Comparison of the different versions of NCCN guidelines for predicting margin-negative resection of pancreatic cancer in patients undergoing upfront surgery.

OBJECTIVES: The purpose of this study was to compare the different versions of the National Comprehensive Cancer Network (NCCN) guidelines for defining resectability of pancreatic ductal adenocarcinoma (PDAC) in predicting margin-negative (R0) resection, and to assess inter-reader agreement. METHODS: This retrospective study included 283 patients (mean age, 65.1 years ± 9.4 [SD]; 155 men) who underwent upfront pancreatectomy for PDAC between 2017 and 2019. Two radiologists independently determined the resectability on preoperative CT according to the 2017, 2019, and 2020 NCCN guidelines. The sensitivity and specificity for R0 resection were analyzed using a multivariable logistic regression analysis with generalized estimating equations. Inter-reader agreement was assessed using kappa statistics. RESULTS: R0 resection was accomplished in 239 patients (84.5%). The sensitivity and specificity averaged across two readers were, respectively, 76.6% and 29.5% for the 2020 guidelines, 74.1% and 32.9% for the 2019 guidelines, and 72.6% and 34.1% for the 2017 guidelines. Compared with the 2020 guidelines, both 2019 and 2017 guidelines showed significantly lower sensitivity for R0 resection (p ≤ .009). Specificity was significantly higher with the 2017 guidelines (p = .043) than with the 2020 guidelines. Inter-reader agreements for determining the resectability of PDCA were strong (k ≥ 0.83) with all guidelines, being highest with the 2020 guidelines (k = 0.91). CONCLUSION: The 2020 NCCN guidelines showed significantly higher sensitivity for prediction of R0 resection than the 2017 and 2019 guidelines.

Phase II study of nivolumab in patients with genetic alterations in DNA damage repair and response who progressed after standard treatment for metastatic solid cancers (KM-06).

BACKGROUND: Immune-modulating antibodies targeting programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) have demonstrated promising antitumor efficacy in various types of cancers, especially highly mutated ones. Genetic alterations in DNA damage response and repair (DDR) genes can lead to genetic instability, often accompanied by a high tumor mutation burden (TMB). However, few studies have validated the aberration of DDR genes as a predictive biomarker for response to immune-modulating antibodies. METHODS: The KM-06 open-label, multicenter, single-arm, phase II trial evaluated the safety and efficacy of nivolumab in refractory solid cancers with DDR gene mutations assessed by clinically targeted sequencing. Nivolumab (3 mg/kg) was administered every 2 weeks until disease progression, unacceptable toxicity, or for 24 months. The primary endpoint was the objective response rate (ORR) as per RECIST V.1.1 criteria. RESULTS: A total of 48 patients were enrolled in the study (median age 61, 58.3% male). The most common cancer type was colorectal cancer (41.7%), followed by prostate and biliary tract cancer (8.3% each). Eight patients achieved a partial response as their best overall response, resulting in an ORR of 17.8%. The disease control rate was 60.0%. The median progression-free survival was 2.9 months. Treatment-related adverse events of any grade and grade ≥3 occurred in 44 (91.7%) and 4 (8.3%) patients, respectively. Clinically targeted sequencing data inferred both TMB and microsatellite instability (MSI). Using a TMB cut-off of 12 mut/Mb, there were significant differences in overall survival (p=0.00035), progression-free survival (p=0.0061), and the best overall response (p=0.05). In the RNA sequencing analysis, nivolumab responders showed activation of the interleukin signaling pathway. Patients who experienced early progression presented high epithelial-mesenchymal transition signaling pathway activation. The responders exhibited a marked increase in PD-1-/Ki67+CD8 T cells at the early stage of treatment (C3D1) compared with non-responders (p=0.03). CONCLUSIONS: In this phase II trial, nivolumab demonstrated moderate efficacy and manageable toxicity in patients with solid cancer harboring DDR gene mutations. A high TMB (>12 mut/Mb) and MSI score (>2.5) determined through clinically target sequencing presented significant discriminatory power for the nivolumab response. TRIAL REGISTRATION NUMBER: NCT04761744.

Prognostic value of CT-based radiomics in grade 1-2 pancreatic neuroendocrine tumors.

