PharmacogeneticsPharmacogenetics in Breast in Breast PharmacogeneticsPharmacogenetics in Breast in Breast CancerCancer

Giampietro Gasparini, MDGiampietro Gasparini, MDSan Filippo Neri Hospital- Rome

TUMOR PATIENT

Few histotypes • Age and PSFew histotypesbut

more molecular

• Age and PS

• Comorbidities

• Compliancediseases

Pharmacogenomics

• Ethnics

• GeneticsDRUG

gPharmacokinetics Pharmacodynamics

Pharmacogeneticsg

MOLECULAR PORTRAIT OF BREAST CANCERHER2HER2 NormalNormal Luminal BLuminal B

Basal likeBasal likeHER2HER2 NormalNormal Luminal BLuminal B

Luminal ALuminal A

Above average

Average

Below average

Not determined

476 cDNA clones

85 arrays

R lt….. ResultsIdentification of 5 breast cancer subgroups with different prognosisIdentification of nonresponder patientsIdentification of nonresponder patients

….but

No predictive activity of absolute value (different outcome in selected “responsive” patients)

No predictive determinants of toxicity

??

Drugs Work Better for Some People Drugs Work Better for Some People Than for OthersThan for Others

Factors influencingdifferent response:different response:

Pharmacokinetics

Environmental

Genetics

Toward Personalized Therapy

• Tumor genetic profiling

• Pharmacogenetics

• Therapeutic drug monitoring• Therapeutic drug monitoring

PharmacokineticsPredictive indicatorsDrug-to-drug interactions

Pharmacogenetics

Pharmacogenetics may help in understanding some of the differences in therapeutic activity and

toxicity of anticancer drugs

Pharmacology GenomeDrugsDrugs

GenomeGenome

Interethnic Genetic Differences in Activity and Toxicity to Anticancer Drugs

Polymorphisms : Structural modifications of DNA with a frequency ≥ 1%DNA with a frequency ≥ 1%

Pharmacogenetics Study of Patient’s Genetic Polymorphisms

90% SNPs (Single nucleotide polymorphisms)

10%Insertions and deletionsTandem repeats/microsatellites

short (2bp)long (20-30 o più)o g ( 0 30 o p ù)

Genetic duplicationsPseudogenes

Untrascripted esonic sequences with high homology withUntrascripted esonic sequences with high homology with expressed genes

Are the Results of Clinical Trials Really Valid Worldwide?

• Ethnic-related incidence of tumors or toxicityGastric cancer: Japanise vs CaucasianGefinitib pulmonary toxicity: Japanise vsGefinitib pulmonary toxicity: Japanise vs Caucasian

• Ethnic-related frequency of SNPsq yTamoxifenTaxanes

Careful interpretation of Can we change the clinical results in a single

ethnic populationparadigm of clinical trials by

using pharmacogenetics?

The New Proposed Paradigm for Clinical TrialsThe New Proposed Paradigm for Clinical Trials

Perform the trial in different ethnic populations

Make the optimal biological characterization of eachMake the optimal biological characterization of each single tumor

++Pharmacogenetics

Variability to drug activity/toxicity

Tailored therapy

PharmacogeneticsPharmacogenetics(Haplotype and Genotype)

Transport (MDR1, MRP2, RFC,…)Metabolism → phase I (CYP)

→ phase II (UGT, GST,…)Target (EGFR, VEGFR, TS,) DrugDrug

Metabolite

Metabolite Drug

Target

TOXICITY

Pharmacogenetic and Pharmacogenomic Diagnostic Biomarkers (approved by FDA)

TOXICITYUGT1A1*28 (irinotecan) [Invader Assay]

TPMT (6-MP, azatioprine) [pro-Predict Py]

CYP2C9 ( f i ) [A li hi ]CYP2C9 (warfarin) [Amplichip]

CYP2D6 (atomoxetina) [Amplichip]

EFFICACY/RESISTANCEEFFICACY/RESISTANCE

CYP2D6 (tamoxifen) [ [ AmpliChipCYP450Test ] ]

