Prognostic implications of adaptive immune features in MMR-proficient colorectal liver metastases classified by histopathological growth patterns

Abstract Adjuvant systemic chemotherapy (CTx) is widely administered in patients with colorectal liver metastases (CRLM). Histopathological growth patterns (HGPs) are an independent prognostic factor for survival after complete resection. This study evaluates whether HGPs can predict the effectiveness of adjuvant CTx in patients with resected CRLM. Two main types of HGPs can be distinguished; the desmoplastic type and the non-desmoplastic type. Uni- and multivariable analyses for overall survival (OS) and disease-free survival (DFS) were performed, in both patients treated with and without preoperative chemotherapy. A total of 1236 patients from two tertiary centers (Memorial Sloan Kettering Cancer Center, New York, USA; Erasmus MC Cancer Institute, Rotterdam, The Netherlands) were included (period 2000–2016). A total of 656 patients (53.1%) patients received preoperative chemotherapy. Adjuvant CTx was only associated with a superior OS in non-desmoplastic patients that had not been pretreated (adjusted hazard ratio (HR) 0.52, 95% confidence interval (CI) 0.37–0.73, p < 0.001), and not in desmoplastic patients (adjusted HR 1.78, 95% CI 0.75–4.21, p = 0.19). In pretreated patients no significant effect of adjuvant CTx was observed, neither in the desmoplastic group (adjusted HR 0.83, 95% CI 0.49–1.42, p = 0.50) nor in the non-desmoplastic group (adjusted HR 0.96, 95% CI 0.71–1.29, p = 0.79). Similar results were found for DFS, with a superior DFS in non-desmoplastic patients treated with adjuvant CTx (HR 0.71, 95% CI 0.55–0.93, p < 0.001) that were not pretreated. Adjuvant CTx seems to improve OS and DFS after resection of non-desmoplastic CRLM. However, this effect was only observed in patients that were not treated with chemotherapy.

Download Full-text

Distinguishing pure histopathological growth patterns of colorectal liver metastases on CT using deep learning and radiomics: a pilot study

Clinical & Experimental Metastasis ◽

10.1007/s10585-021-10119-6 ◽

2021 ◽

Vol 38 (5) ◽

pp. 483-494

Author(s):

Martijn P. A. Starmans ◽

Florian E. Buisman ◽

Michel Renckens ◽

François E. J. A. Willemssen ◽

Sebastian R. van der Voort ◽

...

Keyword(s):

Liver Metastases ◽

Colorectal Liver Metastases ◽

Cross Validation ◽

Area Under The Curve ◽

Growth Patterns ◽

Learning Approaches ◽

Intra Class Correlation ◽

Preoperative Prediction ◽

Intra Class Correlation Coefficient ◽

Histopathological Growth Patterns

AbstractHistopathological growth patterns (HGPs) are independent prognosticators for colorectal liver metastases (CRLM). Currently, HGPs are determined postoperatively. In this study, we evaluated radiomics for preoperative prediction of HGPs on computed tomography (CT), and its robustness to segmentation and acquisition variations. Patients with pure HGPs [i.e. 100% desmoplastic (dHGP) or 100% replacement (rHGP)] and a CT-scan who were surgically treated at the Erasmus MC between 2003–2015 were included retrospectively. Each lesion was segmented by three clinicians and a convolutional neural network (CNN). A prediction model was created using 564 radiomics features and a combination of machine learning approaches by training on the clinician’s and testing on the unseen CNN segmentations. The intra-class correlation coefficient (ICC) was used to select features robust to segmentation variations; ComBat was used to harmonize for acquisition variations. Evaluation was performed through a 100 × random-split cross-validation. The study included 93 CRLM in 76 patients (48% dHGP; 52% rHGP). Despite substantial differences between the segmentations of the three clinicians and the CNN, the radiomics model had a mean area under the curve of 0.69. ICC-based feature selection or ComBat yielded no improvement. Concluding, the combination of a CNN for segmentation and radiomics for classification has potential for automatically distinguishing dHGPs from rHGP, and is robust to segmentation and acquisition variations. Pending further optimization, including extension to mixed HGPs, our model may serve as a preoperative addition to postoperative HGP assessment, enabling further exploitation of HGPs as a biomarker.

Download Full-text