Impact of pretreatment dihydropyrimidine dehydrogenase genotype-guided fluoropyrimidine dosing on chemotherapy associated adverse events.

Consensus guidelines exist for genotype-guided fluoropyrimidine dosing based on variation in the gene dihydropyrimidine dehydrogenase (DPYD). However, these guidelines have not been widely implemented in North America and most studies of pretreatment DPYD screening have been conducted in Europe. Given regional differences in treatment practices and rates of adverse events (AEs), we investigated the impact of pretreatment DPYD genotyping on AEs in a Canadian context. Patients referred for DPYD genotyping prior to fluoropyrimidine treatment were enrolled from December 2013 through November 2019 and followed until completion of fluoropyrimidine treatment. Patients were genotyped for DPYD c.1905+1G>A, c.2846A>T, c.1679T>G, and c.1236G>A. Genotype-guided dosing recommendations were informed by Clinical Pharmacogenetics Implementation Consortium guidelines. The primary outcome was the proportion of patients who experienced a severe fluoropyrimidine-related AE (grade ≥3, Common Terminology Criteria for Adverse Events version 5.0). Secondary outcomes included early severe AEs, severe AEs by toxicity category, discontinuation of fluoropyrimidine treatment due to AEs, and fluoropyrimidine-related death. Among 1394 patients, mean (SD) age was 64 (12) years, 764 (54.8%) were men, and 47 (3.4%) were DPYD variant carriers treated with dose reduction. Eleven variant carriers (23%) and 418 (31.0%) noncarriers experienced a severe fluoropyrimidine-related AE (p = 0.265). Six carriers (15%) and 284 noncarriers (21.1%) experienced early severe fluoropyrimidine-related AEs (p = 0.167). DPYD variant carriers treated with genotype-guided dosing did not experience an increased risk for severe AEs. Our data support a role for DPYD genotyping in the use of fluoropyrimidines in North America.

Genetic and clinical predictors of arthralgia during letrozole or anastrozole therapy in breast cancer patients.

PURPOSE: Female patients with breast cancer frequently develop arthralgia when treated with aromatase inhibitors (AI). Although the mechanism of AI-induced arthralgia is unknown, potential biomarkers have been identified. The purpose of this study was to investigate the clinical and genetic predictors of AI-induced arthralgia in a prospective cohort of patients with estrogen receptor-positive breast cancer. METHODS: One hundred and ninety-six patients were enrolled at initiation of AI therapy with either letrozole or anastrozole. Patients completed two validated self-report questionnaires assessing pain, stiffness, and physical function at baseline, and repeated the questionnaires at two and at six months after the initiation of treatment with an AI. Germline DNA of all patients was genotyped for seven single-nucleotide polymorphisms (SNPs) previously identified by genetic screens and genome-wide association studies as associated with AI-induced arthralgia. RESULTS: More than 50% of the study group experienced arthralgia symptoms. Genetic analysis revealed that four SNPs, in CYP19A1 (rs4775936) and ESR1 (rs9322336, rs2234693, rs9340799), were associated with the development of arthralgia (adjusted P = 0.016, 0.018, 0.017, 0.047). High body mass index (BMI) was also associated with the development of arthralgia symptoms (adjusted P = 0.001). Patients prescribed letrozole were significantly more likely to develop arthralgia than patients on anastrozole (P = 0.018), and also more likely to discontinue AI therapy due to arthralgia. The CYP19A1 (rs4775936) SNP was significantly associated with discontinuation of therapy due to intolerable arthralgia. CONCLUSIONS: Our results suggested that BMI and AI drug (letrozole versus anastrozole) were clinical predictors of arthralgia, while genetic variants rs4775936, rs9322336, rs2234693, and rs9340799 were genetic predictors of AI-induced arthralgia. Significantly, rs4775936 was also a predictor of discontinuation of therapy.

Attenuation of bile acid-mediated FXR and PXR activation in patients with Crohn's disease.

