GRASPNET: Fast spatiotemporal deep learning reconstruction of golden-angle radial data for free-breathing dynamic contrast-enhanced magnetic resonance imaging.

The purpose of the current study was to develop a deep learning technique called Golden-angle RAdial Sparse Parallel Network (GRASPnet) for fast reconstruction of dynamic contrast-enhanced 4D MRI acquired with golden-angle radial k-space trajectories. GRASPnet operates in the image-time space and does not use explicit data consistency to minimize the reconstruction time. Three different network architectures were developed: (1) GRASPnet-2D: 2D convolutional kernels (x,y) and coil and contrast dimensions collapsed into a single combined dimension; (2) GRASPnet-3D: 3D kernels (x,y,t); and (3) GRASPnet-2D + time: two 3D kernels to first exploit spatial correlations (x,y,1) followed by temporal correlations (1,1,t). The networks were trained using iterative GRASP reconstruction as the reference. Free-breathing 3D abdominal imaging with contrast injection was performed on 33 patients with liver lesions using a T1-weighted golden-angle stack-of-stars pulse sequence. Ten datasets were used for testing. The three GRASPnet architectures were compared with iterative GRASP results using quantitative and qualitative analysis, including impressions from two body radiologists. The three GRASPnet techniques reduced the reconstruction time to about 13 s with similar results with respect to iterative GRASP. Among the GRASPnet techniques, GRASPnet-2D + time compared favorably in the quantitative analysis. Spatiotemporal deep learning enables reconstruction of dynamic 4D contrast-enhanced images in a few seconds, which would facilitate translation to clinical practice of compressed sensing methods that are currently limited by long reconstruction times.

Updates on LI-RADS Treatment Response Criteria for Hepatocellular Carcinoma: Focusing on MRI.

As the incidence of hepatocellular carcinoma (HCC) and subsequent treatments with liver-directed therapies rise, the complexity of assessing lesion response has also increased. The Liver Imaging Reporting and Data Systems (LI-RADS) treatment response algorithm (LI-RADS TRA) was created to standardize the assessment of response after locoregional therapy (LRT) on contrast-enhanced CT or MRI. Originally created based on expert opinion, these guidelines are currently undergoing revision based on emerging evidence. While many studies support the use of LR-TRA for evaluation of HCC response after thermal ablation and intra-arterial embolic therapy, data suggest a need for refinements to improve assessment after radiation therapy. In this manuscript, we review expected MR imaging findings after different forms of LRT, clarify how to apply the current LI-RADS TRA by type of LRT, explore emerging literature on LI-RADS TRA, and highlight future updates to the algorithm. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 2.

Treatment Response and Clinical Outcomes of Well-Differentiated High-Grade Neuroendocrine Tumors to Lutetium-177-DOTATATE.

INTRODUCTION: Lutetium-177 (177Lu)-DOTATATE received FDA approval in 2018 to treat somatostatin receptor-positive gastroenteropancreatic neuroendocrine tumors (NETs). Little data are available on response and outcomes for well-differentiated (WD) high-grade (HG) NETs treated with 177Lu-DOTATATE. MATERIALS AND METHODS: Patients with WD HG NETs treated with 177Lu-DOTATATE at MSK from 2018 to 2020 were identified. Demographics, response (RECIST 1.1), and progression-free survival (PFS) were determined. Next-generation sequencing (NGS) was performed in the archival tumor. RESULTS: Nineteen patients, all with progressive, heavily treated disease, were identified. Sites of tumor origin were: pancreas (74%), small bowel (11%), rectum (11%), and lung (5%); median Ki-67 was 32% (range 22-56). Thirteen patients (68%) completed all four 177Lu-DOTATATE cycles. Best response (N = 18 evaluable) was: 5/18 (28%) partial response, 8/18 (44%) stable disease, and 5/18 (28%) disease progression. Median PFS was 13.1 months (95% CI: 8.7-20.9). Most common treatment-related toxicities were thrombocytopenia (9 patients, 47%; G3/4, 1 patient, 5%), anemia (7 patients, 37%; G3/4, 2 patients, 11%), leukopenia (6 patients, 32%; G3/4, 0 patients), and liver function test elevation (4 patients, 21%; G3/4, 0 patients). NGS results were available from 13/19 tumors (68%). The most observed alterations were in MEN1 (6/13, 46%) and DAXX (4/13, 31%). No RB1 alterations identified. CONCLUSION: We observed a meaningful disease control rate of 72% during treatment of WD HG NETs with 177Lu-DOTATATE. In this heavily pre-treated population, more than half of patients received all four treatment cycles with toxicities largely bone marrow-related. As would be expected in WD NETs, the vast majority had alterations in chromatin remodeling genes and no RB1 alterations.

