ABSTRACT
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for the COVID-19 pandemic.According to the CDC, RT-PCR in respiratory samples is the gold standard for confirming the disease, although it has practical limitations as time-consuming procedures and a high rate of false-negative results. Based on data collected at Careggi Hospital from April 7th-30th 2020,we aim to assess the accuracy of a COVID-19 diagnosis through classification methods based on blood tests and information collected at the ED. 971 pts with pre-specified features of suspected COVID-19 were enrolled;physicians prospectively dichotomized patients in COVID-19 likely/unlikely based on clinical features plus results of bedside imaging.Considering the limits of each method to classify a case COVID-19 positive, further evaluation was performed to form the COVID-19 final diagnosis, established after independent clinical review of 30-day follow-up data. Several classifiers were implemented, both parametric (Logistic Regression, LR;Quadratic Discriminant Analysis, QDA) and non-parametric (Random Forest, RF;Support Vector Machine;Neural Networks;K-nearest neighbour;Naive Bayes). Log transform was applied to some of the covariates and results compared with non transformed data.The dataset was divided in training and validation sets.Results based on validation sample show an AUC>0.8 for all classifiers. Best results are obtained applying RF, LR and QDA to a rebalanced sample using the SMOTE techniques on the log transformed data, showing an AUC of 0.890 (LR),0.896 (QDA) and 0.864 (RF). In parallel, best Sens and Spec are obtained via the above methods, the highest chieved by the LR (Sens 0.696;Spec 0.877). The rather high rate of false negative seems to be a feature inherently characterizing this classification problem.Good discriminatory power was shown for: WBC, Neut, AST, LDH, PCR, Na, IL-6 plus symptoms' information. Parametric models have the additional advantage of allowing a scientific interpretation.The performance of the classifiers with respect to the physician's gestalt and data validation are ongoing. The proposed classifiers show a good level of Sens.To improve Spec, a 3-level classification can be implemented;this tool can help in taking decisions when time and resources are scarce.
ABSTRACT
Background: The world, and Italy on the front lines, is experiencing a major medical emergency due to the SARSCoV-2 outbreak. Cancer patients are one of the potentially most vulnerable cohorts of people, but data about their management are still limited. Patients and methods: In this monocentric retrospective study we included all SARS-CoV-2 oncological patients accepted at the Onco-COVID Unit at San Luigi Gonzaga Hospital, one of the few oncological departments dedicated to cancer patients with SARS-Cov-2 infection in Italy. Clinical data were obtained from medical records available until April 24th, 2020. Results: 20 cancer patients were included. The mean (±SD) age of the patients was 66±14 years, 80% were men. Eight (40%) developed infection in their communities and 12 (60%) during the hospitalization. Lung cancer was the most frequent type of cancer (12, 60%), followed by blood/bone marrow cancer (3, 15%). Eight patients (40%) were symptomatic for COVID-19 at the time of diagnosis and symptoms began 2 (±2) days before. The most common were shortness of breath and diarrhea. Fever was present in 7 patients (35%). Among the 12 asymptomatic patients, 8 (67%) became symptomatic during the hospitalization (mean time of symptoms onset 4 days + 4). C-reactive protein increase was detected in 15 (75%) patients, high lactate dehydrogenase levels in 13 (65%), lymphocytopenia and thrombocytopenia in 6 (30%) and 4 (20%), respectively. Seven patients (35%) were on active anti-tumor treatment, 3 (43%) received anti-tumor therapy within two weeks before SARSCoV-2 positivity, and 2 (29%) continued oncological treatment (TKIs and chemotherapy) after the infection diagnosis. Nine (45%) patients were prescribed hydroxychloroquine and 5 (25%) antiviral therapy with lopinavir/ ritonavir or darunavir/ritonavir. Ten (50%) patients died within a mean of 11 days (+ 8) from the diagnosis of COVID-19 infection. Five patients (25%) have been discharged from the hospital, 4 (20%) of them with the indication to best supportive care and 1(5%) to active antitumor treatment. Conclusions: Our series confirms the high mortality rate among cancer patients with COVID-19. The presence of asymptomatic cases suggests that typical symptoms and fever may not be the only useful parameters to suspect COVID-19 in oncological patients. Our Onco-Covid unit suggests the importance of a tailored and holistic approach for cancer patients, even in a challenging situation like SARS-CoV-2 pandemic.