Bile acids are endogenous ligands of nuclear receptors pregnane X (PXR) and farnesoid X (FXR). PXR and FXR regulate pathways that are impaired in inflammatory bowel disease (IBD). Decreases in PXR and FXR activity are documented in IBD; however reasons for this are unknown. We aimed to assess the effect of Crohn's disease (CD) on the plasma bile acid composition in vivo and the resultant impact on PXR and FXR activation. A cross-sectional study evaluated the plasma concentrations of 12 bile acids in addition to 4ß-hydroxycholesterol (4ßOHC), an in vivo probe of the PXR target-gene cytochrome 3A4 (CYP3A4) and the FXR target-gene, fibroblast growth factor (FGF) 19 in individuals with (n = 74) and without (n = 71) CD. An in vitro model was used to assess the impact of CD-specific changes in the plasma bile acid composition on PXR and FXR activation. Decreases in glycochenodeoxycholic acid, taurocholic acid and lithocholic acid were seen in CD with increases in glycodeoxycholic acid and glycocholic acid relative to the total plasma bile acid profile. In vitro, increasing concentrations of bile acids applied in the same ratio as seen in the study cohorts resulted in decreased activation of both PXR and FXR in the CD model. In vivo, plasma 4ßOHC (CD = 18.68 ng/ml ± 13.02 ng/ml, non-CD = 46.38 ng/ml ± 40.70 ng/ml, p ≤ 0.0001) and FGF19 (CD = 0.276 pg/L ± 0.189 pg/L, non-CD = 0.485 pg/L ± 0.42 pg/L, p = 0.0002) concentrations were lower in CD versus controls. Ultimately, CD-specific changes in the plasma bile acid composition lead to reduced activation of FXR and PXR target genes in vitro and in vivo.

Targeted next generation sequencing as a tool for precision medicine.

BACKGROUND: Targeted next-generation sequencing (NGS) enables rapid identification of common and rare genetic variation. The detection of variants contributing to therapeutic drug response or adverse effects is essential for implementation of individualized pharmacotherapy. Successful application of short-read based NGS to pharmacogenes with high sequence homology, nearby pseudogenes and complex structure has been previously shown despite anticipated technical challenges. However, little is known regarding the utility of such panels to detect copy number variation (CNV) in the highly polymorphic cytochrome P450 (CYP) 2D6 gene, or to identify the promoter (TA)7 TAA repeat polymorphism UDP glucuronosyltransferase (UGT) 1A1*28. Here we developed and validated PGxSeq, a targeted exome panel for pharmacogenes pertinent to drug disposition and/or response. METHODS: A panel of capture probes was generated to assess 422 kb of total coding region in 100 pharmacogenes. NGS was carried out in 235 subjects, and sequencing performance and accuracy of variant discovery validated in clinically relevant pharmacogenes. CYP2D6 CNV was determined using the bioinformatics tool CNV caller (VarSeq). Identified SNVs were assessed in terms of population allele frequency and predicted functional effects through in silico algorithms. RESULTS: Adequate performance of the PGxSeq panel was demonstrated with a depth-of-coverage (DOC) ≥ 20× for at least 94% of the target sequence. We showed accurate detection of 39 clinically relevant gene variants compared to standard genotyping techniques (99.9% concordance), including CYP2D6 CNV and UGT1A1*28. Allele frequency of rare or novel variants and predicted function in 235 subjects mirrored findings from large genomic datasets. A large proportion of patients (78%, 183 out of 235) were identified as homozygous carriers of at least one variant necessitating altered pharmacotherapy. CONCLUSIONS: PGxSeq can serve as a comprehensive, rapid, and reliable approach for the detection of common and novel SNVs in pharmacogenes benefiting the emerging field of precision medicine.

Predictors of cisplatin-induced ototoxicity and survival in chemoradiation treated head and neck cancer patients.

