Table 3. Association of the diabetes Genetic Risk Score (GRS) with breast cancer outcomes.
Quartile of GRS | Total Mortality | Recurrence/Breast Cancer-Specific Mortality | ||||
---|---|---|---|---|---|---|
No. participants | No. of Event | HR (95% CI) | No. participants | No. of Event | HR (95% CI) | |
All Participants | ||||||
Q1 | 1588 | 311 | 1 (reference) | 1438 | 291 | 1 (reference) |
Q2 | 1587 | 291 | 0.94 (0.80,1.10) | 1439 | 276 | 0.94 (0.80,1.11) |
Q3 | 1589 | 295 | 0.97 (0.83,1.14) | 1431 | 271 | 0.94 (0.80,1.11) |
Q4 | 1582 | 311 | 0.99 (0.84,1.16) | 1447 | 278 | 0.92 (0.78,1.08) |
P for trend | 0.99 | 0.33 | ||||
By ER status | ||||||
Women with ER- breast cancer | ||||||
Q1 | 508 | 90 | 1 (reference) | 472 | 93 | 1 (reference) |
Q2 | 497 | 102 | 1.33 (1.00,1.77) | 454 | 102 | 1.29 (0.97,1.72) |
Q3 | 521 | 96 | 1.05 (0.78,1.40) | 476 | 90 | 0.88 (0.66,1.19) |
Q4 | 509 | 123 | 1.38 (1.05,1.82) | 476 | 109 | 1.13 (0.86,1.50) |
P for trend | 0.08 | 0.90 | ||||
Women with ER+ breast cancer | ||||||
Q1 | 926 | 175 | 1 (reference) | 840 | 156 | 1 (reference) |
Q2 | 954 | 144 | 0.76 (0.61,0.95) | 870 | 137 | 0.82 (0.65,1.04) |
Q3 | 931 | 155 | 0.92 (0.74,1.15) | 849 | 143 | 0.96 (0.76,1.21) |
Q4 | 934 | 152 | 0.86 (0.69,1.07) | 858 | 142 | 0.88 (0.70,1.11) |
P for trend | 0.45 | 0.53 | ||||
By history of diabetes | ||||||
Breast cancer patients with diabetes | ||||||
Q1 | 107 | 17 | 1 (reference) | 97 | 16 | 1 (reference) |
Q2 | 126 | 15 | 0.87 (0.42,1.80) | 120 | 15 | 0.77 (0.33,1.80) |
Q3 | 157 | 25 | 1.26 (0.67,2.37) | 144 | 27 | 1.37 (0.73,2.59) |
Q4 | 232 | 60 | 2.22 (1.28,3.88) | 210 | 47 | 1.51 (0.82,2.75) |
P for trend | <0.01 | 0.03 | ||||
Breast cancer patients without diabetes | ||||||
Q1 | 1360 | 230 | 1 (reference) | 1225 | 219 | 1 (reference) |
Q2 | 1319 | 203 | 0.90 (0.74,1.08) | 1188 | 191 | 0.87 (0.72,1.06) |
Q3 | 1300 | 205 | 0.95 (0.79,1.15) | 1161 | 191 | 0.92 (0.75,1.11) |
Q4 | 1203 | 177 | 0.83 (0.68,1.01) | 1096 | 168 | 0.81 (0.66,0.99) |
P for trend | 0.12 | 0.07 |
NOTE: All models adjusted for age at diagnosis (treated as continuous, using a restricted cubic spline with 5 knots) and TNM-stage. P values for interaction between ER status and weighted GRS were as follows: P = 0.28 for total mortality and P = 0.71 for recurrence or breast cancer-specific mortality; P values for interaction between diabetes status and weighted GRS were as follows: P<0.01 for total mortality and P<0.01 for recurrence or breast cancer-specific mortality.