BACKGROUND: Surgically resected grade 1-2 (G1-2) pancreatic neuroendocrine tumors (PanNETs) exhibit diverse clinical outcomes, highlighting the need for reliable prognostic biomarkers. Our study aimed to develop and validate CT-based radiomics model for predicting postsurgical outcome in patients with G1-2 PanNETs, and to compare its performance with the current clinical staging system. METHODS: This multicenter retrospective study included patients who underwent dynamic CT and subsequent curative resection for G1-2 PanNETs. A radiomics-based model (R-score) for predicting recurrence-free survival (RFS) was developed from a development set (441 patients from one institution) using least absolute shrinkage and selection operator-Cox regression analysis. A clinical model (C-model) consisting of age and tumor stage according to the 8th American Joint Committee on Cancer staging system was built, and an integrative model combining the C-model and the R-score (CR-model) was developed using multivariable Cox regression analysis. Using an external test set (159 patients from another institution), the models' performance for predicting RFS and overall survival (OS) was evaluated using Harrell's C-index. The incremental value of adding the R-score to the C-model was evaluated using net reclassification improvement (NRI) and integrated discrimination improvement (IDI). RESULTS: The median follow-up periods were 68.3 and 59.7 months in the development and test sets, respectively. In the development set, 58 patients (13.2%) experienced recurrence and 35 (7.9%) died. In the test set, tumors recurred in 14 patients (8.8%) and 12 (7.5%) died. In the test set, the R-score had a C-index of 0.716 for RFS and 0.674 for OS. Compared with the C-model, the CR-model showed higher C-index (RFS, 0.734 vs. 0.662, p = 0.012; OS, 0.781 vs. 0.675, p = 0.043). CR-model also showed improved classification (NRI, 0.330, p < 0.001) and discrimination (IDI, 0.071, p < 0.001) for prediction of 3-year RFS. CONCLUSIONS: Our CR-model outperformed the current clinical staging system in prediction of the prognosis for G1-2 PanNETs and added incremental value for predicting postoperative recurrence. The CR-model enables precise identification of high-risk patients, guiding personalized treatment planning to improve outcomes in surgically resected grade 1-2 PanNETs.

A nationwide real-world study for evaluation of effectiveness and safety of T-DM1 in patients with HER2-positive metastatic breast cancer in Korea (KCSG BR19-15).

Purpose: This study aimed to investigate clinical practices and factors related to the outcomes of T-DM1 use in patients with HER2-positive metastatic breast cancer (mBC). Methods: We included patients with HER2-positive mBC who received T-DM1 as a palliative therapy between August 2017 and December 2018. The safety and outcomes of T-DM1, including overall response rate (ORR), progression-free survival (PFS), and overall survival (OS), were evaluated. A Cox proportional hazards model was used to estimate the hazard ratio and 95% confidence interval (CI) for mortality or progression to HER2-positive mBC. Results: In total, 824 patients were enrolled during the study period. The mean age of patients was 58 years, and 516 (62.6%) patients relapsed after curative treatment. Excluding a history of endocrine therapy, 341 (41.4%) patients previously received none or first-line chemotherapy, 179 (21.7%) received second-line therapy, and 303 (36.9%) received third-or later-line chemotherapy before T-DM1 therapy. During a median follow-up of 16.8 months, the ORR was 35%, the median PFS was 6.6 months, and the median OS was not reached. The clinical factors associated with the hazard of progression were age (<65 years), poor performance status (⩾2), advanced line of palliative chemotherapy (⩾2), prior pertuzumab use, and treatment duration of palliative trastuzumab (<10 months). Common grade 3-4 adverse events were thrombocytopenia (n = 107, 13.2%), neutropenia (n = 23, 2.8%), anemia (n = 21, 2.6%), and elevated liver enzyme (n = 20, 2.5%). Hypokalemia (⩽3.0 mmol/L) and any-grade bleeding events occurred in 25 (3.1%) and 94 (22.6%) patients, respectively. Conclusion: This is the first nationwide real-world study of T-DM1 use in patients with HER2-positive mBC in Korea. The effectiveness and toxicity profiles of T-DM1 in real-world practice were comparable to those in randomized trials. Moreover, patient factors and previous anti-HER2 therapy could predict the outcomes of T-DM1 therapy.

Suboptimal performance of LI-RADS v2018 on gadoxetic acid-enhanced MRI for detecting hepatocellular carcinoma in liver transplant candidates.

OBJECTIVES: To evaluate the diagnostic performance for hepatocellular carcinoma (HCC) detection of the Liver Imaging Reporting and Data System (LI-RADS) version 2018 on gadoxetic acid-enhanced MRI, comparing liver transplant candidates (LT group) with patients who underwent surgical resection (SR group), and to determine significant clinical factors for diagnostic performance of LI-RADS v2018. METHODS: Patients who underwent gadoxetic acid-enhanced MRI and subsequent SR or LT for HCC were retrospectively included between January 2019 and December 2020. The sensitivity and specificity of LI-RADS LR-5 for HCC were compared between the two groups using generalized estimating equations. The accuracy of patient allocation according to the Milan criteria was calculated for the LT group. Univariable and multivariable logistic regression analyses were performed to determine significant clinical factors associated with the sensitivity of LI-RADS. RESULTS: Of the 281 patients, 237 were assigned to the SR group, and 44 were assigned to the LT group. The LT group showed significantly lower per-patient (48.5% vs. 79.6%, p < .001) and per-lesion sensitivity (31.0% vs. 75.9%, p < .001) than the SR group, whereas no significant difference in both per-patient (100.0% vs. 91.7%, p > .99) and per-lesion specificities (100.0% vs. 94.1%, p > .99). The accuracy of patient allocation was 50.0%. Sensitivity was significantly lower in patients with a smaller lesion size (p < .001), a larger lesion number (p = .002), and a higher Child-Pugh score (p = .009). CONCLUSION: LI-RADS v2018 on gadoxetic acid-enhanced MRI might be insufficient in liver transplant candidates and other diagnostic imaging tests should be considered in patients with these significant clinical factors. CLINICAL RELEVANCE STATEMENT: In liver transplant candidates with a smaller lesion size, a larger lesion number, and a higher Child-Pugh score, imaging tests other than gadoxetic acid-enhanced MRI may be clinically useful to determine the transplant eligibility. KEY POINTS: • The sensitivity of the Liver Imaging Reporting and Data System (LI-RADS) was lower in liver transplant candidates than in those who underwent surgical resection. • With the use of gadoxetic acid-enhanced MRI, the accuracy of patient allocation for liver transplantation on the basis of the Milan criteria was suboptimal. • The sensitivity of LI-RADS v2018 was significantly associated with lesion size, lesion number, and Child-Pugh classification.

