Interobserver Variability Study Research Articles

Interobserver variability studies in diagnostic imaging: a methodological systematic review.

To review the methodology of interobserver variability studies; including current practice and quality of conducting and reporting studies. Interobserver variability studies between January 2019 and January 2020 were included; extracted data comprised of study characteristics, populations, variability measures, key results, and conclusions. Risk of bias was assessed using the COSMIN tool for assessing reliability and measurement error. Seventy-nine full-text studies were included covering various imaging tests and clinical areas. The median number of patients was 47 (IQR:23-88), and observers were 4 (IQR:2-7), with sample size justified in 12 (15%) studies. Most studies used static images (n = 75, 95%), where all observers interpreted images for all patients (n = 67, 85%). Intraclass correlation coefficients (ICC) (n = 41, 52%), Kappa (κ) statistics (n = 31, 39%) and percentage agreement (n = 15, 19%) were most commonly used. Interpretation of variability estimates often did not correspond with study conclusions. The COSMIN risk of bias tool gave a very good/adequate rating for 52 studies (66%) including any studies that used variability measures listed in the tool. For studies using static images, some study design standards were not applicable and did not contribute to the overall rating. Interobserver variability studies have diverse study designs and methods, the impact of which requires further evaluation. Sample size for patients and observers was often small without justification. Most studies report ICC and κ values, which did not always coincide with the study conclusion. High ratings were assigned to many studies using the COSMIN risk of bias tool, with certain standards scored 'not applicable' when static images were used. The sample size for both patients and observers was often small without justification. For most studies, observers interpreted static images and did not evaluate the process of acquiring the imaging test, meaning it was not possible to assess many COSMIN risk of bias standards for studies with this design. Most studies reported intraclass correlation coefficient and κ statistics; study conclusions often did not correspond with results.

Reproducibility and repeatability of biventricular function/volume and strain parameters by 2D and 4D stress echocardiography in adult patients with repaired TOF

Abstract Funding Acknowledgements Type of funding sources: None. Background The use of 2D and 4D during stress echocardiography to undertake complex measures in complex patients like patients with repaired TOF is challenging and the validity of these measures is not known yet. Methods For test-retest variability, 20 patients with repaired TOF with no or mild pulmonary regurgitation were selected randomly and underwent a cardiopulmonary exercise test (CPET) with echocardiography. Intra-observer variability study was performed for all 20 patients by the same observer. Interobserver variability study was performed for 5 patients by different experienced observer. Intraclass correlation coefficient (ICC), and coefficients of variation (COV) were used to quantify reproducibility and variability. Results For 2D measures, better reproducibility was observed for semiautomated 2D strain measures than 2D functional measures for biventricular systolic function at baseline and during the stress (ICC>.90 vs >.70, p<0.001), with least COV was observed (COV<10%). 4D semiautomated volumetric measures demonstrated less reproducibility during stress with highest COV was observed for 4D RV volume parameters (COV, 35%), followed by 4D LV volume parameters (COV, 27%). CPET had an excellent agreement of all measures (ICC >.90). with very low COV (<10%). Conclusions Semiautomated echo measures outperformed manual measures during stress echocardiography and can be performed with acceptable reproducibility. Variability is at highest for 4D semiautomated measures despite good reproducibility, while lowest variability was observed for 2D semiautomated measures of myocardial deformation.

MRI tagging of colonic chyme mixing in healthy subjects: Inter-observer variability and reliability of the measurement with time.

Magnetic resonance imaging (MRI) tagging techniques have been applied to the GI tract to assess bowel contractions and content mixing. We aimed to evaluate the dependence of a tagging measurement (for assessing chyme mixing) on inter-observer variability in both the ascending colon (AC) and descending colon (DC) and to investigate the temporal variation and hence reliability of the colonic tagging technique by acquiring multiple measurements over time on healthy participants. Two independent datasets of healthy adults were used for the retrospective inter-observer variability (Study 1: 13 datasets and Study 2: 31 datasets), and ten participants were scanned for the prospective temporal variation study following a 1 L mannitol oral preparation. All colonic tagging data were acquired on 3 T MRI scanners. The mean and the standard deviation (SD) maps were generated pixel-by-pixel using custom-written software in MATLAB. The colonic regions of interest were defined using MIPAV software. Bland-Altman plots and scatter plots were used for the inter-observer variability. The mean and SD of all repeated measures for each subject were calculated along with a one-way ANOVA to test for variations with time. Scatter plots and Bland-Altman plots showed a large range of data with low variation and small limits of agreements (<5% CoV). The intraclass correlation coefficient of inter-rater reliability was excellent and 0.97 or above for the AC and DC measurements for both datasets. The temporal variation study shows that there was no significant difference found between the multiple measures with time (p = 0.53, one-way repeated measures ANOVA test). MRI tagging technique can provide an assessment of colonic chyme mixing. The inter-observer study data showed high inter-rater agreement. The temporal variation study showed some individual variations with time suggesting multiple measurements may be needed to increase accuracy.

