key: cord-0839610-akf73otw authors: Elkrief, Arielle; Desilets, Antoine; Papneja, Neha; Cvetkovic, Lena; Groleau, Catherine; Lakehal, Yahia Abdelali; Shbat, Layla; Richard, Corentin; Malo CCRP, Julie; Belkaid, Wiam; Cook, Erin; Doucet, Stéphane; Tran, Thai Hoa; Jao, Kevin; Daaboul, Nathalie; Bhang, Eric; Loree, Jonathan M.; Miller, Wilson H.; Vinh, Donald C.; Bouganim, Nathaniel; Batist, Gerald; Letendre, Caroline; Routy, Bertrand title: High mortality among hospital-acquired COVID-19 infection in patients with cancer: a multicentre observational cohort study date: 2020-09-03 journal: Eur J Cancer DOI: 10.1016/j.ejca.2020.08.017 sha: 0b7cf4805c44726dda410a960ac3d86ab7ff917d doc_id: 839610 cord_uid: akf73otw INTRODUCTION: Studies suggest that patients with cancer are more likely to experience severe outcomes from COVID-19. Therefore, cancer centers have undertaken efforts to care for patients with cancer in COVID-free units. Nevertheless, the frequency and relevance of nosocomial transmission of COVID-19 in patients with cancer remain unknown. The goal of this study was to determine the incidence and impact of hospital-acquired COVID-19 in this population and identify predictive factors for COVID-19 severity in patients with cancer. METHODS: Patients with cancer and a laboratory-confirmed diagnosis of COVID-19 were prospectively identified using provincial registries and hospital databases between March 3(rd) and May 23(rd), 2020 in the provinces of Quebec and British Columbia in Canada. Patient’s baseline characteristics including age, sex, comorbidities, cancer type, and type of anti-cancer treatment were collected. The exposure of interest was incidence of hospital-acquired infection defined by diagnosis of SARS-CoV-2 ≥ 5 days after hospital admission for COVID-unrelated cause. Co-primary outcomes were death or composite outcomes of severe illness from COVID-19 such as hospitalization, supplemental oxygen, intensive-care unit (ICU) admission and/or mechanical ventilation. RESULTS: A total of 252 patients (N=249 adult, and N=3 pediatric) with COVID-19 and cancer were identified, and the majority were residents of Quebec (N=233). One-hundred-and-six patients (42.1%) received active anti-cancer treatment in the last 3 months prior to COVID-19 diagnosis. During a median follow-up of 25 days, 33 (13.1%) required admission to the ICU, and 71 (28.2%) died. Forty-seven (19.1%) had a diagnosis of hospital-acquired COVID-19. Median overall survival was shorter in those with hospital-acquired infection, compared to a contemporary community-acquired population (27 days vs unreached, HR 2.3, 95% CI 1.2-4.4, p=0.0006). Multivariate analysis demonstrated that hospital-acquired COVID-19, age, ECOG status, and advanced stage of cancer were independently associated with death. INTERPRETATION: Our study demonstrates a high rate of nosocomial transmission of COVID-19, associated with increased mortality in both univariate and multivariate analysis in the cancer population, reinforcing the importance of treating patients with cancer in COVID-free units. We also validated that age and advanced cancer were negative predictive factors for COVID-19 severity in patients with cancer. Introduction: Studies suggest that patients with cancer are more likely to experience severe outcomes from COVID-19. Therefore, cancer centers have undertaken efforts to care for patients with cancer in COVID-free units. Nevertheless, the frequency and relevance of nosocomial transmission of COVID-19 in patients with cancer remain unknown. The goal of this study was to determine the incidence and impact of hospitalacquired COVID-19 in this population and identify predictive factors for COVID-19 severity in patients with cancer. Methods: Patients with cancer and a laboratory-confirmed diagnosis of COVID-19 were prospectively identified using provincial registries and hospital databases between March 3 rd and May 23 rd , 2020 in the provinces of Quebec and British Columbia in Canada. Patient's baseline characteristics including age, sex, comorbidities, cancer type, and type of anti-cancer treatment were collected. The exposure of interest was incidence of hospital-acquired infection defined by diagnosis of SARS-CoV-2 ≥ 5 days after hospital admission for COVID-unrelated cause. Co-primary outcomes were death or composite outcomes of severe illness from COVID-19 such as hospitalization, supplemental oxygen, intensive-care unit (ICU) admission and/or mechanical ventilation. Results: A total of 252 patients (N=249 adult, and N=3 pediatric) with COVID-19 and cancer were identified, and the majority were residents of Quebec (N=233). Onehundred-and-six patients (42.1%) received active anti-cancer treatment in the last 3 months prior to COVID-19 diagnosis. During a median follow-up of 25 days, 33 ( Despite these advances in the understanding of COVID-19 and cancer, the frequency of hospital-acquired infection in this population at high-risk for severe disease remains unknown. Data from Wuhan demonstrated a 7.1% rate of nosocomial transmission of COVID-19 among 918 patients with COVID-19 10 , while a