Diagnostic Accuracy of Qualitative and Quantitative Computed Tomography Analysis for Diagnosis of Pathological Grade and Stage in Upper Tract Urothelial Cell Carcinoma.

OBJECTIVE: The aim of this study was to compare grade and stage of upper tract urothelial cell carcinoma (UCC) using computed tomography. MATERIALS AND METHODS: With institutional review board approval, 48 patients with 49 UCC (44 high grade and 5 low grade, 26 ≤ T1 and 23 ≥ T2) underwent nephroureterectomy and preoperative computed tomography between 2013 and 2015. Two blinded radiologists assessed for tumor appearance (filling defect/mass or wall thickening/stricture), margin (smooth or spiculated/irregular), texture (homogeneous, heterogeneous), hydronephrosis, and calcification. A third blinded radiologist established consensus. A fourth blinded radiologist measured size and first-order histogram texture features. Comparisons were performed using χ test, multivariable logistic regression, and receiver operator characteristic analysis. RESULTS: There was no difference in size of tumors compared by grade or stage (P = 0.80 and 0.13, respectively).Among subjective variables, only tumor texture was significantly different between low- and high-grade UCC (P = 0.03; κ = 0.45). Tumors characterized as spiculated/irregular margin (P = 0.003; 0.30) and heterogeneous (P < 0.001; κ = 0.45) were associated with T2 disease or higher.Entropy was greater in higher grade (6.23 ± 0.46 vs 5.72 ± 0.28) and T2 disease or higher (6.40 ± 0.33 vs 5.95 ± 0.48), (P = 0.03 and 0.02, respectively) with no differences in Kurtosis or Skewness (P > 0.05). Area under the receiver operator characteristic curve for entropy to diagnose high-grade and T2 tumors or higher was 0.83 (confidence interval, 0.64-1.0) and 0.79 (confidence interval 0.59-0.98), respectively. CONCLUSIONS: Heterogeneity, assessed qualitatively and quantitatively, is accurate for diagnosis of higher grade and stage of disease in upper tract UCC. Spiculated/irregular margins are also associated with T2 disease or higher.

Comparison of Prostate Imaging Reporting and Data System versions 1 and 2 for the Detection of Peripheral Zone Gleason Score 3 + 4 = 7 Cancers.

OBJECTIVE: The objective of our study was to compare Prostate Imaging Reporting and Data System version 1 (PI-RADSv1) and Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) for the detection of peripheral zone (PZ) Gleason score 3 + 4 = 7 cancers. MATERIALS AND METHODS: Forty-seven consecutive patients with 52 PZ Gleason score 3 + 4 = 7 cancers that were 0.5 cm3 or larger underwent radical prostatectomy (RP) and 3-T MRI between 2012 and 2015. Two blinded radiologists (readers 1 and 2) retrospectively assigned PI-RADSv1 sequence (T2-weighted imaging, DWI, dynamic contrast-enhanced MRI [DCE-MRI]) and sum scores and PI-RADSv2 assessment categories. A third blinded radiologist (reader 3) measured apparent diffusion coefficient (ADC) ratio (ADC of tumor / ADC of normal PZ) using RP-MRI maps. Sensitivity, false-positive rate, and overall accuracy were compared using McNemar test. Pearson correlation was performed. RESULTS: Using PI-RADSv1, reader 1 detected 86.5% (45/52) of the cancers and reader 2, 76.9% (40/52) of the cancers. Using PI-RADSv2, reader 1 detected 78.9% (41/52) and reader 2, 67.3% (35/52). Reader 1 detected 7.7% (4/52) and reader 2 detected 9.6% (5/52) more tumors using PI-RADSv1 due to T2-weighted imaging score ≥ 4 or DCE-MRI score ≥ 3. Sensitivity was higher for PI-RADSv1 (p = 0.01 and 0.03, readers 1 and 2). False-positive rates were higher with PI-RADSv1 than with PI-RADSv2 (1.8% vs 0.9% for reader 1; 3.6% vs 1.8% for reader 2) without significant differences in false-positive rate (p = 0.41 and 0.25) or overall accuracy (p = 0.06 and 0.23). PI-RADSv1 sum scores correlated strongly with PI-RADSv2 categories (B = 0.78-0.93, p < 0.0001). The mean ADC ratio was 0.61 ± 0.14 mm2/s with no difference between visible and nonvisible tumors (p = 0.06-0.5). Interobserver agreement was moderate for PI-RADSv2 (κ = 0.41) and ranged from slight to substantial for PI-RADSv1 (T2-weighted imaging, κ = 0.32; DWI, κ = 0.52; DCE-MRI, κ = 0.13). CONCLUSION: There was no difference in overall detection of cancers comparing PI-RADSv1 and PI-RADSv2; however, PI-RADSv1 sequence scores on T2-weighted imaging and DCE-MRI detected approximately 10% more tumors that were otherwise underestimated on DWI and using PI-RADSv2 decision-tree rules.