Global uncertainty in the diagnosis of neurological complications of SARS-CoV-2 infection by both neurologists and non-neurologists: An international inter-observer variability study

IntroductionUniform case definitions are required to ensure harmonised reporting of neurological syndromes associated with SARS-CoV-2. Moreover, it is unclear how clinicians perceive the relative importance of SARS-CoV-2 in neurological syndromes, which risks under- or over-reporting. MethodsWe invited clinicians through global networks, including the World Federation of Neurology, to assess ten anonymised vignettes of SARS-CoV-2 neurological syndromes. Using standardised case definitions, clinicians assigned a diagnosis and ranked association with SARS-CoV-2. We compared diagnostic accuracy and assigned association ranks between different settings and specialties and calculated inter-rater agreement for case definitions as “poor” (κ ≤ 0.4), “moderate” or “good” (κ > 0.6). Results1265 diagnoses were assigned by 146 participants from 45 countries on six continents. The highest correct proportion were cerebral venous sinus thrombosis (CVST, 95.8%), Guillain-Barré syndrome (GBS, 92.4%) and headache (91.6%) and the lowest encephalitis (72.8%), psychosis (53.8%) and encephalopathy (43.2%). Diagnostic accuracy was similar between neurologists and non-neurologists (median score 8 vs. 7/10, p = 0.1). Good inter-rater agreement was observed for five diagnoses: cranial neuropathy, headache, myelitis, CVST, and GBS and poor agreement for encephalopathy. In 13% of vignettes, clinicians incorrectly assigned lowest association ranks, regardless of setting and specialty. ConclusionThe case definitions can help with reporting of neurological complications of SARS-CoV-2, also in settings with few neurologists. However, encephalopathy, encephalitis, and psychosis were often misdiagnosed, and clinicians underestimated the association with SARS-CoV-2. Future work should refine the case definitions and provide training if global reporting of neurological syndromes associated with SARS-CoV-2 is to be robust.

Interobserver variation of clinical oncologists compared to therapeutic radiographers (RTT) prostate contours on T2 weighted MRI.

The implementation of MRI-guided online adaptive radiotherapy has enabled extension of therapeutic radiographers' roles to include contouring. An offline interobserver variability study compared five radiographers' and five clinicians' contours on 10 MRIs acquired on a MR-Linac from 10 patients. All contours were compared to a "gold standard" created from an average of clinicians' contours. The median (range) DSC of radiographers' and clinicians' contours compared to the "gold standard" was 0.91 (0.86-0.96), and 0.93 (0.88-0.97) respectively illustrating non-inferiority of the radiographers' contours to the clinicians. There was no significant difference in HD, MDA or volume size between the groups.

Evaluation of therapeutic radiographer contouring for magnetic resonance image guided online adaptive prostate radiotherapy