Observation versus Resection for Small Asymptomatic Pancreatic Neuroendocrine Tumors: A Matched Case-Control Study.

OBJECTIVE: To analyze the natural history of small asymptomatic pancreatic neuroendocrine tumors (PanNET) and to present a matched comparison between groups who underwent either initial observation or resection. Management approach for small PanNET is uncertain. METHODS: Incidentally discovered, sporadic, small (<3 cm), stage I-II PanNET were analyzed retrospectively between 1993 and 2013. Diagnosis was determined either by pathology or imaging characteristics. Intention-to-treat analysis was applied. RESULTS: A total of 464 patients were reviewed. Observation was recommended for 104 patients (observation group), and these patients were matched to 77 patients in the resection group based on tumor size at initial imaging. The observation group was significantly older (median 63 vs. 59 years, p = 0.04) and tended towards shorter follow-up (44 vs. 57 months, p = 0.06). Within the observation group, 26 of the 104 patients (25 %) underwent subsequent tumor resection after a median observation interval of 30 months (range 7-135). At the time of last follow-up of the observation group, the median tumor size had not changed (1.2 cm, p = 0.7), and no patient had developed evidence of metastases. Within the resection group, low-grade (G1) pathology was recorded in 72 (95 %) tumors and 5 (6 %) developed a recurrence, which occurred after a median of 5.1 (range 2.9-8.1) years. No patient in either group died from disease. Death from other causes occurred in 11 of 181 (6 %) patients. CONCLUSIONS: In this study, no patient who was initially observed developed metastases or died from disease after a median follow-up of 44 months. Observation for stable, small, incidentally discovered PanNET is reasonable in selected patients.

Tumor-associated Neutrophils and Malignant Progression in Intraductal Papillary Mucinous Neoplasms: An Opportunity for Identification of High-risk Disease.

OBJECTIVES: To evaluate the association of tumor-associated neutrophils (TANs) with malignant progression in intraductal papillary mucinous neoplasms (IPMNs) and to study the cyst fluid from these lesions for biomarkers of the inflammation-carcinogenesis association. BACKGROUND: There is a strong link between TANs and malignant progression. Inflammatory mediators released by these cells may be a measurable surrogate marker of this progression. METHODS: We evaluated 78 resected IPMNs (2004-2013). Lesions were divided into the low-risk (low- and intermediate-grade dysplasia: n = 48) and high-risk (high-grade dysplasia and invasive carcinoma: n = 30) groups. TANs were assessed and categorized (negative, low, and high). A multiplexed assay was performed to evaluate 87 different cyst fluid proteins, including cyst fluid inflammatory markers (CFIMs), as possible surrogate markers for parenchymal inflammation. RESULTS: Significant positive correlation between grade of dysplasia and TANs was found. High levels of TANs were identified in 2%, 33%, and 89% of the lesions when stratified by grade of dysplasia into low/intermediate-grade dysplasia, high-grade dysplasia, and invasive carcinoma, respectively (P < 0.001). Higher grades of dysplasia were also found to have positive correlation with 29 of the measured proteins, of which 23 (79%) were CFIMs. Higher levels of TANs correlated with higher levels of 18 CFIMs, of which 16 (89%) were also found to be associated with higher grades of dysplasia. CONCLUSIONS: In this study, TANs were strongly associated with malignant progression in IPMNs. Measurement of CFIMs may be a surrogate marker for IPMN progression and allow for the identification of high-risk disease.

