Abstract
Lung neuroendocrine neoplasms (lung NENs) are categorised by morphology, defining a classification sometimes unable to reflect ultimate clinical outcome. Subjectivity and poor reproducibility characterise diagnosis and prognosis assessment of all NENs. Here, we propose a machine learning framework for tumour prognosis assessment based on a quantitative, automated and repeatable evaluation of the spatial distribution of cells immunohistochemically positive for the proliferation marker Ki-67, performed on the entire extent of high-resolution whole slide images. Combining features from the fields of graph theory, fractality analysis, stochastic geometry and information theory, we describe the topology of replicating cells and predict prognosis in a histology-independent way. We demonstrate how our approach outperforms the well-recognised prognostic role of Ki-67 Labelling Index on a multi-centre dataset comprising the most controversial lung NENs. Moreover, we show that our system identifies arrangement patterns in the cells positive for Ki-67 that appear independently of tumour subtyping. Strikingly, the subset of these features whose presence is also independent of the value of the Labelling Index and the density of Ki-67-positive cells prove to be especially relevant in discerning prognostic classes. These findings disclose a possible path for the future of grading and classification of NENs.
Highlights
Neuroendocrine neoplasms of the lung represent about 14% of primary lung neoplasms [1] and 25% of all neuroendocrine tumours [2]
Our dataset was composed of 31 whole slide images (WSIs) of surgical resection tis‐ sue from as many different patients, collected from medical centres located in Bari, Pisa, Turin and Varese (Italy), Brussels (Belgium) and Nice (France)
To perform the prediction of the prognostic class based on the assessed proliferation activity, we evaluated two machine learning models: k-Nearest Neighbours (k-NN), for k = 3, and Support Vector Machines (SVM)
Summary
Neuroendocrine neoplasms of the lung (lung NENs) represent about 14% of primary lung neoplasms [1] and 25% of all neuroendocrine tumours [2]. An increasing number of clues from both clinical and molecular standpoints hint to the existence of a grey area between these two histological categories, with tumours presenting intermediate characteristics: AC-like morphology, LCNEC-like proliferative activity and prognosis and mixed genomic profiles [5,6,7,8,9,10,11,12,13]. These findings suggest a classification of lung NENs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
