Transcript VHL

Roma, 22 febbraio 2013
Highlights
in the management of renal cell carcinoma
Clinical and Molecular Predictive
factors to molecularly targeted
agents: what we know so far..
Enrico Ricevuto, Eleonora Palluzzi
Oncologia Medica
Ospedale San Salvatore
Università dell’Aquila
Renal cell carcinoma
Evolution of medical treatment

Markers
None
“One fit (unfit) all”
Renal cell carcinoma
Evolution of medical treatment


Parameters
None
Bio-Clinical
 Patient
 Tumor
 Drugs
“One fit (unfit) all”
“One fit some”
(>10%)
fitness (age, comorbidities)
prognostic risk
prediction (safety/toxicity, efficacy)
ccRCC
Predictive markers of target therapy

Bio-Clinical
 Hypertension (>90 mm/hg DBP)
 LDH
 Hypothiroidism (increased TSH)
Hypertension
Biomarker of Efficacy with Sunitinib
B Rini, J Natl Cancer Inst 2011; 103:763-773.
Diastolic blood pressure
Biomarker of efficacy with axitinib in solid tumors
OS with landmark at 8 weeks.
B Rini, Clin Cancer Res; 17(11); 3841–9.2011
Serum LDH
Biomarker with Temsirolimus
Andrew J Armstrong et al, J Clin Oncol 30:3402-3407.
Hypothyroidism (increased TSH)
Biomarker of activity with TKI in solid tumors
Objective Remission According to Response Evaluation Criteria in Solid Tumors Based on
Increased Thyroid-Stimulating Hormone Levels
Schmidinger M, Cancer 2011
Hypothyroidism (increased TSH)
Biomarker of efficacy with TKI in solid tumors
Schmidinger M, Cancer 2011.
Renal cell carcinoma
Evolution of medical treatment



Markers
None
“One fit (unfit) all”
Clinical
“One fit some”
(>10%)
Monogene
“One fit few”
(1-10%)
 VHL
 Other genetic alterations
 Heterogeneity (tumor/metastasis)
 VHL/HIF epigenetics alterations
Renal Cell carcinoma
Different diseases




Different histology
Different genes
Different clinical courses
Different response to therapy
Renal Cell Carcinoma
Pathology






Clear cell (75-85%)
 Proximal tubule origin
 Abnormalities in chromosome 3p
Chromophilic (15%)
 85% of these are diagnosed as stage I tumors
 Also proximal tubule in origin, but 3p is normal
 Trisomy 12, 16, 20 can be seen
Chromophobic (5%)
Oncocytic (uncommon)
 usually not aggressive
 Collecting duct origin
 11q13 rearrangements in some cases
Collecting duct (Bellini’s duct) tumors – very rare
Unclassifiable (<3%) – worse prognosis
Renal Cell Carcinoma
Morphology and Genetics
Renal Cell carcinoma
Disease of cell metabolism: Biomolecular Complexity
Pathways that respond to metabolic stress or nutrient stimulation
VHL
oxygen and iron sensing
MET
LKB1-AMPK energy sensing
FLCN
binds AMPK and might interact with the cellular energy and nutrient
sensing
TSC1
TSC2
downstream of AMPK and negatively regulates mTOR in response to
cellular energy deficit
FH
central role in the mitochondrial tricarboxylic acid cycle
SDH
coupled to energy production through oxidative phosphorylation
pVHL targets hypoxia-inducible factor (HIF)-α
for ubiquitin-mediated degradation
Sporadic ccRCC
Genetic alterations of the VHL gene




