Skip to main content

Table 1 Conditional logistic regression model with breast cancer subtype as main outcome

From: Circulating gamma-glutamyl transferase and development of specific breast cancer subtypes: findings from the Apolipoprotein Mortality Risk (AMORIS) cohort

GGT U/L

Number of hormone receptor-positive cases

Number of controls

OR (95% CI)

Number of hormone receptor-negative cases

Number of controls

OR (95% CI)

ER status versus control

 GGT log

5939

59390

1.14 (1.08–1.19)

1295

12950

1.11(1.01–1.23)

 0–11.40

1310

13100

1 (Ref)

296

2960

1 (Ref)

 11.40–15.00

1424

14240

1.08(1.00–1.17)

308

3080

1.04(0.88–1.23)

 15.00–21.60

1585

15850

1.11(1.03–1.20)

340

3400

1.08(0.92–1.28)

 ≥21.60

1620

16200

1.25(1.15–1.35)

351

3510

1.21(1.02–1.43)

 Ptrend

  

<0.0001

  

0.03

PR status versus control

 GGT log

4938

49380

1.18(1.12–1.24)

2207

22070

1.06(0.98–1.15)

 0–11.40

1068

10680

1 (Ref)

512

5120

1 (Ref)

 11.40–15.00

1175

11750

1.09(1.00–1.19)

531

5310

1.04(0.91–1.18)

 15.00-21.60

1324

13240

1.16(1.06–1.26)

580

5800

1.02(0.90–1.16)

 ≥21.60

1371

13710

1.33(1.22–1.45)

584

5840

1.11(0.97–1.26)

 Ptrend

  

<0.0001

  

0.18

  1. Hormone receptor-positive or -negative cases referred to cancer subtypes based on individual ER or PR status. Controls without breast cancer were the referent groups. All models were adjusted for age at diagnosis, socioeconomic status, education, parity, CCI, and interval between measurement and cancer diagnosis or control selection date