The Pharmacogenomics Journal (2016) 16, 47–53 © 2016 Macmillan Publishers Limited All rights reserved 1470-269X/16 www.nature.com/tpj
ORIGINAL ARTICLE
Effect of the CYP3A5 and ABCB1 genotype on exposure, clinical response and manifestation of toxicities from sunitinib in Asian patients YL Teo1,2, HL Wee1, XP Chue1, NM Chau3, M-H Tan3, R Kanesvaran3, HL Wee1, HK Ho1 and A Chan1,2 The objective of this study was to determine the effect of the CYP3A5 and ATP binding cassette subfamily B member 1 (ABCB1) single-nucleotide polymorphisms on the disposition of sunitinib and SU12662, on clinical response, and on the manifestation of toxicities in Asian metastatic renal cell carcinoma patients. At week 4 of each treatment cycle, toxicities and plasma steady-state levels were assessed. Clinical response was assessed after two cycles. Genotyping was performed by using the PCR restriction fragment length polymorphism method. The CC genotype for ABCB1 was associated with a higher sunitinib exposure (76.81 vs 56.55 ng ml − 1, P = 0.03), higher risk of all-grade rash (RR 3.00, 95% CI 1.17–7.67) and mucositis (RR 1.60, 95% CI 1.10–2.34) and disease progression than compared with the CT/TT genotype. There was a lack of association observed between the CYP3A5 polymorphism and exposure, response and toxicities. The polymorphism of ABCB1 (C3435T) has an important role in the manifestation of toxicities and drug exposure, but not polymorphism of CYP3A5. The Pharmacogenomics Journal (2016) 16, 47–53; doi:10.1038/tpj.2015.13; published online 17 March 2015
INTRODUCTION Sunitinib is a multitargeted tyrosine kinase inhibitor, approved for the treatment of various cancers, including metastatic renal cell carcinoma (mRCC).1 In a multicentered randomized phase 3 trial that compared sunitinib with cytokine therapy (the previous gold standard), sunitinib increased the progression-free survival from 5 to 11 months,2 leading to its current recommendation as first-line therapy for mRCC.3 Despite the efficacy of sunitinib, at the approved dosage frequent dose modifications, including dose reductions (32–46%) and discontinuations (38%), have been reported as a result of systemic toxicities.2,4,5 In particular, several of these toxicities, such as hand–foot skin reaction and hematological toxicities, have been observed more frequently in Asians. Differences in the safety profile of sunitinib in Asian and non-Asian patients allude to the possible role of genetic variability, although environmental differences cannot be excluded.6 Disposition of sunitinib involves transportation through the ATP binding cassette subfamily B member 1 (ABCB1) transport protein and subsequent metabolism by the cytochrome P450 3A4 (CYP3A4) enzyme to its principal equipotent metabolite, SU12662. Both sunitinib and SU12662 display comparable activity in biochemical tyrosine kinase and cellular proliferation assays. One study has proposed a combined sunitinib and SU12662 concentration of more than 50 ng ml − 1 and less than 100 ng ml − 1 as the pharmacokinetic target for efficacy and safety.7 Similar to other tyrosine kinase inhibitors, sunitinib exhibits extensive variations in plasma concentration following standard dosage regimens, with wide interpatient variability observed.8 The variability observed with tyrosine kinase inhibitors may be contributed by genetic heterogeneity of drug targets, patients’
pharmacogenetic background (for example, CYP450 and ABC drug transporter polymorphisms), treatment adherence and environmental factors that influence pharmacokinetics.6 Thus, the disposition of sunitinib, which may be affected by variability of proteins in its pharmacokinetic pathway, could have a role in explaining the differences in response and in toxicities observed.9 Highly polymorphic genes that has a significant role in sunitinib’s mechanism may also have a role in explaining the variability in sunitinib exposure.10 One enzyme of interest is the CYP3A5 enzyme. Although both CYP3A5 and CYP3A4 enzymes share substrate specificity, their relative importance in overall CYP3A-mediated metabolism may differ between substrates.11 The single-nucleotide polymorphism (SNP) in CYP3A5 is of importance as the mutant CYP3A5*3 allele, which is associated with the defective CYP3A5 enzyme, is not uncommon in Asians.12 The defective CYP3A5 enzyme may cause an accumulation of the parent drug, which has been shown to be more dermatotoxic than its metabolite.13 Coupled with the fact that Asian race and low body weight decreases sunitinib clearance8, the defective CYP3A5 enzyme in Asians may exacerbate toxicities by further accumulating the parent drug. P-glycoprotein is a transmembrane efflux pump encoded by the ABCB1 gene. It is expressed in the intestines and liver and is involved in the oral absorption and biliary secretion of drugs. Although polymorphism with the ABCB1 gene (C3435T) is synonymous and there is no change in the amino acid of the resulting proteins, studies have demonstrated and suggested that it is likely to affect overall expression and function of the Pglycoprotein.14 It has been proposed that the TT genotype could lead to increase activity or affinity of the efflux pump for sunitinib,
1 Department of Pharmacy, Faculty of Science, National University of Singapore, Singapore, Singapore; 2Department of Pharmacy, National Cancer Centre Singapore, Singapore, Singapore and 3Department of Medical Oncology, National Cancer Centre Singapore, Singapore, Singapore. Correspondence: Dr A Chan or HK Ho, Department of Pharmacy, Faculty of Science, National University of Singapore, 18 Science Drive 4, Singapore 117543, Singapore. E-mail:
[email protected] or
[email protected] Received 8 September 2014; revised 18 November 2014; accepted 28 January 2015; published online 17 March 2015
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
48 hence leading to lower sunitinib levels. The TT genotype was also associated with a delay in dose reductions15 and increased sunitinib clearance.16 The rates of SNPs differ across various populations, and their influence on the variability of the safety profile observed with sunitinib remains to be determined. A focused population study was conducted to explore the effect of CYP3A5 and ABCB1 SNPs on the disposition of sunitinib and SU12662, on clinical response, and on the manifestation of toxicities in Asian mRCC patients.
