Faculty Bibliography

PURPOSE:Natural language processing (NLP) applied to radiology reports can help identify clinically relevant M1 subcategories of patients with colorectal cancer (CRC). The primary purpose was to compare the overall survival (OS) of CRC according to American Joint Committee on Cancer TNM staging and explore an alternative classification. The secondary objective was to estimate the frequency of metastasis for each organ. METHODS:Retrospective study of CRC who underwent computed tomography (CT) chest, abdomen, and pelvis between July 1, 2009, and March 26, 2019, at a tertiary cancer center, previously labeled for the presence or absence of metastasis by an NLP prediction model. Patients were classified in M0, M1a, M1b, and M1c (American Joint Committee on Cancer), or an alternative classification on the basis of the metastasis organ number: M1, single; M2, two; M3, three or more organs. Cox regression models were used to estimate hazard ratios; Kaplan-Meier curves were used to visualize survival curves using the two M1 subclassifications. RESULTS:Nine thousand nine hundred twenty-eight patients with a total of 48,408 CT chest, abdomen, and pelvis reports were included. On the basis of NLP prediction, the median OS of M1a, M1b, and M1c was 4.47, 1.72, and 1.52 years, respectively. The median OS of M1, M2, and M3 was 4.24, 2.05, and 1.04 years, respectively. Metastases occurred most often in liver (35.8%), abdominopelvic lymph nodes (32.9%), lungs (29.3%), peritoneum (22.0%), thoracic nodes (19.9%), bones (9.2%), and pelvic organs (7.5%). Spleen and adrenal metastases occurred in < 5%. CONCLUSION:NLP applied to a large radiology report database can identify clinically relevant metastatic phenotypes and be used to investigate new M1 substaging for CRC. Patients with three or more metastatic disease organs have the worst prognosis, with an OS of 1 year.

PURPOSE:To evaluate an MRI-based radiomic texture classifier alone and combined with radiologist qualitative assessment in predicting pathological complete response (pCR) using restaging MRI with internal training and external validation. METHODS:Consecutive patients with locally advanced rectal cancer (LARC) who underwent neoadjuvant therapy followed by total mesorectal excision from March 2012 to February 2016 (Memorial Sloan Kettering Cancer Center/internal dataset, n = 114, 41% female, median age = 55) and July 2014 to October 2015 (Instituto do Câncer do Estado de São Paulo/external dataset, n = 50, 52% female, median age = 64.5) were retrospectively included. Two radiologists (R1, senior; R2, junior) independently evaluated restaging MRI, classifying patients (radiological complete response vs radiological partial response). Model A (n = 33 texture features), model B (n = 91 features including texture, shape, and edge features), and two combination models (model A + B + R1, model A + B + R2) were constructed. Pathology served as the reference standard for neoadjuvant treatment response. Comparison of the classifiers' AUCs on the external set was done using DeLong's test. RESULTS:Models A and B had similar discriminative ability (P = 0.3; Model B AUC = 83%, 95% CI 70%-97%). Combined models increased inter-reader agreement compared with radiologist-only interpretation (κ = 0.82, 95% CI 0.70-0.89 vs k = 0.25, 95% CI 0.11-0.61). The combined model slightly increased junior radiologist specificity, positive predictive value, and negative predictive values (93% vs 90%, 57% vs 50%, and 91% vs 90%, respectively). CONCLUSION:We developed and externally validated a combined model using radiomics and radiologist qualitative assessment, which improved inter-reader agreement and slightly increased the diagnostic performance of the junior radiologist in predicting pCR after neoadjuvant treatment in patients with LARC.

Accurate evaluation of tumor response to treatment is critical to allow personalized treatment regimens according to the predicted response and to support clinical trials investigating new therapeutic agents by providing them with an accurate response indicator. Recent advances in medical imaging, computer hardware, and machine-learning algorithms have resulted in the increased use of these tools in the field of medicine as a whole and specifically in cancer imaging for detection and characterization of malignant lesions, prognosis, and assessment of treatment response. Among the currently available imaging techniques, magnetic resonance imaging (MRI) plays an important role in the evaluation of treatment assessment of many cancers, given its superior soft-tissue contrast and its ability to allow multiplanar imaging and functional evaluation. In recent years, deep learning (DL) has become an active area of research, paving the way for computer-assisted clinical and radiological decision support. DL can uncover associations between imaging features that cannot be visually identified by the naked eye and pertinent clinical outcomes. The aim of this review is to highlight the use of DL in the evaluation of tumor response assessed on MRI. In this review, we will first provide an overview of common DL architectures used in medical imaging research in general. Then, we will review the studies to date that have applied DL to magnetic resonance imaging for the task of treatment response assessment. Finally, we will discuss the challenges and opportunities of using DL within the clinical workflow.