Background and PurposeThe implementation of MRI-guided online adaptive radiotherapy has facilitated the extension of therapeutic radiographers’ roles to include contouring, thus releasing the clinician from attending daily treatment. Following undergoing a specifically designed training programme, an online interobserver variability study was performed. Materials and Methods117 images from six patients treated on a MR Linac were contoured online by either radiographer or clinician and the same images contoured offline by the alternate profession. Dice similarity coefficient (DSC), mean distance to agreement (MDA), Hausdorff distance (HD) and volume metrics were used to analyse contours. Additionally, the online radiographer contours and optimised plans (n = 59) were analysed using the offline clinician defined contours. After clinical implementation of radiographer contouring, target volume comparison and dose analysis was performed on 20 contours from five patients. ResultsComparison of the radiographers’ and clinicians’ contours resulted in a median (range) DSC of 0.92 (0.86 – 0.99), median (range) MDA of 0.98 mm (0.2–1.7) and median (range) HD of 6.3 mm (2.5–11.5) for all 117 fractions. There was no significant difference in volume size between the two groups. Of the 59 plans created with radiographer online contours and overlaid with clinicians’ offline contours, 39 met mandatory dose constraints and 12 were acceptable because 95 % of the high dose PTV was covered by 95 % dose, or the high dose PTV was within 3 % of online plan. A clinician blindly reviewed the eight remaining fractions and, using trial quality assurance metrics, deemed all to be acceptable.Following clinical implementation of radiographer contouring, the median (range) DSC of CTV was 0.93 (0.88–1.0), median (range) MDA was 0.8 mm (0.04–1.18) and HD was 5.15 mm (2.09–8.54) respectively. Of the 20 plans created using radiographer online contours overlaid with clinicians’ offline contours, 18 met the dosimetric success criteria, the remaining 2 were deemed acceptable by a clinician. ConclusionRadiographer and clinician prostate and seminal vesicle contours on MRI for an online adaptive workflow are comparable and produce clinically acceptable plans. Radiographer contouring for prostate treatment on a MR-linac can be effectively introduced with appropriate training and evaluation. A DSC threshold for target structures could be implemented to streamline future training.

Automatic 3D MRI-Ultrasound Registration for Image Guided Arthroscopy

Registration of partial view intra-operative ultrasound (US) to pre-operative MRI is an essential step in image-guided minimally invasive surgery. In this paper, we present an automatic, landmark-free 3D multimodal registration of pre-operative MRI to 4D US (high-refresh-rate 3D-US) for enabling guidance in knee arthroscopy. We focus on the problem of initializing registration in the case of partial views. The proposed method utilizes a pre-initialization step of using the automatically segmented structures from both modalities to achieve a global geometric initialization. This is followed by computing distance maps of the procured segmentations for registration in the distance space. Following that, the final local refinement between the MRI-US volumes is achieved using the LC2 (Linear correlation of linear combination) metric. The method is evaluated on 11 cases spanning six subjects, with four levels of knee flexion. A best-case error of 1.41 mm and 2.34∘ and an average registration error of 3.45 mm and 7.76∘ is achieved in translation and rotation, respectively. An inter-observer variability study is performed, and a mean difference of 4.41 mm and 7.77∘ is reported. The errors obtained through the developed registration algorithm and inter-observer difference values are found to be comparable. We have shown that the proposed algorithm is simple, robust and allows for the automatic global registration of 3D US and MRI that can enable US based image guidance in minimally invasive procedures.