Genotyping of SNPs
This was a prospective, longitudinal study conducted at the National Cancer Centre Singapore between June 2011 and October 2013. The study was approved by the SingHealth Centralized Institutional Review Board. The procedures were in accordance with the ethical standards of the responsible committee on human experimentation.
This study assessed ABCB1 SNPs (C3435T, rs1045642) and CYP3A5 SNPs (CYP3A5*1*3, rs776746) relevant to sunitinib transport and metabolism. DNA was extracted from the buffy coat layer using DNeasy Blood and Tissue Kit (Qiagen, Venlo, The Netherlands), performed according to manufacturer’s instructions. Genotyping was conducted with PCR restriction fragment length polymorphism. For CYP3A, the forward primer was 5′-CTT TAA AGA GCT CTT TTG TCT CTC A-3′ and the reverse primer was 5′-CCA GGA AGC CAG ACT TTG AT-3′.20 For ABCB1, the forward primer was 5′-TGT TTT CAG CTG CTT GAT GG-3’ and the reverse primer was 5′-AAG GCA TGT ATG TTG GCC TC-3′.21 PCR reaction conditions consisted of initial denaturation at 95 °C for 7 min, denaturation at 95 °C for 30 s, annealing at 60 °C for 20 s, elongation at 68 °C for 10 s, final elongation at 68 °C for 7 min and cooling at 12 °C. For the CYP3A5 and ABCB1 SNPs, digestion of the amplified product was performed with DdeI and Sau3A1 (New England BioLabs, USA), respectively. Digested DNA fragments were then loaded on a 2.5% w/v agarose gel in 1 × TAE buffer for 30 min at 120 V, followed by visualization of bands using Molecular Imager Gel DocTM XR System (Bio-Rad Laboratories, USA).
Patients and follow-up
Definitions and end points
All adult patients with a confirmed diagnosis of mRCC and who are newly starting on sunitinib were invited to participate. Patients need not be treatment-naive. Written informed consent was sought from the potential subjects prior to the inception of the study. Patients were excluded if they are unable to provide written informed consent, if they were not receiving sunitinib or if they were currently receiving sunitinib. During weeks 3–4 of each treatment cycle, when the steady state is achieved, a blood sample was drawn to assess the steady-state plasma levels of sunitinib and SU12662. Patients were followed up for up to a maximum of three cycles. Because of the outpatient setting of this study and taking into account the convenience and compliance of patients, and the feasibility of monitoring, only one sample was taken for each cycle.
Patients will be genotyped if they provided at least one blood sample for the whole study, regardless of the number of cycles completed. Two mutually exclusive groups were defined based on patients’ genotype. For CYP3A5 SNP, patients were categorized into *1*1/*1*3 genotype vs *3*3 genotype. For ABCB1 SNP, patients were categorized into CC genotype vs CT/TT genotype. The maximum Common Terminology Criteria for Adverse Events grade that was observed for each form of toxicity across the cycles was used for analysis. Average exposure between cycles was used to compare these two groups of patients. If a patient had completed only one cycle, then only data from one cycle will be used. Exposure levels were compared between the two groups (*1*1/*1*3 vs *3*3 and CC vs CT/TT). Patients who had provided blood samples for pharmacokinetic analysis and genotyping will be included in the analysis between genotype and drug exposure. Patients who completed at least two cycles, had their clinical response assessed at the end of the two cycles of sunitinib and had provided blood samples for genotyping will be included in the analysis for the association of clinical response with SNPs. Patients will be included in the analysis of the relative risk of toxicities if both genotype and toxicities data were available.
MATERIALS AND METHODS Study design
Data collection The blood sample obtained from each patient was used for both pharmacokinetic analysis and genotyping. Treatment-associated toxicity was assessed in conjunction with the scheduled blood sample, and graded according to Common Terminology Criteria for Adverse Events version 4.0.2.17 Assessment of clinical response was performed through a computed tomography scan at the end of the second cycle and was objectively evaluated by the investigators using the Response Evaluation Criteria in Solid Tumors.18 Treatment-related information was collected during every clinical visit. The following data was collected through patient interviews, case notes and electronic medical records: demographics, disease characteristics, medical history, treatment records and investigations, such as diagnostic scans and laboratory tests results. In the event when blood sampling was declined, toxicity, clinical response and other relevant study data were collected wherever possible.
