Systemic pro-inflammatory response identifies patients with cancer with adverse outcomes from SARS-CoV-2 infection: the OnCovid Inflammatory Score

Dettorre, Gino M; Dolly, Saoirse; Loizidou, Angela; Chester, John; Jackson, Amanda; Mukherjee, Uma; Zambelli, Alberto; Aguilar-Company, Juan; Bower, Mark; Sng, Christopher C T; Salazar, Ramon; Bertuzzi, Alexia; Brunet, Joan; Mesia, Ricard; Sita-Lumsden, Ailsa; Seguí, Elia; Biello, Federica; Generali, Daniele; Grisanti, Salvatore; Seeva, Pavetha; Rizzo, Gianpiero; Libertini, Michela; Maconi, Antonio; Moss, Charlotte; Russell, Beth; Harbeck, Nadia; Vincenzi, Bruno; Bertulli, Rossella; Ottaviani, Diego; Liñan, Raquel; Marrari, Andrea; Carmona-García, M Carmen; Chopra, Neha; Tondini, Carlo Alberto; Mirallas, Oriol; Tovazzi, Valeria; Fotia, Vittoria; Cruz, Claudia Andrea; Saoudi-Gonzalez, Nadia; Felip, Eudald; Roqué, Ariadna; Lee, Alvin J X; Newsom-Davis, Tom; García-Illescas, David; Reyes, Roxana; Wong, Yien Ning Sophia; Ferrante, Daniela; Scotti, Lorenza; Marco-Hernández, Javier; Ruiz-Camps, Isabel; Patriarca, Andrea; Rimassa, Lorenza; Chiudinelli, Lorenzo; Franchi, Michela; Santoro, Armando; Prat, Aleix; Gennari, Alessandra; Van Hemelrijck, Mieke; Tabernero, Josep; Diamantis, Nikolaos; Pinato, David J

BACKGROUND: Patients with cancer are particularly susceptible to SARS-CoV-2 infection. The systemic inflammatory response is a pathogenic mechanism shared by cancer progression and COVID-19. We investigated systemic inflammation as a driver of severity and mortality from COVID-19, evaluating the prognostic role of commonly used inflammatory indices in SARS-CoV-2-infected patients with cancer accrued to the OnCovid study. METHODS: In a multicenter cohort of SARS-CoV-2-infected patients with cancer in Europe, we evaluated dynamic changes in neutrophil:lymphocyte ratio (NLR); platelet:lymphocyte ratio (PLR); Prognostic Nutritional Index (PNI), renamed the OnCovid Inflammatory Score (OIS); modified Glasgow Prognostic Score (mGPS); and Prognostic Index (PI) in relation to oncological and COVID-19 infection features, testing their prognostic potential in independent training (n=529) and validation (n=542) sets. RESULTS: We evaluated 1071 eligible patients, of which 625 (58.3%) were men, and 420 were patients with malignancy in advanced stage (39.2%), most commonly genitourinary (n=216, 20.2%). 844 (78.8%) had ≥1 comorbidity and 754 (70.4%) had ≥1 COVID-19 complication. NLR, OIS, and mGPS worsened at COVID-19 diagnosis compared with pre-COVID-19 measurement (p<0.01), recovering in survivors to pre-COVID-19 levels. Patients in poorer risk categories for each index except the PLR exhibited higher mortality rates (p<0.001) and shorter median overall survival in the training and validation sets (p<0.01). Multivariable analyses revealed the OIS to be most independently predictive of survival (validation set HR 2.48, 95% CI 1.47 to 4.20, p=0.001; adjusted concordance index score 0.611). CONCLUSIONS: Systemic inflammation is a validated prognostic domain in SARS-CoV-2-infected patients with cancer and can be used as a bedside predictor of adverse outcome. Lymphocytopenia and hypoalbuminemia as computed by the OIS are independently predictive of severe COVID-19, supporting their use for risk stratification. Reversal of the COVID-19-induced proinflammatory state is a putative therapeutic strategy in patients with cancer.