Evaluation of apparent diffusion coefficient and MR volumetry as independent associative factors for extra-prostatic extension (EPE) in prostatic carcinoma.

BACKGROUND: To assess mean apparent diffusion coefficient (ADC) and MR-derived tumor volume (Vt) as associative factors for extra-prostatic extension (EPE) in prostate cancer (PCa). METHODS: With institutional review board approval, 73 consecutive patients diagnosed with PCa at trans-rectal ultrasound biopsy underwent preoperative multi-parametric (T2W+DWI+DCE) 3 Tesla MRI before radical prostatectomy between 2012 and 2014; 52% (38/73) patients had EPE. Clinical parameters including: age, prostate serum antigen (PSA), digital rectal examination (DRE) and percentage positive cores (PPC) were recorded. Two blinded radiologists subjectively evaluated for EPE using PI-RADS with T2W-MRI. A third blinded radiologist recorded: mean ADC (mm(2) /s) of tumor and tumor volume on ADC and T2W (derived from planar volumetry). VtMAX (the largest volume on ADC or T2W) was documented. Multivariate and receiver operator characteristic analyses were performed. RESULTS: There were no significant differences in age, DRE, or Gleason score between groups (P = 0.52, 0.06, 0.61, 0.36). PSA approached significance being higher with EPE (12.9 ± 12.6 versus 8.2 ± 7.4; P = 0.06). PPC was higher with EPE (60.9 ± 21.9% versus 38.3 ± 21.6%; P < 0.01) with an area under the curve (AUC) of 0.78 and sensitivity/specificity = 75.7/75% when PPC ≥ 45%. AUC for T2W-MRI was 0.46-0.51 with sensitivity/specificity = 40.0-42.9/48.6-57.1% (R1, R2). Inter-observer agreement was fair, k = 0.39. There was no difference in mean ADC between groups (0.89 ± 0.25 versus 0.88 ± 0.19 [EPE] mm(2) /s), P = 0.70. T2W-Vt, ADC-Vt, and VtMAX were larger with EPE (5.1 ± 7.4, 5.8 ± 6.5, 6.3 ± 7.4 cm(3) versus 1.6 ± 1.8, 1.8 ± 1.3, 2.1 ± 1.8), P < 0.01. VtMAX AUC was 0.77 with sensitivity/specificity = 78.4/73.5% when VtMAX ≥ 2.1 cm(3) which outperformed all other parameters (P > 0.05) except PPC (P = 0.6) for the diagnosis EPE. CONCLUSION: MR volumetry and percentage of positive core biopsies are associated with EPE; whereas, in this study, other clinical and MR parameters including mean ADC and subjective T2W-MR analysis were not useful for assessment of EPE.

Angiomyolipoma (AML) without visible fat: Ultrasound, CT and MR imaging features with pathological correlation.