Abstract P1-02-02: Examination of low Her2 expression in breast cancer

Abstract Introduction: Antibody-drug conjugates (ADC) are designed to effectively deliver cytotoxic agents directly to malignant cells. Trastuzumab deruxtecan (Tdx), an ADC of trastuzumab, an enzyme-cleavable linker, and a cytotoxic topoisomerase I inhibitor, has been shown to have antitumor activity in patients with breast cancer with low levels of HER2. However, current companion diagnostic tests for HER2-targetting therapies, immunohistochemistry (IHC) and fluorescent in situ hybridization (FISH), were optimized for high (gene amplified) levels of HER2. We hypothesize that the current common assays used in clinic do not efficiently differentiate between patients whose cancers have “0” HER2 expression and “1+” HER2 expression and thus could miss patients who would benefit from treatment with this drug. Methods: Here, we evaluated two years of HER2 surveys from the College of American Pathologists’ (CAP) Proficiency Testing Surveys for HER2 expression in breast cancer from 2019 and 2020. Participating laboratories received two tissue microarrays (TMAs) of 10 breast cancer cores, each Laboratories stained for HER2 using the standard IHC assay used in their labs. The scores were returned to the CAP as part of their quality assessment program. Each survey dataset covers the scores from around 1400 labs of 20 cores each as well as supplemental questions regarding the methodology used. We summarized the relative frequency and distribution of each score given to every core. A second independent analytic dataset was selected from the archives of the Department of Pathology at Yale, from breast biopsies in 2018. The set, enriched in HER2 2+ and 3+ cases, was read by eighteen board-certified pathologists, most with over 5 years’ experience, participating in an interobserver variability study. Hematoxylin and eosin (H&amp;E) and HER2 IHC digitally scanned images of 170 independent cases were provided. Pathologists scored the cases as 0,1, 2, or 3+. Fisher’s exact test was used to compare the 0/1+ concordant cases to 2+/3+ cases. All tests were two-sided at a significant level 0.05. Statistical analysis was performed using Graphpad Prism Version 9.0.1 and the dplyr package in R Version 1.0.143. This study was approved by Yale Human Investigation IRB protocol ID 9505008219. Results: We found that 65% of the 80 cores evaluated in the CAP survey (52/80) had a concordance rate ≥90%. This high concordance was limited to scores of 0 and 3+. The lowest concordance was found between 0 versus 1+. Of the 80 cores, 56 were considered negative (HER2 score of 0 or 1). In 25% of those cores there was &amp;lt; 70% concordance (n=15; 6 in 2019 and 9 in 2020). Analytic concordance was assessed in the independent, Yale cohort where we found that of the 170 cases, 92 were read as 0 by at least one pathologist. Of these 92, 24 were concordant (26%), defined as a ≥90% agreement. In comparison, 45/170 were read as 3+ by at least one pathologist. Again,. using a 90%definition of concordance, 26 of 45 cases (58%) were concordant. Comparison of 0/1+ concordant cases versus 2+/3+ concordant cases showed a significant difference (χ2 = 12.07, p&amp;lt;0.0005). Conclusions: Assessment of laboratory performance of around 1400 CAP labs using common current HER2 assays on CAP survey specimens, there is significant discordance in the evaluation of 0 vs. 1+ cases. In a separate selected breast biopsy cohort examined by 18 breast pathologists, we showed that discordance between scores of 0 vs 1+ is significantly larger than that between 2+ and 3+. Given the efficacy of T-DXd, we believe patients may be mis-assigned for treatment or no treatment if the decision depends on performance of the standard current HER2 assays. Citation Format: Aileen I Fernandez, Matthew Liu, Andrew Bellizzi, Jane Brock, Oluwole Fadare, Krisztina Hanley, Malini Harigopal, Julie M. Jorns, M. Gabriela Kuba, Amy Ly, Mirna Podoll, Kimmie Rabe, Mary Ann Sanders, Kamaljeet Singh, Olivia L Snir, Rinda Soong, Shi Wei, Hannah Wen, Serena Wong, Esther Yoon, Lajos Pusztai, Emily Reisenbichler, David L. Rimm. Examination of low Her2 expression in breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-02-02.

A multicenter study of interobserver variability in pathologic diagnosis of papillary breast lesions on core needle biopsy with WHO classification

BackgroundPapillary breast lesions (PBLs) comprise diverse entities from benign and atypical lesions to malignant tumors. Although PBLs are characterized by a papillary growth pattern, it is challenging to achieve high diagnostic accuracy and reproducibility. Thus, we investigated the diagnostic reproducibility of PBLs in core needle biopsy (CNB) specimens with World Health Organization (WHO) classification.MethodsDiagnostic reproducibility was assessed using interobserver variability (kappa value, κ) and agreement rate in the pathologic diagnosis of 60 PBL cases on CNB among 20 breast pathologists affiliated with 20 medical institutions in Korea. This analysis was performed using hematoxylin and eosin (H&E) staining and immunohistochemical (IHC) staining for cytokeratin 5 (CK5) and p63. The pathologic diagnosis of PBLs was based on WHO classification, which was used to establish simple classifications (4-tier, 3-tier, and 2-tier).ResultsOn WHO classification, H&E staining exhibited ‘fair agreement’ (κ = 0.21) with a 47.0% agreement rate. Simple classifications presented improvement in interobserver variability and agreement rate. IHC staining increased the kappa value and agreement rate in all the classifications. Despite IHC staining, the encapsulated/solid papillary carcinoma (EPC/SPC) subgroup (κ = 0.16) exhibited lower agreement compared to the non-EPC/SPC subgroup (κ = 0.35) with WHO classification, which was similar to the results of any other classification systems.ConclusionsAlthough the use of IHC staining for CK5 and p63 increased the diagnostic agreement of PBLs in CNB specimens, WHO classification exhibited a higher discordance rate compared to any other classifications. Therefore, this result warrants further intensive consensus studies to improve the diagnostic reproducibility of PBLs with WHO classification.

Feasibility and Performance of Elastin Trichrome as a Primary Stain in Colorectal Cancer Resection Specimens: Results of an Interobserver Variability Study.