LI-RADS threshold growth based on tumor growth rate can improve the diagnosis of hepatocellular carcinoma ≤ 3.0 cm.

OBJECTIVE: Despite the revision of threshold growth (TG) in the Liver Imaging Reporting and Data System (LI-RADS) version 2018, the appropriate time period between the two examinations for TG has not been determined. We compared the accuracy of LI-RADS with TG based on tumor growth rate for the diagnosis of hepatocellular carcinoma (HCC) with that of LI-RADS v2018 based on the original TG. METHODS: Patients who underwent preoperative MRI for focal solid lesions (≤ 3.0 cm) were retrospectively evaluated. Three readers measured the size of each lesion on prior CT/MRI and index MRI, with tumor growth rate defined as the percent change in lesion size per month. In addition to the original TG (≥ 50% size increase within ≤ 6 months), the modified TG based on tumor growth rates ≥ 10%/month (TG-10%), ≥ 20%/month (TG-20%), and ≥ 30%/month (TG-30%) were evaluated. The accuracies of these evaluation methods for LI-RADS category 5 HCC were compared using generalized estimation equations. RESULTS: A total of 508 lesions from 370 patients were evaluated. Compared with LI-RADS v2018 with the original TG, the accuracy of LI-RADS with TG-10% was significantly higher (85.0% vs. 80.7%, p < .001), whereas the accuracies of LI-RADS with TG-20% (81.3% vs. 80.7%, p = .404) and TG-30% (79.3% vs. 80.7%, p = .052) were not significant. The sensitivity of LI-RADS with TG-10% was higher than that of LI-RADS v2018 (79.0% vs. 72.5%, p < .001), whereas their specificities were not significantly different (96.6% vs. 96.6%, p > .999). CONCLUSION: TG-10% improved the sensitivity of LI-RADS by detecting additional hepatocellular carcinomas underestimated due to short-term follow-up. CLINICAL RELEVANCE STATEMENT: Threshold growth based on tumor growth rate can be clinically useful in the diagnosis of hepatocellular carcinoma, by improving the sensitivity of LI-RADS. KEY POINTS: • The diagnostic accuracy of Liver Imaging Reporting and Data System (LI-RADS) v2018 was not significantly affected by the time interval between prior and index assessments of threshold growth. • In the 334 hepatocellular carcinomas, the frequency of threshold growth was significantly higher using tumor growth rate ≥ 10%/month (TG-10%) than original threshold growth (53.3% vs. 18.0%, p < .001). • Compared with LI-RADS v2018 with the original threshold growth, LI-RADS with TG-10% had significantly higher accuracy (85.0% vs. 80.7%, p < .001) and sensitivity (79.0% vs. 72.5%, p < .001) but a similar specificity (96.6% vs. 96.6%, p > .999).

Large Duct Pancreatic Ductal Adenocarcinoma: A Morphological Variant of Pancreatic Ductal Adenocarcinoma With Distinct CT and MRI Characteristics.

OBJECTIVE: To investigate the imaging characteristics of large duct pancreatic ductal adenocarcinoma (LD-PDAC) on computed tomography (CT) and magnetic resonance imaging (MRI). MATERIALS AND METHODS: Thirty-five patients with LD-PDAC (63.2 ± 9.7 years) were retrospectively evaluated. Tumor morphology on CT and MRI (predominantly solid mass vs. solid mass with prominent cysts vs. predominantly cystic mass) was evaluated. Additionally, the visibility, quantity, shape (oval vs. branching vs. irregular), and MRI signal intensity of neoplastic cysts within the LD-PDAC were investigated. The radiological diagnoses rendered for LD-PDAC in radiology reports were reviewed. RESULTS: LD-PDAC was more commonly observed as a solid mass with prominent cysts (45.7% [16/35] on CT and 37.1% [13/35] on MRI) or a predominantly cystic mass (20.0% [7/35] on CT and 40.0% [14/35] on MRI) and less commonly as a predominantly solid mass on CT (34.3% [12/35]) and MRI (22.9% [8/35]). The tumor morphology on imaging was significantly associated with the size of the cancer gland on histopathological examination (P = 0.020 [CT] and 0.013 [MRI]). Neoplastic cysts were visible in 88.6% (31/35) and 91.4% (32/35) of the LD-PDAC cases on CT and MRI, respectively. The cysts appeared as branching (51.6% [16/35] on CT and 59.4% [19/35] on MRI) or oval shapes (45.2% [14/35] on CT and 31.2% [10/35] on MRI) with fluid-like MRI signal intensity. In the radiology reports, 10 LD-PDAC cases (28.6%) were misinterpreted as diseases other than typical PDAC, particularly intraductal papillary mucinous neoplasms. CONCLUSION: LD-PDAC frequently appears as a solid mass with prominent cysts or as a predominantly cystic mass on CT and MRI. Radiologists should be familiar with the imaging features of LD-PDAC to avoid misdiagnosis.