OBJECTIVES: Cisplatin-induced ototoxicity is a common permanent consequence of curative chemoradiation for locally advanced head and neck squamous cell carcinoma (HNSCC). Predictors of ototoxicity in HNSCC were examined. MATERIALS AND METHODS: In this prospective, observational cohort study, 206 adult HNSCC patients underwent audiometric testing at baseline, during and after treatment with cisplatin-based chemoradiation. Ototoxicity was defined as ≥grade 2 audiometric change from baseline (CTCAE v4.02). Relationships between clinical and pharmacogenetic (TPMT, COMT, ACYP2, CTR1, OCT2, MATE1, ABCC2, ABCC3, and ABCG2) covariates and ototoxicity, progression-free (PFS) and overall survival (OS) were assessed by Cox regression. RESULTS: Weekly cisplatin resulted in lower ototoxicity risk while PFS and OS were similar compared to high dose cisplatin (P = 0.00035; HR = 0.18; 95% CI, 0.07-0.46). COMT (rs9332377) carriers had higher ototoxicity risk (P = 0.00556; HR = 1.72; 95% CI, 1.17-2.52) while MATE1 (rs2289669) A/A carriers were protected from ototoxicity (P = 0.01062; HR = 0.46; 95% CI, 0.26-0.84). Absence of the protective MATE1 allele among those who carry the risk allele in COMT predicted increased ototoxicity risk, (P = 0.00414; HR = 3.22; 95% CI, 1.45-7.17 and P = 0.00022; HR = 4.89; 95% CI, 2.11-11.36). Survival outcomes did not differ between carriers of protective or risk alleles. CONCLUSIONS: Weekly cisplatin dosing, COMT and MATE1 are predictors of ototoxicity without affecting treatment efficacy. COMT and MATE1 genotyping and weekly dosing may be a potential strategy for mitigating cisplatin-induced ototoxicity in HNSCC.

Letrozole concentration is associated with CYP2A6 variation but not with arthralgia in patients with breast cancer.

PURPOSE: The aromatase inhibitor (AI) letrozole is a first-line drug in the adjuvant treatment of breast cancer in postmenopausal women. Adherence to AI therapy, including letrozole, remains problematic due to the development of debilitating AI-induced arthralgia. Letrozole is metabolized in the liver by CYP2A6. It remains unknown if plasma letrozole levels or CYP2A6 genetic variation is associated with the development of arthralgia. METHODS: We enrolled 126 female breast cancer patients initiated on letrozole therapy and prospectively collected blood samples at baseline and two follow-up time points to determine letrozole plasma concentrations and CYP2A6 genotype. At each visit, participants completed two validated questionnaires to assess the severity of arthralgia symptoms. RESULTS: More than half (55%) of patients experienced a significant increase in their arthralgia symptoms after initiation of treatment. The clinical variables of body mass index (P = 0.0003) and age (P = 0.0430) were negatively and positively associated with plasma letrozole concentrations, respectively. CYP2A6 genotype was significantly associated with letrozole levels (P < 0.0001), and increased plasma letrozole levels were observed in patients with CYP2A6 reduced-function genotypes. Plasma levels of letrozole and CYP2A6 genotype were not significantly associated with a change in pain score from baseline. CONCLUSIONS: CYP2A6 genotype was a significant predictor of letrozole plasma levels, but was not associated with the development of arthralgia.

CYP2D6 genotype and endoxifen plasma concentration do not predict hot flash severity during tamoxifen therapy.

PURPOSE: Tamoxifen is frequently prescribed to prevent breast cancer recurrence. Tamoxifen is a prodrug and requires bioactivation by CYP2D6. Tamoxifen use is often limited by adverse effects including severe hot flashes. There is paucity of prospectively collected data in terms of CYP2D6 genotype and measured tamoxifen, 4-hydroxytamoxifen and endoxifen concentrations in relation to hot flash severity during tamoxifen therapy. METHODS: We conducted a longitudinal prospective study of breast cancer patients on tamoxifen (n = 410). At each visit, blood samples were collected, and patients completed a standardized hot flash survey (n = 1144) that reflected hot flash severity during the 7 days prior to the visit. Plasma concentrations of tamoxifen, 4-hydroxytamoxifen, and endoxifen were measured using liquid chromatography-tandem mass spectrometry and genotyping was carried out for CYP2D6. A linear mixed-effects regression analysis assessed the association of covariates in relation to the hot flash severity score (HFSS). RESULTS: Median age at first assessment was 50 years with 61.9% of patients considered peri-menopausal. Most patients (92.2%) experienced hot flash symptoms with 51.0% having low HFSS (0-4) and 7.32% experiencing HFSS > 25. Age was significantly associated with hot flash severity, with patients aged 45-59 more likely to have higher HFSS. Neither duration of tamoxifen therapy nor observed tamoxifen, endoxifen and 4-hydroxy tamoxifen plasma concentration predicted hot flash severity. Genetic variation in CYP2D6 or CYP3A4 was not predictive of hot flash severity. CONCLUSIONS: Hot flash severity during tamoxifen therapy can not be accounted for by CYP2D6 genotype or observed plasma concentration of tamoxifen, 4-hydroxytamoxifen, or endoxifen.

