The purpose of this study was to develop and validate a deep learning (DL)-based radiomics model to predict the response to chemotherapy in colorectal liver metastases (CRLM).

In this retrospective study, we enrolled 192 patients diagnosed with CRLM who received first-line chemotherapy and were followed by response assessment. Tumor response was identified according to the Response Evaluation Criteria in Solid Tumors (RECIST). Contrast-enhanced multidetector computed tomography (MDCT) images were fed as inputs of the ResNet10-based DL radiomics model, and the possibility of response was predicted as the output. The final combined DL radiomics model was constructed by integrating the response-related clinical factors and the developed DL radiomics signature. A time-independent validation cohort (n = 48) was extracted from the 192 patients to evaluate the DL model with area under the receiver operating characteristic curve (AUC), specificity, and sensitivity. Meanwhile, a traditional radiomics model was constructed using least absolute shrinkage and selection operator (lasso) as comparisons with the DL-based model.

According to RECIST criteria, 131 patients were identified as responders with complete response, partial response, and stable disease, while 61 patients were nonresponders with progression disease. The selected predictive clinical factor turned out to be the carcinoembryonic antigen (CEA) level with AUC of 0.489 (95% confidence interval [CI], 0.380-0.599) and 0.558 (95% CI, 0.374-0.741) in the training and validation cohorts, respectively. The DL-based model provided better performance than the traditional classifier-based radiomics model with significantly higher AUC (training: 0.903 [95% CI, 0.851-0.955] vs 0.745 [95% CI, 0.659-0.831]; validation: 0.820 [95% CI, 0.681-0.959] vs 0.598 [95% CI, 0.422-0.774]). The combination of DL-based model with the CEA level provided slightly increased performance with AUC of 0.935 [95% CI, 0.897-0.973] in the training cohort and 0.830 [95% CI, 0.688-0.973] in the validation cohort.

The developed DL-based radiomics model could improve the efficiency to predict the response to chemotherapy in CRLM, which may assist in subsequent personalized treatment decision-making in CRLM management.

OBJECTIVE. The purpose of this article is to evaluate the enhanced rim on the portal venous phase (PVP) on MDCT as a predictor of 1-year progression-free survival (PFS) and response to bevacizumab-based chemotherapy in patients with colorectal liver metastases (CRLM). MATERIALS AND METHODS. We retrospectively identified 111 patients with primary unresectable CRLM treated with bevacizumab-based chemotherapy at two institutions between 2012 and 2018. Pretreatment contrast-enhanced MDCT images were reviewed and data on clinical characteristics were collected from the electronic medical records. Univariable and multivariable analyses were conducted to assess several imaging features and clinical characteristics as potential predictors of 1-year PFS and objective response rate (ORR). RESULTS. After 1 year of follow-up, liver metastatic tumor progression was detected in 52 patients (46.8%) after bevacizumab-based chemotherapy. A log-rank test showed that enhanced rim on PVP (chi-square test, 5.862; p = 0.015) and the occurrence of liver resection surgery (chi-square test, 7.836; p = 0.005) were significant predictors of 1-year PFS. Multivariable analysis showed that enhanced rim on PVP images was an independent predictor of 1-year PFS (hazard ratio, 0.510; 95% CI, 0.282-0.926; p = 0.027) and ORR (odds ratio, 4.694; p < 0.001). CONCLUSION. The presence of an enhanced rim on PVP MDCT is an independent predictor of survival and response to bevacizumab-based chemotherapy among patients with CRLM.

Purpose: Developing an MRI-based radiomics model to effectively and accurately predict the predominant histopathologic growth patterns (HGPs) of colorectal liver metastases (CRLMs). Materials and Methods: In this study, 182 resected and histopathological proven CRLMs of chemotherapy-naive patients from two institutions, including 123 replacement CRLMs and 59 desmoplastic CRLMs, were retrospectively analyzed. Radiomics analysis was performed on two regions of interest (ROI), the tumor zone and the tumor-liver interface (TLI) zone. Decision tree (DT) algorithm was used for radiomics modeling on each MR sequence, and fused radiomics model was constructed by combining the radiomics signature of each sequence. The clinical and combination models were developed through multivariate logistic regression method. The performance of the developed models was assessed by receiver operating characteristic (ROC) curves with indicators of area under curve (AUC), accuracy, sensitivity, and specificity. A nomogram was constructed to evaluate the discrimination, calibration, and usefulness. Results: The fused radiomicstumor and radiomicsTLI models showed better performance than any single sequence and clinical model. In addition, the radiomicsTLI model exhibited better performance than radiomicstumor model (AUC of 0.912 vs. 0.879) in internal validation cohort. The combination model showed good discrimination, and the AUC of nomogram was 0.971, 0.909, and 0.905 in the training, internal validation, and external validation cohorts, respectively. Conclusion: MRI-based radiomics method has high potential in predicting the predominant HGPs of CRLM. Preoperative non-invasive identification of predominant HGPs could further explore the ability of HGPs as a potential biomarker for clinical treatment strategy, reflecting different biological pathways.

Hepatic metastases from colorectal carcinoma frequently recur after resection and hepatic micrometastases most likely are important in the development of such recurrences. The objectives of the current study were to assess the feasibility of the immunohistochemical detection of hepatic micrometastases from colorectal carcinoma and to determine their clinical significance.

Fifty-three patients underwent curative hepatic resection for colorectal carcinoma metastases. Multiple tissue sections were cut from the advancing margin of the largest hepatic metastasis in each patient and were stained with an antibody against cytokeratin-20 to detect hepatic micrometastases, which were defined as discrete microscopic cancerous lesions surrounding the dominant metastasis.

Normal hepatocytes and intrahepatic bile duct epithelia stained negative for cytokeratin-20 in all patients, whereas the largest hepatic tumors stained positive in 46 patients (86.8%). Among the 46 patients with hepatic tumors that were positive for cytokeratin-20, hepatic micrometastases were found immunohistochemically in 32 patients (69.6%). The presence of hepatic micrometastases was associated with a larger number of macroscopic hepatic metastases (P = 0.047) and patients with hepatic micrometastases were found to demonstrate a higher probability of intrahepatic recurrence (P = 0.003) compared with those patients without hepatic micrometastases. In addition, patients with hepatic micrometastases demonstrated a worse survival (10-year survival rate of 21.9%) compared with those patients without hepatic micrometastases (10-year survival rate of 64.3%) (P = 0.017).

Immunohistochemical detection of hepatic micrometastases is feasible in patients with colorectal carcinoma liver metastases. Hepatic micrometastasis indicates widespread hepatic involvement and thus predicts an increased risk of intrahepatic recurrence after hepatic resection and a poorer patient prognosis.