Statistical analysis Descriptive statistics was utilized to summarize patients’ characteristics, incidence of toxicities and prevalence of SNPs. Mann–Whitney U test was used for determination of the relationship between SNPs and exposure. Χ2-test or Fishers’ exact test was performed to determine the association between SNPs with clinical response and toxicities. Genotype frequencies were also tested for Hardy–Weinberg equilibrium. Given the explorative nature of this study, the P-values were not corrected for multiple testing. All tests of statistical significance were two-sided with p-values of less than 0.05 and analysis was conducted with the IBM SPSS statistics 21 (Armonk, NY, USA).
Pharmacokinetic analysis Plasma samples were analyzed according to the validated methodology.19 The clearance (Cl) and volume of distribution (Vd), were estimated by the application of equations reported by Houk et al.8 Other pharmacokinetic parameters such as elimination rate constant (k) and subsequently maximum and minimum concentration at the steady state (Cmax,ss and Cmin,ss, respectively) were estimated according to equations from a one-compartmental model because of the lack of sampling points. As blood samples were obtained during ‘steady state’, peak and trough concentration was assumed to be the same for each dosing interval. After the plasma concentrations of sunitinib and SU12662 had been determined, and taking into account the time of blood collection (number of hours since the last sunitinib dose), the elimination rate constant (k) and the dosing interval (τ, 24 h), the Cmax,ss and the Cmin,ss were then determined. As sunitinib and metabolite demonstrates a linear dose– concentration relationship at doses up to 100 mg[ref. 1], all plasma levels were normalized to 37.5 mg to compensate for differences in doses between cycles and between patients. Total plasma drug (sunitinib and SU12662) concentration and sunitinib to metabolite ratio (SM ratio) were also estimated. The Pharmacogenomics Journal (2016), 47 – 53
RESULTS Patient demographics and disease characteristics A total of 36 patients were recruited for this study. Five patients did not start sunitinib therapy, therefore 31 patients started cycle 1, with 25 blood samples collected for genotyping. However, as one patient provided the blood sample before steady-state drug levels are achieved, this sample was excluded from the pharmacokinetic analysis. Five patients discontinued therapy after one cycle, and twenty-six patients continued with the second cycle. Blood samples for cycle 2 were obtained from 21 patients (Figure 1). The mean (± s.d.) age and weight of the patients were 58.9 ± 10-years old and 66.1 ± 13.5 kg, respectively. The majority of patients were male (80.6%) and of Chinese ethnicity (86.1%). More than half of the patients (63.9%) had presence of comorbidity and about two-thirds of patients (69.4%) had previously undergone nephrectomy. Pulmonary metastases were most commonly observed (75.0%). The majority of the patients (85.7%) were © 2016 Macmillan Publishers Limited
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
49 Table 1.
Patient characteristics (n = 36) n (%)
Age, years (mean ± s.d.) Weight, kg (mean ± s.d.) (range)
58.9 ± 10.0 66.1 ± 13.5 (32.4–93.2)
Gender Male Female
29 (80.6) 7 (19.4)
Ethnicity Chinese Malay Indian Presence of comorbidities Previous nephrectomy
31 3 2 23 25
(86.1) (8.3) (5.6) (63.9) (69.4)
Site of metastasis Lung Bone Adrenals Brain Liver Other
27 14 6 5 3 1
(75.0) (38.9) (16.7) (13.9) (8.3) (2.8)
Number of MSKCC risk factorsa 0 (Favorable) 1–2 (Intermediate) ⩾ 3 (Poor)
12 (34.3) 18 (51.4) 5 (14.3)
Abbreviation: MSKCC, Memorial Sloan–Kettering Cancer Centre. aRisk factors associated with shorter survival according to Memorial Sloan– Kettering Cancer Centre (MSKCC) risk classification.
Figure 1.
Distribution of patients.
classified as favorable or intermediate using the Memorial Sloan– Kettering Cancer Centre (MSKCC) risk factor model (Table 1). Toxicities observed with sunitinib therapy Sunitinib-associated toxicities were commonly experienced by the participating patients. The most common all-grade toxicities were hypertension (91.3%), fatigue (76.9%), anemia (73.3%), altered taste (73.1%), mucositis (73.1%), dry skin (65.4%) and hand–foot skin reaction (65.4%). The most common grade 2 and above toxicities were mucositis (65.3%), hand–foot skin reaction (61.5%) and hypertension (60.9%). Allelic and genotype frequencies The frequency of the CYP3A5*1 allele was 40% and frequency of the ABCB1 CC allele was 66%. Allelic frequencies were in Hardy– Weinberg equilibrium. The allelic frequencies were also largely similar to a previous paper conducted in the Singaporean population, where the reported frequency of CYP3A5*1 allele was 34%12 and frequency of the ABCB1 CC allele was 44%.22 The difference in the frequencies of the ABCB1 allele could be because of the small sample size in our study. For the CYP3A5 SNP, the genotype frequencies for *1*1, *1*3 and *3*3 were 20% (n = 5), 40% (n = 10) and 40% (n = 10), respectively. For ABCB1 SNP, the allele frequencies for CC, CT and TT were 36% (n = 9), 60% (n = 15) and 4% (n = 1), respectively. © 2016 Macmillan Publishers Limited
Exposure levels and SNPs There was no difference between the dose-normalized plasma concentrations of sunitinib, metabolite and total levels between all cycles (P40.05) (Table 2). When exposure levels were compared between the CYP3A5 *1*1/*1*3 genotype and the *3*3 genotype, there was no difference in the normalized sunitinib levels, total levels and SM ratio between the two groups for CYP3A5. On the other hand, normalized metabolite SU12662 levels were significantly higher in the *3*3 genotype than the other group (13.74 vs 9.82 ng ml − 1, P = 0.05). Similar trends were observed with the actual levels (Table 3). Patients who were CC genotype for ABCB1 were observed to have higher normalized sunitinib exposure than those who were CT/TT genotype (76.81 vs 56.55 ng ml − 1, P = 0.03). Likewise, the CC genotype group demonstrated higher SM ratio when the two groups were compared (7.89 vs 5.25, P = 0.02). However, no difference was observed between the two groups for total levels and metabolite levels. Similar trends were observed with the actual levels (Table 3).
