A radiomic biomarker for prognosis of resected colorectal cancer liver metastases generalizes across MRI contrast agents

Introduction

Contrast-enhanced MRI is routinely performed as part of preoperative work-up for patients with Colorectal Cancer Liver Metastases (CRLM). Radiomic biomarkers depicting the characteristics of CRLMs in MRI have been associated with overall survival (OS) of patients, but the reproducibility and clinical applicability of these biomarkers are limited due to the variations in MRI protocols between hospitals.

Methods

In this work, we propose a generalizable radiomic model for predicting OS of CRLM patients who received preoperative chemotherapy and delayed-phase contrast enhanced (DPCE) MRIs prior to hepatic resection. This retrospective two-center study included three DPCE MRI cohorts (n=221) collected between January 2006 and December 2012. A 10-minute delayed Gd-DO3A-butrol enhanced MRI discovery cohort was used to select features based on robustness across contrast agents, correlation with OS and pairwise Pearson correlation, and to train a logistic regression model that predicts 3-year OS.

Results

The model was evaluated on a 10-minute delayed Gd-DO3A-butrol enhanced MRI validation cohort (n=121), a 20-minute delayed Gd-EOB-DTPA (n=72) cohort from the same institute, and a 5-minute delayed Gd-DTPA cohort (n=28) from an independent institute. Two features were selected: minor axis length and dependence variance. The radiomic signature model stratified high-risk and low-risk CRLM groups in the Gd-DO3Abutrol (HR = 6.29, p = .007), Gd-EOB-DTPA (HR = 3.54, p = .003) and Gd-DTPA (HR = 3.16, p = .04) validation cohorts.

Discussion

While most existing MRI findings focus on a specific contrast agent, our study shows the potential of MRI features to be generalizable across main-stream contrast agents at delayed phase.

Challenges and opportunities for artificial intelligence in oncological imaging

Imaging plays a key role in oncology, including the diagnosis and detection of cancer, determining clinical management, assessing treatment response, and complications of treatment or disease. The current use of clinical oncology is predominantly qualitative in nature with some relatively crude size-based measurements of tumours for assessment of disease progression or treatment response; however, it is increasingly understood that there may be significantly more information about oncological disease that can be obtained from imaging that is not currently utilized. Artificial intelligence (AI) has the potential to harness quantitative techniques to improve oncological imaging. These may include improving the efficiency or accuracy of traditional roles of imaging such as diagnosis or detection. These may also include new roles for imaging such as risk-stratifying patients for different types of therapy or determining biological tumour subtypes. This review article outlines several major areas in oncological imaging where there may be opportunities for AI technology. These include (1) screening and detection of cancer, (2) diagnosis and risk stratification, (3) tumour segmentation, (4) precision oncology, and (5) predicting prognosis and assessing treatment response. This review will also address some of the potential barriers to AI research in oncological imaging.

Variables Influencing Radiology Volume Recovery During the Next Phase of the Coronavirus Disease 2019 (COVID-19) Pandemic

The coronavirus disease 2019 (COVID-19) pandemic has reduced radiology volumes across the country as providers have decreased elective care to minimize the spread of infection and free up health care delivery system capacity. After the stay-at-home order was issued in our county, imaging volumes at our institution decreased to approximately 46% of baseline volumes, similar to the experience of other radiology practices. Given the substantial differences in severity and timing of the disease in different geographic regions, estimating resumption of radiology volumes will be one of the next major challenges for radiology practices. We hypothesize that there are six major variables that will likely predict radiology volumes: (1) severity of disease in the local region, including potential subsequent "waves" of infection; (2) lifting of government social distancing restrictions; (3) patient concern regarding risk of leaving home and entering imaging facilities; (4) management of pent-up demand for imaging delayed during the acute phase of the pandemic, including institutional capacity; (5) impact of the economic downturn on health insurance and ability to pay for imaging; and (6) radiology practice profile reflecting amount of elective imaging performed, including type of patients seen by the radiology practice such as emergency, inpatient, outpatient mix and subspecialty types. We encourage radiology practice leaders to use these and other relevant variables to plan for the coming weeks and to work collaboratively with local health system and governmental leaders to help ensure that needed patient care is restored as quickly as the environment will safely permit.