Liver Imaging Reporting and Data System version 2018 for diagnosing hepatocellular carcinoma in alcoholic liver cirrhosis and virus-related cirrhosis.

PURPOSE: We aimed to evaluate and compare the diagnostic performance of Liver Imaging Reporting and Data System (LI-RADS) v2018 for hepatocellular carcinoma (HCC) ≤ 3.0 cm on gadoxetic acid-enhanced MRI according to the etiology of cirrhosis. METHODS: Thirty-eight patients with alcoholic liver cirrhosis (ALC) and 37 with hepatitis C virus-related cirrhosis (HCV) who underwent preoperative MRI and subsequent surgical resection or transplantation were included. For comparison groups, patients with hepatitis B virus-related cirrhosis (HBV) were included by 1:1 matching with HCV and ALC groups according to age, lesion size, and Child-Pugh classification. The imaging characteristics of background liver and focal lesions were analyzed. The diagnostic performance of LI-RADS was compared between HCV and HBV groups, and between ALC and HBV groups. RESULTS: ALC group showed significantly higher frequency of hepatic steatosis (25.8 % vs. 6.1 %, p =.04) and lower frequency of nonperipheral washout on portal venous-phase in HCC (63.2 % vs. 97.1 %, p <.001) compared with HBV group. ALC group showed significantly lower sensitivity than HBV group (52.6 % vs. 88.6 %, p<.001). No significant differences in diagnostic performance were found between HCV and HBV groups. In ALC group, hepatobiliary-phase hypointensity provided significantly higher sensitivity (76.3 % vs. 52.6 %, p =.008). CONCLUSION: The sensitivity of LI-RADS for diagnosing HCC ≤ 3.0 cm was significantly lower in the ALC group than in the HBV group.

A Randomized Phase III Study of Patients With Advanced Gastric Adenocarcinoma Without Progression After Six Cycles of XELOX (Capecitabine Plus Oxaliplatin) Followed by Capecitabine Maintenance or Clinical Observation.

PURPOSE: Oxaliplatin, a component of the capecitabine plus oxaliplatin (XELOX) regimen, has a more favorable toxicity profile than cisplatin in patients with advanced gastric cancer (GC). However, oxaliplatin can induce sensory neuropathy and cumulative, dose-related toxicities. Thus, the capecitabine maintenance regimen may achieve the maximum treatment effect while reducing the cumulative neurotoxicity of oxaliplatin. This study aimed to compare the survival of patients with advanced GC between capecitabine maintenance and observation after 1st line XELOX chemotherapy. MATERIALS AND METHODS: Sixty-three patients treated with six cycles of XELOX for advanced GC in six hospitals of the Catholic University of Korea were randomized 1:1 to receive capecitabine maintenance or observation. The primary endpoint was progression-free survival (PFS), analyzed using a two-sided log-rank test stratified at a 5% significance level. RESULTS: Between 2015 and 2020, 32 and 31 patients were randomized into the maintenance and observation groups, respectively. After randomization, the median number of capecitabine maintenance cycles was 6. The PFS was significantly higher in the maintenance group than the observation group (6.3 vs. 4.1 months, P=0.010). Overall survival was not significantly different between the 2 groups (18.2 vs. 16.5 months, P=0.624). Toxicities, such as hand-foot syndrome, were reported in some maintenance group patients. Maintenance treatment was a significant factor associated with PFS in multivariate analysis (hazard ratio, 0.472; 95% confidence interval, 0.250-0.890; P=0.020). CONCLUSIONS: After 6 cycles of XELOX chemotherapy, capecitabine maintenance significantly prolonged PFS compared with observation, and toxicity was manageable. Maintenance treatment was a significant prognostic factor associated with PFS. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT02289547.

LI-RADS version 2018 for hepatocellular carcinoma < 1.0 cm on gadoxetate disodium-enhanced magnetic resonance imaging.