ABSTRACT
Background: During COVID-19 pandemic, timely diagnosis of SARS-CoV-2 infection was crucial, especially in pts with cancer. Real-time polymerase chain reaction (RT-PCR) on nasopharyngeal swab (NPS) is hampered by ≈30% of false negatives. Clinical and radiological features may identify potentially infected cases in presence of negative test. Materials and methods: We retrospectively retrieved records from 30 pts admitted to our Onco-Covid Unit. Clinical (fever, cough, respiratory failure) and radiological (ground glass opacities-GGO with or without lung consolidation) criteria were assessed. NPS RT-PCR was performed (VIASURE SARS COV-2 RT-PCR Detection Kit) at admission and at 48 hours. Pts underwent laboratory and radiological assessments (chest x-ray, bedside lung ultrasonography, thorax CT scan). Other sources of infection were ruled out (blood cultures, pneumococcal/legionella urinary antigen tests) as well as radiological differential diagnoses (e.g. disease progression). Results: From March 21st to May 9th 2020, 9 NPS negative pts with both clinical and radiological features suggestive for COVID-19 were identified. Mean age was 65.1 (31-78), 4 were female, all with ECOG PS 1. 4 pts had COPD, 8 were stage IV. All pts were on active antitumor treatment. Most common symptoms were dyspnea (n 8), fever (n 5), dry cough (n 3);radiological features include: GGOs alone (n 6), consolidation (n 1), consolidation + GGOs (n 1). 3 pts had baseline lymphopenia, 7 high lactate dehydrogenase, 8 high C-reactive protein. All pts presented with respiratory failure: PaO2/FiO2 ratio <200 (n 3), 200-300 (n 5), > 300 (n 1). All pts received antibiotics (azithromycin + ceftriaxone 3;piperacillin/tazobactam 6), glucocorticoids, O2-therapy: nasal cannula (n 3), Venturi mask (n 2), non-invasive ventilation (n 4). 4 pts died and 5 were discharged from the hospital, 4 with the indication to active antitumor treatment and 1 to best supportive care. Conclusions: High suspicion index is necessary in NSCLC pts with respiratory symptoms during COVID-19 pandemic as NPS may not identify all infected pts and the number of 'gray cases' is expected to increase in Phase II. Clinical and radiological findings correlation is pivotal in this subgroup.
ABSTRACT
Background: At the end of 2019, a novel viral pneumonia, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was described in China. Since then, Italy has soon become one of the most affected countries. In this challenging situation, the oncological community was called to protect cancer patients (pts), especially those affected by lung cancer, considered one of the most vulnerable population due to older age, multiple comorbidities and type of infection. This study aimed to investigate the clinical management of NSCLC pts, in order to provide a reliable picture of real-word practice during the COVID-19 outbreak. Materials and Methods: A 29-questions survey focusing on the clinical management and therapeutic indications for NSCLC pts during COVID-19 pandemic was sent to 95 medical/thoracic oncologists across different Italian regions. Results: From April 12th to May 2nd, 79 responses were received, with an overall response rate of 83%. The majority (77.3%) of oncologists declared a significant change in the outpatient management of NSCLC pts. The number of consultations in case of suspected NSCLC decreased in about half of cases (46.8%), with a major reduction when considering the number of pts coming from the emergency department for a first oncological evaluation (60% of cases). The total number of pts with any stage, newly diagnosed NSCLC, within the observational period, was reported to be lower than the pre-pandemic era by the 56% of oncologists. For pts candidates to adjuvant chemotherapy and concurrent chemoradiation, the therapeutic indications followed guidelines in 62% and 72% of cases, respectively. As regards the metastatic disease, the majority of oncologists confirmed their clinical indication to first-line treatment. The collected data revealed major changes in the second line therapeutic options, most related to timing and schedules of administration. Lung cancer pts' accrual in clinical trials dramatically has fallen for 79% of oncologists and follow-up consultations were mostly managed by telemedicine. Conclusions: This survey showed that Italian oncologists are determined to follow the available guidelines for the clinical management of NSCLC pts during this emergency time, while more complicate is to deal with clinical trials. In this tough landscape, telemedicine provides a valid support to facilitate patient-healthcare interactions.