The morphologic diagnosis of colorectal carcinoma (CRC) is typically straight forward. However, there are certain subtypes of CRC that pose diagnostic challenges for daily practice due to sometimes overlapping morphologic and immunohistochemical features. These subtypes include poorly differentiated adenocarcinoma NOS, in the absence of conventional morphology (PDA-NOS), large cell neuroendocrine carcinoma (LCNEC), medullary carcinoma (MC), undifferentiated carcinoma (UC) and lymphoepithelioma-like carcinoma (LELC). This study aims to see if there is a survival difference between poorly differentiated variants of CRC, as well as other clinicopathological features that may affect prognosis. Additionally, we analyzed interobserver agreement among gastrointestinal pathologists (GP) at our institution in subclassifying poorly differentiated CRC. All consecutive patients with the diagnoses of PDA-NOS, MC, LCNEC, UC and LELC between July 2018 and July 2023 were included. Cox proportional regression test was used for multivariate analysis, while log-rank and Kaplan-Meier tests were used for univariate and survival analyses. Out of the same cohort of patients, 58 samples identified and reviewed by 3 GI-subspecialty-trained pathologists who were asked to assign the cases as PDA-NOS, LCNEC, MC, UC and LELC. Interobserver agreement was analyzed using Fleiss Kappa. Of the total 77 patients, 63 were PDA-NOS, 3 were LCNEC, 6 were MC, 4 were UC and 1 was LELC patients. Multivariate analysis using Cox proportional regression showed that tumor size (p = 0.001, HR = 1.22, 95% CI 1.08-1.38), patient age (p = 0.001, HR 1.73, 95% CI 1.24-2.40), and M stage (p = 0.02, HR 2.22, 95% CI 1.14-4.32) were significantly associated with worse OS. For the 58 cases analyzed, 3 GP agreed on 42 (72%) cases. The most common diagnosis was PDA-NOS and for 33 (57%) agreement was unanimous. There was moderate agreement (k 0.41-0.60) between all 3 GP. Our study evaluated the challenges associated with histological evaluation of colon cancers with poorly differentiated morphologies. Among the diagnoses considered in the study, MC and LCNEC had different prognostic implications compared to PDA-NOS and UC. Additionally, our GP showed moderate interobserver agreement, indicating that some level of variability in diagnosing poorly differentiated CRC subtypes may be inevitable.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.humpath.2024.105710DOI Listing

Publication Analysis

Top Keywords

interobserver agreement
12
pda-nos lcnec
12
lcnec lelc
12
colorectal carcinoma
8
differentiated crc
8
cox proportional
8
proportional regression
8
multivariate analysis
8
differentiated
6
carcinoma
6

Similar Publications

Background/aim: To explore transfer learning (TL) techniques for enhancing vertical root fracture (VRF) diagnosis accuracy and to assess the impact of artificial intelligence (AI) on image enhancement for VRF detection on both extracted teeth images and intraoral images taken from patients.

Materials And Methods: A dataset of 378 intraoral periapical radiographs comprising 195 teeth with fractures and 183 teeth without fractures serving as controls was included. DenseNet, ConvNext, Inception121, and MobileNetV2 were employed with model fusion.

View Article and Find Full Text PDF

Advances in the Application of Artificial Intelligence in the Ultrasound Diagnosis of Vulnerable Carotid Atherosclerotic Plaque.

Ultrasound Med Biol

January 2025

Department of Ultrasound, The Second Affiliated Hospital of Fujian Medical University, Quanzhou, China; Departments of Medical Imaging, Quanzhou Medical College, Quanzhou, China. Electronic address:

Vulnerable atherosclerotic plaque is a type of plaque that poses a significant risk of high mortality in patients with cardiovascular disease. Ultrasound has long been used for carotid atherosclerosis screening and plaque assessment due to its safety, low cost and non-invasive nature. However, conventional ultrasound techniques have limitations such as subjectivity, operator dependence, and low inter-observer agreement, leading to inconsistent and possibly inaccurate diagnoses.

View Article and Find Full Text PDF

This study aimed to evaluate the efficacy of the single-energy metal artifact reduction (SEMAR) algorithm in reducing metal artifacts and enhancing image quality in contrast-enhanced computed tomography (CT) for patients undergoing endovascular aneurysm repair (EVAR) with coil embolization. Thirty-eight patients (mean age 81.0 ± 6 years; 31 men, 7 women) who underwent contrast-enhanced CT following EVAR and internal iliac artery coil embolization between September 2022 and May 2023 were retrospectively analyzed.

View Article and Find Full Text PDF

Objectives: To produce standards of references for quality control and assess the reproducibility of fetal ultrasound Doppler measurements commonly used for blood flow assessment in fetal growth.

Methods: Women with singleton normal pregnancies were prospectively recruited at University College London Hospital, UK, between 24 and 41 weeks. Umbilical artery (UA), middle cerebral artery (MCA), and their pulsatility indices (PI), resistance indices (RI) and ratios such as cerebro-placental (CPR) and umbilical cerebral ratio (UCR) were obtained twice by two sonographers in training or after completion of training, blind to each other's measurements.

View Article and Find Full Text PDF

Purpose: Pre-procedural imaging is critical for transcatheter mitral valve repair planning in patients with mitral valve disease. As differences among various measurement techniques for valve evaluation are still poorly understood, we sought to assess the intra- and interobserver agreement of complex measurements derived from a prototype mitral evaluation tool (Siemens) and a commercially available tool (CVI42) using both saddle- and D-shaped mitral annulus techniques.

Materials And Methods: Multiphasic cardiac computed tomography angiography data were loaded into each software.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!