Expression levels of Her2/neu (trastuzumab)

Mutations of K-RAS (cetuximab, panitumumab )

Chromosome Philadelphia/ Bcr-abl (imatinib)

C-kit (imatinib)C kit (imatinib)

Not approvedEGFR mutations (gefitinib, erlotinib)

Tamoxifen: The Key-Pharmacogenetic Step

CYPs cytochrome P450CYPs-cytochrome P450 metabolism

Tamoxifen Biotransformation CH3

OCH2CH2NCH3

c cCH3 CH2

CYP3A4CYP3A4 CYP2B6CYP3A5 Tamoxifen CYP2C9CYP2C19CYP2D6 CYP2D6CYP2C9

CYP2C19CYP1A2 CH3

CH3 OCH2CH2NOCH2CH2N CH3

Hc c

c c CH3 CH2CH3 CH2

OHOH

4-hydroxytamoxifenN-desmethyltamoxifen CYP2B6CYP2C9CYP2C19 CYP3A4

CH3

OCH2CH2NCYP2D6 5 10 nM (a erage 7 nM)CYP2D6 H 5-10 nM (average ~7 nM)c c

CH3 CH2

OH

4-hydroxy-N-desmethyltamoxifenyd o y des e y a o e(Endoxifen)

10-200 nM (average ~ 90 nM)Borges S, et al. Clin Pharmacol Ther. 2006;80:61-74.

Tamoxifen and PolymorphismsP l hi ti i ti f t if t b li iPolymorphic genetic variations of tamoxifen-metabolizingenzymes ( http://www.imm.ki.se./cyp)Concomitant administration of CYP2D6 inhibitors mayConcomitant administration of CYP2D6 inhibitors mayaffect tamoxifen-related clinical outcomesCYP3A4 promoter variant also involved in tamoxifenmetabolismPolymorphism of the SULT1A1 gene (SULT1A*2 variant)ER t (ESR Xb I d ESR2 02) bER genotypes (ESR-XbaI and ESR2-02) may beassociated with tam-associated lipid changes and maycontribute to interindividual variability to tamoxifenb fitbenefits

Tamoxifen PolymorphismsPOLYMORPHISM Ethnical Clinical POLYMORPHISM

CLASSES CYPs Frequency Implication

Poor CYP2D6: CAUCASIANSNonfunctional variants•worse DFS (Goetz MPPoor

metabolizer *3, * 4, * 5, * 6(RARE IN ASIANS AND

BLACK AFRICANS)

worse DFS (Goetz MP, Schroth W)

•Worse PFS (Schroth W)

Intermediate metabolizer

CYP2D6 * 10, *17, * 41

ASIANS(RARE IN EUROPEAN

CAUCASIANS)

Worse recurrence, decreased RFS, worse

PFS (in adyuvantsetting) (Schroth W)g) ( )

ExtensiveCYP2D6 * 2xn CAUCASIANS Favorable RFS, to be

metabolizerCYP2D6 2xn confirmed

Endoxifen, But Not 4-hydroxytamoxifen, , y y ,Degrades The Estrogen Receptor In BreastCancer Cells: A Differential Mechanism Of

Action Potentially Explaining CYP2D6

SABCS 2008

John R. Hawse, Xianglin Wu, Malayannan Subramaniam,

Matthew P. Goetz, Thomas C. Spelsberg, James N. IngleIngle

Hawse JR, et al. Cancer Res. 2009;69(Suppl 2): Abstract 19.Goetz MP, et al. Cancer Res. 2009;69(Suppl 2): Abstract 57.

Relapse-Free Time According to CYP2D6Relapse-Free Time According to CYP2D6 Metabolizer Status

(n=108)

(n=65)(n=65)

(n=16)

HR 4.0PM relative to EM

Goetz MP, et al. Cancer Res. 2009;69(Suppl 2): Abstract 6037.

