[Skip to Navigation]
Sign In
Comment & Response
July 2, 2020

Risk of COVID-19 in Patients With Cancer

Author Affiliations
  • 1Inova Schar Cancer Institute, Fairfax, Virginia
  • 2Department of Medicine, University of Virginia Cancer Center, Charlottesville
  • 3George Washington University School of Medicine and Health Sciences, Washington, DC
JAMA Oncol. 2020;6(9):1471. doi:10.1001/jamaoncol.2020.2583

To the Editor We are grateful to Yu et al1for their efforts to control the coronavirus disease 2019 (COVID-19) pandemic and for taking time to collect and report their findings to advance care for patients elsewhere in the world.1 We greatly appreciated their proposal “that aggressive measures be undertaken to reduce frequency of hospital visits of patients with cancer during a viral epidemic going forward.”

However, basic statistical analysis of the data in Table 2 contradicts their interpretation (and now widespread reporting). In their review of medical records of 1524 patients with cancer admitted from December 30, 2019, to February 17, 2020, to Zhongnan Hospital, they reported 7 COVID-19 cases among the 228 who had non–small cell lung cancer (NSCLC). The authors stated that “patients with NSCLC older than 60 years had a higher incidence of COVID-19 than those aged 60 years or younger (4.3% vs 1.8%).”1 These percentages correspond to 5 of 117 patients older than 60 years and 2 of 111 patients aged 60 years or younger. However, a standard χ2 analysis of this difference between patients in these age groups of 2.5% has 95% CIs of −2.6% to 8.0% with P = .28. So, their data are also consistent with their acknowledgment that “a population study of 1099 patients with COVID-19 did not indicate that age was associated with susceptibility to infection.”2

We share their concern for urgency in managing the pandemic effectively and saving patients’ lives through sharing data and recognize that later in their report they qualified that “a larger sample size in patients with cancer will resolve these potential associations.”1 There is urgency to first-line reports, but the reports are most informative, especially as they are transmitted through the lay press and social media, if usual statistical analyses are applied to the interpretations. This rigor is essential as we move from the study of the epidemiology to the treatment of COVID-19 and its complications. Clearly, patients with cancer are at elevated risk for complications from COVID-19 infection. Health systems and cancer care centers like ours should still continue to establish a series of processes intended to minimize visits and reduce the risk for patients of contracting COVID-19 infection in our region. Especially in the context of this pandemic and “infodemic,” we all should strive for rigorous scientific assessment and reporting of results.

Back to top
Article Information

Corresponding Author: Michael L. Maitland, MD, PhD, Inova Schar Cancer Institute, 8081 Innovation Park Dr, Fairfax, VA 22031 (michael.maitland@inova.org).

Published Online: July 2, 2020. doi:10.1001/jamaoncol.2020.2583

Conflict of Interest Disclosures: Dr Maitland reported being co–principal investigator of a National Institutes of Health R01 Academic-Industry Partnership grant on which a key industry partner/co–principal investigator is an employee of Roche/Genentech. The work is related to new methods to improve phase 2 clinical trials in oncology and is unrelated to rheumatology or IL-6 inhibitors, but Roche/Genentech markets and studies tocilizumab, a candidate therapy for COVID-19–related pneumonitis. No other disclosures were reported.

Yu  J, Ouyang  W, Chua  MLK, Xie  C.  SARS-CoV-2 transmission in patients with cancer at a tertiary care hospital in Wuhan, China.   JAMA Oncol. Published online March 25, 2020. doi:10.1001/jamaoncol.2020.0980PubMedGoogle Scholar
Guan  WJ, Ni  ZY, Hu  Y,  et al; China Medical Treatment Expert Group for Covid-19.  Clinical characteristics of coronavirus disease 2019 in China.   N Engl J Med. 2020;382(18):1708-1720. doi:10.1056/NEJMoa2002032PubMedGoogle ScholarCrossref