OBJECTIVES: We aimed to develop and evaluate a modified Liver Imaging Reporting and Data System (LI-RADS) version 2018 using significant ancillary features for diagnosing hepatocellular carcinoma (HCC) < 1.0 cm on gadoxetate disodium-enhanced magnetic resonance imaging (MRI). METHODS: Patients who underwent preoperative gadoxetate disodium-enhanced MRI for focal solid nodules < 2.0 cm within 1 month of MRI between January 2016 and December 2020 were retrospectively analyzed. Major and ancillary features were compared between HCCs of < 1.0 cm and 1.0-1.9 cm using the chi-square test. Significant ancillary features associated with HCC < 1.0 cm were determined by univariable and multivariable logistic regression analysis. The sensitivity and specificity of LR-5 were compared between LI-RADS v2018 and our modified LI-RADS (applying the significant ancillary feature) using generalized estimating equations. RESULTS: Of 796 included nodules, 248 were < 1.0 cm and 548 were 1.0-1.9 cm. HCC < 1.0 cm less frequently showed an enhancing capsule (7.1% vs. 31.1%, p < .001) and threshold growth (0% vs. 8.3%, p = .007) than HCC of 1.0-1.9 cm. Restricted diffusion was the only ancillary feature significant for diagnosing HCC < 1.0 cm (adjusted odds ratio = 11.50, p < .001). In the diagnosis of HCC, our modified LI-RADS using restricted diffusion had significantly higher sensitivity than LI-RADS v2018 (61.8% vs. 53.5%, p < .001), with similar specificity (97.3% vs. 97.8%, p = .157). CONCLUSION: Restricted diffusion was the only significant independent ancillary feature for diagnosing HCC < 1.0 cm. Our modified LI-RADS using restricted diffusion can improve the sensitivity for HCC < 1.0 cm. KEY POINTS: • The imaging features of hepatocellular carcinoma (HCC) < 1.0 cm differed from those of HCC of 1.0-1.9 cm. • Restricted diffusion was the only significant independent ancillary feature for HCC < 1.0 cm. • Modified Liver Imaging Reporting and Data System (LI-RADS) with the addition of restricted diffusion can improve the sensitivity for HCC < 1.0 cm.

Prognostic Potential of Cyclin D1 Expression in Colorectal Cancer.

Cyclin D1 is mainly known as an oncogenic driver in cancers, and the dysregulated cyclin D1/cyclin-dependent kinase (CDK) 4/6 axis is considered an attractive target for cancer therapy. Recent studies have reported that tumors respond to therapeutic interventions targeting altered cyclin D1 expression via application of the CDK4/6 inhibitor. However, the prognostic and therapeutic contributions of cyclin D1 to colorectal cancer (CRC) remain controversial. Herein, we assessed the associations between cyclin D1 expression and clinicopathological factors, including patients' overall survival (OS) and recurrence-free survival (RFS), in 495 surgically resected primary CRCs. We also examined previous studies for cyclin D1 in CRCs. High expressions of cyclin D1 (cyclin D1High) was observed in 389 CRC cases (78.6%). Cyclin D1High consistently predicted better patient OS and RFS in CRCs. Based on multivariate analysis, cyclin D1High and young age of patients remained as independent prognosticators of higher OS rate, whereas cyclin D1High, females, chemotherapy, absence of nodal metastasis, and lower T-category remained as independent prognosticators of better RFS. Cyclin D1 is commonly overexpressed in CRCs, and its expression can be used as a favorable prognostic indicator in patients with CRCs; this may be important for predicting responses to subsequent CDK4/6 inhibitors.

Impacts of Subtype on Clinical Feature and Outcome of Male Breast Cancer: Multicenter Study in Korea (KCSG BR16-09).

PURPOSE: The treatment of male breast cancer (MBC) has been extrapolated from female breast cancer (FBC) because of its rarity despite their different clinicopathologic characteristics. We aimed to investigate the distribution of intrinsic subtypes based on immunohistochemistry, their clinical impact, and treatment pattern in clinical practice through a multicenter study in Korea. MATERIALS AND METHODS: We retrospectively analyzed clinical data of 248 MBC patients from 18 institutions across the country from January 1995 to July 2016. RESULTS: The median age of MBC patients was 63 years (range, 25 to 102 years). Among 148 intrinsic subtype classified patients, 61 (41.2%), 44 (29.7%), 29 (19.5%), and 14 (9.5%) were luminal A, luminal B, human epidermal growth factor receptor 2, and triple-negative breast cancer, respectively. Luminal A subtype showed trends for superior survival compared to other subtypes. Most hormone receptor-positive patients (166 patients, 82.6%) received adjuvant endocrine treatment. Five-year completion of adjuvant endocrine treatment was associated with superior disease-free survival (DFS) in patients classified with an intrinsic subtype (hazard ratio [HR], 0.15; 95% confidence interval [CI], 0.04 to 0.49; p=0.002) and in all patients (HR, 0.16; 95% CI, 0.05 to 0.54; p=0.003). CONCLUSION: Distribution of subtypes of MBC was similar to FBC and luminal type A was most common. Overall survival tended to be improved for luminal A subtype, although there was no statistical significance. Completion of adjuvant endocrine treatment was associated with prolonged DFS in intrinsic subtype classified patients. MBC patients tended to receive less treatment. MBC patients should receive standard treatment according to guidelines as FBC patients.