OBJECTIVE: To compare imaging findings with histopathology in AML without visible fat (AMLwvf). MATERIAL AND METHODS: With IRB approval, we identified 18 AMLwvf that underwent CT between 2002-2014. A radiologist measured NECT-attenuation, corticomedullary (CM) and nephrographic (NG) enhancement, echogenicity relative to renal cortex (RC) (N = 5), T2W (T2AML/T2RC) signal-intensity (SI), and chemical-shift SI ([SIIN-PHASE - SIOPPOSED-PHASE]/SIIN-PHASE) indices (N = 6). A pathologist re-evaluated 15/18 AMLwvf for 1) < or > 25% adipocytes/high-power-field (HPF), 2) "many or few" blood vessels. Comparisons were performed using chi-square and independent t-tests. RESULTS: 73.3%(11/15) of AMLwvf had <25% adipocytes/HPF and 86.7%(13/15) had "many" blood vessels. NECT-attenuation was 41.8(±6.9) HU. 61.1 %(11/18) of AMLwvf were hyper-attenuating and 38.9%(7/18) iso-attenuating; attenuation was associated with %-adipocytes/HPF, (p = 0.01). CM/NG enhancement were 63.3(±20.8)/51.7(±15.5) HU. 72.2%(13/18) of AMLwvf had wash-out enhancement, with no association with amount of blood vessels at pathology, (p = 0.68). No difference in echogenicity was noted by histology (p > 0.05). All AMLwvf were T2-hypointense (SI ratio = 0.61 [±0.1]). 2/6 AMLwvf showed SI drop on chemical-shift MRI; both were iso-attenuating and were associated with >25% adipocytes/HPF (p = 0.04). CONCLUSIONS: AMLwvf are typically T2-hypointense and hyper-attenuating with wash-out enhancement due to abundant smooth muscle and vessels respectively. Iso-attenuating AMLwvf with microscopic fat on MRI contain more adipocytes/HPF. KEY POINTS: • Five percent of AML do not demonstrate detectable fat on imaging • These AML are hyperattenuating and T2-hypointense due to abundant smooth muscle • These AML show washout enhancement without association to vessel count at histopathology • Iso-attenuating AML with microscopic fat on MRI show >25% adipocytes/HPF • The term "AML without visible fat" is proposed to reduce ambiguity.

Whole-Tumor Quantitative Apparent Diffusion Coefficient Histogram and Texture Analysis to Predict Gleason Score Upgrading in Intermediate-Risk 3 + 4 = 7 Prostate Cancer.

OBJECTIVE: The objective of our study was to evaluate whole-lesion quantitative apparent diffusion coefficient (ADC) for the prediction of Gleason score (GS) upgrading in 3 + 4 = 7 prostate cancer. MATERIALS AND METHODS: Fifty-four patients with GS 3 + 4 = 7 prostate cancer diagnosed at systematic transrectal ultrasound (TRUS)-guided biopsy underwent 3-T MRI and radical prostatectomy (RP) between 2012 and 2014. A blinded radiologist contoured dominant tumors on ADC maps using histopathologic correlation. The whole-lesion mean ADC, ADC ratio (normalized to peripheral zone), ADC histogram, and texture analysis were compared between tumors with GS upgrading and those without GS upgrading using multivariate ROC analyses and logistic regression modeling. RESULTS: Tumors were upgraded to GS 4 + 3 = 7 after RP in 26% (n = 14) of the 54 patients, and tumors were downgraded after RP in none of the patients. The mean ADC, ADC ratio, 10th-centile ADC, 25th-centile ADC, and 50th-centile ADC were similar between patients with GS 3 + 4 = 7 tumors (0.99 ± 0.22, 0.58 ± 0.15, 0.77 ± 0.31, 0.94 ± 0.28, and 1.15 ± 0.24, respectively) and patients with upgraded GS 4 + 3 = 7 tumors (1.02 ± 0.18, 0.55 ± 0.11, 0.71 ± 0.26, 0.89 ± 0.20, and 1.11 ± 0.16) (p > 0.05). Regression models combining texture features improved the prediction of GS upgrading. The combination of kurtosis, entropy, and skewness yielded an AUC of 0.76 (SE = 0.07) (p < 0.001), a sensitivity of 71%, and a specificity of 73%. The combination of kurtosis, heterogeneity, entropy, and skewness yielded an AUC of 0.77 (SE = 0.07) (p < 0.001), a sensitivity of 71%, and a specificity of 78%. CONCLUSION: In this study, whole-lesion mean ADC, ADC ratio, and ADC histogram analysis were not predictive of pathologic upgrading of GS 3 + 4 = 7 prostate cancer after RP. ADC texture analysis improved accuracy.

