1. Role of CYP24A1, VDR and GC gene polymorphisms on deferasirox pharmacokinetics and clinical outcomes.
Sarah Allegra1
Silvia De Francia2, Jessica Cusato1, Arianna Arduino1, Elisa Pirro2, Davide Massano3, Filomena Longo3, Antonio Piga3 and Antonio D'Avolio1.
1. Unit of Infectious Diseases, University of Turin, Department of Medical Sciences, Amedeo di Savoia Hospital, 10149, Turin, Italy.
2. Department of Biological and Clinical Sciences, University of Turin, S. Luigi Gonzaga Hospital, 10043, Orbassano (Turin), Italy.

2. FREQUENT RED BLOOD CELLS TRANFUSIONS
Β-THALASSEMIA MAJOR
FREQUENT RED BLOOD CELLS TRANFUSIONS
IRON
OVERLOAD
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

3. Advantages and disadvantages of different modalities to assess iron overload.
Chelation therapy
It aims to remove iron excess, thus to prevent any further accumulation, improving the survival probability and reducing the risk for developing co-morbidities in transfused patients.
Chelators mobilize iron, making it soluble and excretable through urine or faeces.
Measured through a drawn blood;
Easy and low expensive;
Its levels correlate with total body iron burden;
It is still used to define adequate and inadequate responders to chelation treatment.
whitdrawal
Paul C. Kruger et al, Clin Cancer Res. 2012;18:
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

4. Overview of Iron Chelators
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

5. Deferasirox BACKGROUND AIMS RESULTS CONCLUSIONS Absorption:
Median Tmax of about 0.5 to 1 hours
High bioavailability (approx. 70%)
Distribution:
High protein bound (99%)
Small Vd (14.37±2.69 L)
Metabolism: in liver by
UGT1A1
CYP1A1, CYP1A2 and CYP2D6
Elimination:
In bile (84% of the dose) through MRP2 and BCRP1
In urine (8% of the dose)
Proposed scheme for the disposition of deferasirox in humans after peroral administration. ST, sulfotransferase; Gluc., glucuronic acid conjugate; Sulf., sulfate conjugate.
Waldmeier et al., Drug Metabolism and Disposition. 2010; 38 (5) 
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

6. DFX has a dose-dependent effect, however…
Patients treated with DFX, especially those heavily iron loaded, don’t achieve adequate iron chelation and a negative iron balance, even when receiving DFX doses exceeding 30 mg/kg/day (poor responders).
Others may experience DFX related adverse events at the dose required to maintain the iron burden balance (intolerant patients).
Therefore an high inter-individual variability of DFX exposure may occur, leading to inadequate chelation treatment or to a toxicity increase.
Nisbet-Brown E. et al, Lancet. 2003; 361(9369):
Chirnomas D. et al, Blood. 2009; 114(19):
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

7. Inadequate responders
Ferritin < 1000 ng/mL
Administration ≤ 30 mg/Kg/day.
Inadequate responders
Ferritin > 1500 ng/mL
Administration
> 30 mg/Kg/day.
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

8. BACKGROUND AIMS RESULTS CONCLUSIONS
Efficacy cutoff:
Plasma Ctrough: 20 µg/mL  negatively predict by CYP1A1 rs AA and CYP1A2 rs TT
Plasma AUC: 360 µg/mL/h positively predict by UGT1A3 rs GG and BCRP1 rs CC
Non response cutoff:
Plasma AUC: 250 µg/mL/h negatively
predict by UGT1A3 rs GG
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