Venous invasion (VI) is a powerful prognostic factor in colorectal cancer (CRC) that is widely underreported. The ability of elastin stains to improve VI detection is now recognized in several international CRC pathology protocols. However, concerns related to the cost and time required to perform and evaluate these stains in addition to routine hematoxylin and eosin (H&E) stains remains a barrier to their wider use. We therefore sought to determine whether an elastin trichrome (ET) stain could be used as a "stand-alone" stain in CRC resections, by comparing the sensitivity, accuracy, and reproducibility of detection of CAP-mandated prognostic factors using ET and H&E stains. Representative H&E- and ET-stained slides from 50 CRC resections, including a representative mix of stages and prognostic factors, were used to generate 2 study sets. Each case was represented by H&E slides in 1 study set and by corresponding ET slides from the same blocks in the other study set. Ten observers (3 academic gastrointestinal [GI] pathologists, 4 community pathologists, 3 fellows) evaluated each study set for CAP-mandated prognostic factors. ET outperformed H&E in the assessment of VI with respect to detection rates (50% vs. 28.6%; P<0.0001), accuracy (82% vs. 59%, P<0.0001), and reproducibility (k=0.554 vs. 0.394). No significant differences between ET and H&E were observed for other features evaluated. In a poststudy survey, most observers considered the ease and speed of assessment at least equivalent for ET and H&E for most prognostic factors, and felt that ET would be feasible as a stand-alone stain in practice. If validated by others, our findings support the use of ET, rather than H&E, as the primary stain for the evaluation of CRC resections.

An international study of interobserver variability of "string sign" of pancreatic cysts among experienced endosonographers.

Background and Objectives: No single optimal test reliably determines the pancreatic cyst subtype. Following EUS-FNA, the “string sign” test can differentiate mucinous from nonmucinous cysts. However, the interobserver variability of string sign results has not been studied.Methods: An experienced endosonographer performed EUS-FNA of pancreatic cysts on different patients and was recorded on video performing the string sign test for each. The videos were shared internationally with 14 experienced endosonographers, with a survey for each video: “Is the string sign positive?” and “If the string sign is positive, what is the length of the formed string?” Also asked “What is the cutoff length for string sign to be considered positive?” Interobserver variability was assessed using the kappa statistic (κ).Results: A total of 112 observations were collected from 14 endosonographers. Regarding string sign test positivity, κ was 0.6 among 14 observers indicating good interrater agreement (P < 0.001) while κ was 0.38 when observers were compared to the index endosonographer demonstrating marginal agreement (P < 0.001). Among observations of the length of the string in positive samples, 89.8% showed >5 mm of variability (P < 0.001), indicating marked variability. There was poor agreement on the cutoff length for a string to be considered positive.Conclusion: String sign of pancreatic cysts has a good interobserver agreement regarding its positivity that can help in differentiating mucinous from nonmucinous pancreatic cysts. However, the agreement is poor on the measured length of the string and the cutoff length of the formed string to be considered a positive string sign.

Evaluation of Deep Learning to Augment Image-Guided Radiotherapy for Head and Neck and Prostate Cancers

Personalized radiotherapy planning depends on high-quality delineation of target tumors and surrounding organs at risk (OARs). This process puts additional time burdens on oncologists and introduces variability among both experts and institutions. To explore clinically acceptable autocontouring solutions that can be integrated into existing workflows and used in different domains of radiotherapy. This quality improvement study used a multicenter imaging data set comprising 519 pelvic and 242 head and neck computed tomography (CT) scans from 8 distinct clinical sites and patients diagnosed either with prostate or head and neck cancer. The scans were acquired as part of treatment dose planning from patients who received intensity-modulated radiation therapy between October 2013 and February 2020. Fifteen different OARs were manually annotated by expert readers and radiation oncologists. The models were trained on a subset of the data set to automatically delineate OARs and evaluated on both internal and external data sets. Data analysis was conducted October 2019 to September 2020. The autocontouring solution was evaluated on external data sets, and its accuracy was quantified with volumetric agreement and surface distance measures. Models were benchmarked against expert annotations in an interobserver variability (IOV) study. Clinical utility was evaluated by measuring time spent on manual corrections and annotations from scratch. A total of 519 participants' (519 [100%] men; 390 [75%] aged 62-75 years) pelvic CT images and 242 participants' (184 [76%] men; 194 [80%] aged 50-73 years) head and neck CT images were included. The models achieved levels of clinical accuracy within the bounds of expert IOV for 13 of 15 structures (eg, left femur, κ = 0.982; brainstem, κ = 0.806) and performed consistently well across both external and internal data sets (eg, mean [SD] Dice score for left femur, internal vs external data sets: 98.52% [0.50] vs 98.04% [1.02]; P = .04). The correction time of autogenerated contours on 10 head and neck and 10 prostate scans was measured as a mean of 4.98 (95% CI, 4.44-5.52) min/scan and 3.40 (95% CI, 1.60-5.20) min/scan, respectively, to ensure clinically accepted accuracy. Manual segmentation of the head and neck took a mean 86.75 (95% CI, 75.21-92.29) min/scan for an expert reader and 73.25 (95% CI, 68.68-77.82) min/scan for a radiation oncologist. The autogenerated contours represented a 93% reduction in time. In this study, the models achieved levels of clinical accuracy within expert IOV while reducing manual contouring time and performing consistently well across previously unseen heterogeneous data sets. With the availability of open-source libraries and reliable performance, this creates significant opportunities for the transformation of radiation treatment planning.