Pharmacogenetic (CYP2D6) And GenePharmacogenetic (CYP2D6) And Gene Expression Profiles (HOXB13/IL17BR AndMolecular Grade Index) For Prediction OfMolecular Grade Index) For Prediction Of Adjuvant Endocrine Therapy Benefit In

The ABCSG 8 TrialThe ABCSG 8 Trial

SABCS 2008 MP Goetz, M Ames, M Gnant, M Filipits, H Heinzl,

R Jakesz,R G il C M th H S i V S S S fR Greil, C Marth, H Samonigg, V Suman, S Safgren,

M Kuffel,R Weinshilboum M Erlander X Ma J IngleR Weinshilboum, M Erlander, X Ma, J Ingle

ABCSG Trial 8 Structure Switch point

Tamoxifen Tamoxifen(2 years) (3 years)R

an

Primary dsurgery o

miz Tamoxifen Anastrozolee (2 years) (3 years)( y ) ( y )

Primary endpoint: Switching periodevent-free survivalSequencing period

Jakesz R, et al. Lancet. 2005;366(9484):455-462.

Relapse-Free Time According to CYP2D6 inRelapse Free Time According to CYP2D6 in Women Receiving Adjuvant Tamoxifen

Goetz MP, et al. Cancer Res. 2009;69(Suppl 2): Abstract 6037.

CYP2D6 and Risk of Breast Events

ARM A: TAM for 5 Years(n = 67) Relative Risk to EM P Value

CYP2D6 PM 3 83 (1 27 11 55) 017CYP2D6 PM 3.83 (1.27-11.55) .017

CYP2D6 IM 0.87 (0.44-1.71) .689

ARM B: TAM to Anastrozole (n = 55)

CYP2D6 PM 1.02 (0.21-4.83) .985

CYP2D6 IM 0.81 (0.40-1.61) .538

Adjuvant Tamoxifen and CYP2D6• CYP2D6 associated with recurrenceGoetz et al. 2005,1 20082 (USA)Hawse et al. 20083 (AUSTRIA)Schroth et al. 20074 (Germany)Kiyotani et al. 20085 (Japan)Newman et al. 20086 (UK)Xu et al. 20087 (China)• CYP2D6 not associated with recurrenceWegman et al. 2005,8 20079 (Sweden)

1 Goetz MP et al J Clin Oncol 2005;23(36):9312 9318 2 Goetz MP et al Cancer Res 2009;69(Suppl 2): Abstract1. Goetz MP, et al. J Clin Oncol 2005;23(36):9312-9318. 2. Goetz MP, et al. Cancer Res. 2009;69(Suppl 2): Abstract 57 3. Hawse JR, et al. Cancer Res. 2009;69(Suppl 2): Abstract 19. 4. Schroth W, et al. J Clin Oncol. 2007;25(33):5187-5193. 5. Kiyotani K, et al. Cancer Sci. 2008;99(5):995-999. 6. Newman WG, et al. Clin Cancer Res. 2008;14(18):5913-5918. 7. Xu Y, et al. Ann Oncol. 2008;19(8):1423-1429. 8. Wegman P, et al. Breast Cancer Res. 2005;7(3):R284-R290. 9. Wegman P, et al. Breast Cancer Res. 2007;9(1):R7.

Aromatase Inhibitors: The Key-Pharmacogenetic Stepa acoge et c Step

CYPs-cytochrome P450 enzyme Caromatase by CYP 19A1 or

aromatase genesa o atase ge es

POLYMORPHISMS ENVOLVED IN EXEMESTANE ACTIVITY

CYP19A1gene

CYP19CYP19Ex11+410 G>T(TTTA) t

CYP3A4*1BCYP3A4*1BCYP3A5*3CYP3A5*3 CYP19A1gene

exemestane exemestane

(TTTA)n repeatsEx10+1558 C>TEx7+47C>T

CYP3A5 3CYP3A5 3

RIZ1RIZ1Delezione Pro704

ER1ER1497 C>T

Delezione Pro704

CYP17CYP17Ex1+27T>C (5’UTR)