Food Effect on Rosuvastatin Disposition and Low-Density Lipoprotein Cholesterol.

Rosuvastatin is commonly prescribed for the treatment of hypercholesterolemia and hepatic transporter-mediated accumulation in the liver enhances its efficacy. Current guidelines indicate no preference for fed or fasted rosuvastatin administration. We investigated the association between food intake and rosuvastatin disposition in healthy subjects and low-density lipoprotein cholesterol (LDL-C)-lowering effects among patients taking rosuvastatin. We demonstrate that administration with food resulted in a near 40% reduction of rosuvastatin exposure in healthy Asian (n = 12) and Caucasian (n = 11) subjects. Higher rosuvastatin concentrations in Asian subjects also correlated with higher allele frequency of ABCG2 c.421C>A. In mice, a greater rosuvastatin liver:plasma ratio was noted when administered with food. Among rosuvastatin patients (n = 156), there was no difference in dose needed to reach target LDL-C, measured LDL-C, or lathosterol concentrations, when administered in a fed or fasting state. Therefore, taking rosuvastatin with food could reduce systemic concentrations, and subsequent myopathy risk, without compromising LDL-C-lowering benefit.

Pharmacogenomics Guided-Personalization of Warfarin and Tamoxifen.

The use of pharmacogenomics to personalize drug therapy has been a long-sought goal for warfarin and tamoxifen. However, conflicting evidence has created reason for hesitation in recommending pharmacogenomics-guided care for both drugs. This review will provide a summary of the evidence to date on the association between cytochrome P450 enzymes and the clinical end points of warfarin and tamoxifen therapy. Further, highlighting the clinical experiences that we have gained over the past ten years of running a personalized medicine program, we will offer our perspectives on the utility and the limitations of pharmacogenomics-guided care for warfarin and tamoxifen therapy.

Interpatient Variation in Rivaroxaban and Apixaban Plasma Concentrations in Routine Care.

BACKGROUND: Direct-acting oral anticoagulants (DOACs) are widely prescribed for stroke prevention in patients with atrial fibrillation (AF). An important advantage of DOACs is that routine monitoring of an anticoagulation response is not necessary. Nevertheless, because of their mechanism of action, a DOAC anticoagulation effect can be inferred based on the observed plasma concentration. However, there is a paucity of data relating to observed interpatient variation in DOAC plasma concentrations in the postmarket clinical setting. METHODS: We determined rivaroxaban and apixaban plasma concentrations in patients with AF during routine clinic visits. RESULTS: Among 243 patients (rivaroxaban, n = 94; apixaban, n = 149) enrolled in this study, a 60- and 50-fold interpatient variation in plasma concentration was observed for rivaroxaban and apixaban, respectively. Approximately 12% of patients receiving rivaroxaban and 13% of patients receiving apixaban exceeded the 95th percentile for predicted maximum plasma concentration observed in clinical trials. CONCLUSIONS: In this routine-care setting, rivaroxaban and apixaban plasma concentrations tended to be more variable than those observed in clinical trials. Identification of additional clinical and molecular determinants that more fully predict patients at risk for excessively high or low DOAC concentrations may enable a more precise DOAC dosing regimen for the individual patient.

Identification and Characterization of Trimethylamine-N-oxide Uptake and Efflux Transporters.