S0126 An International Study of Interobserver Variability of “String Sign” of Pancreatic Cysts Among Experienced Endosonographers

S0126 An International Study of Interobserver Variability of “String Sign” of Pancreatic Cysts Among Experienced Endosonographers

Visual and quantitative evaluation of [18F]FES and [18F]FDHT PET in patients with metastatic breast cancer: an interobserver variability study

PurposeCorrect identification of tumour receptor status is important for treatment decisions in breast cancer. [18F]FES PET and [18F]FDHT PET allow non-invasive assessment of the oestrogen (ER) and androgen receptor (AR) status of individual lesions within a patient. Despite standardised analysis techniques, interobserver variability can significantly affect the interpretation of PET results and thus clinical applicability. The purpose of this study was to determine visual and quantitative interobserver variability of [18F]FES PET and [18F]FDHT PET interpretation in patients with metastatic breast cancer.MethodsIn this prospective, two-centre study, patients with ER-positive metastatic breast cancer underwent both [18F]FES and [18F]FDHT PET/CT. In total, 120 lesions were identified in 10 patients with either conventional imaging (bone scan or lesions > 1 cm on high-resolution CT, n = 69) or only with [18F]FES and [18F]FDHT PET (n = 51). All lesions were scored visually and quantitatively by two independent observers. A visually PET-positive lesion was defined as uptake above background. For quantification, we used standardised uptake values (SUV): SUVmax, SUVpeak and SUVmean.ResultsVisual analysis showed an absolute positive and negative interobserver agreement for [18F]FES PET of 84% and 83%, respectively (kappa = 0.67, 95% CI 0.48–0.87), and 49% and 74% for [18F]FDHT PET, respectively (kappa = 0.23, 95% CI − 0.04–0.49). Intraclass correlation coefficients (ICC) for quantification of SUVmax, SUVpeak and SUVmean were 0.98 (95% CI 0.96–0.98), 0.97 (95% CI 0.96–0.98) and 0.89 (95% CI 0.83–0.92) for [18F]FES, and 0.78 (95% CI 0.66–0.85), 0.76 (95% CI 0.63–0.84) and 0.75 (95% CI 0.62–0.84) for [18F]FDHT, respectively.ConclusionVisual and quantitative evaluation of [18F]FES PET showed high interobserver agreement. These results support the use of [18F]FES PET in clinical practice. In contrast, visual agreement for [18F]FDHT PET was relatively low due to low tumour-background ratios, but quantitative agreement was good. This underscores the relevance of quantitative analysis of [18F]FDHT PET in breast cancer.Trial registrationClinicalTrials.gov, NCT01988324. Registered 20 November 2013, https://clinicaltrials.gov/ct2/show/NCT01988324?term=FDHT+PET&draw=1&rank=2.

Geometrical and dosimetric evaluation of breast target volume auto-contouring.