CT-based body composition measures are well established in research settings as prognostic markers in oncologic patients. Numerous retrospective studies have shown the role of objective measurements extracted from abdominal CT images of skeletal muscle, abdominal fat, and bone mineral density in providing more accurate assessments of frailty and cancer cachexia in comparison with traditional clinical methods. Quantitative CT-based measurements of liver fat and aortic atherosclerotic calcification have received relatively less attention in cancer care but also provide prognostic information. Patients with cancer routinely undergo serial CT examinations for staging, treatment response, and surveillance, providing the opportunity for quantitative body composition assessment to be performed as part of routine clinical care. The emergence of fully automated artificial intelligence-based segmentation and quantification tools to replace earlier time-consuming manual and semiautomated methods for body composition analysis will allow these opportunistic measures to transition from the research realm to clinical practice. With continued investigation, the measurements may ultimately be applied to achieve more precise risk stratification as a component of personalized oncologic care.

We map single energy CT (SECT) scans to synthetic dual-energy CT (synth-DECT) material density iodine (MDI) scans using deep learning (DL) and demonstrate their value for liver segmentation. A 2D pix2pix (P2P) network was trained on 100 abdominal DECT scans to infer synth-DECT MDI scans from SECT scans. The source and target domain were paired with DECT monochromatic 70 keV and MDI scans. The trained P2P algorithm then transformed 140 public SECT scans to synth-DECT scans. We split 131 scans into 60% train, 20% tune, and 20% held-out test to train four existing liver segmentation frameworks. The remaining nine low-dose SECT scans tested system generalization. Segmentation accuracy was measured with the dice coefficient (DSC). The DSC per slice was computed to identify sources of error. With synth-DECT (and SECT) scans, an average DSC score of 0.93±0.06 (0.89±0.01) and 0.89±0.01 (0.81±0.02) was achieved on the held-out and generalization test sets. Synth-DECT-trained systems required less data to perform as well as SECT-trained systems. Low DSC scores were primarily observed around the scan margin or due to non-liver tissue or distortions within ground-truth annotations. In general, training with synth-DECT scans resulted in improved segmentation performance with less data.

Pancreatic neuroendocrine neoplasms (PaNENs) are a unique group of pancreatic neoplasms with a wide range of clinical presentations and behaviors. Given their heterogeneous appearance and increasing detection on cross-sectional imaging, it is essential that radiologists understand the variable presentation and distinctions PaNENs display compared to other pancreatic neoplasms. Additionally, some of these neoplasms may be hormonally functional, and it is imperative that radiologists be aware of the common clinical presentations of hormonally active PaNENs. Knowledge of PaNEN pathology and treatments may influence which imaging modality is optimal for each patient. Each imaging modality used for PaNENs has distinct advantages and disadvantages, particularly in different treatment settings. Thus, the focus of this manuscript is to provide an update for the radiologist on PaNEN pathology, imaging, and treatments.

Rectal cancer is a relatively common malignancy in the United States. Magnetic resonance imaging (MRI) of rectal cancer has evolved tremendously in recent years, and has become a key component of baseline staging and treatment planning. In addition to assessing the primary tumor and locoregional lymph nodes, rectal MRI can be used to help with risk stratification by identifying high-risk features such as extramural vascular invasion and can assess treatment response for patients receiving neoadjuvant therapy. As the practice of rectal MRI continues to expand further into academic centers and private practices, standard MRI protocols, and reporting are critical. In addition, it is imperative that the radiologists reading these cases work closely with surgeons, medical oncologists, radiation oncologists, and pathologists to ensure we are providing the best possible care to patients. This review aims to provide a broad overview of the role of MRI for rectal cancer.

PURPOSE:To investigate practice patterns related to MR technique and structured reporting for MRI of rectal cancer at academic centers and private practice groups in the United States. METHODS AND MATERIALS:A survey developed by active members of the Society of Abdominal Radiology Rectal and Anal Cancer Disease Focus Panel was sent to 100 private practice and 189 academic radiology groups. The survey asked targeted questions about practice demographics and utilization, technical MR parameters and reporting practices related to MRI of rectal cancer. The results were analyzed using software in an online survey program. RESULTS:or greater (19/43, 44.2%); the rest were unsure. A substantial portion of respondents do not use intravenous contrast (13/47, 27.7%). Most believe that structured report templates contribute to uniformity of reporting practices (39/47, 83.0%), though there is considerable heterogeneity in usage and included elements. CONCLUSION:There is considerable technical heterogeneity among respondents' answers and reporting practices in MR for rectal cancer, and most of the groups report reading only a modest number of studies per week. Our findings suggest there may be room for improvement in terms of radiologist education for performance and standardization of clinical practice for MR imaging of rectal cancer.

