Significance of Magnetic Resonance Imaging-Assessed Tumor Response for Locally Advanced Rectal Cancer Treated With Preoperative Long-Course Chemoradiation

Evaluation of an Objective MRI-Based Tumor Regression Grade (mrTRG) Score and a Subjective Likert Score for Assessing Treatment Response in Locally Advanced Rectal Cancers-A Retrospective Study

Purpose:  Magnetic resonance imaging (MRI) with the help of MRI-based tumor regression grade (mrTRG) score has been used as a tool to predict pathological tumor regression grade (pTRG) in patients of rectal cancer post-neoadjuvant chemoradiation. Our study aims to evaluate the ability of MRI in assessing treatment response comparing an objective mrTRG score and a subjective Likert score, with a focus on the ability to predict pathologic complete response (pCR). Methods:  Post-treatment MRI studies were retrospectively reviewed for 170 consecutive cases of histopathologically proven rectal cancer after receiving neoadjuvant chemoradiation and prior to surgery by two oncoradiologists blinded to the eventual postoperative histopathology findings. An objective (mrTRG) and a subjective Likert score were assigned to all the cases. Receiver operating characteristic curves were constructed to determine the ability of Likert scale and mrTRG to predict pCR, with postoperative histopathology being the gold standard. The optimal cutoff points on the scale of 1 to 5 were obtained for mrTRG and Likert scale with the greatest sum of sensitivity and specificity using the Youden Index. Results:  The most accurate cutoff point for the mrTRG to predict complete response was 2.5 (using Youden index), with a sensitivity of 69.2%, specificity of 69.6%, positive predictive value (PPV) of 85.6%, negative predictive value (NPV) of 46.4%, and accuracy of 69.3%. The most accurate cutoff for the Likert scale to predict complete response was 3.5, with a sensitivity of 47.5%, specificity of 89.1%, PPV of 91.9%, NPV of 39.4%, and accuracy of 59%. mrTRG had a lower cutoff and was more accurate in predicting pCR compared to Likert score. Conclusion:  An objective mrTRG was more accurate than a subjective Likert scale to predict complete response in our study.

Prospective Correlation of Magnetic Resonance Tumor Regression Grade With Pathologic Outcomes in Total Neoadjuvant Therapy for Rectal Adenocarcinoma

PURPOSE

Total neoadjuvant therapy (TNT) is a newly established standard treatment for rectal adenocarcinoma. Current methods to communicate magnitudes of regression during TNT are subjective and imprecise. Magnetic resonance tumor regression grade (MR-TRG) is an existing, but rarely used, regression grading system. Prospective validation of MR-TRG correlation with pathologic response in patients undergoing TNT is lacking. Utility of adding diffusion-weighted imaging to MR-TRG is also unknown.

METHODS

We conducted a multi-institutional prospective imaging substudy within NRG-GI002 (ClinicalTrials.gov identifier: NCT02921256) examining the ability of MR-based imaging to predict pathologic complete response (pCR) and correlate MR-TRG with the pathologic neoadjuvant response score (NAR). Serial MRIs were needed from 110 patients. Three radiologists independently, then collectively, reviewed each MRI for complete response (mriCR), which was tested for positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity with pCR. MR-TRG was examined for association with the pathologic NAR score. All team members were blinded to pathologic data.

RESULTS

A total of 121 patients from 71 institutions met criteria: 28% were female (n = 34), 84% White (n = 101), and median age was 55 (24-78 years). Kappa scores for T- and N-stage after TNT were 0.38 and 0.88, reflecting fair agreement and near-perfect agreement, respectively. Calling an mriCR resulted in a kappa score of 0.82 after chemotherapy and 0.56 after TNT reflected near-perfect agreement and moderate agreement, respectively. MR-TRG scores were associated with pCR (P < .01) and NAR (P < .0001), PPV for pCR was 40% (95% CI, 26 to 53), and NPV was 84% (95% CI, 75 to 94).

CONCLUSION

MRI alone is a poor tool to distinguish pCR in rectal adenocarcinoma undergoing TNT. However, the MR-TRG score presents a now validated method, correlated with pathologic NAR, which can objectively measure regression magnitude during TNT.