Background and purposeAutomatic delineations are often used as a starting point in the radiotherapy contouring workflow, after which they are manually reviewed and adapted. The purpose of this work was to quantify the geometric differences between automatic and manually edited breast clinical target volume (CTV) contours and evaluate the dosimetric impact of such differences.Materials and methodsEighty-seven automatically generated and manually edited contours of the left breast were retrieved from our clinical database. The automatic contours were obtained with a commercial auto-segmentation toolbox. The geometrical comparison was performed both locally and globally using the Dice score and the 95% Hausdorff distance (HD). Two treatment plans were generated for each patient and the obtained dosimetric differences were quantified using dose-volume histogram (DVH) parameters in the lungs, heart and planning target volume (PTV). An inter-observer variability study with four observers was performed on a subset of ten patients.ResultsA median Dice score of 0.95 and a median 95% HD of 9.7 mm were obtained. Larger breasts were consistently under-contoured. Cranial under-contouring resulted in more than 5% relative decrease in PTV coverage in 15% of the patients while lateroposterior over-contouring increased the lung V20Gy by a maximum of 2%. The inter-observer variability of the PTV coverage was smaller than the difference between PTV coverage achieved by the automatic and the consensus contours.ConclusionsCranial under-contouring resulted in under-treatment, while lateroposterior over-contouring resulted in an increased lung dosage that is clinically irrelevant, showing the need to consider dose distributions to assess the clinical impact of local geometrical differences.

EP1.09-03 Interobserver Variability Study of PD-L1 Immunostaining in Non-Small Cell Lung Cancer

Pathologic assessment of programmed cell death ligand 1 (PD-L1) immunohistochemical staining now have been used for treatment of non-small cerll lung cancer in advanced stage. But, due to misinterpretation or unspecified staining pattern, interpretation of PD-L1 staining may be difficult. This study was to investigate interobserver variability between three different pathologists to see if the staining can be safely assessed in practice. PD-L1 immunostaining (SP263 and SP142) was performed in representative sections of 70 adenocarcinoma cases and 58 squamous cell carcinoma cases which surgically resected at the chonnam National University Hwasun Hospital, Republic of Korea between 2013 to 2015. Three pathologists individually evaluated the percentage of positive tumor cells, scoring each sample applying cutoff levels used in clinical practice: <1% positive tumor cells (score 0), 1–4% (score 1), 5–9% (score 2), 10–49% (score 3), and >50% positive tumor cells (score 4). Three pathologists were in agreement in 121 of 128 (94.5%) of scored cases in the SP263 test and 109 of 128 (85.2%) in the SP142 test. In 26 remaining cases, at least two pathologists were in aggrement of scored cases. Weighted kappa value for interobserver variability between pathologists was respectively 0.892 in the SP263 test and 0.837 in the SP142 test. The number of differently classified cases was significantly higher for the ≥1% cutoff value than every other cutoff value. Also, a significantly better agreement between pathologists was seen using ≥50% as cutoff. No statistically significant differences were seen between adenocarcinoma and squamous cell carcinoma. Interobserver agreement remains an important challenge, and the ≥1% cutoff value seems to be problematic. But, this may be overcome by training, longer experience and optimization of cutoff value.

Adaptation of a deprescription intervention to the medication management of older people living in long-term care facilities

ABSTRACTObjectives: Inappropriate prescription is a problem related to aging and deprescription could be a possible strategy. The aim of this study was to adapt the available evidence on deprescribing to the medication management of older people living in long-term care facilities.Methods: The authors carried out a bibliographic search based on the most prevalent drugs in our context to develop a specific deprescribing intervention. A committee of experts reviewed this intervention which was later validated through an interobserver variability study. Finally, the authors evaluated it in a sample of patients through a controlled before-after study.Results: The authors included 119 subjects living in a public elderly long-term care center and 122 controls living in a center with similar clinical characteristics. The authors evaluated 852 medications, identifying a total of 175 (20.5%) potentially inappropriate medications and 162 (92.5%) drugs were deprescribed (1.4 per patient). The application of the intervention meant an annual saving of 9.525.25 €.Conclusion: The implementation of a deprescribing intervention aimed at the most prevalent drugs prescribed in a particular setting improves the appropriateness of pharmacotherapy in the patients.

