Calabrese F, Pezzuto F, Fortarezza F, et al. - As there is no systematic study evaluating the severity and distribution of morphological lesions associated with SARS‐CoV‐2 and how they relate to clinical, laboratory, and radiological data, researchers sought to objectively determine pathological phenotypes and factors, that, in addition to SARS‐CoV‐2, may affect their occurrence. They comprehensively analyzed lungs from 26 patients who died from SARS‐CoV‐2 acute respiratory failure and implemented robust machine learning techniques to retrieve a global pathological score to differentiate phenotypes with prevalent vascular or alveolar injury. Most cases exhibited vascular injury phenotype which was consistently present as pure form or in combination with alveolar injury. Significantly more frequent tracheal intubation, longer invasive mechanical ventilation, illness duration, intensive care unit, hospital stay and lower tissue viral quantity were reported in correlation with phenotypes with more severe alveolar injury. In addition, this phenotype had more frequent superimposition of infections, tumors, and aspiration pneumonia. Some clinical features at admission (body mass index, white blood cells, D‐dimer, lymphocyte and platelet counts, fever, respiratory rate, and PaCO2) were identified via random forest algorithm to stratify patients into different clinical clusters and potential pathological phenotypes (a web‐app for score assessment has also been developed https://r‐ubesp.dctv.unipd.it/shiny/AVI‐Score/). SARS‐CoV‐2 positive patients frequently exhibit association of alveolar injury with other factors in addition to viral infection. Recognition of phenotypic patterns at admission may allow a better stratification of patients, ultimately favouring the most relevant management.