A multiple-time-scale comparative study for the added value of magnetic resonance imaging-based radiomics in predicting pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer

Objective

Radiomics based on magnetic resonance imaging (MRI) shows potential for prediction of therapeutic effect to neoadjuvant chemoradiotherapy (nCRT) in locally advanced rectal cancer (LARC); however, thorough comparison between radiomics and traditional models is deficient. We aimed to construct multiple-time-scale (pretreatment, posttreatment, and combined) radiomic models to predict pathological complete response (pCR) and compare their utility to those of traditional clinical models.

Methods

In this research, 165 LARC patients undergoing nCRT followed by surgery were enrolled retrospectively, which were divided into training and testing sets in the ratio of 7:3. Morphological features on pre- and posttreatment MRI, coupled with clinical data, were evaluated by univariable and multivariable logistic regression analysis for constructing clinical models. Radiomic parameters were derived from pre- and posttreatment T2- and diffusion-weighted images to develop the radiomic signatures. The clinical-radiomics models were then generated. All the models were developed in the training set and then tested in the testing set, the performance of which was assessed using the area under the receiver operating characteristic curve (AUC). Radiomic models were compared with the clinical models with the DeLong test.

Results

One hundred and sixty-five patients (median age, 55 years; age interquartile range, 47-62 years; 116 males) were enrolled in the study. The pretreatment maximum tumor length, posttreatment maximum tumor length, and magnetic resonance tumor regression grade were selected as independent predictors for pCR in the clinical models. In the testing set, the pre- and posttreatment and combined clinical models generated AUCs of 0.625, 0.842, and 0.842 for predicting pCR, respectively. The MRI-based radiomic models performed reasonably well in predicting pCR, but neither the pure radiomic signatures (AUCs, 0.734, 0.817, and 0.801 for the pre- and posttreatment and combined radiomic signatures, respectively) nor the clinical-radiomics models (AUCs, 0.734, 0.860, and 0.801 for the pre- and posttreatment and combined clinical-radiomics models, respectively) showed significant added value compared with the clinical models (all P > 0.05).

Conclusion

The MRI-based radiomic models exhibited no definite added value compared with the clinical models for predicting pCR in LARC. Radiomic models can serve as ancillary tools for tailoring adequate treatment strategies.

MRI restaging of rectal cancer: The RAC (Response-Anal canal-CRM) analysis joint consensus guidelines of the GRERCAR and GRECCAR groups

PURPOSE

To develop guidelines by international experts to standardize data acquisition, image interpretation, and reporting in rectal cancer restaging with magnetic resonance imaging (MRI).

MATERIALS AND METHODS

Evidence-based data and experts' opinions were combined using the RAND-UCLA Appropriateness Method to attain consensus guidelines. Experts provided recommendations for reporting template and protocol for data acquisition were collected; responses were analysed and classified as "RECOMMENDED" versus "NOT RECOMMENDED" (if ≥ 80% consensus among experts) or uncertain (if < 80% consensus among experts).

RESULTS

Consensus regarding patient preparation, MRI sequences, staging and reporting was attained using the RAND-UCLA Appropriateness Method. A consensus was reached for each reporting template item among the experts. Tailored MRI protocol and standardized report were proposed.

CONCLUSION

These consensus recommendations should be used as a guide for rectal cancer restaging with MRI.

Comparison of tumor regression grade and clinical stage based on MRI image as a selection criterion for non-radical management after concurrent chemoradiotherapy in locally advanced rectal cancer: a multicenter, retrospective, cross-sectional study

PURPOSE

There has been no comparative study on the clinical value of magnetic resonance tumor regression grade (mrTRG)1-2 and ycT0-1N0 for the prediction of ypT0-1N0 after concurrent chemoradiotherapy (CCRT) for rectal cancer. We compared the diagnostic performance between mrTRG1-2 and ycT0-1N0 for predicting ypT0-1N0 as a selection criterion for non-radical management after CCRT in locally advanced rectal cancer.

METHODS

This retrospective study enrolled 291 patients from three referral hospitals between January 2018 and March 2020. The diagnostic performance of ycT0-1N0 and mrTRG1-2 for the prediction of ypT0-1N0 was compared in terms of sensitivity, specificity, positive-predictive value, negative-predictive value, and area under the curve (AUC).