9. Low Vitamin D levels in β-thalassemia patients
Interference with Vitamin D skin synthesis
Reduced renal absorption
Subicteric tint and/or the iron induced higher pigmentation
Liver impairment
Siderosis
Limited hepatic hydroxylation
Negative correlation between 25(OH)D3 and serum ferritin
Iron overload induced Vitamin D deficiency
Dimitriadou M, Christoforidis A, Fidani L, Economou M, Vlachaki E, Athanassiou-Metaxa M, et al. A 2-year prospective densitometric study on the influence of Fok-I gene polymorphism in young patients with thalassaemia major. Osteoporos Int Aug 15.
Moulas A, Challa A, Chaliasos N, Lapatsanis PD. Vitamin D metabolites (25-hydroxyvitamin D, 24,25-dihydroxyvitamin D and 1,25-dihydroxyvitamin D) and osteocalcin in beta-thalassaemia. Acta Paediatr Jun;86(6):594-9.
Napoli N, Carmina E, Bucchieri S, Sferrazza C, Rini GB, Di Fede G. Low serum levels of 25-hydroxy vitamin D in adults affected by thalassemia major or intermedia. Bone Jun;38(6):
Wood JC, Claster S, Carson S, Menteer JD, Hofstra T, Khanna R, et al. Vitamin D deficiency, cardiac iron and cardiac function in thalassaemia major. Br J Haematol Jun;141(6):891-4.

10. Vitamin D metabolism
We investigated whether polymorphisms within the genes:
CYP27B1 (2838 and -1260),
CYP24A1 (22776, 3999 and 8620),
GC (1296)
VDR (TaqI, FokI, BsmI, Cdx2 and ApaI ),
could play a role in
DFX pharmacokinetics
outcome cut-offs prediction.
Gene name
Role on Vitamin D metabolism
Protein Function
CYP2R1
ACTIVATION
25-hydroxylase
CYP27A1
CYP27B1
1α-hydroxylase
VDR
RECEPTOR
Receptor GC
BINDING PROTEIN
Transporter
CYP24A1
INACTIVATOR
24-hydroxylase
BACKGROUND
AIMS
RESULTS
CONCLUSIONS
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

11. Patients & METHODS
We performed a monocentric cohort study in β-thalassemic patients treated at Hemoglobinopathies Centre of San Luigi Gonzaga University Hospital in Orbassano (Turin).
Inclusion criteria were:
β-thalassemia disease with transfusional iron overload;
DFX treatment for at least 6 months with a self-reported adherence of 90%;
Age above 18 years old.
Study protocol (“Studio dei determinanti farmacogenetici nella farmacocinetica e nella risposta clinica del deferasirox”, registration number 79/2012) was approved by the local Ethics Committee. A written informed consent for the study was obtained from each subject or it was signed by natural/biological father or mother of a child with full parental legal rights.
Pharmacokinetic analysis
Plasma DFX concentrations were determined from samples obtained at the end of dosing interval (Ctrough) and after 0, 2, 4, 6 and 24 hours drug administration.
A fully validated high performance liquid chromatography coupled with an ultraviolet detection (HPLC-UV) method for the quantification of plasma DFX levels.
Pharmacogenetic analysis
DNA was extracted using the QIamp DNA Mini Kit (Quiagen).
Purified and eluted DNA was directly used for the real-time PCR (BIORAD) reaction. The allelic discrimination analysis was performed using the TaqMan assays (Applied Biosystems).
Statistical analyses
Kruskal-Wallis and Mann-Whitney tests have been used to compare pharmacokinetic parameters and SNPs, considering the level of statistical significance (p value <0.05).
Any predictive power of the considered variables was evaluated through univariate and multivariate linear regression analyses. Factors (β coefficient; IC, interval of confidence at 95%) with p value <0.2 in univariate analysis were considered in multivariate analysis (p value <0.05).

12. BACKGROUND AIMS RESULTS CONCLUSIONS
Variable
All patients (N=99)
AUC patients (N=58)
Gender
Male, n (%)
53 (53.5)
34 (58.6)
Female, n (%)
46 (46.5)
24 (41.4)
Age (years)
Median (IQR)
34 (18-53)
Body mass index (BMI) Kg/m2
22 (15-32)
21.95 (15-32)
Ethnicity
Caucasian, n (%)
93 (93.9)
55 (94.8)
Other, n (%)
6 (6.1)
3 (5.2)
Serum ferritin ng/mL
1139 ( )
( )
Efficacy
Responders, n (%)
27 (27.3)
16 (27.6)
Non responders, n (%)
72 (72.7)
42 (72.4)
DFX plasmatic AUC < 250 μg/mL/h
26 (44.8)
DFX plasmatic AUC > 360 μg/mL/h )
DFX dose mg/day/Kg
29 ( )
28 (17-40)
DFX plasmatic concentration at the end of dosing interval (Ctrough) μg/mL
10.36 ( )
 4.98 ( )
DFX plasmatic AUC μg/mL/h
( )
DFX plasmatic volume of distribution mL
( )
DFX plasmatic half-life hours
7.51 (22.51)
DFX plasmatic maximum concentration μg/mL
22.51 ( )
Time to reach DFX maximum concentration hs
4 (2-6)
Demographic, clinical and pharmacokinetic characteristics of β-thalassemia patients (N=99) and of AUC patients subgroup (N=58).
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

