrouping, patients in the LR group showed better OS (χ2= 9.809, P= 0.002) and TTP fa2= 5.840, P= 0.016) than those in the TACE group (Fig 2A & 2D). Subsequently, LR and TACE patients were divided based on the wavelet-2-H and wavelet-2-V median at filter 1.0 into two subgroups (+or-). OS and TTP showed a significant difference among the four subgroups, resepectively (χ2= 24.292, P< 0.001; χ2= 10.969, P= 0.012).LR+patients showed the best survival, followed by LR-, TACE+, and TACE-. Similar results were noted when LR was separated by wavelet-2-H (filter 1.0) and TACE was separated by wavelet-2-V (filter 0). Filter 1.0 was the best filter, as it showed significant results in both the LR and TACE groups, which was consistent with the findings of published studies. The reason for this result might be that textural features at filter 0 tend to reflect radiologists’impressions of image quality, which could be influenced by image noise. By using filters at larger scales (filter 1.0,5 pixels), subjective bias might be alleviated, and underlying biologic heterogeneity could be enhanced.In all patients, for OS, univariate Cox regression showed that BCLC, corona, and subgrouping had P-values< 0.10, and the multivariate Cox regression models confirmed that subgrouping was the only factor that was significantly associated with OS (P= 0.012). For TTP, univariate Cox regression showed that the presence of a capsule, corona and subgrouping had P-values< 0.10, and the multivariate Cox regression models confirmed that the capsule was the only factor that was significantly associated with the TTP (P= 0.021).Untill now, the feasibility of texture features in patient stratification and determination of the most suitable therapy (LR or TACE) was partly confirmed; however, further validation was still considered necessary, identified textural parameters among the subgroups were compared.We noted that LR+ patients treated using TACE would exhibit a survival similar to TACE-patients and worse than TACE+ patients, with a severe compromise in overall survival. If LR-patients are treated by TACE, their survival would be similar to that of TACE+ patients and better than that of TACE-patients, without compromise of OS. If TACE+ patients are treated by LR, their OS would be similar to that of LR-patients and worse than that of LR+ patients, with no extension of survival. If TACE-patients are treated by LR, their survival would be similar to that of LR+ patients and better than that of LR-patients, and their OS would be considerably improved. Therefore, LR was recommended for LR+ and TACE-patients, whereas TACE was preferred for LR-and TACE+patients.CONCLUSION Our results showed that, for patients with a single HCC>5 cm without extrahepatic metastasis, corona and mosaic patterns are indicators of limited LR efficacy. When both of the features are present, TACE can be used instead of LR with no negative influence on survival.Identification and quantification of tumor heterogeneity by computed tomography (CT) textural analysis shows promise in enhancing prognostic accuracy and facilitating therapeutic decision making. Independent of tumor size, CT textural features showed positive and negative correlations with survival after LR and TACE, respectively. Although further validation is needed, texture analysis demonstrated the feasibility of using HCC patient stratification for determining the suitability of LR vs. TACE. |
PDF Full Text Request