Automated classification of cancer morphology from Italian pathology reports using Natural Language Processing techniques: A rule-based approach.

Pathology reports represent a primary source of information for cancer registries. Hospitals routinely process high volumes of free-text reports, a valuable source of information regarding cancer diagnosis for improving clinical care and supporting research. Information extraction and coding of textual unstructured data is typically a manual, labour-intensive process. There is a need to develop automated approaches to extract meaningful information from such texts in a reliable and accurate way. In this scenario, Natural Language Processing (NLP) algorithms offer a unique opportunity to automatically encode the unstructured reports into structured data, thus representing a potential powerful alternative to expensive manual processing. However, notwithstanding the increasing interest in this area, there is still limited availability of NLP approaches for pathology reports in languages other than English, including Italian, to date. The aim of our work was to develop an automated algorithm based on NLP techniques, able to identify and classify the morphological content of pathology reports in the Italian language with micro-averaged performance scores higher than 95%. Specifically, a novel, domain-specific classifier that uses linguistic rules was developed and tested on 27,239 pathology reports from a single Italian oncological centre, following the International Classification of Diseases for Oncology morphology classification standard (ICD-O-M). The proposed classification algorithm achieved successful results with a micro-F1 score of 98.14% on 9594 pathology reports in the test dataset. This algorithm relies on rules defined on data from a single hospital that is specifically dedicated to cancer, but it is based on general processing steps which can be applied to different datasets. Further research will be important to demonstrate the generalizability of the proposed approach on a larger corpus from different hospitals.

Imaging Biomarkers of Oral Dysplasia and Carcinoma Measured with In Vivo Endoscopic Optical Coherence Tomography.

Optical coherence tomography is a noninvasive imaging technique that provides three-dimensional visualization of subsurface tissue structures. OCT has been proposed and explored in the literature as a tool to assess oral cancer status, select biopsy sites, or identify surgical margins. Our endoscopic OCT device can generate widefield (centimeters long) imaging of lesions at any location in the oral cavity-but it is challenging for raters to quantitatively assess and score large volumes of data. Leveraging a previously developed epithelial segmentation network, this work develops quantifiable biomarkers that provide direct measurements of tissue properties in three dimensions. We hypothesize that features related to morphology, tissue attenuation, and contrast between tissue layers will be able to provide a quantitative assessment of disease status (dysplasia through carcinoma). This work retrospectively assesses seven biomarkers on a lesion-contralateral matched OCT dataset of the lateral and ventral tongue (40 patients, 70 sites). Epithelial depth and loss of epithelial-stromal boundary visualization provide the strongest discrimination between disease states. The stroma optical attenuation coefficient provides a distinction between benign lesions from dysplasia and carcinoma. The stratification biomarkers visualize subsurface changes, which provides potential for future utility in biopsy site selection or treatment margin delineation.

Diffuse tumors: Molecular determinants shared by different cancer types.

Most cancer types have both diffuse and non-diffuse subtypes, which have rather distinct morphologies, namely scattered tiny tumors vs. one solid tumor, and different levels of aggressiveness. However, the causes for forming such distinct subtypes remain largely unknown. Using the diffuse and non-diffuse gastric cancers (GCs) as the illustrative example, we present a computational study based on the transcriptomic data from the TCGA and GEO databases, to address the following questions: (i) What are the key molecular determinants that give rise to the distinct morphologies between diffuse and non-diffuse cancers? (ii) What are the main reasons for diffuse cancers to be generally more aggressive than non-diffuse ones of the same cancer type? (iii) What are the reasons for their distinct immunoactivities? And (iv) why do diffuse cancers on average tend to take place in younger patients? The study is conducted using the framework we have previously developed for elucidation of general drivers cancer formation and development. Our main discoveries are: (a) the level of (poly-) sialic acids deployed on the surface of cancer cells is a significant factor contributing to questions (i) and (ii); (b) poly-sialic acids synthesized by ST8SIA4 are the key to question (iii); and (c) the circulating growth factors specifically needed by the diffuse subtype dictate the answer to question (iv). All these predictions are substantiated by published experimental studies. Our further analyses on breast, prostate, lung, liver, and thyroid cancers reveal that these discoveries generally apply to the diffuse subtypes of these cancer types, hence indicating the generality of our discoveries.

Molecular bases of morphologically diffused tumors across multiple cancer types.

Gastric cancer has two distinct subtypes: the diffuse (DGC) and the intestinal (IGC) subtypes. Morphologically, the former each consists of numerous scattered tiny tumors while the latter each has one or a few solid biomasses. The former tends to be more aggressive and takes place in younger patients than the latter. While these have long been documented, little is known about the underlying causes. Our hypothesis is that the level of sialic acid (SA) accumulation on the cancer cell surfaces is a key reason for the observed differences. Our transcriptomic data-based analyses provide evidence that (i) DGCs tend to deploy more SAs on cancer cell surfaces than IGCs; (ii) this gives rise to considerably stronger cell-cell electrostatic repulsion in DGCs due to the negative charge that each SA carries; and (iii) such repulsion drives stronger cell protrusion and metastasis. Similar observations as well as our transcriptomic data-based predictions hold for multiple other cancer types, namely breast, lung, prostate plus liver and thyroid cancers, each known to have diffuse-like vs. non-diffused subtypes as well as more aggressive behaviors like DGCs vs. IGCs. Hence, we speculate that the discovery presented here applies not only to gastric cancer but multiple and even potentially all cancer types having diffuse-like and non-diffused subtypes.

Patterns in melanocytic lesions: impact of the geometry on growth and transport inside the epidermis.

In glabrous skin, nevi and melanomas exhibit pigmented stripes during clinical dermoscopic examination. They find their origin in the basal layer geometry which periodically exhibits ridges, alternatively large (limiting ridges) and thin (intermediate ridges). However, nevus and melanoma lesions differ by the localization of the pigmented stripes along furrows or ridges of the epidermis surface. Here, we propose a biomechanical model of avascular tumour growth which takes into account this specific geometry in the epidermis where both kinds of lesions first appear. Simulations show a periodic distribution of tumour cells inside the lesion, with a global contour stretched out along the ridges. In order to be as close as possible to clinical observations, we also consider the melanin transport by the keratinocytes. Our simulations show that reasonable assumptions on melanocytic cell repartition in the ridges favour the limiting ridges of the basal compared with the intermediate ones in agreement with nevus observations but not really with melanomas. It raises the question of cell aggregation and repartition of melanocytic cells in acral melanomas and requires further biological studies of these cells in situ.