13. BACKGROUND AIMS RESULTS CONCLUSIONS
Analyzed SNPs characteristics: gene, variant name (SNP name), rs number, position, alleles, observed heterozygous (ObsHET), predicted heterozygous (predHET), Hardy-Weinberg equilibrium χ2 (HW p-value) and minimum allele frequency (MAF), considering all our cohort (N=99) and AUC patients subgroup (N=58).
ALL PATIENTS (N=99)
AUC PATIENTS (N=58)
Gene
SNP name rs
Position
Alleles
ObsHET
PredHET
HW p-value
MAF
CYP27B1
+2838 rs
C>T
0.323
0.397
0.1038
0.273
0.345
0.348 1
0.224
-1260 rs
G>T
0.333
0.382
0.2846
0.258
0.362
0.357
0.233
CYP24A1
+8620 rs
T>C
0.545
0.494
0.4309
0.444
0.500
0.496
0.457
+22776
rs927650
A>G
0.495
0.490
0.431
0.499
0.4015
0.474
+3999 rs
0.424
0.478
0.3406
0.394
0.475
0.6182
0.388
VDR
ApaI rs
C>A
0.455
0.497
0.4871
0.460
0.466
0.7564
TaqI
rs731236
0.485
0.4947
0.414
0.467
0.371
FokI rs
0.354
0.418
0.1769
0.298
0.379
0.400
0.8835
0.276
BsmI rs
G>A
0.434
0.3280
0.429
0.482
0.9537
0.405
Cdx2 rs
0.343
0.422
0.0960
0.303
0.259
0.0771
VDBP
+1296
rs7041
T>G
0.6130
0.369
0.135
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

14. Effect of VDR, CYP24A1, CYP27B1 and GC SNPs on plasma DFX concentration at the end of dosing interval (Ctrough)
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

15. MULTIVARIATE REGRESSION ANALYSIS
Effect of VDR, CYP24A1, CYP27B1 and GC SNPs on plasma DFX area under the concentration curve (AUC)
MULTIVARIATE REGRESSION ANALYSIS
FACTOR
p;β (IC95%)
CYP24A GG
0.031;
( ; )
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

16. Effect of VDR, CYP24A1, CYP27B1 and GC SNPs on plasma DFX half life (t1/2)
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

17. MULTIVARIATE REGRESSION ANALYSIS
Effect of VDR, CYP24A1, CYP27B1 and GC SNPs on plasma DFX minimum serum concentration (Cmin)
MULTIVARIATE REGRESSION ANALYSIS
FACTOR
p;β (IC95%)
CYP24A TT
0.036;
(-23.98; -0.81)
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

18. Factors, in univariate and multivariate logistic regression analyses,
able to predict DFX AUC efficacy cut-off (360 µg/mL/h).
UNIVARIATE
MULTIVARIATE
FACTOR
GENOTYPE
p;β (IC95%)
Age
0.639; (0.27; 2.62)
Gender
0.623; (0.26; 2.36)
BMI
0.955; (0.33; 2.85)
Ethnicity
0.999; 0 (0-0)
DFX dose (mg/Kg/day)
0.342; (0.58; 4.89)
CYP27B
TT 
0.689; 1.25 (0.42; 3.73)
CYP27B
GT/TT  
0.623; (0.45; 3.83)
CYP24A1 8620
 GG
0.049; (0.06; 1.00)
0.210; (0.08; 1.74)
CYP24A
CT/TT 
0.207; (0.16; 1.50)
CYP24A1 3999
TC/CC 
0.124; (0.08; 1.35)
0.061; (0.05; 1.07)
FokI
0.402; (0.27; 25.14)
Cdx2
AG/GG 
0.146; (0.04; 1.59)
0.022; (0.01; 0.71)
ApaI
AA 
0.790; (0.26; 2.76)
TaqI
CC 
0.423; (0.09; 2.73)
BsmI
0.572; (0.15; 2.85)
GC 1296
TG/GG 
0.027; (0.09; 0.87)
0.010; (0.05; 0.66)
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