Clinical response and SNPs Twenty-six patients completed at least two treatment cycles of sunitinib. Of the 26 patients, 4 (15.4%) achieved a partial response, 15 (57.7%) had a stable disease, while 7 (26.9%) had progressive disease. There was no association between clinical response and CYP3A5 genotype. However, the CC genotype of ABCB1 was associated with disease progression (P = 0.05). Patients expressing the CC genotype were 4.6 times more likely to experience disease progression compared with the patients expressing CT/TT genotype (Table 4). The Pharmacogenomics Journal (2016), 47 – 53
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
50 Table 2.
Exposure across cycles Cycle 1 (n = 25)
Cycle 2 (n = 20)
Cycle 3 (n = 12)
P
Cmin,ss(ng ml − 1) Sunitinib Metabolite Total
62.60 (43.15, 85.43) 11.12 (8.05, 17.85) 83.03 (50.75, 104.04)
65.00 (49.88, 87.15) 11.16 (7.08, 13.01) 68.85 (53.90, 91.92)
64.94 (50.11, 102.66) 11.97 (8.69, 21.94) 71.17 (60.22, 99.39)
0.84 0.81 0.87
Cmax,ss(ng ml − 1) Sunitinib Metabolite Total
96.05 (67.70, 128.11) 13.00 (9.45, 20.39) 119.93 (76.14, 152.30)
99.94 (74.51, 128.46) 14.89 (11.96, 20.18) 119.51 (87.94, 159.00)
98.76 (74.61, 152.68) 13.93 (10.08, 25.50) 116.89 (87.14, 167.22)
0.85 0.75 0.85
Table 3.
Exposure levels and SNPs
CYP3A5 Sunitinib, normalized Sunitinib, actual Metabolite, normalized Metabolite, actual Total, normalized Total, actual SM Ratio
*1*1/*1*3 (n = 15) 59.20 59.29 9.82 9.58 68.27 64.97 5.50
ABCB1 Sunitinib, normalized Sunitinib, actual Metabolite, normalized Metabolite, actual Total, normalized Total, actual SM Ratio
(45.66, 85.58) (39.79, 75.97) (8.29, 15.81) (5.98, 12.15) (53.95, 101.38) (47.13, 85.55) (4.10, 7.93)
74.67 70.15 13.74 13.74 89.65 85.93 5.32
CC (n = 8) 76.81 76.81 13.16 12.16 89.15 89.15 7.89
P
*3*3 (n = 9) (56.55, 87.19) (53.62, 87.19) (11.67, 18.90) (10.08, 18.90) (75.10, 115.75) (65.46, 115.75) (4.00, 6.64)
0.36 0.33 0.05 0.05 0.09 0.08 0.42 P
CT/TT (n = 16)
(64.49, 101.46) (64.49, 101.46) (9.07, 17.05) (9.07, 16.64) (80.05, 114.17) (80.05, 114.17) (5.40, 9.10)
56.55 49.85 11.13 10.40 69.10 58.87 5.25
(45.18, 76.65) (39.84, 63.57) (8.98, 15.29) (7.57, 13.35) (54.31, 98.45) (48.10, 83.74) (4.07, 6.31)
0.02 0.003 0.85 0.54 0.11 0.04 0.02
Abbreviation: SM ratio, Sunitinib to metabolite ratio. Note: All values are average exposure (Cmin,ss, ng ml − 1) between the cycles, normalized to 37.5 mg and reported as median (interquartile range).