study from the UK found a rate of 20% 11 . These alarming rates occurred in the absence of optimized J o u r n a l P r e -p r o o f targeted infection control and prevention (ICP) measures 12 . Due to these reports, as well as the high level of contact that patients with cancer have with the healthcare system (e.g. frequent blood tests, treatments, and hospitalizations for treatment-related complications), cancer centers rapidly enforced ICP measures in order to limit exposure of these patients to COVID-19. There is therefore an urgent need to describe the impact of hospital-acquired COVID-19 infection in patients with cancer in order to reinforce stringent infection control strategies to protect this vulnerable population. The goal of the present study was to characterize the incidence of hospital-acquired transmission in Canadian cancer centres across two provinces, and further define predictive factors for COVID-19 severity in the cancer patient population. We conducted an observational cohort study of patients with cancer and a laboratory-confirmed diagnosis of COVID-19. Patients were identified using provincial registries and hospital databases between March 3 rd and May 23 rd , 2020. Patients were also prospectively identified by their treating oncologist (or hematologist) and referred for study inclusion. This study was conducted across 8 Canadian institutions in Quebec and British Colombia and was approved by the institutional ethics committee at each site (Ethics number: MP-02-2020-8911 and H20-00892). The inclusion criteria for this study were patients with a laboratory-confirmed case of COVID-19 with any history of invasive malignancy (either solid or hematologic). All clinical data were extracted from chart review. Patients' baseline characteristics were recorded and included age, gender, ECOG status prior to COVID-19 illness, smoking history, tumor type, presence of co-morbidities and concomitant medications, including active anti-cancer treatment at the time of COVID-19 illness. The exposure of interest was incidence of hospital-acquired infection, defined by a diagnosis of COVID-19 ≥ 5 days after admission to the hospital for a COVID-unrelated cause in a non-COVID unit according to the National public health definition 13 . Acquisition in long-term care facilities were not considered hospital-acquired. Primary J o u r n a l P r e -p r o o f outcome was death from any-cause or and co-primary outcomes were composite outcomes of severe illness from COVID-19 such as supplemental oxygen, intensivecare unit (ICU) admission and/or mechanical ventilation. Clinical outcomes, including mortality, were ascertained through patient's medical records reviewing following hospital discharge. Index-date was set from time of COVID-19 detection until date of last follow-up or death. Descriptive analyses were performed for age, sex, cancer type, stage, active cancer therapy and type of therapy, cancer status, smoking history, and presence of comorbidities according to type of COVID-19 infection (hospital-acquired vs communityacquired), and baseline patient and tumor characteristics were compared using the Chisquared test. There were no corrections for multiple comparisons. Active anti-cancer therapy was defined as either 1) cytotoxic chemotherapy or 2) all other therapies (targeted agents, endocrine therapy, immunotherapy, radiotherapy) received within 3 months of COVID-19 diagnosis. Survival curves were estimated using the Kaplan-Meier method and compared with the log-rank test (Mantel-Cox method) in a univariate analysis. Multivariate analysis was performed using multivariable Cox regression model to determine hazard ratios (HR) and 95% confidence intervals (CIs) for death adjusting for other clinicopathologic features. All tests were two-sided and statistical significance was set at a p-value <0.05. All statistical analyses were conducted using the GraphPad Prism and R softwares. A total of 252 patients (N=249 adult, and N=3 pediatric) with COVID-19 and cancer were identified, and the majority were residents of the province of Quebec (N=233). Patients' baseline characteristics at time of COVID-19 diagnosis are presented in Table 1 . J o u r n a l P r e -p r o o f and presence of co-morbidities, were well balanced between the two groups ( Table 1) . Furthermore, routine blood tests were not statistically significant different between the J o u r n a l P r e -p r o o f nosocomial and non-nosocomial cases in terms of neutrophil count (p=0.08), lymphocyte count (p=0.29), neutrophil-lymphocyte ratio (p=0.23) or CRP (p=0.48). With a median follow-up of 25 days, 71 patients (28.2%) died at last follow-up. Thirty-three patients (13.1%) required ICU admission ( Table 2) . (Figure 1A) . At last follow-up, the rate of death remained higher for patients with hospital-acquired COVID-19 (p=0.002) ( Figure 1B ). Hospitalacquired N = 47 Of note, only 10 (4.0%) patients received treatment for COVID-19 (N=8 with hydroxychloroquine, n=1 with lopinavir/ritonavir, and n=1 with interleukin-6 inhibitor). We then performed Kaplan Meier analyses for the other factors