RESULTS

Sixty-eight patients (23.4%) achieved ypT0-1N0. Nineteen patients (6.5%) had ycT0-1N0, and 91 patients (31.2%) had mrTRG1-2. For predicting ypT0-1N0, ycT0-1N0 had a sensitivity of 16.2% (95% confidence interval [CI]: 8.36‒27.10) and positive-predictive value of 57.9% (95% CI: 36.57‒76.63), while mrTRG1-2 had a sensitivity of 58.8% (95% CI: 46.23‒70.63) and positive-predictive value of 44.0% (95% CI: 36.46‒51.74). When predicting ypT0-1N0, mrTRG1-2 showed a higher AUC (0.680, 95% CI: 0.604‒0.756) than ycT0-1N0 (0.563, 95% CI: 0.481‒0.645) (P < 0.001).

CONCLUSION

mrTRG1-2 might be a better indicator than ycT0-1N0 for the selection of non-radical management of advanced rectal cancer post-CCRT. However, additional diagnostic tools are required for predicting ypT0-1N0 because mrTRG1-2 or yc stage on MRI has insufficient evidence for diagnosing ypT0-1N0.

Diagnostic performance of MRI and endoscopy for assessing complete response in rectal cancer after neoadjuvant chemoradiotherapy: a systematic review of the literature

BACKGROUND

The diagnostic performance of magnetic resonance imaging (MRI) modalities and/or endoscopy for assessing complete response in rectal cancer after neoadjuvant chemoradiotherapy (nCRT) is unclear.

PURPOSE

To summarize existing evidence on the diagnostic performance of diffusion-weighted MRI, perfusion-weighted MRI, T2-weighted MR tumor regression grade, and/or endoscopy for assessing complete tumor response after nCRT.

MATERIAL AND METHODS

MEDLINE and Embase databases were searched. The PRISMA guidelines were followed. Sensitivity, specificity, negative predictive, and positive predictive values were retrieved from included studies.

RESULTS

In total, 81 studies were eligible for inclusion. Evidence suggests that combined use of MRI and endoscopy tends to improve the diagnostic performance compared to single imaging modality. The positive predictive value of a complete response varies substantially between studies. There is considerable heterogeneity between studies.

CONCLUSION

Combined re-staging tends to improve diagnostic performance compared to single imaging modality, but the vast majority of studies fail to offer true clinical value due to the study heterogeneity.

MRI-based delta-radiomics are predictive of pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer

RATIONALE AND OBJECTIVES

To investigate the capability of delta-radiomics to predict pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer (LARC).

MATERIALS AND METHODS

This retrospective study enrolled 165 consecutive patients with LARC (training set, n = 116; test set, n = 49) who received nCRT before surgery. All patients underwent pre- and post-nCRT MRI examination from which radiomics features were extracted. A delta-radiomics feature was defined as the percentage change in a radiomics feature from pre- to post-nCRT MRI. A data reduction and feature selection process including the least absolute shrinkage and selection operator algorithm was performed for building T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) delta-radiomics signature. Logistic regression was used to build a T2WI and DWI combined radiomics model. Receiver operating characteristic analysis was performed to assess diagnostic performance. Delong method was used to compare the performance of delta-radiomics model with that of magnetic resonance tumor regression grade (mrTRG).

RESULTS

Twenty-seven of 165 patients (16.4%) achieved pCR. T2WI and DWI delta-radiomics signature, and the combined model showed good predictive performance for pCR. The combined model achieved the highest areas under the receiver operating characteristic curves of 0.91 (95% confidence interval: 0.85-0.98) and 0.91 (95% confidence interval: 0.83-0.99) in the training and test sets, respectively (significantly greater than those for mrTRG; training set, p < 0.001; test set, p = 0.04).

CONCLUSION

MRI-based delta-radiomics can help predict pCR after nCRT in patients with LARC with better performance than mrTRG.