Evaluation of T1-Weighted MRI to Detect Intratumoral Hemorrhage Within Papillary Renal Cell Carcinoma as a Feature Differentiating From Angiomyolipoma Without Visible Fat.

OBJECTIVE: The objective of the present study is to determine whether hemorrhage within papillary renal cell carcinoma (RCC) can be detected using T1-weighted MRI and to ascertain whether it can be used to differentiate papillary RCC from angiomyolipoma (AML) without visible fat. MATERIALS AND METHODS: A retrospective case-control study compared 11 AMLs without visible fat with 58 papillary RCCs smaller than 5 cm that were evaluated using MRI between 2003 and 2015. Two blinded radiologists subjectively evaluated MR images to identify the presence of intratumoral hemorrhage on the basis of a decrease in signal intensity (SI) on in-phase, compared with opposed-phase, chemical-shift MRI and also on the basis of the SI of the lesion compared with that of the renal cortex on fat-suppressed T1-weighted MRI. A third radiologist established consensus and measured the ratio of the SI of the lesion to that of the renal cortex (hereafter referred to as the "SI ratio") on T2-weighted MRI; the SI loss index, as calculated using the equation [(SItumorIP - SItumorOP) / SItumorOP] × 100, where IP denotes the in-phase image and OP denotes the opposed-phase image; and the SI ratio on fat-suppressed T1-weighted MRI. Analyses were performed using tests of association and ROCs. RESULTS: When AMLs without visible fat were compared with papillary RCCs, no statistically significant difference in the T2-weighted SI ratio was noted (p = 0.08). Papillary RCCs had a lower mean (± SD) SI loss index (-3.7% ± 17.3%; range, -51.3% to 31.3%) than did AMLs without visible fat (37.8% ± 76.1%; range, -15.6% to 184.4%) (p < 0.001). A mean SI loss index of less than -16% resulted in an AUC of 0.71 (95% CI, 0.52-0.91), with a sensitivity and specificity of 22.8% and 100%, respectively, for the diagnosis of papillary RCC. After consensus review, none of the AMLs without visible fat and 16 of the 58 papillary RCCs (27.6%) were found to have a decrease in SI on subjective analysis (p = 0.06, κ = 0.60). Between groups, no differences were noted in the SI ratio on fat-suppressed T1-weighted MRI (p = 0.58) or in the SI observed on subjective analysis of fat-suppressed T1-weighted MRI (p = 0.20, κ = 0.48). CONCLUSION: The presence of intratumoral hemorrhage within papillary RCC is a specific feature that differentiates papillary RCCs from AMLs without visible fat. Subjective analysis may be more clinically appropriate than chemical-shift MRI because of limitations in the quantitative measurement of T2* signal with the use of chemical-shift MRI.

Nested and microcystic variants of urothelial carcinoma displaying immunohistochemical features of basal-like urothelial cells: an immunohistochemical and histopathogenetic study.

Nested/microcystic (NV/MV) urothelial carcinoma (UC) variants are associated with mild cytologic atypia and commonly present at high-stage disease. The histopathogenesis is investigated using urothelial basal cell markers. Archival 14 NV/MV and three inverted papilloma (IP) were immunostained for CD44, cytokeratin 5 (CK5), CK34bE12 and p63. Twenty consecutive cases of invasive high-grade UC including 14 superficial and 6 muscle-invasive UC cases were used as control. Immunostaining was scored as high for staining of full or more than 50% thickness of the epithelial nest or epithelium and low for lesser immunoreactivity and negative reactivity. All 14 NV/MV, 3 IP and 6 control cases showed a high score of immunoreactivity for CK5, CD44, CK34bE12 and focally for p63. The remaining control cases showed a high score of immunoreactivity for CK34bE12, while negative or low for CK5, CD44 and p63. In conclusion, immunoreactivity CK5 and CD44 commonly immunostained NV/MV and some invasive high grade UC. Other basal cell markers (CK34bE12 and p63) appear to be non specific or non sensitive. NV and MV and some UC likely represent a subset of UC displaying immunohistochemical features of urothelial basal cells. They had tendency of endophytic growth and early invasion despite the innocuous cytologic appearance.

