In the initial search, a total of 8.806 articles were found, with 1.357 in Web of Science, 2.346 in PubMed, 1.689 in Embase, 2.956 in Scopus, and 458 in medRxiv databases. After a preliminary review and the elimination of duplicates, 5.328 papers were secreened and chosen for further evaluation. A total of 58 studies [28-85] (30 in Europe, 16 in Asia, 11 in America, and 1 in Africa) and 709.908 participants (31.732 cancer patients) were included in this systematic review and meta-analysis after applying the inclusion / exclusion criteria. A flow diagram demonstrating the selection process is available in Figure 1, and the major parameters of the included studies are presented in Table 1. The quality scores of the included studies ranged from 6 to 9. The quality risk assessment of the relevant articles are shown in Supplementary Table S2. Furthermore, a bubble chart showing the distribution of studies by years is visually presented in Figure S7.
Cancer incidence in SARS-CoV-2 infected patients
Data were analyzed from a total of 55 studies [28-50,52-69,71-79,81-85] on the incidence of cancer in SARS-CoV-2 infected participants (689.462 total participant, 31.066 with cancer). The pooled incidence of cancer in SARS-CoV-2 infected patients is presented in Figure 2. The pooled ES of incidence in cancer patients was calculated as 8% (95% CI: 8%-9%). The cancer incidence in SARS-CoV-2 infected patients was higher than the global cancer incidence (approximately 0.2%) [86]. The incidence differences between countries were also examined. Among the included studies, the highest incidence was in France (16.912%, 5.939 / 89.952); the lowest incidence was found in Nigeria and Brazil (0.024%, 7.406 / 322.816; 0.004%, 14 / 2.848) (Supplementary Table S3). There was no significant publication bias in the analysis results (P > 0.05) (Supplementary Figure S2). In our analysis, a significant level of heterogeneity was determined among the studies (df = 54, I2 = 99%, p < 0.001). Sensitivity analyzes were performed by extracting each study separately. No significant change was observed in the analysis results. Thus, the robustness of the analysis results was confirmed by sensitivity analysis.
Mortality in SARS-CoV-2 infected cancer and non-cancer patients
A total of 42 studies [29-35,37-42,49-56,58-61,65-68,70-73,76-80,82-85] were included in the analysis to compare the mortality rates of cancer and non-cancer patients infected with SARS-CoV-2. There were a total of 557.053 participants, of whom 21.599 were cancer patients. According to the analysis results, cancer is a serious risk factor for mortality among patients infected with SARS-CoV-2 (RR = 2.26, 95% CI: 1.94– 2.62, P < 0.001, Figure 3). Mortality rates between continents were also evaulated as subgroup analysis and presented in Figure 4. Mortality in cancer patients infected with SARS-CoV-2 varies between continents, with the highest mortality rate in the Asian continent (RR = 2.92, CI: 2.42 - 3.53) and with the lowest in the European Continent (RR = 2.21, 95% CI: 1.69 - 2.89, p < .001). No noticeable publication bias and obvious asymmetry was observed among the included studies (Supplementary Figure S3 and S5). We found significant heterogeneity in this study results as seen in Figure 3 (df = 41, I2 = 96%, p < 0.001). Sensitivity analyses were conducted by subtracting each of the studies. No significant change was observed in the analysis results.
ICU admission rates in SARS-CoV-2 infected cancer and non-cancer patients
ICU admission rates of a total of 22.671 SARS-CoV-2 infected cancer patients and 532.161 non-cancer patients were analyzed from 22 eligible studies [30,34,36,37,40,42,44,47,49-51, 56,58-61,65,67,70,72]. The rate of ICU admission in patients with cancer was significantly higher than in individuals without cancer (RR = 1.45, 95% CI: 1.28 - 1.64, p < 0.001; heterogeneity: df = 21, I2 = 87%, p < 0.001) (Figure 5). It was determined that there was no publication bias according to the symmetry of the funnel plot and Egger's linear regression test (Supplementary Figure S4). ICU admission in cancer patients infected with SARS-CoV-2 varies between continents, with the highest ICU admission rate in the Asian continent (RR = 2.26, CI: 1.80 - 2.83) and with the lowest in the European Continent (RR = 1.13, 95% CI: 0.86 - 1.48, p < 0.001) with no publication bias (Supplementary Figure S1 and S6). Although there is a significant heterogeneity in Europe and America (df = 8, I2 = 78%, p < 0.001; df = 5, I2 = 86%, p < 0.001, respectively), no significant heterogeneity was observed in the Asian continent (df = 3, I2 = 0%, p = 0.61).