Prognostic value of tumor-to-parenchymal contrast enhancement ratio on portal venous-phase CT in pancreatic neuroendocrine neoplasms.

OBJECTIVES: We aimed to evaluate the prognostic value of tumor-to-parenchymal contrast enhancement ratio on portal venous-phase CT (CER on PVP) and compare its prognostic performance to prevailing grading and staging systems in pancreatic neuroendocrine neoplasms (PanNENs). METHODS: In this retrospective study, data on 465 patients (development cohort) who underwent upfront curative-intent resection for PanNEN were used to assess the performance of CER on PVP and tumor size measured by CT (CT-Size) in predicting recurrence-free survival (RFS) using Harrell's C-index and to determine their optimal cutoffs to stratify RFS using a multi-way partitioning algorithm. External data on 184 patients (test cohort) were used to validate the performance of CER on PVP in predicting RFS and overall survival (OS) and compare its predictive performance with those of CT-Size, 2019 World Health Organization classification system (WHO), and the 8th American Joint Committee on Cancer staging system (AJCC). RESULTS: In the test cohort, CER on PVP showed C-indexes of 0.83 (95% confidence interval [CI], 0.74-0.91) and 0.84 (95% CI, 0.73-0.95) for predicting RFS and OS, respectively, which were higher than those for the WHO (C-index: 0.73 for RFS [p = .002] and 0.72 for OS [p = .004]) and AJCC (C-index, 0.67 for RFS [p = .002] and 0.58 for OS [p = .002]). CT-Size obtained C-indexes of 0.71 for RFS and 0.61 for OS. CONCLUSIONS: CER on PVP showed superior predictive performance on postoperative survival in PanNEN than current grading and staging systems, indicating its potential as a noninvasive preoperative prognostic tool. KEY POINTS: • In pancreatic neuroendocrine neoplasms, the tumor-to-parenchymal enhancement ratio on portal venous-phase CT (CER on PVP) showed acceptable predictive performance of postoperative outcomes. • CER on PVP showed superior predictive performance of postoperative survival over the current WHO classification and AJCC staging system.

Value of threshold growth as a major diagnostic feature of hepatocellular carcinoma in LI-RADS.

BACKGROUND & AIMS: The Liver Reporting and Data System (LI-RADS) version 2018 simplified the definition of threshold growth to '≥50% size increase in a mass in ≤6 months'. However, the diagnostic value of threshold growth for hepatocellular carcinoma (HCC) remained unclear. We evaluated the value of threshold growth, as defined by LI-RADS v2018, in diagnosing HCCs. METHODS: Patients who underwent preoperative gadoxetate disodium-enhanced MRI because of the presence of LI-RADS category 2, 3, or 4 rather than category 5 on prior CT/MRI between January 2017 and December 2020 were retrospectively evaluated. Pathologic or clinical diagnoses were used as reference standards. Imaging features were evaluated by three readers according to LI-RADS v2018. The frequency and diagnostic odds ratio of threshold growth were calculated. The diagnostic performance of LI-RADS category 5 was separately evaluated when threshold growth was and was not considered a major feature, and results were compared using generalized estimation equations. Subgroups of patients who underwent CT/MRI during the previous 3-6 months were analyzed. RESULTS: Analysis of 340 observations in 243 patients found that the frequency of threshold growth was 18.8% and it gradually increased over time. Threshold growth was significantly associated with HCC (diagnostic odds ratio 5.2; 95% CI 2.1-12.7; p <0.001). Use of threshold growth as a major feature significantly increased sensitivity in both the overall (66.4% vs. 57.3%, p <0.001) and subgroup (73.4% vs. 58.2%, p <0.001) cohorts, but had no effect on specificity in either the overall (97.5% vs. 98.3%, p = 0.319) or subgroup (95.9% vs. 98.0%, p = 0.323) cohorts. CONCLUSION: The revised threshold growth of LI-RADS v2018 was significantly associated with HCC. Use of threshold growth as a major diagnostic feature of HCC can improve the sensitivity of LI-RADS v2018. IMPACT AND IMPLICATIONS: We found that the revised threshold growth in the Liver Imaging Reporting and Data System version 2018 (LI-RADS v2018) was a significant predictor of hepatocellular carcinoma (HCC). The use of threshold growth as a major imaging feature of HCC significantly increased the sensitivity of LI-RADS v2018, especially small HCCs (≤3.0 cm), compared with its non-use. Because these small HCCs are eligible for curative treatments, the additional detection of small HCCs is clinically meaningful.

Deep Learning-based Detection of Solid and Cystic Pancreatic Neoplasms at Contrast-enhanced CT.