Clinical features and outcome of primary pancreatic lymphoma.

INTRODUCTION: Primary pancreatic lymphoma (PPL) is a rare tumor that is often misdiagnosed. Clinicopathologic features, optimal therapy, and outcomes are not well defined. We reviewed our institutional experience with PPL. METHODS: Search of our institutional database identified that between 1987-2012, 21,760 patients with lymphoma and 11,286 patients with a primary pancreatic tumor were evaluated. There were 44 patients with pathologically confirmed PPL. Clinical data were obtained by chart review and survival distributions were estimated using the Kaplan-Meier method and compared using the log-rank test. RESULTS: At baseline, LDH was elevated in 55 % of the patients, CA 19-9 in 25 %, and CEA in 20 %. Imaging characteristics included large, unresectable tumors (67 %), and lymphadenopathy inferior to the renal vein (50 %). Twenty-three patients underwent surgery for resection (5), diagnosis (13), or palliation (5). Chemotherapy alone achieved a 75 % complete response rate. Eight patients experienced relapse, 88 % of which occurred at distant sites. Median overall survival was 6.1 years and 10-year disease-specific survival (DSS) was 69 %. Patients with a low risk International Prognostic Index (IPI) and those with a follicular histologic subtype demonstrated 5-year DSS of 100 %. CONCLUSIONS: Chemotherapy for PPL results in a high complete response rate and long DSS, which is similar to nodal non-Hodgkin's lymphoma (NHL). A favorable outcome is expected for IPI low risk patients and follicular histologic subtype. Systemic therapy should generally be the initial therapy when the diagnosis is known. Prolonged follow up is recommended to detect relapses. Surgery alone should be reserved for non-curative intent (i.e. diagnostic or palliative).

FOLFIRINOX Induction Therapy for Stage 3 Pancreatic Adenocarcinoma.

BACKGROUND: Reports show that FOLFIRINOX therapy for pancreatic ductal adenocarcinoma (PDAC) results in objective response rates two to threefold higher than those of other regimens. This study aimed to assess response and resection rates for locally unresectable (stage 3) patients initially treated with induction FOLFIRINOX. METHODS: The institutional cancer database was queried for patients treated with induction FOLFIRINOX therapy between 2010 and 2013. Patients were included in the study if they were treated at the authors' institution for stage 3 PDAC (locally unresectable) that had been adjudicated at a weekly multidisciplinary tumor board. RESULTS: The study identified 101 patients. The median age was 64 years (range 37-81 years), and the median follow-up period was 12 months (range 3-37 months). The patients received a median of six cycles (range 1-20 cycles) of induction FOLFIRINOX. No grade 4 or 5 toxicity was recorded. At the initial restaging (median of 3 months after diagnosis), 23 patients (23 %) had developed distant metastases, 15 patients (15 %) had undergone resection, and 63 patients (63 %) had proceeded to chemoradiation. In the group of 63 patients who had proceeded to chemoradiation (median of 9 months after diagnosis), an additional 16 patients (16 %) had undergone resection, and 5 patients (5 %) had developed metastases. A partial radiographic response was observed in 29 % of all the patients, which was associated with ability to perform resection (p = 0.004). The median overall survival time was 11 months for the group that progressed with FOLFIRINOX and 26 months for the group that did not progress. CONCLUSION: Nearly one third of the patients who had been initially identified as having stage 3 pancreatic carcinoma and had been treated with FOLFIRINOX responded radiographically and underwent tumor resection.

Interreader and inter-test agreement in assessing treatment response following transarterial embolization for hepatocellular carcinoma.