Trimethylamine-N-oxide (TMAO) is a recently identified predictor of cardiovascular and chronic kidney disease. TMAO is primarily generated through gut-microbiome mediated conversion of dietary choline and carnitine to TMA, which is converted to TMAO by hepatic flavin monooxygenase 3 (FMO3) and subsequently undergoes renal elimination. We investigated the role of uptake and efflux drug transporters in TMAO disposition in vitro and in vivo. After screening a large array of uptake transporters, we show organic cation transporter 2 (OCT2) is the key transporter for TMAO cellular uptake. In Oct1/2 knockout mice, we observed increased plasma TMAO levels with reduced renal retention, suggesting the importance of Oct2 in facilitating the uptake of TMAO into renal tubular cells in vivo. Multiple transporters of the ATP-binding cassette (ABC) family, including ABCG2 (BCRP) and ABCB1 (MDR1), were capable of TMAO efflux. In human subjects, clinical, dietary, and pharmacogenetic covariates were evaluated for contribution to TMAO levels in a cohort of dyslipidemic patients (n = 405). Interestingly, genetic variation in ABCG2, but not other transporters, appeared to play a role in modulating TMAO exposure.

Profound reduction in tamoxifen active metabolite endoxifen in a breast cancer patient treated with rifampin prior to initiation of an anti-TNFα biologic for ulcerative colitis: a case report.

BACKGROUND: Tamoxifen, a common anti-estrogen breast cancer medication, is a prodrug that undergoes bioactivation via cytochrome P450 enzymes, CYP2D6 and to a lesser degree, CYP3A4 to form the active metabolite endoxifen. With an increasing use of oral anti-cancer drugs, the risk for drug-drug interactions mediated by enzyme inhibitors and inducers may also be expected to increase. Here we report the first case demonstrating a potent drug-drug interaction in a real-world clinical setting between tamoxifen and rifampin in a breast cancer patient being treated concurrently for ulcerative colitis. CASE PRESENTATION: We describe a patient on adjuvant tamoxifen therapy for breast cancer that was prescribed rifampin for TB prophylaxis prior to initiation of an anti-tumor necrosis factor (TNF)-α agent due to worsening ulcerative colitis. This 39 year old Caucasian woman had been followed by our personalized medicine clinic where CYP2D6 genotyping and therapeutic monitoring of tamoxifen and endoxifen levels had been carried out. The patient, known to be a CYP2D6 intermediate metabolizer, had a previous history of therapeutic endoxifen levels. Upon admission to hospital for a major flare of her ulcerative colitis a clinical decision was made to initiate an anti-TNFα biological agent. Due to concerns regarding latent TB, rifampin as an anti-mycobacterial agent was initiated which the patient was only able tolerate for 10 days. Interestingly, her plasma endoxifen concentration measured 2 weeks after cessation of rifampin was sub-therapeutic at 15.8 nM and well below her previous endoxifen levels which exceeded 40 nM. CONCLUSION: Rifampin should be avoided in patients on tamoxifen therapy for breast cancer unless continued tamoxifen efficacy can be assured through endoxifen monitoring. Drug-drug interactions can pose a significant risk of sub-therapeutic benefit in tamoxifen patients.

Trimethylamine-N-oxide: A Novel Biomarker for the Identification of Inflammatory Bowel Disease.

BACKGROUND: The gastrointestinal (GI) microbiome is recognized for potential clinical relevance in inflammatory bowel disease (IBD). Data suggest that there is a disease-dependent loss of microbial diversity in IBD. Trimethylamine-N-oxide (TMAO) is generated by GI anaerobes through the digestion of dietary phosphatidylcholine and carnitine in a microbial-mammalian co-metabolic pathway. IBD-related changes in the gut microbiome may result in disease-specific changes in TMAO plasma concentrations. AIM: To determine whether TMAO plasma levels in IBD are altered compared to controls and whether they correlate with disease presence or activity. METHODS: Liquid chromatography-tandem mass spectrometry was used to measure TMAO, choline, and carnitine plasma levels in 479 subjects (373 non-IBD controls, 106 IBD). Subjects were also genotyped for the flavin monooxygenase (FMO)3 variants, E158K and E308G. RESULTS: Plasma TMAO levels were 2.27 µM lower in the IBD population compared to the control population (p = 0.0001). Lower TMAO levels were similarly seen in active ulcerative colitis (UC) (1.56 µM) versus inactive disease (3.40 µM) (p = 0.002). No difference was seen in active Crohn's disease (CD) versus inactive CD. No intergroup variation existed in plasma TMAO levels based on FMO3 genotype. Choline levels were higher in IBD, while carnitine levels were similar between the two groups, suggesting that lower TMAO levels in IBD were not due to dietary differences. CONCLUSIONS: Decreased TMAO levels are seen in IBD compared to a non-IBD population. These data suggest that TMAO may have potential as a biomarker to support IBD diagnosis as well as to assess disease activity in UC.

