Introduction: Breast cancer is the most frequently diagnosed cancer among women in both rndeveloped and developing countries. Its treatment involves a multi-modality approach torneradicate residual cancer and prevent recurrent disease. Large inter-individual variations in thernphamacokinetics and treatment outcome of cyclophosphamide and tamoxifen have beenrnreported. In addition to other patient factors such as age, sex and etc, pharmacogeneticrnvariations have been implicated to potentially influence the pharmacokinetics of anticancerrndrugs resulting in suboptimal effects and/or unpredictable toxicity. This thesis contributes tornbetter increase our understanding of the implication of pharmacogentic variations in therntreatment of breast cancer.rnrnObjective: The aim of this thesis was to investigate the effect of genetic polymorphism of drugrnmetabolizing enzymes and transporter proteins on the pharmacokinetics and pharmacodynamicsrnof cyclophosphamide and pharmacokinetics of tamoxifen metabolites among Ethiopian breastrncancer patients. rnrnMethods: The study was conducted in breast cancer patients who were on CPA basedrnchemotherapy (chemotherapy group) and on tamoxifen (tamxifen group). Drug and metaboliternquantification was performed using HPLC (cyclophosphamide) or LC/MS/MS (Tam and itsrnmetabolites). Genotyping of candidate genes encoding drug metabolizing enzymes andrntransporter proteins, relevant to cyclophosphamide and tamoxifen metabolism and transportrnincluding CYP2B6, CYP2D6, CYP2C9, CYP2C19, CYP2J2, CYP3A5, POR, and ABCB1 andrnUGT2B15, was done using TaqManrnTMrn allele specific PCR. CYP2D6 gene copy number was alsorndetermined by TaqManrnTMrn Copy Number Assay. Among chemotherapy group, hematologicalrntoxicity (neutropenia, anemia, and thrombocytopenia) were also monitored throughoutrnchemotherapy cycle. A population PK-PD modeling of cyclophosphamide was performed usingrnnon linear mixed effect modeling (NONMEM) software. rnrnResult: In study I, the incidence of chemotherapy induced grade 3 or 4 hematological toxicityrnwas 51.0% (95% confidence interval (CI) = 44.54 â€“ 57.46%). Most of the hematologic toxicitiesrnwere neutropenic toxicity (50.2%). CYP2J2*7 variant allele and low baseline white blood cellsrncount (grade 1 leukocytopenia), or low neutrophils count (grade 1 or 2 neutropenia), werernassociated with increased risk of grade 3 or 4 hematologic toxicity in the subsequent cycles.rnPatients carrying CYP2C9 *2 or *3 alleles were associated with lower incidence of hematologicrntoxicity. In study II, the overall actual average RDI was 81.9%. The proportion of patients whornreceived reduced RDI < 85% of the standard/planned dose intensity was 56.6%. Thernindependent risk predictors of reduced RDI were CYP2J2*7 allele, BMI < = 18.4 kg/mrnrn(underweight), low baseline leukocyte count, and low baseline neutrophils count. On the otherrnhand, the odds of receiving reduced RDI was lower in patients with CYP2B6 *6/*6 genotype. Inrnstudy 3, the PK analysis showed that cyclophosphamide was described by one compartmentrnpharmacokinetic model with population clearance, and apparent volume of distribution of 5.41rnL/hr and 46.5 L, respectively. The inter-individual variability (IIV) in clearance was 51% (38.9%rnand 49.3% for 500 and 600 mg/mrn2 rnCPA regimen group, respectively). CYP3A5 and CYP2C9rngenotypes, body surface area and cyclophosphamide dosing regimen were identified asrnsignificant predictors of PK parameters. Plasma cyclophosphamide exposure moderately and rnpositively predicts neutropenic toxicity. In study IV, the proportion of patients with lowrnendoxifen concentration (below 5.9 ng/mL) was 35.8%. Large inter-individual variability inrnendoxifen concentration and MRrn was observed (coefficient of variation 74.6% and 59%,rnrespectively). An increase in CYP2D6 activity score (AS) was associated with a correspondingrnincrement in endoxifen concentration and metabolic ratios (MRrnE/NDMrnE/NDMrn, MRrn).rnABCB1 was also significantly associated with MRrnE/NDM rn(p = 0.042) and MRrnE/4-HTrn(p = 0.015).rnCarriers of G allele for ABCB1 had a lower metabolic ratios for MRrn(0.022 vs 0.041; p =rn0.042) and MRrnE/4-HT rnE/NDM rn(2.52 vs 3.02, p = 0.016) compared to the wild type allele (ABCB1, allele A).rnMoreover, CYP2D6 diplotype explained 29% of the variability in endoxifen concentration andrn46% of the variability in MRrn Similarly, CYP2D6 phenotype explained 26.3% of thernvariability in endoxifen concentration and 40.9% of the variability in MRrnE/NDM.rn. rnrnIn conclusion, we report high rates of chemotherapy-induced hematological toxicities causingrnlarger proportion of patients to receive inadequate RDI. Patients carrying CYP2J2 *7 allele andrnwith low baseline WBC or ANC are at a higher risk for chemotherapy induced hematologicrntoxicities and receiving reduced RDI. Such patients require prior support and close follow uprnduring chemotherapy. CYP3A5, and CYP2C9 genotypes, CPA dosage regimen, BMI, and BSA,rninfluence the PK of CPA. Plasma CPA exposure moderately and positively predicts neutropenicrntoxicity. On the other hand, CYP2D6, POR and ABCB1 explain the interindividual variations ofrnendoxifen. Tamoxifen therapy guided by CYP2D6 genotyping in clinical practice may assistrntreatment decision-making for optimal patient benefit. rnrnE/NDMrn and MRrnE/4-HT rn4HT/Tam