19. Cmin (p=0.011)
significantly influenced
CYP24A TT
CYP24A C>T (rs927650), 3999 T>C (rs ) and 8620 A>G (rs ) variants are located in intronic region; therefore, they could create an alternative splice site which competes with the normal one during the RNA processing; this results in a part of mature messenger RNA with improperly spliced intron sequences.
22776 C>T : in literature there are not evidence about this SNP and drugs pharmacokinetic;
3999 T>C is located in a promoter associated region, thus it probably operates by changing gene expression. C allele resulted in enhanced mRNA expression and calcitriol-inactivation, leading to decreased VDR activity;
8620 GG genotype resulted as negative predictor of plasma ETA Cmax, suggesting an enhanced drug metabolism, leading to its minor extracellular concentration.
negatively predicted
Cmin (p=0.036)
significantly influenced
Ctrough (p=0.010)
Cmin (p=0.40)
CYP24A CC
Ctrough (p=0.047)
AUC (p=0.021)
t1/2 (p=0.031)
Cmin (p=0.010)
significantly influenced
CYP24A GG
Cdx2 is located in the binding site of transcription factor Cdx2. Its A to G base substitution eliminates the Cdx binding site and reduces transcriptional activity of VDR to 70% of the A allele.
GC 1296 T>G (rs7041, Glu432Asp) SNP could result in altered rates of transcription, changes in mRNA stability or in a self-clearance of the protein.
negatively predict ed
AUC (p=0.031)
Cdx2 AG/GG
GC 1296 TG/GG
AUC efficacy cut-off
negatively predicted
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

20. CONCLUSION PERSONALIZED MEDICINE
Inter-individual variation in drug metabolism and in pathways involved in its efficacy and toxicity mechanisms are the major contributor to treatment response and safety.
To our knowledge, this is the first study to date that focus on role of
vitamin D pharmacogenetics and iron chelation therapy.
Larger studies, incorporating vitamin D serum levels, are warranted.
In our opinion, the genetic factors should be taken in consideration in the routine clinical practice for the optimization of drug therapies.
PERSONALIZED MEDICINE
• Shift the emphasis in medicine from reaction to prevention
• Direct the selection of optimal therapy and reduce trial-and-error prescribing • Help avoid adverse drug reactions • Increase patient adherence to treatment • Improve quality of life
• Reveal additional or alternative uses for medicines and drug candidates • Help control the overall cost of health care
BACKGROUND
AIMS
RESULTS
CONCLUSIONS

21. “Personalized medicine is our chance to revolutionize health care, but it will require a team effort by innovators, entrepreneurs, regulators, payers, and policymakers.”
Brook Byers

22. ACKNOWLEDGEMENT Dr. Antonio D’Avolio Prof. Antonio Piga
Prof. Giovanni Di Perri
Dr. Silvia De Francia
Dr. Jessica Cusato
Dr. Filomena Longo
Dr. Arianna Arduino
Dr. Elisa Pirro
Dr. Davide Massano
The Laboratory Team
Dr. Amedeo De Nicolò
Dr. Giovanna Fatiguso
Dr. Fabio Favata
Dr. Luca Paglietti
Dr. Valeria Avataneo
Dr. Alessandra Ariaudo
Dr. Marco Simiele
Dr. Debora Pensi
Dr. Cristina Tomasello
Dr. Mauro Sciandra
…and all the students
Laboratory of Clinical Pharmacology and Pharmacogenetics
UNI EN ISO 9001:2008 Certified Laboratory