Profound reduction in the tamoxifen active metabolite endoxifen in a patient on phenytoin for epilepsy compared with a CYP2D6 genotype matched cohort.

Tamoxifen is a prodrug, requiring cytochrome P450 enzyme-mediated metabolism to form the active metabolite endoxifen. We identified a case of drug-drug interaction involving tamoxifen and phenytoin, associated with a markedly lower endoxifen level than predicted. The patient is a 49-year-old woman, genotyped as a cytochrome P450 2D6 (CYP2D6) extensive metabolizer, chronically taking phenytoin for a seizure disorder. The plasma endoxifen level 2 months after starting tamoxifen was 4.72 nmol/l, the lowest level we have seen in our clinic among patients with CYP2D6 extensive metabolizer genotypes (n=195). To our knowledge, this is the first report documenting the extent of induction in terms of both tamoxifen and endoxifen levels during concomitant phenytoin therapy, and this effect would likely result in loss of therapeutic benefit from tamoxifen. Phenytoin should therefore not be used concurrently with tamoxifen for extended periods of time unless a therapeutic drug (endoxifen) monitoring strategy is utilized.

Tamoxifen-associated hot flash severity is inversely correlated with endoxifen concentration and CYP3A4*22.

Tamoxifen use is often limited in some patients due to adverse effects including severe hot flash symptoms. Tamoxifen undergoes hepatic bioactivation by CYP2D6 and CYP3A4 to form the active metabolite endoxifen. It remains unclear whether the extent of attained endoxifen level or genetic polymorphisms in drug metabolizing enzymes is associated with the frequency and severity of hot flashes. We conducted a prospective study using self-reported surveys to assess tamoxifen side effects experienced during the week prior to clinic visits of 132 female breast cancer patients on tamoxifen therapy, and hot flash severity scores were tabulated. At the time of clinic visit, blood samples were obtained to determine tamoxifen and its metabolite levels and to determine CYP2D6 and CYP3A4 genotypes. The majority of participants (77 %) experienced hot flashes, with 11 % experiencing severe or very severe symptoms. We observed an inverse correlation between endoxifen concentration and hot flash severity score following adjustment for age, BMI, and menopausal status in patients with non-zero scores (p < 0.001). Interestingly, CYP2D6 genotype was not significantly associated with hot flash scores in patients on no known inhibitory medications. However, CYP3A4*22 carriers were less likely to have hot flashes with an odds ratio of 8.87 (p < 0.01) even when compared to a cohort with similar endoxifen levels. Our data demonstrate that patients with higher endoxifen levels tended to predict lower hot flash severity scores. Importantly, this is the first study to show CYP3A4*22 genotype as an independent predictor of hot flash severity during tamoxifen therapy.

Developmental competence and the induction of ectopic proboscises in Drosophila melanogaster.

Developmental competence is the response of a cell(s) to information. Determination of adult labial identity in Drosophila requires Proboscipedia (PB) and Sex combs reduced (SCR); however, co-ectopic expression of PB and SCR is not sufficient for induction of ectopic adult labial identity, because the developmental information supplied by PB and SCR is suppressed. The evolutionarily conserved LASCY, DYTQL, NANGE motifs, and the C-terminal domain of SCR are sequence elements that mediate some, or all, of the suppression of ectopic proboscis determination. Therefore, the developmentally competent primordial proboscis cells provide an environment devoid of suppression, allowing PB and SCR to determine proboscis identity. SCR derivatives lacking suppression sequences weakly induce ectopic proboscis transformations independently of PB, suggesting that SCR may be the activity required for induction of adult labial identity, as is the case for larval labial identity. A possible explanation for PB independence of SCR in determination of adult and embryonic labial identity is PB operates as a competence factor that switches SCR from determining T1 identity to labial identity during metamorphosis. Lastly, labial determination is not conserved between SCR and murine HOXA5, suggesting that SCR has acquired this activity during evolution.