LOH 3p 24-25
Intragenic mutations
Hypermethylation
Biallelic loss
78-96%
51-71%
5-20%
50-75%
Sukosd et al, Canc Res’03, 63, 455
Kondo et al, Gene Chrom Cancer’02, 34, 58-68
Banks et al, Canc Res’06, 66, 2000-11
Spectrum of VHL mutations
cumulative data of 1244 mutations reported in the literature
Young A et al, Clin Canc Res’09, 15; 7582
RCC
Spectrum of VHL gene alterations
Young A et al, Clin Canc Res’09, 15; 7582
VHL genetic alterations
Prognostic relevance
Young A et al, Clin Canc Res’09, 15; 7582
Choueiri et al, J Urol. 2008; Rini BI, et al. BJU Int. 2006
Choueiri et al, J Urol. 2008; Rini BI, et al. BJU Int. 2006
Klatte et al, Clin Canc Res’07
Metastatic ccRCC
Heterogeneity
Gerlinger M et al, N Engl J Med 2012;366:883-92
Metastatic ccRCC
Heterogeneity
Gerlinger M et al, N Engl J Med 2012;366:883-92
Expanded HIF signal output activates mediators
of metastasis
CXCR4 expression correlates with poor prognosis and metastasis
in ccRCC and is inducted by VHL loss.
Vanharanta S et al, Nature Medicine 2013
DNA demethylation allows CYTIP expression in
metastatic ccRCC
Vanharanta S et al, Nature Medicine 2013
Current renal cell carcinoma biomarker
initiatives
1) EuroTARGET
2) SCOTRRCC
3) Predict Consortium
4) TCGA
5) CAGEKID
Vasudev et al. BMC Medicine 2012, 10:112
Renal cell carcinoma
Evolution of medical treatment




Markers
None
“One fit (unfit) all”
Clinical
“One fit some”
(>10%)
Monogene
“One fit few”
(1-10%)
Multigenes
“One fit one”
(<1%)
 Multiple biopsies
 Genetic and epigenetic alterations




Multiple drugs
anti-angiogenesis
mTOR-inh
Need of drugable and actionable targets
mRCC
Angiogenesis inhibitors
Sonpavde G Exp Opin Invest Drugs 2008
mTOR inh image
Why should we need clinical
and biological markers?






Individual tumor heterogeneity
differential clinical outcome
aggressiveness
efficacy
OS
differential biology
Biological heterogeneity


Genetic alterations
VHL
M
M
1
2
3
4
M
M
Metastatic ccRCC
Heterogeneity
Gerlinger M et al, N Engl J Med 2012;366:883-92
Open question

Multiple biopsies
Hand-foot syndrome (HFS) as a potential
biomarker of efficacy in patients (pts) with
metastatic renal cell carcinoma (mRCC) treated
with sunitinib.
Methods: Analyses included pooled data from 770 pts who received single-
agent SU as 50 mg/d on a 4-week-on/2-week-off schedule (n=544; 71%)
or 37.5 mg continuous once-daily dosing (n=226; 29%). Median PFS and
OS were estimated by Kaplan-Meier methods and compared between pts
with vs without HFS using a log-rank test. ORR was compared by Pearson's
chi-square test. Tumor response was assessed by investigators and
adverse events were recorded regularly. Multivariate and time-dependent
covariate analyses were performed. Results: Of 770 pts, 179 (23%)
developed any-grade HFS, compared with 591 (77%) who did not. Most
instances of HFS (63%) initially occurred during the first 3 treatment
cycles. Pts who developed HFS had significantly better ORR (55.6% vs.
32.7%), PFS (14.3 vs. 8.3 mo), and OS (38.3 vs. 18.9 mo) than pts who did
not develop HFS (p<0.0001). In a multivariate analysis, SU-associated HFS
remained a significant independent predictor of both PFS and OS (and of
OS by time-dependent covariate analysis).
Conclusions: In mRCC pts, SU-associated HFS was significantly and
independently associated with improved clinical outcomes. Overall, pts who
did not develop HFS still had substantial benefit from SU. However, the
presence of HFS identified a subset of pts that manifested highly favorable
efficacy results with SU. These findings suggest that development of HFS
Michealson MD, J Clin Oncol 2011; 29
Renal Cell carcinoma
Disease of cell metabolism
Pathways that respond to metabolic stress or nutrient stimulation
VHL
oxygen and iron sensing
MET
LKB1-AMPK energy sensing
FLCN binds AMPK and might interact with the cellular energy and
nutrient sensing
TSC1 downstream of AMPK and negatively regulates mTOR in
response to TSC2 cellular energy deficit
FH
central role in the mitochondrial tricarboxylic acid cycle
SDH coupled to energy production through oxidative phosphorylation
ccRCC
Prognostic and Predictive Biomarkers

DNA
RNA
Protein
FISH

IHC


Elisa
Mutations
Western
VHL
HIF1 alfa
X
X

CAIX
X

VEGF
X

sVEGFR-2
X

TIMP-1
X

Ras p21
X
Bui, Clin Canc res’03
Atkins et al, Clin Canc Res’05
VHL genotype in ccRCC





