Genetic variability in drug transport, metabolism or DNA repair affecting toxicity of chemotherapy in ovarian cancer

Table 3 Association between genetic variants and hematologic toxicity

3A: Significant correlations with anemia
	All patients N = 290 (%)	Patients with anemia gr 3–4 N = 57 (19.6%)	Patients without anemia gr 3–4 N = 233 (80.3%)	Unadjusted OR (95%CI)	*p-value	Adjusted OR (95% CI)	**Corrected p value
ABCB1 rs1128503	CC 94 (32.4)	13 (22.8)	81 (34.8)	1.58 (1.03; 2.42)	0.035	1.71 (1.07; 2.71)	0.023
	CT 147 (50.7)	30 (52.6)	117 (50.2)
	TT 49 (16.9)	14 (24.6)	35 (15.0)
ABCC2 rs12762549	CC 80 (27.6)	25 (43.8)	55 (23.6)	0.55 (0.36; 0.83)	0.005	0.51 (0.33; 0.81)	0.004
	CG132 (45.5)	22 (38.6)	110 (47.2)
	GG 76 (26.2)	10 (17.5)	66 (28.3)
ABCA1 rs363717	AA 86 (29.6)	10 (17.5)	76 (32.6)	1.31 (1.98; 2.99)	0.001	2.08 (1.32; 3.27)	0.002
	GA 131 (45.2)	23 (40.3)	108 (33.5)
	GG 73 (25.2)	24 (42.1)	49 (15.2)
ERCC1 rs11615	TT 133 (45.9)	18 (31.6)	115 (49.3)	1.58 (1.06-2.35)	0.024	1.61 (1.04-2.50)	0.031
	TC 114 (39.3)	28 (49.1)	86 (36.9)
	CC 42 (14.5)	11 (19.3)	31 (13.3)
3B: Significant correlations with thrombocytopenia (TCP)
	All patients	Patients with TCP gr 3 – 4	Patients without TCP gr 3 – 4	Unadjusted OR	*p-value	Adjusted OR	**Corrected p value
	N = 290 (%)	N = 57 (19.6%)	N = 233 (80.3%)	(95%CI)	*p-value	(95% CI)	**Corrected p value
CYP3A4 rs4986910	TT 280 (96.5)	51(89.5)	229(98.3)	5.61 (1.46; 21.64)	0.012	4.99 (1.22; 20.31)	0.025
	CT 9 (3.1)	5(8.8)	4(1.7)
	CC 0 (0)	0	0

OR: Odds Ratio using wild type as reference category. *Uncorrected p values were calculated using binary logistic regression without correction for covariates. Per-allele ORs and 95% CIs are shown. There were missing genotypes for rs12762549 (n = 2), rs11615 (n = 1) and rs4986910 (n = 1). **Corrected p values were obtained using a logistic regression for the presence or absence of anemia/thrombocytopenia/febrile neutropenia while including the following covariates: genetic variant, age, BMI, AUC of carboplatin, number of administered cycles, and use of ESA for anemia or use of CSF for febrile neutropenia. In the regression for anemia, age, BMI, administered AUC of carboplatin or number of administered cycles were not identified as significant covariates (p = 0.576, p = 0.614 and p = 0.317, p = 0.481), whereas use of ESA was significant (p = 0.034). In the regression for grade 3–4 thrombocytopenia, age and AUC of administered carboplatin were a significant covariate (p = 0.023 and p = 0.014), but BMI or number of administered cycles were not (p = 0.571 and p = 0.243). In the regression for grade 4 neutropenia, BMI and age were significant covariates (p = 0.043 and p = 0.041), while administered AUC and number of administered cycles were not (p = 0.607 and p = 0.321).

ISSN: 2050-6511