Table 1.
Summary of meta-analyses by combining all the data on associations of the use of statin and the incidence of cancers.
| Cancer Type | No of Studies |
No of Total Participants |
Random Effects (RR, 95%CI) |
P (Random) |
Fixed Effects (RR, 95%CI) |
P (Fixed) | Largest Effect§ (RR, 95%CI) |
D/N/I | Egger | I2 (P) † | 95% PI (Random Effects) |
95% PI (Fixed Effects) |
Small Study Effects |
Concordant Direction | Evidence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bladder cancer | 13 | 1,266,218 | 1.07 (0.95–1.21) | 0.282 | 1.12 (1.07–1.19) | <0.001 | 1.08 (0.99–1.19) | 0/11/2 | 0.851 | 62.6 (0.001) | 0.76–1.51 | 0.81–1.56 | No | Yes | Non-significant |
| Breast cancer | 62 | 3,884,629 | 0.91 (0.85–0.97) | 0.004 | 1.00 (0.97–1.02) | 0.724 | 1.04 (0.98–1.11) | 12/44/3 | 0.023 | 79.6 (<0.001) | 0.63–1.32 | 0.69–1.44 | Yes | No | Weak |
| Colorectal cancer | 59 | 13,855,147 | 0.92 (0.88–0.95) | <0.001 | 0.94 (0.93–0.96) | <0.001 | 0.88 (0.81–0.95) | 15/33/3 | 0.106 | 71.5 (<0.001) | 0.76–1.11 | 0.78–1.14 | No | Yes | Weak |
| Endometrial cancer | 15 | 878,885 | 0.94 (0.82–1.07) | 0.349 | 1.02 (0.97–1.08) | 0.423 | 1.05 (0.95–1.15) | 4/11/0 | 0.043 | 54.9 (<0.001) | 0.66–1.34 | 0.73–1.43 | Yes | Yes | Non-significant |
| Esophageal cancer | 27 | 3,158,414 | 0.70 (0.63–0.78) | <0.001 | 0.85 (0.71–0.89) | <0.001 | 0.68 (0.52–0.88) | 15/12/0 | 0.115 | 60.7 (<0.001) | 0.46–1.05 | 0.50–1.11 | No | Yes | Suggestive * |
| Gastric cancer | 16 | 5,396,224 | 0.74 (0.60–0.90) | 0.004 | 0.84 (0.79–0.88) | <0.001 | 0.97 (0.74–1.26) | 5/11/0 | 0.325 | 90.8 (<0.001) | 0.33–1.62 | 0.39–1.78 | No | No | Weak |
| Gynecological cancer | 23 | 928,721 | 0.89 (0.78–1.02) | 0.087 | 1.00 (0.93–1.06) | 0.899 | 1.05 (0.95–1.15) | 4/19/0 | 0.003 | 43.7 (0.014) | 0.62–1.29 | 0.70–1.41 | Yes | Yes | Non-significant |
| Hematological cancer | 34 | NA | 0.89 (0.82–0.96) | 0.005 | 0.86 (0.81–0.90) | <0.001 | NA ** | 7/26/1 | 0.161 | 46.7 (0.002) | 0.60–1.20 | 0.64–1.15 | No | - | Suggestive |
| Kidney cancer | 11 | 4,052,120 | 0.91 (0.70–1.17) | 0.457 | 0.94 (0.88–1.00) | 0.034 | 1.08 (0.99–1.18) | 2/9/0 | 0.722 | 88.7 (<0.001) | 0.39–2.09 | 0.43–2.05 | No | Yes | Non-significant |
| Leukemia | 9 | 1174 | 0.85 (0.74–0.98) | 0.031 | 0.83 (0.74–0.92) | 0.001 | 0.74 (0.62–0.87) | 2/7/0 | 0.120 | 25.0 (0.220) | 0.63–1.16 | 0.62–1.10 | No | Yes | Suggestive |
| Liver cancer | 27 | 2,622,626 | 0.58 (0.52–0.66) | <0.001 | 0.65 (0.62–0.68) | <0.001 | 0.52 (0.41–0.66) | 22/5/0 | 0.117 | 83.8 (<0.001) | 0.33–1.03 | 0.38–1.13 | No | Yes | Suggestive * |
| Lung cancer | 33 | 8,833,965 | 0.89 (0.80–0.99) | 0.036 | 0.82 (0.80–0.84) | <0.001 | 1.03 (0.94–1.21) | 5/28/0 | 0.265 | 94.9 (<0.001) | 0.51–1.57 | 0.47–1.42 | No | No | Weak |
| Lymphoma | 16 | 8863 | 0.85 (0.73–0.99) | 0.042 | 0.86 (0.80–0.92) | <0.001 | 0.96 (0.83–1.11) | 6/9/1 | 0.850 | 69.1 (<0.001) | 0.52–1.40 | 0.54–1.39 | No | No | Weak |
| Melanoma | 24 | 434,680 | 0.94 (0.86–1.03) | 0.204 | 0.94 (0.88–1.00) | 0.063 | 0.94 (0.88–1.00) | 3/21/0 | 0.836 | 26.0 (0.121) | 0.74–1.19 | 0.60–1.46 | No | No | Non-significant |
| Myeloma | 5 | 609 | 0.89 (0.53–1.51) | 0.674 | 0.89 (0.73–1.09) | 0.251 | 0.83 (0.61–1.12) | 2/2/1 | 0.983 | 81.0 (<0.001) | 0.14–5.73 | 0.17–4.78 | No | Yes | Non-significant |
| Pancreatic cancer | 20 | 2,832,052 | 0.89 (0.75–1.06) | 0.207 | 0.91 (0.86–0.97) | 0.003 | 1.10 (0.81–1.49) | 1/18/1 | 0.927 | 79.0 (<0.001) | 0.46–1.71 | 0.49–1.71 | No | Yes | Non-significant |
| Prostate cancer | 44 | NA | 0.94 (0.90–0.99) | 0.017 | 1.02 (1.00–1.04) | 0.056 | NA ** | 18/42/4 | 0.002 | 74.5 (<0.001) | 0.71–1.24 | 0.78–1.33 | Yes | - | Weak |
| Non-melanoma skin cancer | 17 | 1,240,281 | 1.07 (1.00–1.16) | 0.063 | 1.09 (1.06–1.13) | <0.001 | 1.09 (1.06–1.13) | 1/11/5 | 0.768 | 58.5 (0.001) | 0.88–1.31 | 0.90–1.32 | No | No | Non-significant |
D/N/I: Decreasing risk/No difference/Increasing risk; RR: Relative risk; CI: Confidence interval; PI: Prediction interval. § Relative risk (95% Confidence interval) of the largest study in each meta-analysis. † I2 metric of inconsistency (95% confidence interval of I2) and p-value of the Cochran Q test for evaluation of heterogeneity. * Suggestive level of evidence due to the greater number of studies that decrease risk in which a high heterogeneity is due to differences in the effect size of the association. ** Largest effect of study of hematologic and prostate cancer were not assessible due to lack of number of participants data in individual studies.