Late Gadolinium Hyperintensity of Suspected Colorectal Liver Metastases on Gadofosveset-Enhanced Magnetic Resonance Imaging: A Predictor of Benignity and a Potential Problem-Solving Tool

PURPOSE

Late gadolinium hyperintensity (LGH) is sometimes seen in colorectal liver metastasis (CRLM) and represents a significant diagnostic pitfall due to overlap with LGH in benign hemangiomas; therefore, the objective of this study was to determine the prevalence of LGH and the ability of LGH to differentiate between CRLM and benign lesions with intravascular (gadofosveset) vs extracellular contrast agents (gadobutrol).

METHODS

Patients with known colorectal cancer and suspected liver lesions were prospectively recruited into this institutional review board-approved, single institution study and received magnetic resonance imaging of the liver with gadofosveset and gadobutrol. The prevalence of LGH for CRLMs and solid benign lesions was determined. Receiver operating characteristics curves were determined for the presence of LGH as a predictor of benignity. The utility of LGH to differentiate between CRLM and solid benign lesions using gadofosveset vs gadobutrol was compared using the generalized estimating equation.

RESULTS

Twenty-five patients with 131 solid focal liver lesions were recruited. The prevalence of LGH of CRLMs was 11.2% (95% confidence interval [CI]: 0.5%-21.8%) with gadofosveset vs 63.7% (95% CI: 45.7%-81.7%) with gadobutrol. The area under the receiver operating characteristic curve for the presence of LGH as a predictor of benignity was 0.86 using gadofosveset vs 0.75 using gadobutrol. Both LGH (P = .003) and the interaction of contrast agent and LGH (P = .003) statistically significantly differentiated CRLM from benign lesions.

CONCLUSION

LGH is more common with extracellular than with intravascular contrast agents and is statistically significantly associated with benign lesions rather than metastases.

Tumor enhancement of colorectal liver metastases on preoperative gadobutrol-enhanced MRI at 5 minutes post-contrast injection is associated with overall survival post-hepatectomy

Previous studies have demonstrated that target tumor enhancement (TTE) of colorectal cancer liver metastases (CRCLM) on 10-minute delayed phase gadobutrol-enhanced MRI is associated with overall survival post-hepatectomy. The purpose of this study was to determine whether TTE of CRCLM on preoperative gadobutrol-enhanced MRI measured at 5-minute delayed phase is also associated with overall survival. We performed a single-institution, REB-approved, retrospective study of 121 patients with CRCLM who had received a clinical gadobutrol-enhanced MRI after treatment with chemotherapy and prior to liver surgery between January 1, 2006 and December 31, 2012. The TTE of the colorectal liver metastases (CRLM) on 5-minute delayed phase was determined. Kaplan-Meier and Cox-regression survival analyses were used in order to determine the association between TTE on 5-minute delayed phase and overall survival, after adjusting for known prognostic variables. TTE of chemotherapy-treated CRLM on gadobutrol-enhanced MRI at 5-minute post-contrast injection is associated with overall survival post-hepatectomy. On Kaplan-Meier survival analysis, there was a significant difference in overall survival between strong and weak TTE groups (log-rank P=0.009) with 74.4% survival at 36 months in the strong TTE group compared to only 44.6% in the weak TTE group. On Cox-regression analysis, the adjusted hazard ratio of death for patients with low TTE was 0.40 (95% CI: 0.18-0.90, P=0.026), after adjusting for known prognostic variables. This study provides preliminary evidence that tumor enhancement of CRLM at 5 minutes post-contrast injection on gadobutrol-enhanced MRI may provide preoperative prognostic information. This may be helpful for risk stratification of patients for surgery.

Delayed tumour enhancement on gadoxetate-enhanced MRI is associated with overall survival in patients with colorectal liver metastases

OBJECTIVES

To determine whether tumour enhancement on preoperative delayed-phase gadoxetate-enhanced MRI can predict long-term survival in patients with colorectal liver metastases (CRCLM) post-hepatectomy.

MATERIALS AND METHODS

Sixty-five patients who received a preoperative gadoxetate-enhanced MRI prior to liver resection for CRCLM from January 1, 2010, to December 31, 2012, were included in this retrospective study. Target tumour enhancement (TuEn) was calculated as the mean percentage increase in SNR from precontrast to 10-min or 20-min delayed phase for up to two target lesions. Per-patient TuEn was stratified into weak and strong enhancement based on the cut-off determined by the Youden Index for 3-year survival. Kaplan-Meier and Cox regression analyses were used to determine whether tumour enhancement could predict overall survival independent of potential confounders (clinical risk score).