Re-staging and follow-up of rectal cancer patients with MR imaging when "Watch-and-Wait" is an option: a practical guide

In the past nearly 20 years, organ-sparing when no apparent viable tumour is present after neoadjuvant therapy has taken an increasingly relevant role in the therapeutic management of locally-advanced rectal cancer patients. The decision to include a patient or not in a “Watch-and-Wait” program relies mainly on endoscopic assessment by skilled surgeons, and MR imaging by experienced radiologists. Strict surveillance using the same modalities is required, given the chance of a local regrowth is of approximately 25–30%, almost always surgically salvageable if caught early. Local regrowths occur at the endoluminal aspect of the primary tumour bed in almost 90% of patients, but the rest are deep within it or outside the rectal wall, in which case detection relies solely on MR Imaging. In this educational review, we provide a practical guide for radiologists who are, or intend to be, involved in the re-staging and follow-up of rectal cancer patients in institutions with an established “Watch-and-Wait” program. First, we discuss patient preparation and MR imaging acquisition technique. Second, we focus on the re-staging MR imaging examination and review the imaging findings that allow us to assess response. Third, we focus on follow-up assessments of patients who defer surgery and confer about the early signs that may indicate a sustained/non-sustained complete response, a rectal/extra-rectal regrowth, and the particular prognosis of the “near-complete” responders. Finally, we discuss our proposed report template.

MR tumor regression grade for pathological complete response in rectal cancer post neoadjuvant chemoradiotherapy: a systematic review and meta-analysis for accuracy

OBJECTIVES

To determine the diagnostic accuracy of magnetic resonance tumor regression grade (mrTRG) for pathological complete response (pCR) and its correlation with pathological findings.

METHODS

Original studies that investigated the correlation of mrTRG with pathological tumor regression grade and pathological T stage were identified in MEDLINE and EMBASE up until August 31, 2018, according to PRISMA guidelines. The search terms included colorectal cancer, chemoradiation therapy, magnetic resonance imaging, and response or regression. Meta-analytic summary sensitivity and specificity for pathologic complete response (pCR) and pathologic T1 or lower than T1 stage (≤ypT1) were calculated using a bivariate random-effects model. The sensitivity and specificity were calculated in both mrTRG 1 and mrTRG 1 or 2, respectively.

RESULTS

Six studies with 916 patients were included. The meta-analytic summary sensitivity and specificity of mrTRG 1 for pCR were 32.3% (95% CI, 18.2-50.6%) and 93.5% (95% CI, 91.5-95.1%), while for ≤ypT1 they were 31.8% (95% CI, 16.2-53.0%) and 94.7% (95% CI, 91.9-96.5%). On the contrary, sensitivity and specificity of mrTRG 1 or 2 for pCR were 69.9% (95% CI, 60.2-78.1%) and 62.2% (95% CI, 56.2-67.8%), while those for ≤ypT1 were 71.4% (95% CI, 61.6-79.6%) and 67.7% (95% CI, 59.8-74.7%).

CONCLUSIONS

mrTRG 1 showed high specificity for pCR and ≤ypT1, but suboptimal sensitivity. mrTRG 1 or 2 showed higher sensitivity for pCR and ≤ypT1, but lower specificity. Because of the suboptimal sensitivity of mrTRG 1, it might be limited as a criterion for less aggressive treatment after neoadjuvant chemoradiotherapy.

KEY POINTS

• Magnetic resonance tumor regression grade 1 shows high specificity for pCR and ≤ypT1, but suboptimal sensitivity. • Magnetic resonance tumor regression grade 1 or 2 shows higher sensitivity for pCR and ≤ypT1, but lower specificity than magnetic resonance tumor regression grade 1 alone.

Diagnostic performance of magnetic resonance to assess treatment response after neoadjuvant therapy in patients with locally advanced rectal cancer

PURPOSE

Our study aimed to evaluate the diagnostic performance of rectal magnetic resonance imaging (MRI) for local restaging in patients with non-metastatic locally advanced rectal cancer (LARC) after neoadjuvant chemoradiotherapy (CRT) using surgical histopathology of total mesorectal excision as the reference standard.

METHODS

Ninety-five patients with LARC who underwent rectal MRI after CRT between January 2014 and December 2016 were included. Accuracy, sensitivity, specificity, positive, and negative predictive value for local staging regarding T-stage, N-stage, circumferential resection margin, and MRI tumor regression grade (ymriTRG) were calculated, and inter-test agreements were assessed.

