Combining Radiomics and Blood Test Biomarkers to Predict the Response of Locally Advanced Rectal Cancer to Chemoradiation

Abstract

Background/Aim: A noninvasive method for predicting a patient's response to neoadjuvant chemoradiotherapy (nCRT) for locally advanced rectal cancer would be useful because this would help determine the subsequent treatment strategy. Two types of noninvasive biomarkers have previously been studied, based on radiomics and based on blood test parameters. We hypothesized that a combination of both types would provide a better predictive power, and this has not previously been investigated. Patients and Methods: Data from 135 patients with locally advanced rectal cancer who underwent nCRT were retrospectively allocated into training and validation cohorts in a 2:1 ratio. Sixty-five radiomics features were extracted from tumors segmented on T2-weighted magnetic resonance images. An elastic net was applied to generate four models for discerning the patients with good responses to nCRT based on radiomics features (model R), blood biomarkers (model B), both (model RB), and a linear combination of models R and B (model R+B). Results: Among 65 radiomics features, 17 were selected as robust features for model development. The AUC values of model R, model B, model RB, and model R+B achieved 0.751, 0.627, 0.785, and 0.711 in the training cohort (n=90), and 0.705, 0.603, 0.679, and 0.705 in validation cohort (n=45), respectively. In the entire cohort, models RB and R+B demonstrated a significantly better performance than model B but not R. There was no correlation between the scores of models R and B (p=0.76). Radiomics features had a greater influence than blood biomarkers on models RB and R+B. Conclusion: A non-redundancy between radiomics features and blood-based biomarkers was observed. Furthermore, radiomics features are more valuable in terms of predicting response to nCRT. The importance of combining noninvasive biomarkers in future investigations is highlighted. Neoadjuvant chemoradiotherapy (nCRT) is the standard treatment for locally advanced rectal cancer (LARC). The patient's response to nCRT is associated with disease outcome (1) and is often used to determine the subsequent 2955 This article is freely accessible online.

Performance and Dimensionality of Pretreatment MRI Radiomics in Rectal Carcinoma Chemoradiotherapy Prediction

“…This finding aligns with the previous report that radiomics features add value when combined with qualitative MRI features [20] while most studies did not provide such comparison. However, some studies have reported clinicopathological scores outperforming models that incorporate radiomics features [35].…”

Radiogenomic and Deep Learning Network Approaches to Predict KRAS Mutation from Radiotherapy Plan CT

“…In these cases, pretreatment imaging biomarkers may have a role in predicting genomic profile. Some of the published radiomics studies on colorectal cancer mainly focused on chemoradiation response (12)(13)(14) or the related prognostic factors (11), and only a small number of studies performed their investigation on CT imaging modality. Yang et al (9) developed a support vector machine (SVM) model based on 346 radiomic features derived from pretreatment contrast-enhanced CT.…”

“…This method has been evaluated to predict genotype or phenotype in breast cancer (3), renal cell carcinoma (4), glioma (5), and advanced or metastatic solid tumors treated with immunotherapy (6). In colorectal cancer, several studies have investigated the radiogenomic approach with various imaging modalities to predict KRAS mutation (7)(8)(9)(10), prognosis (11), and treatment response (12)(13)(14). Most imaging modalities of those studies were rectal MRI or PET for accurate tumor segmentation.…”

Radiomics prediction of response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer

“…The nomogram demonstrated accurate prediction of a GR and non-GR in both training and validation cohorts with AUC values of 0.970 and 0.949, respectively. In another study conducted in a single center, Jeon et al [86] developed a clinical-radiomics model based on T2WI images and blood biomarkers with an AUC of 0.785, which effectively distinguished between patients who did and did not achieve a GR. Additionally, according to Jeon et al [86], both blood biomarkers and radiomics features provide useful information for prediction, with the latter having a higher relative predictive power.…”