OBJECTIVE:To interrogate the mesorectal fat using MRI radiomics feature analysis in order to predict clinical outcomes in patients with locally advanced rectal cancer. METHODS:This retrospective study included patients who underwent neoadjuvant chemoradiotherapy for locally advanced rectal cancer from 2009 to 2015. Three radiologists independently segmented mesorectal fat on baseline T2-weighted axial MRI. Radiomics features were extracted from segmented volumes and calculated using CERR software, with adaptive synthetic sampling being employed to combat large class imbalances. Outcome variables included pathologic complete response (pCR), local recurrence, distant recurrence, clinical T-category (cT), post-treatment T category (ypT), and post-treatment N category (ypN). A maximum of eight most important features were selected for model development using support vector machines and fivefold cross-validation to predict each outcome parameter via elastic net regularization. Diagnostic metrics of the final models were calculated, including sensitivity, specificity, PPV, NPV, accuracy, and AUC. RESULTS:The study included 236 patients (54 ± 12 years, 135 men). The AUC, sensitivity, specificity, PPV, NPV, and accuracy for each clinical outcome were as follows: for pCR, 0.89, 78.0%, 85.1%, 52.5%, 94.9%, 83.9%; for local recurrence, 0.79, 68.3%, 80.7%, 46.7%, 91.2%, 78.3%; for distant recurrence, 0.87, 80.0%, 88.4%, 58.3%, 95.6%, 87.0%; for cT, 0.80, 85.8%, 56.5%, 89.1%, 49.1%, 80.1%; for ypN, 0.74, 65.0%, 80.1%, 52.7%, 87.0%, 76.3%; and for ypT, 0.86, 81.3%, 84.2%, 96.4%, 46.4%, 81.8%. CONCLUSION/CONCLUSIONS:Radiomics features of mesorectal fat can predict pathological complete response and local and distant recurrence, as well as post-treatment T and N categories. KEY POINTS/CONCLUSIONS:• Mesorectal fat contains important prognostic information in patients with locally advanced rectal cancer (LARC). • Radiomics features of mesorectal fat were significantly different between those who achieved complete vs incomplete pathologic response (accuracy 83.9%, 95% CI: 78.6-88.4%). • Radiomics features of mesorectal fat were significantly different between those who did vs did not develop local or distant recurrence (accuracy 78.3%, 95% CI: 72.0-83.7% and 87.0%, 95% CI: 81.6-91.2% respectively).

OBJECTIVES/OBJECTIVE:Describe the cumulative incidence (CUIN) of peritoneal carcinomatosis (PC) and survival in patients presenting with advanced rectal cancer at staging pelvic MRI. METHODS:From 2013 to 2018, clinicopathologic records of patients with pretreatment rectal MRI clinical (c)T3c, cT3d, cT4a, and cT4b primary rectal adenocarcinoma were retrospectively reviewed by two radiologists. Standard MRI descriptors and pathologic stages were recorded. Recurrence-free (RFS) and overall survival (OS) were estimated using the Kaplan-Meier method. Development of PC was explored using competing risk analysis. Differences in survival were compared using the log-rank test. Gray's test was used to test for differences in CUIN of PC. RESULTS:Three hundred forty-three patients (147 women; median age, 56 years) had MRI stages cT3cd, n = 170; cT4a, n = 40; and cT4b, n = 133. Median follow-up among survivors was 27 months (0.36-70 months). For M1 patients, OS differed only by cT stage (2-year OS: cT3 88.1%, cT4a 79.1%, cT4b 64.7%, p = 0.045). For M0 patients, OS and RFS differed only by pathological (p)T stage. We observed a statistically significant difference in the cumulative incidence of PC by cT stage (2-year CUIN: cT3 3.2%, cT4a 8.5%, cT4b 1.6%, p = 0.01), but not by pT stage. Seventy-nine patients (23%) presented with metastatic disease (M1), eight with PC (2.3%). Overall, eight patients presented with PC (cT4a: n = 4, other stages: n = 4) and 22 developed PC (cT4a: n = 5, other stages: n = 17). CONCLUSIONS:PC is uncommon in rectal cancer. MRI-based T stage exhibited an overall association with the cumulative incidence of PC, and descriptively, cT4a stage appears to have the highest CUIN. KEY POINTS/CONCLUSIONS:• In a retrospective study of 343 patients with rectal cancer undergoing baseline MRI and clinical follow-up, we found that peritoneal carcinomatosis was rare. • We observed a significant overall association between PC at presentation and cT stage that appeared to be driven by the higher proportion of cT4a patients presenting with PC. • Among patients that did not present with PC, we observed a significant overall association between time to PC and cT stage that may be driven by the higher cumulative incidence of PC in cT4a patients.