RESULTS

The proportion surviving at 3 years was 85.1% in patients with strong TuEn at 10 min vs. 56.5% in those with weak TuEn at 10 min (p = 0.001). The proportion surviving at 3 years was 79.4% in patients with strong TuEn at 20 min vs. 58.7% in those with weak TuEn at 20 min (p = 0.011). After adjusting for potential confounders, the hazard ratio of death was 0.24 (p = 0.009) in patients who had weak TuEn at 10 min and 0.32 (p = 0.018) in patients who had weak TuEn at 20 min.

CONCLUSIONS

Strong delayed tumour enhancement seen on gadoxetate-enhanced MRI is associated with overall survival in patients with CRCLM post-hepatectomy and may be useful for preoperative risk stratification.

KEY POINTS

• Delayed tumour enhancement of colorectal liver metastases on gadoxetate-enhanced MRI is associated with survival post-hepatectomy • Delayed tumour enhancement of colorectal liver metastases on gadoxetate-enhanced MRI can be measured at both 10 min and 20 min post-contrast injection.

Late gadolinium enhancement of colorectal liver metastases post-chemotherapy is associated with tumour fibrosis and overall survival post-hepatectomy

Purpose

To determine whether late gadolinium MRI enhancement of colorectal liver metastases (CRCLM) post-chemotherapy is associated with tumour fibrosis and survival post-hepatectomy.

Materials and methods

The institutional review board approved this retrospective cohort study and waived the requirement for informed consent. A cohort of 121 surgical patients who received preoperative MRI after chemotherapy between 2006-2012 was included in this study. Target tumour enhancement (TTE), defined as the mean contrast-to-noise ratio of up to two target lesions on late-phase gadobutrol-enhanced MRI, was determined by two independent raters. The average TTE was correlated with tumour fibrosis on post-hepatectomy specimens using Spearman correlation and with survival post-hepatectomy using Kaplan-Meier and Cox regression. Inter-rater reliability was determined using relative intra-class correlation coefficients.

Results

In the surgical cohort (mean age: 63.0 years; male: 58%), TTE was associated with tumour fibrosis (r = 0.43, p < 0.001). Strong TTE was associated with improved survival compared to weak TTE (3-year survival: 88.4% vs. 58.8%, p = 0.003) with a hazard ratio of 0.32 (95% CI: 0.14-0.75, p = 0.008), after taking into account known prognostic variables. Inter-rater reliability was very good with a relative intraclass correlation of 0.84 (95% CI: 0.77-0.89).

Conclusion

Late gadolinium MRI enhancement of CRCLM post-chemotherapy is associated with tumour fibrosis and survival.

Key Points

• MRI enhancement of colorectal liver metastases is associated with survival post-hepatectomy

• MRI enhancement of chemotherapy-treated colorectal liver metastases correlates with tumour fibrosis

• Measuring late MRI enhancement using target tumour enhancement is reliable

Late stage complicated atheroma in low-grade stenotic carotid disease: MR imaging depiction--prevalence and risk factors

PURPOSE

To determine if complicated plaque can be found by using magnetic resonance (MR) imaging-depicted intraplaque hemorrhage (IPH), even among symptomatic patients with low-grade (≤50%) carotid stenosis.

MATERIALS AND METHODS

The institutional ethics review board approved this retrospective study and waived requirements for written informed consent. Symptomatic patients with bilateral 0%-50% carotid stenosis referred for carotid MR imaging were considered. Risk factors (age, sex, hypertension, diabetes, hyperlipidemia, myocardial infarction, atrial fibrillation, smoking, coronary artery disease, and cerebrovascular disease), medications (antihypertensive drugs, diabetes drugs, statins, and aspirin), and the brain side causing symptoms were recorded. MR-depicted IPH prevalence in the carotid arteries ipsilateral and contralateral to the symptomatic side was compared by using the Fisher exact test. Multivariable regression was used to compare the MR-depicted IPH prevalence, while adjusting for risk factors and medications.

RESULTS

A total of 217 patients (434 carotid arteries) were included. MR-depicted IPH was found in 13% (31 of 233) of carotid arteries ipsilateral and 7% (14 of 201) of arteries contralateral to symptoms (P < .05). Male sex (P < .05) and increasing age (P < .05) were associated with MR-depicted IPH after controlling for risk factors and medications.

CONCLUSION

Complicated carotid atheroma can be found among symptomatic patients with low-grade (≤50%) stenosis, and this is associated with male sex and increasing age. MR-depicted IPH may be useful to stratify risk for patients with low-grade carotid stenosis.