CYP1B1*3CYP1B1*3OO497 C>T

256 A>G(TA)n repeats

ER2ER21082 G>A

COMT COMT Ex4-12G>A

UGT1A1*28UGT1A1*28

1730A>G(CA)n repeats

AI P l hAIs Polymorphysms> 80 CYP 19 polymorphysms resulting in 44 haplotypes 80 C 9 po y o p ys s esu t g ap otypesfrom each of the 4 ethnic groups have been identified.Arg39 variant OF CYP19A1: Present in 6.7% Han Chinese American (rare in other ethnic groups);Cys264 variant OF CYP19A1: Higher frequencies in Han Chinese Americans and African Americans (11 7%Chinese Americans and African Americans (11.7%-22.5%) than Caucasian or Mexican Americans (2.5%-5%)CYP19A1 3’-untranslated region variant: Associated to higher RR and TTP in postmenopausal ER+ metastatic disease treated with letrozoledisease treated with letrozole

ROLE OF THE CYP19 Ex11+410G>T ANDROLE OF THE CYP19 Ex11+410G>T AND OTHER GENETIC POLYMORPHISMS ON

RESPONSE TO EXEMESTANE AS FIRST LINERESPONSE TO EXEMESTANE AS FIRST LINE TREATMENT IN PATIENTS WITH METASTATIC

BREAST CARCINOMABREAST CARCINOMA

Giuseppe Toffoli, MD, Director Pharmacology CRO Aviano

Giampietro Gasparini, Study Coordinator CIPOMO

STUDY DESIGNPatients enrollment and treatmentPatients enrollment and treatment

394 patients will be enrolled forpharmacogenetic analysis.

In 100 patients will be done also PK analysis

Th t 25 / /d At l t 8Therapy: exemestane 25 mg/po/day. At least 8weeks on therapy are required for evaluation.

Taxanes: The Key-Pharmacogenetic StepStep

Metabolism (CYP3A4, CYP3A5)

Transport-efflux (ABCB1) gene polymorphisms

Taxanes Docetaxel is metabolized by CYP3A4/5 ,while paclitaxel by y , p yCYP2C8/CYP3A4

There is controversy whether polymorphisms in the ABCB1 di P l t i l t t ffi f tgene, encoding P-glycoprotein, correlate to efficacy of taxanes

In a prospective study with paclitaxel in metastatic breast cancer the genotype ABCB1 2677 GG showed a significantcancer the genotype ABCB1 2677 GG showed a significant correlation with resistance1

In a retrospective analysis of the CALCB 9871 study (109 p y y (patients) no difference was found between Caucasian and African-American patients in terms of docetaxel toxicity regarding polymorphisms CYP3A4, CYP3A5, and ABCB12

1. Chang H, et al. Ann Oncol. 2008;20(2):272-277. 2. Lewis LD, et al. Clin Cancer Res. 2007;13(11):3302-3311..

TaxanesInterethnic differences in docetaxel PK and toxicities exist between south east populations (Chinese vs Indian vs Malays)1south-east populations (Chinese vs Indian vs Malays)1

CYP3A4*1B is a promoter polymorphism with different ethnic incidences: 45% in African-American, 2-9% in Caucasian; rare in Asian2AsianCaucasian patients harboring both CYP3A4*1B +/- CYP3A5*1°alleles have higher docetaxel clearance3,4 (I C i ti t t t d ith d t l th l hiIn Caucasian patients treated with docetaxel, the polymorphism C1236T in the ABCB1 gene was significantly related to decreased clearance (-25%)5

D t l i d d th h t l i l t i iti d OSDocetaxel-induced neuropathy, hematological toxicities, and OS are linked to ABCB1 allelic variants6

1 H SY t l Ph i J 2008 8(2) 139 146 2 St d H t l Ph i 2008 8(7) 8031. Hor SY, et al. Pharmacogenomics J. 2008;8(2):139-146. 2. Steed H, et al. Pharmacogenomics. 2008;8(7):803-815. 3. Baker SD, et al. Clin Pharmacol Ther. 2009;85(2):155-163. 4. Tran A, et al. Clin Pharmacol Ther. 2006;79(6):570-580. 5. Bosch TM, et al. Clin Cancer Res. 2006;12(19):5786-5793. 6. Sissung TM, et al. Clin Cancer Res. 2008;14(14):4543-4549.

