“…For the studies assessing three polymorphisms (-1082A/G, -819T/C and -592A/C) [ 32 , 37 ], two (-1082A/G and -592A/C) [ 31 ], only one such as -1082A/G [ 29 , 30 , 33 - 35 , 38 ] or -819T/C [ 36 , 39 ] polymorphism and cancer risk but no other IL-10 gene polymorphisms, the genotypes distribution in the controls were deviated from HWE, thus, these studies were excluded in the final analysis. Sixteen studies were also deviated from HWE, but the genotypes distribution in the controls of eight studies [ 18 , 64 - 67 , 70 , 73 , 76 ] were consistent with that expected from the HWE for both -819T/C and -592A/C polymorphisms, five [ 81 , 84 , 86 , 87 , 90 ] for the -592A/C polymorphism and three [ 41 , 48 , 54 ] for the -1082A/G polymorphism, thus, these studies were included in the final analysis. For those studies [ 18 , 45 , 65 ] with the same control subjects, the control numbers were calculated once in the total number.…”