Uncovering the dosage-dependent roles of Arid1a in gastric tumorigenesis for combinatorial drug therapy.

Gastric cancer (GC) is one of the most common deadly cancers in the world. Although patient genomic data have identified AT-rich interaction domain 1A (ARID1A), a key chromatin remodeling complex subunit, as the second most frequently mutated gene after TP53, its in vivo role and relationship to TP53 in gastric tumorigenesis remains unclear. Establishing a novel mouse model that reflects the ARID1A heterozygous mutations found in the majority of human GC cases, we demonstrated that Arid1a heterozygosity facilitates tumor progression through a global loss of enhancers and subsequent suppression of the p53 and apoptosis pathways. Moreover, mouse genetic and single-cell analyses demonstrated that the homozygous deletion of Arid1a confers a competitive disadvantage through the activation of the p53 pathway, highlighting its distinct dosage-dependent roles. Using this unique vulnerability of Arid1a mutated GC cells, our combined treatment with the epigenetic inhibitor, TP064, and the p53 agonist, Nutlin-3, inhibited growth of Arid1a heterozygous tumor organoids, providing a novel therapeutic option for GC.

An Algorithmic Approach for Assessment of Mediastinal Lesions Using Conventional Transbronchial Needle Aspiration and Endoscopic Ultrasonography in a Single Procedure.

BACKGROUND: In the era of endobronchial/esophageal ultrasound (EBUS-TBNA/EUS-FNA), many centers forgo conventional transbronchial needle aspiration (C-TBNA) in favour of EBUS-TBNA/EUS-FNA despite no conclusive evidence showing better yields with EBUS-TBNA/EUS-FNA. OBJECTIVES: Assess the feasibility of an algorithmic approach for mediastinal sampling beginning with C-TBNA utilizing rapid onsite cytologic evaluation. METHODS: Descriptive analysis of 92 consecutive patients referred for adenopathy that underwent C-TBNA and subsequent EBUS-TBNA/EUS-FNA if C-TBNA was negative or nondiagnostic. RESULTS: 92 procedures were analyzed. In 50 (54.3%) of cases, C-TBNA alone was sufficient. EBUS-TBNA was performed after C-TBNA in 27 (29.3%) of cases and EUS-FNA in 33 (35.9%) of cases. The yield was 92.9% for C-TBNA, 92.5% for EBUS-TBNA, and 89.7% for EUS-FNA. There were no statistically significant differences in yields by LN station (P = 0.51), the relationship between yield and LN size (P = 0.37), or time difference in procedures following the algorithm compared to EBUS/EUS only procedures (33.7 minutes versus 32.4 minutes on average [95% CI for difference: -9.1 to 11.7], P = 0.80). CONCLUSIONS: An algorithmic approach to assess the mediastinum using C-TBNA initially is feasible without sacrificing yield or procedure times. C-TBNA was sufficient for diagnosis in 54.3% of cases and can be efficiently taught in an IP training program.

Plasmacytoid urothelial carcinoma (PUC): Imaging features with histopathological correlation.

