X Liang1, C Yang2. 1. Clinical Epidemiology and Biostatistics Department, Children's Hospital of Chongqing Medical University, Chongqing, China. 2. Department of Surgical Oncology, National Clinical Research Center for Child Health and Disorders, Ministry of Education Key Laboratory of Child Development and Disorders, China International Science and Technology Cooperation Base of Child Development and Critical Disorders, Children's Hospital of Chongqing Medical University, Chongqing, China.
Madam — Liang and colleagues [1] first reported a nationwide analysis with cancerpatients in SARS-CoV-2 infection in China. However, there are some aspects worth reinterpreting that may cause misleading conclusions.First, the proportion of COVID-19patients with cancer in this cohort was not equal to the incidence of cancer in all COVID-19 cases. Moreover, detection signal bias may exist in cancerpatients, as they may pay more attention to their health condition and are more likely to seek medical help in the early stages of any disease, which may increase the detection rate in cancerpatients.Most importantly, age is a very important confounding factor. The mean age of cancerpatients (63.1 years) was significantly higher than that of those without cancer (48.7 years). Moreover, when focused on the 18 cases with cancer, the mean age with severe events was even higher than in those without severe events (71.89 years versus 54.33 years). There is clear evidence from other studies that older patients are more likely to be infected and have more serious conditions and death [2,3]. More male patients (male 12 versus female 6) in cancer cohorts may also response for the high proportion of severe events in cancerpatients, as female patients might have relatively mild symptoms. The logistic regression analysis further shows that age and gender have a significant impact on severity, thus it is necessary to carry out age stratification analysis or paired analysis according to age and gender. The results should be reported after adjusting for these two factors.Collectively, there were several confounding factors and defects in the statistical methods of this paper, and the effectiveness of the results may be uncertain. More rigid designed studies are needed to depict the full spectrum of cancerpatients in SARS-CoV-2 infection.