Bevacizumab: The Key-Bevacizumab: The Key-Pharmacogenetic Step

VEGF as the Therapeutic Target

Association of Vascular Endothelial Growth Factor and Vascular Endothelial Growth Factor Receptor-2 Genetic Polymorphisms With

Outcome in a Trial of Paclitaxel Compared With Paclitaxel Plus Bevacizumab in Advanced BreastPaclitaxel Plus Bevacizumab in Advanced Breast

Cancer: ECOG 2100

Bryan P. Schneider, Molin Wang, Milan Radovich, George W. Sledge, Sunil Badve, Ann Thor, David A. Flockhart, Bradley Hancock, Nancy Davidson,

Julie Gralow, Maura Dickler, Edith A. Perez, Melody Cobleigh, Tamara Shenkier, Susan Edgerton, Kathy D. Miller

Schneider BP, et al. J Clin Oncol. 2008;26(28):4672-4678.

Kaplan Meier Curve for Overall Survival (OS) in Experimental Arm by Genotype;Kaplan-Meier Curve for Overall Survival (OS) in Experimental Arm by Genotype; (A) Vascular Endothelial Growth Factor (VEGF)-2578 C/A; (B) VEGF-1154 G/A

The VEGF-2578 AA genotype andthe VEGF-1154 AA genotypepredicted a favorable median OS forpredicted a favorable median OS forpatients in the combination arm butdid not predict an improved medianOS for patients in the control arm andOS for patients in the control arm anddid not predict a superior PFS or RRfor either arm. There was a significantincremental benefit from eachincremental benefit from eachaddition of the VEGF-1154 A allele.

Schneider BP, et al. J Clin Oncol. 2008;26(28):4672-4678.

RESULTS• Patients with VEGF-2578 AA and VEGF-1154 AA genotype have asuperior OS compared with patients with alternative genotypes

• The allele frequency of VEGF-2578 AA and VEGF-1154 AA in whiteq ypopulation is frequent and ranges from 33% to 49%

• The VEGF-634 CC and VEGF-1498 TT genotypes correlated with lessThe VEGF 634 CC and VEGF 1498 TT genotypes correlated with lessgrade 3 or 4 hypertension (0% and 8%, respectively) as compared to thecombined alternate genotypes (P = .005 and P = .022, respectively)

• Patients with grade 3 or 4 hypertension have a superior OS as comparedto pts with no hypertension (38.7 months vs 25.3 months, respectively;P = .002))

Schneider BP, et al. J Clin Oncol. 2008;26(28):4672-4678.

CRITICISMS• The combination of paclitaxel/bevacizumab showed prolonged PFS (11.8 monthsvs 5 9 months) and RR (36 9% vs 21 2%) as compared to paclitaxel alone (E2100vs 5.9 months) and RR (36.9% vs 21.2%) as compared to paclitaxel alone (E2100Study), but….

• No significant difference was observed on OS (26.7 months vs 25.2g (months; P = .16)

• The grade 3/4 toxicity was higher in the experimental arm (hypertension:g y g p ( yp14.8% vs 0%; proteinuria: 3.6% vs 0%; cerebrovascular ischemia. 1.9%vs 0%)

• The RR and PFS observed in the control arm are worse than that usuallyobserved in first-line therapy: Why?

• Retrospective analysis of pharmacogenetics on tumor specimens andnot as usually performed on blood samples

Schneider BP, et al. J Clin Oncol. 2008;26(28):4672-4678.

TAKE HOME MESSAGE• The results on pharmacogenetic studies are promising butThe results on pharmacogenetic studies are promising butmainly obtained in retrospective analysis

• Pharmacogenetic studies emphasize the relevance ofethnical diversities

• It is reasonable to hypothesize that pharmacogeneticanalysis coupled with molecular characterization of eachy psingle tumor may lead to improved personalized therapy

D t d t b lid t d i ti li i l t i l• Data need to be validated in prospective clinical trials