Incidence of toxicities and SNPs There was no significant association between CYP3A5 SNPs and the incidence of toxicities, with the exception of anemia. Patients with the *1*1/*1*3 genotype had a lower risk of all-grade anemia than patients with the *3*3 genotype (RR 0.47, 95% CI 0.27–0.80). However, this was not observed for grade 2 and above anemia (Table 5). A significant association between ABCB1 SNP and incidence of all-grade rash was observed, with the incidence of rash being three times higher in patients who were homozygous CC wild type than in patients who were heterozygous CT or mutant TT (RR 3.00, 95% CI 1.17–7.67). CC genotype were also at a higher risk for both all-grade (RR 1.60, 95% CI 1.10–2.34) and grade 2 and above (RR 2.00, 95% CI 1.23–3.27) mucositis. There was no significant association between ABCB1 SNPs and incidence of other toxicities. DISCUSSION This is the first study to investigate the effect of SNPs on drug exposure and the consequent effect on toxicity and clinical response in Asian patients. It demonstrated that the ABCB1 polymorphism may affect sunitinib levels and the risk for toxicities, but that there was a lack of evidence associating the polymorphism of CYP3A5 to the same outcomes. Patients expressing the CC genotype for ABCB1 achieved higher sunitinib levels, and consequently higher SM ratio, than patients who were CT/TT genotype. The former group of patients was more likely than the latter group to develop rash and mucositis in the course of their The Pharmacogenomics Journal (2016), 47 – 53
Table 4.
Clinical response and SNPs
CYP3A5 *1*1/*1*3 (n = 14) *3*3 (n = 9) ABCB1 CC (n = 7) CT/TT (n = 16)
PD
PR+SD
RR (95% CI)
P
5 (35.7) 1 (11.1)
9 (64.3) 8 (88.9)
3.21 (0.45–23.21)
0.34
PD
PR+SD
RR (95% CI)
P
4 (57.1) 2 (12.5)
3 (42.9) 14 (87.5)
4.57 (1.08–19.42)
0.05
Abbreviations: CI, confidence interval; PD, disease progression; PR, partial response; RR, relative risk; SD, stable disease, SNP, single-nucleotide polypeptide.
sunitinib therapy. However, the CC genotype was also associated with disease progression (P = 0.05), with patients in this group 4.6 times more likely to experience disease progression than CT/TT patients. Metabolism of sunitinib was unaffected by variations in the CYP3A5 SNP. This may be attributed to the redundancy between CYP3A5 and CYP3A4 enzymes as the relative metabolizing capacity of CYP3A4 for sunitinib may greatly exceed, and adequately compensate for, any variability in the CYP3A5 enzyme. As sunitinib possesses an intermediate hepatic extraction ratio, the dependence of the clearance on P450 activity may not be as strong as a low hepatic extraction ratio drug. It has been proposed that sunitinib is a much better substrate of CYP3A4 compared with © 2016 Macmillan Publishers Limited
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
51 Table 5.
Relative risk of toxicities and SNPs CYP3A5 *1*1/ *1*3 (n = 15) vs *3*3 (n = 9) RR (95% CI)
All-grade a
Dermatological toxicity Dry skin HFSR Rash Pruritus
0.98 1.00 1.32 0.90
ABCB1 CC (n = 8) vs CT/TT (n = 16)
Grade 2 and above
All-grade
Grade 2 and above
(0.72–1.32) (0.56–1.79) (0.68–2.55) (0.35–2.35) NA
0.86 1.20 1.20 0.30
(0.52–1.41) (0.27–5.29) (0.61–2.38) (0.03–2.86) NA
1.23 1.20 1.20 3.00 2.00
(0.97–1.56) (0.69–2.08) (0.69–2.08) (1.17–7.67) (0.52–7.77)
1.09 2.00 1.33 1.00
(0.65–1.83) (0.52–7.77) (0.74–2.40) (0.11–9.44) NA
1.20 0.60 3.00 2.10 1.20
(0.61–2.38) (0.15–2.36) (0.41–21.76) (0.55–7.99) (0.13–11.43)
0.67 0.44 0.36 1.33
NA (0.32–1.41) (0.12–1.59) (0.11–1.26) (0.52–3.41)
0.73 0.40 1.00 0.57 4.00
(0.34–1.57) (0.06–2.88) (0.23–4.35) (0.15–2.15) (0.42–37.78)
2.00 2.67 2.00 2.00 3.00
(0.67–5.98) (0.78–9.15) (0.14–27.99) (0.34–11.70) (0.62–14.49)
2.00 (0.14–27.99) NA NA NA NA
Hematological toxicityb Anemia Leucopenia Neutropenia Thrombocytopenia
0.47 2.70 0.96 0.90
NA (0.27–0.80) (0.74–9.81) (0.45–2.04) (0.35–2.35)
Hepatotoxicityc Transaminitisd Increase in TB Increase in ALT Increase in AST
1.00 0.80 0.60 0.60 0.40
(0.31–3.22) (0.23–2.79) (0.04–8.46) (0.10–3.55) (0.08–1.96)
Gastrointestinal Mucositis
0.94 (0.59–1.50)
1.00 (0.56–1.79)
1.60 (1.10–2.34)
2.00 (1.23–3.27)
Constitutional Fatigue
0.83 (0.56–1.21)
0.60 (0.10–3.55)
0.92 (0.58–1.47)
N.A.
Cardiac Increase in BP
1.14 (0.88–1.49)
2.00 (0.77–5.18)
1.08 (0.93–1.25)
1.67 (0.88–3.14)
Neurology Altered taste
0.68 (0.42–1.08)
1.00 (0.31–3.22)
1.09 (0.65–1.83)
0.29 (0.04–1.94)
NA NA NA NA NA
Abbreviations: ALT, alanine transaminase; AST, aspartate transaminase; BP, blood pressure; CI, confidence interval; HFSR, hand–foot skin reaction; RR, relative risk; TB, total bilirubin. aIncludes dry skin, HFSR, rash and pruritus. bIncludes anemia, leucopenia, neutropenia and thrombocytopenia. cIncludes elevation of TB, ALT and AST. dIncludes elevation of ALT and AST.