Background Deep learning (DL) may facilitate the diagnosis of various pancreatic lesions at imaging. Purpose To develop and validate a DL-based approach for automatic identification of patients with various solid and cystic pancreatic neoplasms at abdominal CT and compare its diagnostic performance with that of radiologists. Materials and Methods In this retrospective study, a three-dimensional nnU-Net-based DL model was trained using the CT data of patients who underwent resection for pancreatic lesions between January 2014 and March 2015 and a subset of patients without pancreatic abnormality who underwent CT in 2014. Performance of the DL-based approach to identify patients with pancreatic lesions was evaluated in a temporally independent cohort (test set 1) and a temporally and spatially independent cohort (test set 2) and was compared with that of two board-certified radiologists. Performance was assessed using receiver operating characteristic analysis. Results The study included 852 patients in the training set (median age, 60 years [range, 19-85 years]; 462 men), 603 patients in test set 1 (median age, 58 years [range, 18-82 years]; 376 men), and 589 patients in test set 2 (median age, 63 years [range, 18-99 years]; 343 men). In test set 1, the DL-based approach had an area under the receiver operating characteristic curve (AUC) of 0.91 (95% CI: 0.89, 0.94) and showed slightly worse performance in test set 2 (AUC, 0.87 [95% CI: 0.84, 0.89]). The DL-based approach showed high sensitivity in identifying patients with solid lesions of any size (98%-100%) or cystic lesions measuring 1.0 cm or larger (92%-93%), which was comparable with the radiologists (95%-100% for solid lesions [P = .51 to P > .99]; 93%-98% for cystic lesions ≥1.0 cm [P = .38 to P > .99]). Conclusion The deep learning-based approach demonstrated high performance in identifying patients with various solid and cystic pancreatic lesions at CT. © RSNA, 2022 Online supplemental material is available for this article.

Preoperative data-based deep learning model for predicting postoperative survival in pancreatic cancer patients.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis even after curative resection. A deep learning-based stratification of postoperative survival in the preoperative setting may aid the treatment decisions for improving prognosis. This study was aimed to develop a deep learning model based on preoperative data for predicting postoperative survival. METHODS: The patients who underwent surgery for PDAC between January 2014 and May 2015. Clinical data-based machine learning models and computed tomography (CT) data-based deep learning models were developed separately, and ensemble learning was utilized to combine two models. The primary outcomes were the prediction of 2-year overall survival (OS) and 1-year recurrence-free survival (RFS). The model's performance was measured by area under the receiver operating curve (AUC) and was compared with that of American Joint Committee on Cancer (AJCC) 8th stage. RESULTS: The median OS and RFS were 23 and 10 months in training dataset (n = 229), and 22 and 11 months in test dataset (n = 53), respectively. The AUC of the ensemble model for predicting 2-year OS and 1-year RFS in the test dataset was 0.76 and 0.74, respectively. The performance of the ensemble model was comparable to that of the AJCC in predicting 2-year OS (AUC, 0.67; P = 0.35) and superior to the AJCC in predicting 1-year RFS (AUC, 0.54; P = 0.049). CONCLUSION: Our ensemble model based on routine preoperative variables showed good performance for predicting prognosis for PDAC patients after surgery.

Phase II study to investigate the efficacy of trastuzumab biosimilar (Herzuma®) plus treatment of physician's choice (TPC) in patients with heavily pretreated HER-2+ metastatic breast cancer (KCSG BR 18-14/KM10B).

We investigated the efficacy and safety of a trastuzumab biosimilar, Herzuma®, in combination with treatment of physician's choice (TPC) in patients with HER2+ metastatic breast cancer (MBC) who had failed at least two HER2 directed chemotherapies.

Simplified LI-RADS for Hepatocellular Carcinoma Diagnosis at Gadoxetic Acid-enhanced MRI.