CYP3A4 and seasonal variation in vitamin D status in addition to CYP2D6 contribute to therapeutic endoxifen level during tamoxifen therapy.

Tamoxifen is a widely utilized adjuvant anti-estrogen agent for hormone receptor-positive breast cancer, known to undergo CYP2D6-mediated bioactivation to endoxifen. However, little is known regarding additional genetic and non-genetic determinants of optimal endoxifen plasma concentration. Therefore, 196 breast cancer patients on tamoxifen were enrolled in this prospective study over a 24-month period. Blood samples were collected for pharmacogenetic and drug-level analysis of tamoxifen and metabolites. Regression analysis indicated that besides CYP2D6, the recently described CYP3A4*22 genotype, seasonal variation, and concomitant use of CYP2D6-inhibiting antidepressants were significant predictors of endoxifen concentration. Of note, genetic variation explained 33 % of the variability while non-genetic variables accounted for 13 %. Given the proposed notion of a sub-therapeutic endoxifen concentration for predicting breast cancer recurrence, we set the therapeutic threshold at 18 nM, the 20th percentile for endoxifen level among enrolled patients in this cohort. Nearly 70 % of CYP2D6 poor metabolizers as well as extensive metabolizers on potent CYP2D6-inhibiting antidepressants exhibited endoxifen levels below 18 nM, while carriers of CYP3A4*22 were twofold less likely to be in sub-therapeutic range. Unexpectedly, endoxifen levels were 20 % lower during winter months than mean levels across seasons, which was also associated with lower vitamin D levels. CYP3A4*22 genotype along with sunshine exposure and vitamin D status may be unappreciated contributors of tamoxifen efficacy. The identified covariates along with demographic variables were integrated to create an endoxifen concentration prediction algorithm to pre-emptively evaluate the likelihood of individual patients falling below the optimal endoxifen concentration.

Determination of clinically therapeutic endoxifen concentrations based on efficacy from human MCF7 breast cancer xenografts.

Tamoxifen is a widely prescribed adjuvant anti-estrogen agent for estrogen receptor-positive breast cancer. Tamoxifen is known to undergo CYP2D6-mediated bioactivation to the active metabolite endoxifen. Endoxifen concentrations exhibit high interindividual variability, contributing to either sub-optimal tamoxifen efficacy or side effects in subsets of patients. However, the relationship between endoxifen exposure and tumor growth inhibition has not been well-characterized and little is known regarding the optimal in vivo endoxifen plasma level required for tumor inhibition. Pharmacokinetics-Pharmacodynamics (PK-PD) modeling was carried out to characterize the relationship between endoxifen concentration and tumor growth inhibition (TGI) in dose-ranging experiments in the human MCF7 xenograft bearing mouse model. Subsequently, simulations using human PK were used to determine the efficacious clinically relevant endoxifen concentration required to produce optimal tumor suppression. Based on the PK-PD model and simulations using clinical PK/concentration data of endoxifen, C stasis (100 % TGI) is observed at 53 nM, a concentration attained by many tamoxifen-treated patients. Importantly, PK-PD simulations indicate that mean steady-state levels observed in CYP2D6 extensive metabolizers are expected to result in optimal tumor suppression while mean concentrations observed in poor metabolizers are predicted to result in suboptimal TGI. Our study is the first to characterize the in vivo PK-PD relationship for endoxifen where clinically observed endoxifen concentrations are associated, in an exposure-dependent manner, with % TGI measured in a xenograft model. It is anticipated that endoxifen concentration achieved in individual patients is the limiting factor for achieving optimal tumor growth suppression.