CYP3A5, and therefore the contribution of CYP3A5 to the metabolism of sunitinib may be low.11 A recent study has also suggested that CYP3A5 SNPs do not affect the clearance of sunitinib and SU12662.16 This supports our findings that the CYP3A5 polymorphism may not be an important contributor to the disposition of sunitinib. One other study has demonstrated that the presence of a copy of the *1 allele in CYP3A5 is associated with an increased risk of dose reduction due to toxicity.23 For that reason, we conducted an additional analysis to evaluate whether CYP3A5*1 carriers were more likely to have their dose reduced, but did not observe such an association. Although CYP3A5 plays a minimal role in the disposition of sunitinib, several studies have highlighted the contributions of CYP3A4 in sunitinib exposure. Midazolam exposure was found to be highly correlated with sunitinib exposure and this accounted for 51% of the observed interpatient variability in sunitinib pharmacokinetics.24 A further study in a Caucasian population has also found that the clearance of sunitinib is affected by the CYP3A4*22 polymorphism.16 These findings further support CYP3A4 as the main contributor to sunitinib metabolism and the reduced dependence on CYP3A5. However, the CYP3A4*22 polymorphism was not detected in Asians so it cannot explain the differences in exposure in Asians.25 In addition, patients with the *3*3 polymorphism possessed higher median metabolite levels than the *1*1/*1*3 group. This can be attributed to one patient in the former group displaying an extremely high metabolite level, which skewed the results. In this patient, the level was 144 ng ml − 1, whereas the metabolite level in the rest of the patients ranged from 4 to 26 ng ml − 1. When the outlier was excluded from the *3*3 group, the statistical difference was no longer present. © 2016 Macmillan Publishers Limited
Our results demonstrated that the ABCB1 CC genotype is correlated with a higher sunitinib level and SM ratio. The functional effect of the ABCB1 polymorphism on P-glycoprotein has been heavily debated,26 with the 3435 TT genotype associated with both increased27–29 and decreased14,30–32 expression of P-glycoprotein. The lack of a clear conclusion could be reflected by the limitations of the studies, such as retrospective design, small sample size and lack of population stratification and so on.26 It has been shown that the TT, TC and CC genotypes for ABCB1 affect sunitinib clearance by +7, +3 and − 11.5%, respectively.16 This is in line with our results in which the CC group displayed higher levels of sunitinib, possibly because of a decrease in its clearance. Patients exhibiting the CC genotype had an increased risk for allgrade rash (three-fold increase), all-grade mucositis (1.6-fold increase) and grade 2 and above mucositis (two-fold increase) compared with the CT/TT patients. This is not surprising, as the accumulation of sunitinib in the former group of patients could have resulted in the higher exposure and occurrence of toxicity. This is further supported by our previous findings in which sunitinib exhibited a greater dermatological toxicity than its metabolite.13 Despite the higher levels of exposure observed, the CC group was more likely to experience disease progression than the CT/TT genotype. Nevertheless, our previous study demonstrated that 84–92% of the patients managed to achieve the minimum target concentration of 50 ng ml − 1, which may explain the lack of association between clinical response and exposure levels (unpublished data). There was also no statistically significant difference between the CC and the CT/TT groups in terms of clinical characteristics, such as the MSKCC risk factors, age and The Pharmacogenomics Journal (2016), 47 – 53
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
52 weight, suggesting that there may be other factors, such as drug resistance, influencing the clinical response of sunitinib. As tumor type (mRCC, GIST and other solid tumor) accounted for a major portion of the variability in the clearance of sunitinib and its metabolite,8 our study only included patients with mRCC. The limitations of our study included limited sampling points and small sample size. Studies conducted in Korea and Japan have demonstrated an association between the ABCG2 polymorphism, toxicities such as thrombocytopenia, hand–foot skin reaction and hypertension,33 and the pharmacokinetics of sunitinib.34 The polymorphic effects of ABCG2, which encodes for the breast cancer resistance protein, have also been very recently identified as a significant covariate for the prediction of sunitinib clearance.34 To confirm such associations, future studies should be performed in our local South-East Asian population, which is made up of different ethnicities from Korea and Japan. As all the patients in our study used the AD regimen and demonstrated a lower incidence and severity of toxicities, the effects of SNPs on severe toxicities may have been downplayed. Future studies involving a larger sample should be conducted to confirm these findings. If the role of the ABCB1 polymorphism on exposure and toxicities is confirmed, genotyping for ABCB1 and therapeutic drug monitoring should be considered in clinical practice to aid in the personalization of drug therapy. Even with the AD sunitinib, our patients managed to achieve the sufficient exposure to sunitinib and experience frequent toxicities, thus further dose reductions may be necessary in this group of patient. Thus, therapeutic drug monitoring may complement clinical evaluation by providing additional information on efficacy, adherence and toxicity. Furthermore, since patients of CC genotype had a higher likelihood of disease progression and toxicities, cautious dose titration with the aid of therapeutic drug monitoring can help to ensure that sunitinib levels are within the therapeutic range, as subtherapeutic levels would further increase the risk for disease progression. By ensuring that the optimal dose is prescribed to the right patient according to the genotype, we can achieve maximum efficacy with minimal toxicity. In conclusion, polymorphism of ABCB1 may be associated with drug exposure and manifestation of toxicities in Asian mRCC patients. Although CYP3A5 is highly polymorphic in Asian population, the CYP3A5 polymorphism may be excluded as a cause of high drug exposure and/or toxicity as this enzyme has a minimal role in the disposition of sunitinib. CONFLICT OF INTEREST Drs Kanesvaran, Wee and Chan serve as consultants to GlaxoSmithKline (Kanesvaran and Chan), Pfizer (Hwee Lin Wee), Novartis Asia Pacific (Hwee Lin Wee) and Abbvie Pte (Hwee Lin Wee). The remaining authors declare no conflict of interest.