PURPOSE: To evaluate interreader and inter-test agreement in applying size- and necrosis-based response assessment criteria after transarterial embolization (TAE) for hepatocellular carcinoma (HCC), applying two different methods of European Association for the Study of the Liver (EASL) criteria. METHODS: Seventy-four patients (median age, 67 years) from a prospectively accrued study population were included in this retrospective study. Four radiologists independently evaluated CT data at 2-3 (1st follow-up, FU) and 10-12 (2nd FU) weeks after TAE and assessed treatment response using size-based (WHO, RECIST) and necrosis-based (mRECIST, EASL) criteria. Enhancing tissue was bidimensionally measured (EASLmeas) and also visually estimated (EASLest). Interreader and inter-test agreements were assessed using intraclass correlation coefficient (ICC) and κ statistics. RESULTS: Interreader agreement for all response assessment methods ranged from moderate to substantial (κ = 0.578-0.700) at 1st FU and was substantial (κ = 0.716-0.780) at 2nd FU. Inter-test agreement was substantial between WHO and RECIST (κ = 0.610-0.799, 1st FU; κ = 0.655-0.782, 2nd FU) and excellent between EASLmeas and EASLest (κ = 0.899-0.918, 1st FU; κ = 0.843-0.877, 2nd FU). CONCLUSION: Size- and necrosis-based criteria both show moderate to excellent interreader agreement in evaluating treatment response after TAE for HCC. Inter-test agreement regarding EASLmeas and EASLest was excellent, suggesting that either may be used. KEY POINTS: • Applying EASL criteria, visual estimation and bidimensional measurements show comparable interreader agreement. • EASL meas and EASL est show substantial interreader agreement for treatment response in HCC. • Agreement was excellent for EASL meas and EASL est after TAE of HCC. • Visual estimation of enhancement is adequate to assess treatment response of HCC.

Distinct CT imaging features of new liver metastases from primary genitourinary cancers.

PURPOSE: To apply natural language processing (NLP) to a large volume of structured radiology reports in the investigation of CT imaging features of new liver metastases from primary genitourinary cancers. METHODS: In this retrospective study, a previously reported NLP model was applied to consecutive structured CT reports from 2016 to 2022 to predict those patients with primary genitourinary cancer who developed liver metastasis. Pathology or imaging follow-up served as the reference standard for validating NLP predictions. Subsequently, diagnostic CTs of the identified patients were qualitatively assessed by two radiologists, whereby several imaging features of new liver metastasis were assessed. Proportions of the assessed imaging features were compared between primary genitourinary cancers using the Chi-square or Fisher's exact test. RESULTS: In 112 patients (mean age = 72 years; 83 males), the majority of new liver metastases were hypovascular (73.2%), well defined (76.6%), homogenous (66.9%), and without necrotic/cystic component (73.2%). There was a higher proportion of iso- to hyperdense liver metastases for primary kidney cancer vs other primary genitourinary cancers (42.5% in kidney cancer; 2.3% in ureter/bladder cancer, 8% in prostate cancer, and 0% in testicular cancer; p < 0.05) and a higher proportion of new liver metastases with ill-defined margin for primary prostate cancer vs other primary genitourinary cancers (44.0% in prostate cancer, 15.0% in kidney cancer, 18.6% in ureter/bladder cancer, and 25.0% in testicular cancer; p < 0.05). CONCLUSION: New liver metastases from primary genitourinary cancers tend to be hypovascular and show several distinct imaging features between different primary genitourinary cancers.

Phase 1 Dose Escalation Study of SBRT Using 3 Fractions for Locally Advanced Pancreatic Cancer.