INTRODUCTION: Plasmacytoid urothelial carcinoma (PUC) is a high-grade variant of conventional urothelial cell carcinoma. This study is the first to describe the imaging findings of PUC, which are previously unreported, using clinical and histopathological correlation. METHODS: With internal review board approval, we identified 22 consecutive patients with PUC from 2007-2014. Clinical parameters, including age, gender, therapy, surgical margins, and long-term outcome, were recorded. Baseline imaging was reviewed by an abdominal radiologist who evaluated for tumour detectability/location/morphology, local staging, and presence/location of metastases. Pelvic peritoneal spread of tumour (defined as >5mm thick soft tissue spreading along fascial planes) was also evaluated. Followup imaging was reviewed for presence of local recurrence or metastases. RESULTS: Median age at presentation was 74 years (range 51-86), with only three female patients. Imaging features of the primary tumour in this study were not unique for PUC. Muscle-invasive disease was present on pathology in 19/22 (86%) of tumours, with distant metastases in 2/22 (9%) at baseline imaging. Pelvic peritoneal spread of tumour was radiologically present in 4/20 (20%) at baseline. During followup, recurrent/residual tumour was documented in 16/22 (73%) patients and 7/16 (44%) patients eventually developed distant metastases. Median time to disease recurrence in patients who underwent curative surgery was three months (range 0-19). CONCLUSIONS: PUC is an aggressive variant of urothelial carcinoma with poor prognosis. Pelvic peritoneal spread of tumour as thick sheets extending along fascial planes may represent a characteristic imaging finding of locally advanced PUC.

Prostatic ductal adenocarcinoma: An aggressive variant that is underdiagnosed and undersampled on transrectal ultrasound (TRUS)-guided needle biopsy.

INTRODUCTION: We sought to determine if prostatic ductal adenocarcinoma is undersampled and/or underdiagnosed at transrectal ultrasound (TRUS)-guided biopsy. METHODS: With institutional review board approval, we searched our pathology database between 2008 and 2014 for patients with a diagnosis of ≥10% ductal adenocarcinoma on radical prostatectomy and available TRUS-guided needle biopsy specimens. Three blinded genitourinary pathologists independently examined the biopsy slides. The presence or absence of ductal adenocarcinoma was determined. Diagnostic accuracy was calculated using consensus diagnosis as the reference standard. Inter-observer agreement was assessed using Cohen's kappa coefficient. RESULTS: Based on consensus review, 66.7% (12/18) biopsy specimens demonstrated ductal adenocarcinoma and 33.3% (6/18) demonstrated conventional acinar prostatic adenocarcinoma. The sensitivity/specificity for each reader (R) was: 83/100% (R1), 100/83% (R2) and 58/83% (R3) and the inter-observer agreement was only fair (K=0.32). Only two of the original needle-biopsy reports correctly identified ductal adenocarcinoma (sensitivity = 17%). The main limitations of the study are the relatively small sample size and the potential for selection bias since we could only examine patients who underwent radical prostatectomy. CONCLUSIONS: Prostatic ductal adenocarcinoma may be undersampled at TRUS-guided biopsy and in this study was under-reported in routine clinical practice. This highlights the importance of increased awareness of ductal adeoncarcinoma and the need for clear diagnostic criteria. These findings have significant clinical impact especially when determining candidacy for active surveillance protocols.

Role of fine-needle aspiration in the surgical management of pancreatic neuroendocrine tumors: utility and limitations in light of the new World Health Organization classification.

CONTEXT: Pancreatic neuroendocrine tumors (Panc-NETs) are rare and tend to get overshadowed by their more prevalent and aggressive ductal adenocarcinoma counterparts. The biological behavior of PancNETs is unpredictable, and thus management is controversial. However, the new World Health Organization classification has significantly contributed to the prognostic stratification of these patients. Concurrently, there have been advances in surgical techniques for benign or low-grade pancreatic tumors. These procedures include minimally invasive and parenchyma-sparing operations such as laparoscopy and enucleation. OBJECTIVE: To report on the utility and limitations of fine-needle aspiration in the preoperative evaluation and management of PancNETs. DESIGN: This was a retrospective review of our institutional tumor database from 2002 to 2012. There were 25 cases of PancNETs that were localized and staged by medical imaging and diagnosed by fine-needle aspiration. RESULTS: Fourteen patients underwent laparotomy, with some requiring only limited surgery; 4 had laparoscopic resections; 4 were serially observed without surgical intervention; and another 3 were inoperable. After a mean follow-up of 37 months, more than half of the patients had no evidence of disease, including most of those who underwent minimally invasive surgery. CONCLUSIONS: Fine-needle aspiration is a useful diagnostic adjunct to medical imaging in the preoperative evaluation and management of PancNETs. However, there are limitations with regard to grading PancNETs using this technique.