Background Although various modifications to the Liver Imaging Reporting and Data System (LI-RADS) at gadoxetic acid-enhanced MRI have been suggested, LI-RADS shows suboptimal sensitivity for hepatocellular carcinoma (HCC) and is perceived to be too complex. Purpose To evaluate clinical usefulness of a simplified LI-RADS for diagnosing HCCs of 30 mm or smaller at gadoxetic acid-enhanced MRI. Materials and Methods Patients who underwent gadoxetic acid-enhanced MRI examination and subsequent resection, transplantation, or biopsy for focal solid nodules of 30 mm or smaller between January 2019 and December 2020 at a single tertiary referral institution were retrospectively analyzed. Two strategies for simplified LI-RADS using one size criterion (≥10 mm) were evaluated (strategy A, using classifications for nodules of 10-19 mm for nodules both 10-19 mm and ≥20 mm; strategy B, using classifications for nodules ≥20 mm for nodules both 10-19 mm and ≥20 mm). Multivariable analysis was performed to determine significant ancillary features for HCC. Generalized estimating equations were used to compare diagnostic performance for LR-5 (definite HCC) between LI-RADS version 2018 and simplified LI-RADS. The time required for LI-RADS category assignment was compared between the two systems with use of a paired t test. Results A total of 645 nodules from 510 patients (mean age ± SD, 60 years ± 10; 393 men) were evaluated. Compared with strategy A, strategy B had a higher sensitivity of 74% (347 of 470 nodules [95% CI: 70, 78]) vs 73% (342 of 470 nodules [95% CI: 69, 77]) (P = .02) with the same specificity of 96% (168 of 175 nodules [95% CI: 92, 98]) vs 96% (168 of 175 nodules [95% CI: 92, 98]) (P > .99). In strategy B, transitional phase hypointensity was an independent ancillary feature for HCC (P = .04) in LR-4 of at least 10 mm with arterial phase hyperenhancement and no other major features. In all 645 nodules, simplified LI-RADS with use of both strategy B and transitional phase hypointensity had a higher sensitivity of 82% (387 of 470 nodules [95% CI: 79, 86]) vs 73% (343 of 470 nodules [95% CI: 69, 77]) (P < .001) than LI-RADS version 2018, without lower specificity (94%, 165 of 175 nodules [95% CI: 90, 97] vs 96%, 168 of 175 nodules [95% CI: 92, 98], P = .08). Compared with LI-RADS version 2018, simplified LI-RADS reduced the time for LI-RADS category assignment (44 seconds ± 23 vs 74 seconds ± 22, P < .001). Conclusion A simplified Liver Imaging Reporting and Data System was found to be clinically useful for diagnosing hepatocellular carcinomas of 30 mm or smaller at gadoxetic acid-enhanced MRI. © RSNA, 2022 Online supplemental material is available for this article.

Pancreatic ductal adenocarcinoma with a predominant large duct pattern has better recurrence-free survival than conventional pancreatic ductal adenocarcinoma: a comprehensive histopathological, immunohistochemical, and mutational study.

Large duct pattern of pancreatic ductal adenocarcinomas (PDACs) comprises occasional large cancer glands (>0.5 mm in size), along with conventional smaller cancer glands. They histologically mimic intraductal papillary mucinous neoplasms. However, the clinicopathologic significance of PDACs with predominant large duct pattern (PLDP) has not been systematically evaluated. A total of 41 cases of PDACs with PLDP, which were defined as irregularly-shaped cancer glands >0.5 mm in size occupied >50% of tumor volume, were enrolled and their clinicopathological, immunohistochemical, and targeted exome-wise mutational characteristics were compared with 298 conventional PDACs. PDACs with PLDP had cancers with larger tumor sizes (P = 0.025), which were more frequently well to moderately differentiation (P < 0.001), with less lymphovascular invasion (P = 0.013) and had a higher T category (P = 0.023) than conventional PDACs. Immunohistochemically, PDACs with PLDP showed similar abnormal p53 (61%) and SMAD4 (59%) expression patterns as conventional PDACs. In addition, PDACs with PLDP showed diffuse MUC1 (88%), MUC5AC (100%), MUC6 (66%), and focal MUC2 (20%) expressions. More frequent ROS1 mutations were observed in PDACs with PLDP. PDAC patients with PLDP had a better overall and recurrence-free survival (OS and RFS; median, 42 and 34 months) than that of patients with conventional PDACs (34 and 16 months) as per univariate (P = 0.037 and P = 0.001) and multivariate (P = 0.031 and P = 0.034) analyses. PDACs with PLDP showed mutational patterns similar to those of conventional PDACs. They had unique histologic features and longer OS and RFS compared to those of conventional PDACs. Therefore, PDACs with PLDP could be considered a histologic subtype of PDACs.

Prediction of margin-negative resection of pancreatic ductal adenocarcinoma following neoadjuvant therapy: Diagnostic performance of NCCN criteria for resection vs CT-determined resectability.

BACKGROUND: Accurate assessment of pancreatic ductal adenocarcinoma (PDAC) resectability after neoadjuvant therapy (NAT) is crucial. Recently, the NCCN introduced criteria for resection of PDAC following NAT. METHODS: We analyzed 127 patients who underwent NAT and pancreatectomy for PDAC between January 2010 and March 2020. CT-determined resectability according to the NCCN guideline and CA 19-9 level was evaluated before and after NAT. Diagnostic performance of the NCCN criteria for margin-negative (R0) resection was investigated and compared with CT alone. RESULTS: R0 resection was achieved in 104 (81.9%) patients. After NAT, there were 30 (23.6%) resectable, 90 (70.9%) borderline resectable, and seven (5.5%) locally advanced tumors. Significantly decreased or stable CA 19-9 levels were noted in 114 (89.8%) patients. The sensitivity and specificity of the NCCN criteria were 87.5% (91/104) and 21.7% (5/23), respectively, which were significantly different from CT including only resectable PDAC (26.9% [28/104] and 91.3% [21/23]; P < .001), but less prominently different from CT including resectable and borderline resectable PDAC (95.2% [99/104]; P = .022 and 8.7% [2/23]; P = .375). CONCLUSIONS: The NCCN criteria for resection following NAT showed high sensitivity and low specificity for predicting R0 resection. It had supplementary benefit over CT alone, mainly in preventing underestimation of R0 resection.