. 2005 Jan 28;7(3):R284–R290. doi: 10.1186/bcr993

Table 2.

Statistics of oestrogen-receptor-positive patients: association between tamoxifen therapy/no tamoxifen therapy (Tam+/Tam-) and distant recurrence rate, stratified according to genotype

Genotype	Tamoxifen therapy	Number of patients^a	Number of recurrences	Recurrence rate ratio (95% confidence interval)	P value
SULT1A11/1^†	Tam-	29	16	1.0
	Tam+	26	9	0.48 (0.21 – 1.12)	0.074
SULT1A1*2^†	Tam-	49	26	1.0
	Tam+	50	24	0.82 (0.47 – 1.43)	0.48
CYP2D61/1^‡	Tam-	55	27	1.0
	Tam+	52	25	0.91 (0.53– 1.57)	0.75
CYP2D64*^‡	Tam-	23	15	1.0
	Tam+	24	6	0.28 (0.11 – 0.74)	0.0089
SULT1A11/1 and/or CYP2D64*^§	Tam-	45	27	1.0
	Tam+	43	15	0.38 (0.19– 0.74)	0.0041
SULT1A12* and CYP2D61/1^§	Tam-	33	18	1.0
	Tam+	33	15	1.22 (0.61– 2.40)	0.57

The relative risks of distant recurrence, calculated for each combined genotype are adjusted for age, tumour size and lymph node status.

^aFollow-up data of two patients were missing.

^†,‡,§The risk ratio was first calculated separately for each genotype and genotype combination. Second, the test for interaction between the risk ratios was performed by Cox regression: ^†P = 0.27, ^‡P = 0.064, and ^§P = 0.018. The risk ratio for patients not receiving tamoxifen (Tam-) is calculated as 1.0.