associated with mortality in the multivariate analysis. In keeping with the multivariate analysis, older age (Figure 2A ), poor ECOG (Figure 2B) , and advanced stage IV disease ( Figure 2C ) were all associated with significantly shorter OS. Patients with cancer in our cohort experienced much higher mortality from COVID-19 compared to the Canadian average (28% vs 9.4% in Quebec, 6.3% in BC) 2 . Moreover, we observed a discrepancy between this high mortality rate and the rate of ICU admission, which may reflect that patients with cancer are less likely to be admitted to the intensive care setting. This finding highlights the importance of healthcare providers discussing advance directives, especially among patients with cancer. The largest study to date from the CCC19 consortium discovered negative prognostic factors associated COVID-19 severity in the cancer population. This study of J o u r n a l P r e -p r o o f Oncology Group (ECOG) performance status, and presence of comorbidities were independent risk factors for severe COVID-19 6 . Our study not only re-identified these negative risk factors (i.e. age, poor ECOG status, advanced stage of cancer), but also identified hospital-acquisition of COVID-19 as being an independent, negative factor. This novel finding clearly needs to be confirmed in other cancer centers, but identifies a patient-extrinsic factor to be considered in the fight against COVID-19. Despite this novel finding, our study has several limitations. Firstly, over 80% of patients in our cohort were hospitalized, and this reflects a selection bias for more severely ill patients. Despite our effort to identify every patient with cancer and a COVID-19 diagnosis through collaboration with microbiology departments and provincial registries (Fonds de recherche en Santé du Québec -FRSQ) which provided us access to comprehensive lists of all tested outpatients, community-acquired cases were likely missed in this study. Also, we could not detect any differences in groups who received treatment for COVID-19 given the small number of patients in this study who were treated with hydroxychloroquine or anti-viral agents. Lastly, we did not correct for factors which could increase hospital-acquired transmission, such as availability of adequate protective personal equipment, ventilation systems, single-patient rooms, and location of COVID-19 outbreaks. Our study in patients with cancer demonstrates a high mortality from COVID-19 in the adult population. This is the first report that describes a high rate of hospitalacquired COVID-19 in patients with cancer, and this was associated with high mortality in both univariate and multivariate analyses. Other independent negative risk factors for COVID-19 severity included age, ECOG status, and advanced cancer stage. Our study reinforces the importance of adherence to stringent infection control guidelines in order to protect vulnerable patients such as those with cancer. This work was partially supported by Astra Zeneca. Dr Routy and Dr Elkrief declare grant support from Astra Zeneca (grant number: N/A). Additional authors have no conflict of interest to declare. Center JHUaMCR. COVID-19 Dashboard by the Center for Systems Science and Engineering Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China Clinical features of COVID-19 in elderly patients: A comparison with young and middle-aged patients Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China Clinical impact of COVID-19 on patients with cancer (CCC19): a cohort study SARS-CoV-2 Transmission in Patients With Cancer at a Tertiary Care Hospital in Wuhan, China Case-Fatality Rate and Characteristics of Patients Dying in Relation to COVID-19 in Italy Patients with Cancer Appear More Vulnerable to SARS-COV-2: A Multicenter Study during the COVID-19 Outbreak. Cancer Discov Nosocomial infection among patients with COVID-19: A retrospective data analysis of 918 cases from a single center in Wuhan Covid-19: Doctors sound alarm over hospital transmissions Early lessons from the frontline of the 2019-nCoV outbreak Quebec INdSPd. COVID-19 Mesures de prévention et contrôle des infections pour les milieux de soins aigus : recommandations intérimaires COVID-19: a challenge for oncology services Proposed protocol to keep COVID-19 out of hospitals Universal Screening for SARS-CoV-2 in Women Admitted for Delivery COVID-19 Guidance: Primary Care Providers in a Community Setting services/diseases/2019-novel-coronavirus-infection/health-professionals/infectionprevention-control-covid-19-second-interim-guidance.html 19. (WHO) WHO. Rational use of personal protective equipment for coronavirus disease (COVID-19) and considerations during severe shortages Hand hygiene and the novel coronavirus pandemic: The role of healthcare workers ☐ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☒The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: This work was partially supported by Astra Zeneca. Dr Routy and Dr Elkrief declare grant support from Astra Zeneca (grant number: N/A).Additional authors have no conflict of interest to declare.J o u r n a l P r e -p r o o f