Tumor budding is a prognostic factor linked to epithelial mesenchymal transition in pancreatic ductal adenocarcinoma. Study report and literature review

BackgroundPancreatic ductal adenocarcinoma (PDAC) has a devastatingly poor prognosis. Surgical resection is undertaken in only 20% of patients. Most of well-known prognostic factors reflect tumor stage more than its biology. So it is important to identify new biological indicators related to survival in order to develop new therapies. ObjectiveTo determine the relation between tumor budding and Epithelial Mesenchymal Transition (EMT) and to evaluate their impact on survival for patients after resection of PDAC. MethodsWe herein report a retrospective study of 50 patients with resected PDAC. Tumor budding, immunohistochemical expression of vimentin and other standard factors were correlated with survival using the Kaplan-Meier method and Cox multivariable survival analysis. For tumor budding assessment, an inter-observer variability study was performed using 100 images of tumor slides stained with Hematoxylin & Eosin and Pan-Cytokeratin. ResultsTumor budding was present in all tumors. A substantial agreement between six pathologists was established in distinguishing high-grade from low-grade budding (κ = 0.6 and 0.73 for H&E and PCK images respectively). High-grade budding was identified in 56% of tumors (28/50). It was an adverse prognostic factor independent of tumor size, resection margins status, nodal status and vascular invasion (p = 0.008). Tumor budding was significantly associated with vimentin expression (p = 0.002). ConclusionsThe association of tumor budding with vimentin expression supported the idea that EMT is a key process in PDAC responsible for progression and drug resistance. Consequently, the elucidation of EMT molecular biology and development of new targeted therapy may improve disease outcome.

Goblet cell carcinoid of the appendix - An interobserver variability study using two proposed classification systems.

Goblet cell carcinoid (GCC) is an uncommon tumor of the vermiform appendix. Due to a broad spectrum of morphological differentiation, subclassification and grading of GCCs remains an area of controversy. Two separate systems have proposed classifying GCC tumors into three (classical GCC; adenocarcinoma ex-GCC, signet ring cell type; adenocarcinoma ex-GCC, poorly differentiated carcinoma type) OR two subgroups (low and high grade GCC) based on morphological criteria. We independently compared the inter-observer variability associated with each classification system. Overall, both systems had moderate interobserver agreement, with the two-tiered system (κ=0.54) performing slightly better than the three-tiered system (κ=0.42). GI-specialist pathologists had substantial agreement for both two and three-tiered systems (κ=0.65 vs. 0.65). Non-GI trained pathologists had lower overall agreement than GI trained pathologists, but their agreement was better using the two-tiered system (κ=0.44) than the three-tiered system (κ=0.22). A sub-analysis of 6 cases with a high rate of discordant classification revealed several challenges that exist in applying current criteria, including differentiating "goblet" vs. "signet ring" cell morphology, applying a 1 mm2 criteria to multifocal non-contiguous glandular and single infiltrating cell architecture, differentiating fibro-inflammatory stroma from desmoplastic stroma, and solid architecture in cases with abundant extracellular mucin, and distinguishing "reactive" nuclear atypia from true "cytologic atypia". Despite these challenges, the study identified better agreement among GI pathologists than non-GI trained pathologists. While GI pathologist review may be helpful, further research on objective classification criteria remains an area of interest.

A pattern‐based risk‐stratification scheme for salivary gland cytology: A multi‐institutional, interobserver variability study to determine applicability

Salivary gland aspiration cytology is useful in the preoperative management of patients but remains challenging, because of the extensive morphologic overlap of some tumors limits the ability to always determine the presence of malignancy. In response to this challenge, there has been increasing drive to develop a risk-based categorization scheme for salivary gland aspirates. Herein, the authors examine the interobserver variability of 1 such pattern and risk-based system. Select smears and cell-block sections of 50 salivary gland aspirates from 2 large academic centers were digitally imaged. These scanned slides were independently and blindly reviewed by 4 cytopathologists, and each aspirate was assigned to 1 of the proposed pattern-based categories if it was considered neoplastic by the observer. Interobserver agreement was scored and aggregated risks of malignancy were calculated for cases with available surgical follow-up. In total, 42 samples (84%) were considered neoplastic by at least 2 observers and were scored for interobserver agreement: 10 of 42 (23.8%) had uniform agreement, 14 of 42 (33.3%) had majority agreement, and 5 of 42 (11.9%) had divided agreement. Only 9 of 42 samples (21.4%) had minimal agreement, and 4 of 42 (9.5%) had no agreement. Condensation of similar categories was able to improve interobserver agreement and still maintain stratified risk of malignancy. The proposed pattern-based risk-stratification scheme, which could be implemented with the forthcoming Milan System, has good overall interobserver agreement and successfully stratifies the risk of malignancy. Some simplification is possible to make the system easier to use and improve interobserver agreement while maintaining stratification of risk. Cancer Cytopathol 2017;125:776-85. © 2017 American Cancer Society.