Structural alteration
Point mutations
Methylation
Prevalence of mutations
Functional relevance
Diagnostic strategy
60%
Gain of function
Direct Sequencing


Scanning for unknown mutations
Clinical implications
Predictive
anti-VEGF
VHL gene
Structural Features



Chromosomal locus
3p24-25

Exons
3

mRNA
4.7 kb.
Proteins
pVHL30
213 aa. (28-30 KD)
pVHL19
160 aa. (18-19 KD)
Hypoxia-inducible factor (HIF)-α
Transcriptional activity: Pathways and Genes














Energy metabolism: increase in glycolytic pathway
Glut-1
Angiogenesis
VEGF
VEGFR
PDGF
Ang-2
FGF
Tie-2
PH regulation
CA IX
Proliferation
TGF-alfa/beta
CXCR
IGF
Apoptosis
p53
NIX
BNIP-3
Erythrocitosis
EPO
Hypoxia-inducible factor (HIF)-α
Transcriptional activity: Pathways and Genes
Energy metabolism: increase in glycolytic pathway
Glut-1
Angiogenesis
VEGF
VEGFR PDGF
Ang-2
FGF
Tie-2
PH regulation
CA IX
Proliferation
TGF-alfa/beta
CXCR
IGF
Apoptosis
p53
NIX
BNIP-3
Erythrocitosis
EPO
Renal Cell Carcinoma
Evidence for VHL initiation



Both sporadic and VHL disease-associated
ccRCC display loss of VHL
HIF activation is found in early renal lesions
including cysts and dysplasias
Features of ccRCC are consistent with
overexpression of HIF target genes and pathways
Predictive Value
Rini et al, BJU’06
Gad et al, Target Onc’07
VHL status and Clinical Outcome to VEGFtargeted Therapy
•182 patients with metastatic RCC who
received initial anti-VEGF therapy at CCF
and UCSF between 2003 and 2006.
•59 patients excluded:
-Missing key data (n=3)
-Pure non-clear cell histology (n=8)
-Insufficient tissue for DNA extraction
(n=12)
-Unavailability of tissue (n=36)
Clinical impact of VHL mutations in
ccRCC
Summary
 VHL mutations in ccRCC
60%



HIF activation as a result of VHL loss is tightly
correlated with tumor phenotype and may play a
direct role in tumor growth
HIF target gene activation is associated with the
earliest stages of renal tumorigenesis and
correlates with survival
VHL mutation correlates with improved TTP in
patients treated with inhibitors of the HIF target
VEGF angiogenesis signaling pathway
Association of FGFR2 Polymorphism with OS (P = 0.01)
Selected genes associated with RFI
Rini, ASCO’10 CSS in JCO’10, 28: 4501
ccRCC – localised disease (I-III)
Pathways Signatures
Genes associated with outcome (RFI)

N.










%
Genes
732
Genes significantly associated with outcome (RFI, OS) 448
(69%)
(unadjusted univariate analysis)
Increased expression associated with better outcome
366
(82%)
Covariate analysis
Significant association with
necrosis
503
Fuhrman grade
494
Pathologic Stage
482
T-size
Rini, ASCO’10
CSS in JCO’10, 28: 4501
492
N-status
ccRCC – localised disease (I-III)
Genes selected for further analysis






Total
72
Associated with RFI after covariate adjustment 29
Most significant before covariate adjustment
By Pathway Cluster Analysis to identify genes
from additional pathway
17
Member of VEGF/mTOR pathways
12
14
Rini, ASCO’10 CSS in JCO’10, 28: 4501
ccRCC – localised disease (I-III)
Genes expressions and Recurrence Risk









Expression
Angiogenesis
L
Immune Response
L

IL6, IL8
H
ECM/Cell adhesion
H
Cell cycle
H
I: Increased
D: Decreased
L: Low
H: High
Recurrence Rate
I
I
I
I
I
Rini, ASCO’10 CSS in JCO’10, 28: 4501
ccRCC – localised disease (I-III)
Molecular Stratification
Risk of Recurrence