PURPOSE: The optimal dose and fractionation of stereotactic body radiation therapy (SBRT) for locally advanced pancreatic cancer (LAPC) have not been defined. Single-fraction SBRT was associated with more gastrointestinal toxicity, so 5-fraction regimens have become more commonly employed. We aimed to determine the safety and maximally tolerated dose of 3-fraction SBRT for LAPC. METHODS AND MATERIALS: Two parallel phase 1 dose escalation trials were conducted from 2016 to 2019 at Memorial Sloan Kettering Cancer Center and University of Colorado. Patients with histologically confirmed LAPC without distant progression after at least 2 months of induction chemotherapy were eligible. Patients received 3-fraction linear accelerator-based SBRT at 3 dose levels, 27, 30, and 33 Gy, following a modified 3+3 design. Dose-limiting toxicity, defined as grade ≥3 gastrointestinal toxicity within 90 days, was scored by National Cancer Institute Common Terminology Criteria for Adverse Events, version 4. The secondary endpoints included cumulative incidence of local failure (LF) and distant metastasis (DM), as well as progression-free and overall survival PFS and OS, respectively, toxicity, and quality of life (QoL) using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (QLQ-C30) and the pancreatic cancer-specific QLQ-PAN26 questionnaire. RESULTS: Twenty-four consecutive patients were enrolled (27 Gy: 9, 30 Gy: 8, 33 Gy: 7). The median (range) age was 67 (52-79) years, and 12 (50%) had a head/uncinate tumor location, with a median tumor size of 3.8 (1.1-11) cm and CA19-9 of 60 (1-4880) U/mL. All received chemotherapy for a median of 4 (1.4-10) months. There were no grade ≥3 toxicities. Two-year rates (95% confidence interval) of LF, DM, PFS, and OS were 31.7% (8.6%-54.8%), 70.2% (49.7%-90.8%), 20.8% (4.6%-37.1%), and 29.2% (11.0%-47.4%), respectively. Three- and 6-month QoL assessment showed no detriment. CONCLUSIONS: For select patients with LAPC, dose escalation to 33 Gy in 3 fractions resulted in no dose-limiting toxicities, no detriments to QoL, and disease outcomes comparable with conventional RT. Further exploration of SBRT schemes to maximize tumor control while enabling efficient integration with systemic therapy is warranted.

Determinants of Survival with Combined HER2 and PD-1 Blockade in Metastatic Esophagogastric Cancer.

PURPOSE: We report updated clinical outcomes from a phase II study of pembrolizumab, trastuzumab, and chemotherapy (PTC) in metastatic esophagogastric cancer in conjunction with outcomes from an independent Memorial Sloan Kettering (MSK) cohort. PATIENTS AND METHODS: The significance of pretreatment 89Zr-trastuzumab PET, plasma circulating tumor DNA (ctDNA) dynamics, and tumor HER2 expression and whole exome sequencing was evaluated to identify prognostic biomarkers and mechanisms of resistance in patients treated on-protocol with PTC. Additional prognostic features were evaluated using a multivariable Cox regression model of trastuzumab-treated MSK patients (n = 226). Single-cell RNA sequencing (scRNA-seq) data from MSK and Samsung were evaluated for mechanisms of therapy resistance. RESULTS: 89Zr-trastuzumab PET, scRNA-seq, and serial ctDNA with CT imaging identified how pre-treatment intrapatient genomic heterogeneity contributes to inferior progression-free survival (PFS). We demonstrated that the presence of intensely avid lesions by 89Zr-trastuzumab PET declines in tumor-matched ctDNA by 3 weeks, and clearance of tumor-matched ctDNA by 9 weeks were minimally invasive biomarkers of durable PFS. Paired pre- and on-treatment scRNA-seq identified rapid clearance of HER2-expressing tumor clones with expansion of clones expressing a transcriptional resistance program, which was associated with MT1H, MT1E, MT2A, and MSMB expression. Among trastuzumab-treated patients at MSK, ERBB2 amplification was associated with improved PFS, while alterations in MYC and CDKN2A/B were associated with inferior PFS. CONCLUSIONS: These findings highlight the clinical relevance of identifying baseline intrapatient heterogeneity and serial ctDNA monitoring of HER2-positive esophagogastric cancer patients to identify early evidence of treatment resistance, which could guide proactive therapy escalation or deescalation.

Incidental liver lesions on baseline breast MRI: Outcomes on subsequent abdominal imaging.