ACKNOWLEDGMENTS We acknowledge the Singapore Cancer Society for the provision of grant support for this work.
REFERENCES 1 Pfizer. Sutent Prescribing Information 2012. 2 Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Rixe O et al. Sunitinib versus Interferon Alfa in metastatic renal-cell carcinoma. N Engl J Med 2007; 356: 115–124. 3 Clinical Practice Guidelines in Oncology, Kidney Cancer. National Comprehensive Cancer Network (NCCN) 2014. Available from www.NCCN.org. 4 Gore ME, Szczylik C, Porta C, Bracarda S, Bjarnason GA, Oudard S et al. Safety and efficacy of sunitinib for metastatic renal-cell carcinoma: an expanded-access trial. Lancet Oncol 2009; 10: 757–763. 5 Yoo C, Kim JE, Lee JL, Ahn JH, Lee DH, Lee JS et al. The efficacy and safety of sunitinib in Korean patients with advanced renal cell carcinoma: high incidence of toxicity leads to frequent dose reduction. Jpn J Clin Oncol 2010; 40: 980–985.
The Pharmacogenomics Journal (2016), 47 – 53
6 Lee SH, Bang YJ, Mainwaring P, Ng C, Chang JWC, Kwong P et al. Sunitinib in metastatic renal cell carcinoma: an ethnic asian subpopulation analysis for safety and efficacy. Asia Pac J Clin Oncol 2014; 10: 237–245. 7 Yu H, Steeghs N, Nijenhuis CM, Schellens JHM, Beijnen JH, Huitema ADR. Practical guidelines for therapeutic drug monitoring of anticancer tyrosine kinase inhibitors: focus on the pharmacokinetic targets. Clin Pharmacokinet 2014; 53: 305–325. 8 Houk BE, Bello CL, Kang D, Amantea M. A population pharmacokinetic metaanalysis of sunitinib malate (SU11248) and its primary metabolite (SU12662) in healthy volunteers and oncology patients. Clin Cancer Res 2009; 15: 2497–2506. 9 Houk BE, Bello CL, Poland B, Rosen LS, Demetri GD, Motzer RJ. Relationship between exposure to sunitinib and efficacy and tolerability endpoints in patients with cancer: Results of a pharmacokinetic/pharmacodynamic meta-analysis. Cancer Chemother Pharmacol 2010; 66: 357–371. 10 Kuehl P, Zhang J, Lin Y, Lamba J, Assem M, Schuetz J et al. Sequence diversity in CYP3A promoters and characterization of the genetic basis of polymorphic CYP3A5 expression. Nat Genet 2001; 27: 383–391. 11 Sugiyama M, Fujita K, Murayama N, Akiyama Y, Yamazaki H, Sasaki Y. Sorafenib and sunitinib, two anticancer drugs, inhibit CYP3A4-mediated and activate CY3A5-mediated midazolam 1'-hydroxylation. Drug Metab Dispos 2011; 39: 757–762. 12 Balram C, Zhou Q, Cheung YB, Lee EJD. CYP3A5*3 and *6 single nucleotide polymorphisms in three distinct Asian populations. Eur J Clin Pharmacol 2003; 59: 123–126. 13 Teo YL, Chong XJ, Chue XP, Chau NM, Tan MH, Kanesvaran R et al. Role of sunitinib and SU12662 on dermatological toxicities in metastatic renal cell carcinoma patients: In vitro, in vivo, and outcomes investigation. Cancer Chemother Pharmacol 2014; 73: 381–388. 14 Hoffmeyer S, Burk O, Von Richter O, Arnold HP, Brockmöller J, Johne A et al. Functional polymorphisms of the human multidrug-resistance gene: Multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proc Natl Acad Sci USA 2000; 97: 3473–3478. 15 Beuselinck B, Lambrechts D, Van Brussel T, Wolter P, Cardinaels N, Joniau S et al. Efflux pump ABCB1 single nucleotide polymorphisms and dose reductions in patients with metastatic renal cell carcinoma treated with sunitinib. Acta Oncol 2014; 53: 1413–1422. 16 Diekstra MHM, Klümpen HJ, Lolkema MPJK, Yu H, Kloth JSL, Gelderblom H et al. Association analysis of genetic polymorphisms in genes related to sunitinib pharmacokinetics, specifically clearance of sunitinib and SU12662. Clin Pharmacol Ther 2014; 96: 81–89. 17 Common Terminology Criteria for Adverse Events (CTCAE). National Institutes of Health, National Cancer Institute 2009. 18 Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228–247. 