Low
Intermediate
High
Increased

Increased

Angiogenesis
Immune Response
KDR1
IL6
EMCN

IL8


Cell-mediated
Cycle
cytotoxic response
CD8A
Cell
TPX2
CX3CL1
BUB1

Invasion

CX CL10
MMP14
Rini, ASCO’10 CSS in JCO’10, 28: 4501
LAMB1
Select RNA gene expression data to
clinical / pathologic variables enhances
prediction of recurrence of localized RCC
Renal Cell Carcinoma
Angiogenesis and VEGF Inhibition
Oudard S et al, Cancer Treat Rev 2012.
Renal Cell Carcinoma
Resistance to VEGF Inhibitors
Casanovas O, et al. Cancer Cell. 2005;8:299-309
Rationale for combined Met and VEGFR inhibition
• Met activation causes cells to proliferate and migrate
• VEGFR activation initiates angiogenesis
• Combining anti-tumour and anti-angiogenic activities may be a valid treatment strategy
University of L’Aquila
Department of
Biotechnological and Applied
Clinical Sciences
Medical Oncology
Corrado Ficorella
Enrico Ricevuto
Katia Cannita
Gemma Bruera
Eleonora Palluzzi
Azzurra Irelli
Valentina Cocciolone
65
Chow LQM, Eckhardt SG. J Clin Oncol. 2007;25(7):884-896
Eskens FALM, AACR; 2008. abst# LB-201
Spectrum and potency of TKIs versus VEGF receptors
C.Sternberg. Sixth European International Kidney Cancer Association, May 6, 2011.
Metastatic renal cell carcinoma
MSKCC prognostic factors





KPS <80
Time from diagnosis to treatment <12 months
Hemogobin
 <lower limit of lab’s reference range
LDH
 > 1.5x upper limit of lab’s reference range
Corrected serum calcium > 10.0 mg/dl
Metastatic renal cell carcinoma
MSKCC Risk groups

Favorable
no poor prognostic factors

Intermediate
1 or 2 poor prognostic factors

Poor
>2 poor prognostic factors
Renal Cell Carcinoma
VEGF Pathway
G Korpanty et al, Journal of Oncology 2010
VEGF Inhibition
Bevacizumab
G Korpanty et al, Journal of Oncology 2010
Metastatic RCC
Identification of biomarkers-Rationale




..may help determine patient prognosis
..identify patients most likely to benefit from
specific treatments
..help monitor response to treatment
..guide clinicians in designing personalized
treatment strategies
Hypertension
Biomarker of Efficacy with Sunitinib
B Rini, J Natl Cancer Inst 2011; 103:763-773.
Hypertension
Biomarker of Efficacy with Sunitinib
B Rini, J Natl Cancer Inst 2011; 103:763-773.
Hypothyroidism in patients with renal cell
carcinoma
Schmidinger M, Cancer 2011.
VHL genotype
Prognostic relevance
Choueiri et al, J Urol. 2008; Rini BI, et al.
HIF-α protein expression
and response to Sunitinib
Patel et al, ASCO’08
RCC cell lines
HIF-α protein expression
Not all VHL
mutations correlate
with increased
HIF-α expression
Patel et al, ASCO’08
Metastatic ccRCC
Heterogeneity
PrognosticSignatureGenes
Gerlinger M et al, N Engl J Med 2012;366:883-92
Mechanism of action of mTOR inhibitors
Everolimus
Temsirolimus
Cen and Amato . OncoTargets and Therapy 2012:5 217–224
Serum LDH
Biomarker with Temsirolimus
Kaplan-Meier estimates for overall survival
distribution by treatment group in the
analyzed population (n 404).
Kaplan-Meier estimates for overall
survival distribution according to
baseline
lactate
dehydrogenase
(LDH) category.
Kaplan-Meier estimates for overall survival
distribution by treatment
group and
lactate dehydrogenase (LDH) levels, for
(upper) normal level of LDH and (lower)
Andrew J increased
ArmstrongLDH.
et al, J Clin Oncol 30:3402-3407.
Metastatic ccRCC
Heterogeneity
Metastatic ccRCC
Heterogeneity
Gerlinger M et al, N Engl J Med 2012;366:883-92
Metastatic ccRCC
Heterogeneity
Gerlinger M et al, N Engl J Med 2012;366:883-92