Table 3. Summary ORs (95% CI) and value of value of the heterogeneity of XRCC1 Arg399Gln polymorphism for studies according to source of controls and cancer type1.
Source of controls | Cancer type3 | No. comparisons(SZ case/control) | Dominant model | Recessive model | Additive model | ||||||||
OR (95% CI) | P h /I 2 | OR (95% CI) | P h /I 2 | OR (95% CI) | P h /I 2 | ||||||||
Population-based studies | Bladder cancer | 3 (628/949) | 1.32 (1.07–1.63) | 0.674/0.0% | 0.80 (0.59–1.08)* | 0.095/57.6% | 0.97 (0.70–1.34) | 0.113/54.1% | |||||
Breast cancer | 29 (22,399/24,221) | 1.04 (0.99–1.08) | 0.136/22.8% | 1.06 (0.97–1.17)* | <0.001/55.4% | 1.07 (0.97–1.17)* | 0.003/47.2% | ||||||
Colorectal cancer | 8 (3,700/7,176) | 1.02 (0.89–1.16)* | 0.089/43.4% | 1.20 (0.85–1.71)* | <0.001/77.7% | 1.20 (0.84–1.71)* | <0.001/77.7% | ||||||
Esophageal cancer | 8 (2,256/4,785) | 0.98 (0.85–1.13)* | 0.080/44.8% | 1.16 (0.97–1.38) | 0.497/0.0% | 1.15 (0.95–1.38) | 0.252/22.2% | ||||||
Gastric cancer | 9 (2,352/5,836) | 0.98 (0.80–1.20)* | <0.001/73.6% | 1.09 (0.92–1.30) | 0.319/13.8% | 1.06 (0.83–1.34)* | 0.065/45.6% | ||||||
Glioma | 3 (1,438/2,340) | 1.02 (0.87–1.18) | 0.627/0.0% | 1.02 (0.75–1.39)* | 0.064/63.7% | 1.18 (0.92–1.50) | 0.976/0.0% | ||||||
Head and neck cancer | 5 (990/2,523) | 1.05 (0.80–1.37) | 0.862/0.0% | 1.01 (0.78–1.30) | 0.923/0.0% | 1.04 (0.88–1.22) | 0.701/0.0% | ||||||
Lung cancer | 18 (5,943/7,925) | 0.94 (0.87–1.01) | 0.116/29.5% | 1.01 (0.90–1.14) | 0.145/27.1% | 0.97 (0.86–1.10) | 0.119/29.8% | ||||||
Pancreatic cancer | 3 (293/919) | 1.15 (0.88–1.50) | 0.665/0.0% | 1.23 (0.82–1.85) | 0.995/0.0% | 1.29 (0.84–1.98) | 0.820/0.0% | ||||||
Prostate cancer | 6 (2,017/2,192) | 0.97 (0.80–1.18) | 0.138/40.1% | 1.23 (0.81–1.86)* | 0.050/57.9% | 1.00 (0.81–1.22) | 0.102/48.3% | ||||||
Skin cancer | 3 (1,456/1,683) | 1.01 (0.86–1.18) | 0.299/17.2% | 0.84 (0.68–1.04) | 0.156/46.2% | 0.86 (0.68–1.09) | 0.132/50.6% | ||||||
Hospital-based studies | Bladder cancer | 17 (5,748/6,444) | 0.99 (0.92–1.06) | 0.512/0.0% | 0.88 (0.79–0.99) | 0.585/0.0% | 0.89 (0.79–1.01) | 0.430/1.8% | |||||
Breast cancer | 24 (6,625/7,774) | 1.11 (0.99–1.25)* | <0.001/62.5% | 1.16 (1.00–1.35)* | 0.014/43.2% | 1.22 (1.02–1.45)* | 0.002/50.7% | ||||||
Cervical cancer | 5 (894/1370) | 0.94 (0.61–1.44)* | 0.003/75.3% | 1.39 (1.02–1.88) | 0.624/0.0% | 1.27 (0.73–2.18)* | 0.071/57.3% | ||||||
Colorectal cancer | 18 (3,914/4,849) | 1.11 (0.95–1.29)* | 0.001/59.2% | 1.23 (1.03–1.42) | 0.249/17.5% | 1.24 (1.06–1.45)* | 0.110/30.9% | ||||||
Esophageal cancer | 6 (910/1,459) | 1.00 (0.78–1.28)* | 0.095/46.7% | 1.09 (0.67–1.76)* | 0.010/66.9% | 1.08 (0.63–1.85)* | 0.005/70.0% | ||||||
Gastric cancer | 6 (1,030/1,446) | 1.01 (0.85–1.19) | 0.578/0.0% | 1.06 (0.77–1.46) | 0.508/0.0% | 1.09 (0.78–1.51) | 0.471/0.0% | ||||||
Glioma | 4 (1,049/1,289) | 2 | <0.001/92.3% | 2 | <0.001/86.9% | 2 | <0.001/92.8% | ||||||
Hepatocellular cancer | 9 (1,621/2,310) | 1.18 (0.92–1.50)* | 0.009/60.5% | 1.13 (0.90–1.42) | 0.978/0.0% | 1.23 (0.96–1.59) | 0.829/0.0% | ||||||
Head and neck cancer | 34 (7,545/9,732) | 0.99 (0.89–1.09)* | <0.001/57.0% | 0.96 (0.83–1.10)* | 0.011/39.2% | 0.96 (0.82–1.13)* | 0.001/48.0% | ||||||
Leukemia | 15 (2,261/2,854) | 1.24 (1.00–1.53)* | <0.001/66.8% | 1.09 (0.91–1.30) | 0.206/22.8% | 1.23 (0.91–1.67)* | 0.009/53.5% | ||||||
Lung cancer | 23 (8,213/8,742) | 1.07 (0.98–1.16)* | 0.062/33.4% | 1.15 (0.98–1.34)* | 0.015/44.7% | 1.17 (0.98–1.40)* | 0.004/50.9% | ||||||
Lymphoma | 3 (399/821) | 1.06 (0.83–1.37) | 0.655/0.0% | 1.35 (0.84–2.18) | 0.797/0.0% | 1.34 (0.82–2.19) | 0.669/0.0% | ||||||
HB | Pancreatic cancer | 3 (954/1,303) | 0.93 (0.78–1.11) | 0.158/45.7% | 0.97 (0.75–1.25) | 0.671/0.0% | 0.94 (0.71–1.24) | 0.407/0.0% | |||||
Prostate cancer | 10 (1,798/1,759) | 1.06 (0.81–1.39)* | 0.001/68.8% | 1.24 (0.94–1.63)* | 0.043/48.3% | 1.28 (1.04–1.58) | 0.184/28.3% | ||||||
Skin cancer | 10 (3,307/3,788) | 0.98 (0.89–1.08) | 0.424/1.5% | 1.01 (0.87–1.17) | 0.294/16.2% | 0.99 (0.85–1.16) | 0.450/0.0% |
all summary ORs were calculated using fixed-effects models. In the case of significant heterogeneity (indicated by *), ORs were calculated using random-effects models;
the results were excluded due to high heterogeneity;
total breast cancer comparisons add up to 53 (which should be 54) because there was one study that we can not determine hospital-based study, population-based study or family-based study. The reason is same with colorectal cancer, prostate cancer and so on; PB, population-based study; HB, hospital-based study; SZ Sample size. The bold values indicate that the results are statistically significant.