Tocilizumab effects in COVID-19 pneumonia: role of CT texture analysis in quantitative assessment of response to therapy

• Published in: Radiologia medica Vol. 126; no. 9; pp. 1170 - 1180
• Main Authors: Masci, Giorgio Maria, Iafrate, Franco, Ciccarelli, Fabio, Pambianchi, Giacomo, Panebianco, Valeria, Pasculli, Patrizia, Ciardi, Maria Rosa, Mastroianni, Claudio Maria, Ricci, Paolo, Catalano, Carlo, Francone, Marco
• Format: Journal Article
• Language: English
• Published: Milan Springer Milan 01.09.2021; Springer Nature B.V
• Subjects: Adult; Aged; Aged, 80 and over; Antibodies, Monoclonal, Humanized - therapeutic use; Attenuation; Biomarkers; Chest Radiology; Coronaviruses; COVID-19; COVID-19 - diagnostic imaging; COVID-19 - drug therapy; COVID-19 - mortality; Diagnostic Radiology; Dimers; Evaluation; Female; Humans; Imaging; Immunosuppressive agents; Interventional Radiology; Kaplan-Meier Estimate; Laboratories; Lesions; Lung - diagnostic imaging; Lung - pathology; Male; Medicine; Medicine & Public Health; Middle Aged; Neuroradiology; Order parameters; Pneumonia; Radiology; Receptors, Interleukin-6 - antagonists & inhibitors; Retrospective Studies; Severe acute respiratory syndrome coronavirus 2; Severity of Illness Index; Survival; Survival Analysis; Texture; Tomography, X-Ray Computed; Treatment Outcome; Ultrasound; COVID-19; Pneumonia; Treatment; Computed tomography; Tocilizumab
• ISSN: 0033-8362; 1826-6983
• DOI: 10.1007/s11547-021-01371-7

Purpose To evaluate CT and laboratory changes in COVID-19 patients treated with tocilizumab, compared to a control group, throughout a combined semiquantitative and texture analysis of images. Materials and methods From March 11 to April 20, 2020, 57 SARS-CoV-2 positive patients were retrospectively compared: group T ( n  = 30) receiving tocilizumab and group non-T ( n  = 27) undergoing only antivirals/antimalarials. Chest-CT and laboratory findings were analyzed before and after treatment. CT evaluation included both semiquantitative scoring and texture analysis of all parenchymal lesions. Survival and recovery analyses were also provided with Kaplan–Meier method. Results In group T, no significant differences were found for CT score after treatment, while several texture features significantly changed, including mean attenuation ( p  < 0.0001), skewness ( p  < 0.0001), entropy ( p  = 0.0146) and higher-order parameters, suggesting considerable fading of parenchymal lesions. PaO 2 /FiO 2 mean value significantly increased after treatment, from 240 ± 93 to 363 ± 107 ( p  = 0.0003), with parallel decrease in inflammatory biomarkers (CRP, D-dimer and LDH). In group non-T, CT scoring, texture and laboratory parameters showed significant worsening at follow-up. Findings were clinically associated with opposite trends between two groups, with reduction of severe cases in group T (from 21/30 to 5/30; p  < 0.0001) as compared to a significant worsening in group non-T (severe cases increasing from 6/27 to 14/27; p  = 0.0473). Probability of discharge was significantly higher in group T ( p  < 0.0001), as well as survival rate, although not statistically significant. Conclusions Our results suggest the potential role of CT texture analysis for assessing response to treatment in COVID-19 pneumonia, using Tocilizumab, as compared to semiquantitative evaluation, providing insight into the intrinsic parenchymal changes.