Study population and setting
The study included 232 patients with cancer and 519 matched non-cancer patients admitted with laboratory-confirmed COVID-19 at nine hospitals in Wuhan, China. Cancer patients were matched to non-cancer patients based on age, sex, hypertension, diabetes, coronary heart disease, kidney damage, cerebrovascular disease, hepatitis, and chronic obstructive pulmonary disease (COPD) using propensity scores. Data were collected from the electronic medical records of each hospital between January 13 and March 18, 2020.
Summary of Main Findings
The median age of patients was 64, and 50% were female. Patients with cancer were more likely to present with severe disease, including higher levels of pro-inflammatory cytokines, infection-related biomarkers, and multiple-organ damage than patients without cancer. Among patients with cancer, older age, advanced tumor stage, elevated concentrations of inflammatory markers (tumor necrosis factor 𝝰, D-dimer, interleukin 6, and N-terminal pro-B-type natriuretic peptide), lower CD4+ T cell count, lower albumin-globulin ratio, and recent treatment with targeted or immunotherapy were associated with COVID-19 severity in adjusted analyses (adjusted for age, sex, comorbidities, cancer type, and anti-tumor treatment). Patients with cancer who had received chemotherapy within two weeks prior to admission were at the highest risk of COVID-19 severity and death.
This was a multi-center study with detailed clinical, laboratory, and radiological data. The study attempted to minimize differences between COVID-19 patients with and without cancer using propensity score matching, and it adjusted for potential confounders in the analyses among cancer patients. The study appeared not to selectively choose patients into the study but instead included all patients with known cancer and COVID-19 who were admitted to the hospitals during the study period. The authors reported that no patients left the study early and that they verified missing and uncertain data.
This study had a relatively small sample size of cancer patients who had cancer in many different sites of their bodies and differing disease characteristics, limiting the power to analyze the results in different sub-populations. The multitude of statistical comparisons (e.g. patients with and without cancer were compared on more than 70 factors in Table 1) increased the probability of chance findings — in other words, some associations may just be a coincidence. Thus, inferences should be drawn cautiously. The explanation of methods for the propensity score matched analysis was also lacking and made it difficult to discern whether the unadjusted comparisons between patients with and without cancer were appropriate; there are likely to be other differences, that were not accounted for, between patients with and without cancer. Likewise, the descriptions of the survival analysis methods and results were insufficient. Patients received a variety of antiviral, antibiotic, and immunomodulatory treatment, which could have influenced disease severity and outcomes.
This study presented extensive comparisons between COVID-19 patients with and without cancer, and identified cancer-related factors associated with disease severity in COVID-19 patients with cancer.
This review was posted on: 4 July 2020