PURPOSE: To examine outcomes of incidental liver lesions on baseline breast magnetic resonance imaging (MRI) that were further evaluated with dedicated abdominal imaging. METHODS: Consecutive breast MRI reports from 2011 to 2016 were retrospectively reviewed to identify incidental liver lesions. Only patients without prior breast MRI, without prior abdominal imaging, and with subsequent abdominal imaging were included. Patient demographics, breast MRI indication, and final liver lesion diagnosis were recorded. RESULTS: Of 131 women (mean age 53.8 years), 94/131 (71.8%) underwent breast MRI for extent of disease evaluation, 25/131 (19.1%) for high-risk screening, 11/131 (8.4%) for implant evaluation, and 1/131 (0.8%) for problem-solving. Of 131 liver lesions (6-80 mm), 117/131 (89.3%) were deemed benign on subsequent abdominal imaging; 10/131 (7.6%) probably benign; and 4/131 lesions (3.1%) were confirmed breast cancer metastases. Metastatic liver lesions identified on breast MRI were more likely for women with a current diagnosis of breast cancer than for women without a current diagnosis of breast cancer: 4.3% vs 0%. Similarly, metastatic liver lesions identified on breast MRI were more likely for those with a higher prognostic stage (2 or 3) vs a lower prognostic stage (0 or 1) or no current breast cancer: 11.1% vs 0%. CONCLUSION: Baseline breast MRIs showing incidental liver lesions showed unsuspected liver metastases only in women with a current diagnosis of clinical stage 2 or 3 breast cancer. This suggests breast MRI indication and clinical staging of current breast cancer, if present, can help plan management and decisions to obtain follow-up of liver lesions.

Gender and racial diversity among plenary session speakers at the Society of Abdominal Radiology Annual Meetings: a five-year assessment.

PURPOSE: To evaluate the gender and racial diversity of plenary session speakers in the annual meetings of Society of Abdominal Radiology (SAR) over 2016 to 2020. MATERIALS AND METHODS: The brochures of the SAR annual meetings were reviewed for plenary session speakers and titles. Publicly available institutional profiles and social media were reviewed by the investigator in order to infer gender and race. Gender assessments were men, women, transgender men, transgender women or gender non-binary. Race was classified as White, Black or African American, American Indians and Alaskan Natives, Asian, Native Hawaiian and Pacific Islander and Multiracial. Statistical analysis was performed using chi square and T-tests. RESULTS: Based on self-reported data, the SAR has 64% male and 36% female members. Over 2016-2020, plenary session speakers were more likely to be men [69.6% (183/263)] than women [30.4% (80/263)] (p-value = 0.0007). No speakers could be reliably identified as transgender, gender non-binary or gender expansive. In 2016, there were 24% women plenary speakers. This proportion was 28% in 2017, 33% in 2018 and 36% in 2019, and 30% in 2020. When assessing racial distribution, white speakers accounted for the majority of plenary speakers, ranging from 61 to 78%. Asians speakers accounted for 22 to 35%. There were no Black and African American, American Indian & Alaskan Native, Native Hawaiian & Pacific Islander plenary speakers (0%). Multiracial speakers were represented from 2018 to 2020, accounting for 2-4% speakers (p-value < 0.0001). CONCLUSIONS: Plenary speakers at SAR Annual Meetings from 2016-2020 were more likely to be men, but with the proportion of women presenters increasing over time. White speakers represented the majority of plenary session speakers, followed by Asians. No plenary session speakers were identified as Black or African American or Native Americans.

Risk Factors for Hypervascularization in Hepatobiliary Phase Hypointense Nodules without Arterial Phase Hyperenhancement: A Systematic Review and Meta-analysis.

RATIONALE AND OBJECTIVES: To perform a systematic review and meta-analysis to determine risk factors for hypervascularization in hepatobiliary phase (HBP) hypointense nodules without arterial phase hyperenhancement (APHE) in patients with hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Pubmed and EMBASE databases were searched up to May 7, 2020. Studies which evaluated radiologic and clinical risk factors for hypervascularization in HBP hypointense nodules without APHE were included. Hazard ratios were meta-analytically pooled using random-effects model. Methodological quality of included studies was assessed using Quality in Prognostic Studies (QUIPS) tool. RESULTS: Sixteen studies with 934 patients were included. HBP hypointense nodules without APHE with baseline size greater than 10 mm, T2 hyperintensity, and restricted diffusion showed risk for hypervascularization with pooled HRs of 2.95 (95% confidence interval [CI], 1.94-4.20), 4.21 (95% CI, 1.15-15.40), 5.83 (95% CI, 1.42-23.95), respectively. Previous HCC history contributed to hypervascularization of the nodules with hazard ratio of 2.06 (95% CI, 1.23-3.44). T1 hyperintensity, intralesional fat, Child-Pugh Class B, sex, alfa-fetoprotein, hepatitis B or C infection were not significant risk factors for hypervascularization (p ≥0.05). Study quality was generally moderate. CONCLUSION: HBP hypointense nodules without APHE on gadoxetic acid-enhanced MRI with baseline size greater than 10 mm, T2 hyperintensity, restricted diffusion and previous hepatocellular carcinoma history pose higher risk for hypervascularization. Proper patient management in patients with HBP hypointense nodules without APHE on gadoxetic acid-enhanced MRI may need to be tailored according to these risk factors.

