Table 1

Multivariate association analysis of immunomodulatory germline variants with melanoma recurrence-free survival

SNPGenePositionGenotype (%)CountsHR (CI 95%)p Value*
GG (0.6)611Ref
rs1562302IL36Bchr2:113810458GA (0.36)3630.70 (0.51 to 95)
AA (0.04)430.48 (0.17 to 1.30)0.026
AA (0.52)527Ref
rs2221903IL21chr4:123538912AG (0.39)3941.19 (0.90 to 1.58)
GG (0.09)1000.52 (0.28 to 0.97)0.016
CC (0.46)471Ref
rs2069840IL6chr7:22768572CG (0.45)4560.59 (0.44 to 0.79)
GG (0.09)930.77 (0.48 to 1.22)0.0022
TT (0.24)247Ref
rs6783IL17Dchr13:21297050TC (0.52)5220.64 (0.46 to 0.90)
CC (0.24)2410.63 (0.42 to 0.95)0.024
CC (0.51)519Ref
rs7145531IL25chr14:23842017CT (0.42)4261.39 (1.05 to 1.86)
TT (0.07)760.55 (0.28 to 1.08)0.0035
CC (0.46)466Ref
rs3811178IL25chr14:23845244CT (0.44)4521.30 (0.98 to 1.74)
TT (0.10)1000.70 (0.41 to 1.21)0.028
AA (0.41)414Ref
rs2292980IRF8chr16:85945076AG (0.48)4900.68 (0.51 to 0.91)
GG (0.11)1140.82 (0.51 to 1.33)0.038
AA (0.52)522Ref
rs12926854IRF8chr16:85951258AG (0.41)4130.67 (0.50 to 0.90)
GG (0.07)720.98 (0.59 to 1.61)0.022
  • The results of multivariate analysis showing the most significant associations (p<0.05) between 72 tagging SNPs in 44 immunomodulatory genes and melanoma recurrence-free survival in 1022 melanoma patients ascertained at NYU Medical Center. The associations, HRs and 95% CIs were calculated using Cox regression models under two degrees of freedom test, adjusted by age at diagnosis, gender, stage, thickness, ulceration and histological subtype.

  • Of note, none of the associations with RFS reached Bonferroni level of significance (p<0.0007).

  • *p Values were determined using two degrees of freedom test.

  • SNP, single nucleotide polymorphism.