19 Etienne-Grimaldi MC, Renee N, Izzedine H, Milano G. A routine feasible HPLC analysis for the anti-angiogenic tyrosine kinase inhibitor, sunitinib, and its main metabolite, SU12662, in plasma. J Chromatogr B Analyt Technol Biomed Life Sci 2009; 877: 3757–3761. 20 Fukuen S, Fukuda T, Maune H, Ikenaga Y, Yamamoto I, Inaba T et al. Novel detection assay by PCR-RFLP and frequency of the CYP3A5 SNPs, CYP3A5* 3 and * 6, in a Japanese population. Pharmacogenetics 2002; 12: 331–334. 21 Cizmarikova M, Wagnerova M, Schonova L, Habalova V, Kohut A, Linkova A et al. MDR1 (C3435T) polymorphism: Relation to the risk of breast cancer and therapeutic outcome. Pharmacogenomics J 2010; 10: 62–69. 22 Balram C, Sharma A, Sivathasan C, Lee EJ. Frequency of C3435T single nucleotide MDR1 genetic polymorphism in an Asian population: phenotypic-genotypic correlates. Br J Clin Pharmacol 2003; 56: 78–83. 23 Garcia-Donas J, Esteban E, Leandro-García LJ, Castellano DE, del Alba AG, Climent MA et al. Single nucleotide polymorphism associations with response and toxic effects in patients with advanced renal-cell carcinoma treated with first-line sunitinib: a multicentre, observational, prospective study. Lancet Oncol 2011; 12: 1143–1150. 24 De Wit D, Gelderblom H, Sparreboom A, Den Hartigh J, Den Hollander M, KönigQuartel JMC et al. Midazolam as a phenotyping probe to predict sunitinib exposure in patients with cancer. Cancer Chemother Pharmacol 2014; 73: 87–96. 25 Elens L, Van Gelder T, Hesselink DA, Haufroid V, Van Schaik RHN. CYP3A4*22: promising newly identified CYP3A4 variant allele for personalizing pharmacotherapy. Pharmacogenomics 2013; 14: 47–62. 26 Leschziner GD, Andrew T, Pirmohamed M, Johnson MR. ABCB1 genotype and PGP expression, function and therapeutic drug response: a critical review and recommendations for future research. Pharmacogenomics J 2007; 7: 154–179. 27 Kim RB, Leake BF, Choo EF, Dresser GK, Kubba SV, Schwarz UI et al. Identification of functionally variant MDR1 alleles among European Americans and African Americans. Clin Pharmacol Ther 2001; 70: 189–199.
© 2016 Macmillan Publishers Limited
Effect of the CYP3A5 and ABCB1 genotype YL Teo et al
53 28 Nakamura T, Sakaeda T, Horinouchi M, Tamura T, Aoyama N, Shirakawa T et al. Effect of the mutation (C3435t) at exon 26 of the MDR1 gene on expression level of MDR1 messenger ribonucleic acid in duodenal enterocytes of healthy Japanese subjects. Clin Pharmacol Ther 2002; 71: 297–303. 29 Illmer T, Schuler US, Thiede C, Schwarz UI, Kim RB, Gotthard S et al. MDR1 gene polymorphisms affect therapy outcome in acute myeloid leukemia patients. Cancer Res 2002; 62: 4955–4962. 30 Siegsmund M, Brinkmann U, Schäffeler E, Weirich G, Schwab M, Eichelbaum M et al. Association of the P-glycoprotein transporter MDR1C3435T polymorphism with the susceptibility to renal epithelial tumors. J Am Soc Nephrol 2002; 13: 1847–1854. 31 Hitzl M, Drescher S, Van der Kuip H, Schäffeler E, Fischer J, Schwab M et al. The C3435T mutation in the human MDR1 gene is associated with altered efflux of the
© 2016 Macmillan Publishers Limited
P-glycoprotein substrate rhodamine 123 from CD56+ natural killer cells. Pharmacogenetics 2001; 11: 293–298. 32 Wang D, Johnson AD, Papp AC, Kroetz DL, Sadée W. Multidrug resistance polypeptide 1 (MDR1, ABCB1) variant 3435C4T affects mRNA stability. Pharmacogenet Genomics 2005; 15: 693–704. 33 Kim HR, Park HS, Kwon WS, Lee JH, Tanigawara Y, Lim SM et al. Pharmacogenetic determinants associated with sunitinib-induced toxicity and ethnic difference in Korean metastatic renal cell carcinoma patients. Cancer Chemother Pharmacol 2013; 72: 825–835. 34 Mizuno T, Fukudo M, Fukuda T, Terada T, Dong M, Kamba T et al. The effect of ABCG2 genotype on the population pharmacokinetics of sunitinib in patients with renal cell carcinoma. Ther Drug Monit 2014; 36: 310–316.
The Pharmacogenomics Journal (2016), 47 – 53