LI-RADS Treatment Response Algorithm: Performance and Diagnostic Accuracy With Radiologic-Pathologic Explant Correlation in Patients With SBRT-Treated Hepatocellular Carcinoma.

PURPOSE: Our purpose was to evaluate the accuracy of LI-RADS Treatment Response Algorithm (LR-TRA) for assessing the viability of hepatocellular carcinoma (HCC) treated with stereotactic body radiation therapy (SBRT), using explant pathology as the gold standard. METHODS AND MATERIALS: This retrospective study included patients who underwent SBRT for locoregional treatment of HCC between 2008 and 2019 with subsequent liver transplantation. Five radiologists independently assessed all treated lesions by using the LR-TRA. Imaging and posttransplant histopathology were compared. Lesions were categorized as either completely (100%) or incompletely (<100%) necrotic, and performance characteristics and predictive values for the LR-TR viable and nonviable categories were calculated for each reader. Interreader reliability was calculated using the Fleiss kappa test. RESULTS: A total of 40 treated lesions in 26 patients (median age, 63 years [interquartile range, 59.4-65.5]; 23 men) were included. For lesions treated with SBRT, sensitivity for incomplete tumor necrosis across readers ranged between 71% and 86%, specificity between 85% and 96%, and positive predictive value between 86% and 92%, when the LR-TR equivocal category was treated as nonviable, accounting for subject clustering. When the LR-TR equivocal category was treated as viable, sensitivity of complete tumor necrosis for lesions treated with SBRT ranged from 88% to 96%, specificity from 71% to 93%, and negative predictive value from 85% to 96%. Interreader reliability was fair (k = 0.22; 95% confidence interval, 0.13-0.33). Although a loss of arterial phase hyperenhancement (APHE) was highly correlated with pathologically nonviable tumor on explant, almost half of the patients with APHE had pathologically nonviable tumor on explant. CONCLUSIONS: LR-TRA v2018 performs well for predicting complete and incomplete necrosis in HCC treated with SBRT. In contrast to other locoregional therapies, the presence of APHE after SBRT does not always indicate viable tumor and suggests that observation may be an appropriate strategy for these patients.

Artificial intelligence in assessment of hepatocellular carcinoma treatment response.

Artificial Intelligence (AI) continues to shape the practice of radiology, with imaging of hepatocellular carcinoma (HCC) being of no exception. This article prepared by members of the LI-RADS Treatment Response (TR LI-RADS) work group and associates, presents recent trends in the utility of AI applications for the volumetric evaluation and assessment of HCC treatment response. Various topics including radiomics, prognostic imaging findings, and locoregional therapy (LRT) specific issues will be discussed in the framework of HCC treatment response classification systems with focus on the Liver Reporting and Data System treatment response algorithm (LI-RADS TRA).

Hepatocellular carcinoma Liver Imaging Reporting and Data Systems treatment response assessment: Lessons learned and future directions.

