MRI Radiomics Model Predicts Pathologic Complete Response of Rectal Cancer Following Chemoradiotherapy

Abstract

GASTROINTESTINAL IMAGINGF or patients with locally advanced rectal cancer (LARC), neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision is currently the standard treatment (1); however, the treatment response varies considerably, from no tumor regression to pathologic complete response (pCR). Following nCRT, approximately 10%-25% of patients achieve pCR, histopathologically showing no residual tumor (2). Considering the surgical complications and excellent long-term outcomes with nCRT alone, a waitand-see strategy with careful regular surveillance could be implemented in a selected group of patients expected to achieve pCR (3-5). Thus, there is an increasing need for the preoperative assessment of pCR to identify candidates for the organ-preserving strategy.Although MRI is the modality of choice for both initial pretreatment evaluation and evaluation of the response to nCRT in patients with rectal cancer, the visual assessment of pCR is still challenging due to the relative scarcity of consensus for an MRI evaluation of the nCRT response (6,7). Consequently, radiomics, providing nonvisual information by extracting many quantitative features from imaging (8,9), has gained popularity to noninvasively predict post-nCRT tumor response in patients with LARC. Previous radiomics models have shown potential results based on pre-and post-nCRT MRI using either T2-weighted imaging (10,11) or multiparametric sequences (12-17). However, these studies had substantial limitations, including an insufficient number of patients for validation and a lack of comparison with MRI interpretation by radiologists. A Background: Preoperative assessment of pathologic complete response (pCR) in locally advanced rectal cancer (LARC) after neoadjuvant chemoradiotherapy (nCRT) is increasingly needed for organ preservation, but large-scale validation of an MRI radiomics model remains lacking.Purpose: To evaluate radiomics models based on T2-weighted imaging and diffusion-weighted MRI for predicting pCR after nCRT in LARC and compare their performance with visual assessment by radiologists. Materials and Methods:This retrospective study included patients with LARC (clinical stage T3 or higher, positive nodal status, or both) who underwent post-nCRT MRI and elective resection between January 2009 and December 2018. Surgical histopathologic analysis was the reference standard for pCR. Radiomic features were extracted from the volume of interest on T2-weighted images and apparent diffusion coefficient (ADC) maps from post-nCRT MRI to generate three models: T2 weighted, ADC, and both T2 weighted and ADC (merged). Radiomics signatures were generated using the least absolute shrinkage and selection operator with tenfold cross-validation. Three experienced radiologists independently rated tumor regression grades at MRI and compared these with the radiomics models' diagnostic outcomes. Areas under the curve (AUCs) of the radiomics models and pooled readers were compared by using the DeLong method.Results: Among 898 patients, 189 (21...

Artificial intelligence with magnetic resonance imaging for prediction of pathological complete response to neoadjuvant chemoradiotherapy in rectal cancer: A systematic review and meta-analysis

“…The 21 included studies were published between 2018 and 2022. More than half of the studies (11/21) were based on a population from China ( 34 – 44 ), three from South Korea ( 45 – 47 ), three from Italy ( 48 – 50 ), two from the USA ( 51 , 52 ), one from Brazil ( 53 ), and one from Belgium ( 54 ). Two studies were prospective, and all the remaining studies (19/21) had a retrospective design.…”

A nomogram for predicting good response after neoadjuvant chemoradiotherapy for locally advanced rectal cancer: a retrospective, double-center, cohort study

“…Previous research mostly defined the study endpoint as pCR after LARC treated with nCRT [9,10], but in practice, there are still difficulties in implementing methods including watch and wait and endoscopic transanalresection, even after achieving clinical complete remission (cCR); most patients still undergo TME surgery. Due to the excellent prognosis, patients defined as having a good response, including pCR and stage yp I patients, are the population with the greatest benefit after nCRT, but this population has rarely been studied, and no reliable predictive models are available.…”

Artificial intelligence in cancer imaging for diagnosis and evaluation of treatment response

“…Distinguishing minimal residual tumors after neoadjuvant chemoradiotherapy and predicting treatment effectiveness remain challenges in CRC [114]. Kim et al [115], in a study aimed at identifying pathological complete response (pCR, defined as ypT0), have shown that a radiomics model based on T2 WI performs better than radiologists (AUC, 0.82 vs. 0.74; sensitivity, 80.0% vs. 15.6%). Several studies [15,116,117] have developed radiomic models to predict pCR (defined as ypT0N0), with AUCs of 0.812-0.89 [116,117].…”