RESULTS

22/95 (23.2%) patients had radiological complete response (rCR), whereas 20/95 (21.1%) had pathological complete response (pCR). Among the patients with pCR, 11/20 (55%) had rCR. Fair agreement was demonstrated between ymriTRG and pathological TRG (ypTRG) (κ = 0.255). The sensitivity and specificity for detection of pCR were 61.1% (95% CI 35.7-82.7) and 89.6% (95% CI 80.6-95.4). For the detection of ypTRG grades 1 and 2, the corresponding values were 67.2% (95% CI 54.3-78.4) and 51.6 (95% CI 33.1-69.8). The accuracy of ymriTRG was 24.2% (95% CI 15.6-32.8). Inter-test agreement in TRG between MRI and pathology was overall fair (κ = 0.255) and slight (κ = 0.179), if TRG 1 + 2.

CONCLUSION

Qualitative assessment on MRI for diagnosing pCR showed moderate sensitivity and high specificity, whereas the diagnosis of TRG had moderate sensitivity and low specificity with slight to fair inter-test agreement when compared with pathological specimens.

Developing a prediction model based on MRI for pathological complete response after neoadjuvant chemoradiotherapy in locally advanced rectal cancer

PURPOSE

The aim of this study was to build an appropriate diagnostic model for predicting pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer (LARC), by combining magnetic resonance imaging (MRI) parameters with clinical factors.

METHODS

Eighty-four patients with LARC who underwent MR examination before and after nCRT were enrolled in this study. MRI parameters including cylindrical approximated tumor volume (CATV) and relative signal intensity of tumor (rT2wSI) were measured; corresponding reduction rates (RR) were calculated; and MR tumor regression grade (mrTRG) and other conventional MRI parameters were assessed. Logistic regression with lasso regularization was performed and the appropriate prediction model for pCR was built up. An external cohort of thirty-six patients was used as the validation group for testing the model. Receiver-operating characteristic (ROC) analysis was used to assess the diagnostic performance.

RESULTS

In the development and the validation group, 17 patients (20.2%) and 11 patients (30.6%), respectively, achieved pCR. Two CATV-related parameters (CATVpost, which is the CATV measured after nCRT and CATVRR), one rT2wSI-related parameter (rT2wSIRR), and mrTRG were the most important parameters for predicting pCR and were retained in the diagnostic model. In the development group, the area under the receiver-operating characteristic curve (AUC) for predicting pCR is 0.88 [95% confidence interval (CI) 0.78-0.97, p < 0.001], with a sensitivity of 82.4% and a specificity of 83.6%. In the validation group, the AUC is 0.84 (95% CI 0.70-0.98, p = 0.001), with a sensitivity of 81.8% and a specificity of 76.0%.

CONCLUSION

A diagnostic model including CATVpost, CATVRR, rT2wSIRR, and mrTRG was useful for predicting pCR after nCRT in patients with LARC and may be used as an effective organ-preservation strategy.

Morphologic predictors of pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer

Purpose

To evaluate the value of morphological parameters that can be obtained conveniently by MRI for predicting pathologically complete response (pCR) in patients with rectal cancer.

Materials and Methods

A cohort of 101 patients was examined using MRI before and after Neoadjuvant chemoradiotherapy (nCRT). Morphological parameters including maximum tumor area (MTA), maximum tumor length (MTL) and maximum tumor thickness (MTT), as well as cylindrical approximated tumor volume (CATV), distance to anal verge (DTA), and the reduction rates were evaluated by two experienced readers independently.

Results

Post-nCRT MTA and MTL, reduction rates and pre-nCRT DTA were proved to be significantly different between pCR and non-pCR with the AUCs of 0.672-0.853. The sensitivity and specificity for assessing pCR were 61.1-89.9% and 59.0-80.7% respectively. No significant correlation between pre-nCRT size measurements and pCR was obtained.

Conclusion

The convenient morphological measurements may be useful for predicting pCR with moderate sensitivity and specificity. Combining these predictors with the aim of building diagnostic model should be explored.