Hepatocellular carcinoma (HCC) is a leading cause of morbidity and mortality worldwide, with rising clinical and economic burden as incidence increases. There are a multitude of evolving treatment options, including locoregional therapies which can be used alone, in combination with each other, or in combination with systemic therapy. These treatment options have shown to be effective in achieving remission, controlling tumor progression, improving disease free and overall survival in patients who cannot undergo resection and providing a bridge to transplant by debulking tumor burden to downstage patients. Following locoregional therapy (LRT), it is crucial to provide treatment response assessment to guide management and liver transplant candidacy. Therefore, Liver Imaging Reporting and Data Systems (LI-RADS) Treatment Response Algorithm (TRA) was created to provide a standardized assessment of HCC following LRT. LI-RADS TRA provides a step by step approach to evaluate each lesion independently for accurate tumor assessment. In this review, we provide an overview of different locoregional therapies for HCC, describe the expected post treatment imaging appearance following treatment, and review the LI-RADS TRA with guidance for its application in clinical practice. Unique to other publications, we will also review emerging literature supporting the use of LI-RADS for assessment of HCC treatment response after LRT.

Convolutional neural networks: an overview and application in radiology.

Convolutional neural network (CNN), a class of artificial neural networks that has become dominant in various computer vision tasks, is attracting interest across a variety of domains, including radiology. CNN is designed to automatically and adaptively learn spatial hierarchies of features through backpropagation by using multiple building blocks, such as convolution layers, pooling layers, and fully connected layers. This review article offers a perspective on the basic concepts of CNN and its application to various radiological tasks, and discusses its challenges and future directions in the field of radiology. Two challenges in applying CNN to radiological tasks, small dataset and overfitting, will also be covered in this article, as well as techniques to minimize them. Being familiar with the concepts and advantages, as well as limitations, of CNN is essential to leverage its potential in diagnostic radiology, with the goal of augmenting the performance of radiologists and improving patient care. KEY POINTS: • Convolutional neural network is a class of deep learning methods which has become dominant in various computer vision tasks and is attracting interest across a variety of domains, including radiology. • Convolutional neural network is composed of multiple building blocks, such as convolution layers, pooling layers, and fully connected layers, and is designed to automatically and adaptively learn spatial hierarchies of features through a backpropagation algorithm. • Familiarity with the concepts and advantages, as well as limitations, of convolutional neural network is essential to leverage its potential to improve radiologist performance and, eventually, patient care.

Imaging features of hepatocellular carcinoma compared to intrahepatic cholangiocarcinoma and combined tumor on MRI using liver imaging and data system (LI-RADS) version 2014.

PURPOSE: To evaluate the prevalence of major and ancillary imaging features from liver imaging reporting and data systems (LI-RADS) version 2014 and their interreader agreement when comparing hepatocellular carcinoma (HCC) to intrahepatic cholangiocarcinoma (ICC) and combined tumor (cHCC-CC). METHODS: The Institutional Review Board approved this HIPAA-compliant retrospective study and waived the requirement for patients' informed consent. Patients with resected HCC (n = 51), ICC (n = 40), and cHCC-CC (n = 11) and available pre-operative contrast-enhanced MRI were included from 2000 to 2015. Imaging features and final LI-RADS category were evaluated by four radiologists. Imaging features were compared by Fisher's exact test and interreader agreements were assessed by κ statistics. RESULTS: None of the features were unique to either HCC or non-HCC. Imaging features that were significantly more common among HCC compared to ICC and cHCC-CC included washout (76%-78% vs. 10%-35%, p < 0.001), capsule (55%-71% vs. 16%-49%, p < 0.05), and intralesional fat (27%-52% vs. 2%-12%, p < 0.002). Features that were more common among ICC and cHCC-CC included peripheral arterial phase hyperenhancement (40%-64% vs. 10%-14%, p < 0.001) and progressive central enhancement (65%-82% vs. 14%-25%, p < 0.001). The interreader agreement was moderate for each of these imaging features (κ = 0.41-0.55). Moderate agreement was also achieved in the assignment of LR-M (κ = 0.53), with an overall sensitivity and specificity for non-HCC malignancy of 86.3% and 78.4%, respectively. CONCLUSION: HCC and non-HCC show significant differences in the prevalence of imaging features defined by LI-RADS, and are identified by radiologists with moderate interreader agreement. Using LI-RADS, radiologists also achieved moderate interreader agreement in the assignment of the LR-M category.