Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Personalised Medicine: experiences in MS
Giancarlo ComiDept. of Neurology & Institute of Experimental Neurology
Università Vita Salute S.Raffaele, Milano
Premio SAPIO Per la Ricerca Italiana
Firenze, 15 Aprile 2013
Personalised medicine is an emerging model that will revolutionise our current healthcare system.
In the last decade, several genomic aberrations were discovered that are now used as predictive markers for treatment with targeted therapeutics.
The technological advances in the last few years, such as the development of high resolution DNA microarrays or second generation sequencers, have led to a dramatic increase in the number of ongoing genomic profiling studies.
Epidermal growth factor receptor (EGFR) tyrosine kinase
inhibitors and their molecular modes of binding to the target
Gonzales de Castro 2013
Bases for individualised treatment in MS
• Complexity and heterogeneity of MS– Polygenic inheritance– Multifaced gene-gene and gene-environment interaction
• Large intraindividual variability of MS courses– Early long term prognostic factors– Short term prognostic factors
• Treatments with different mechanisms of action and different efficacy/safety profile
• Interindividual variability of the response to treatments– Clinical and MRI predictors– Pharmacogenomics
• Multiple treatment algorithms– Induction– Escalation– Combination
“Genetic risk and a primary role for cell-mediated immune mechanisms in multiple sclerosis” Nature 2011
This GWAS study involved 9,772 cases of European descent collected by 23 research groups working in 15 different countries: almost all of the previously suggested associations have been replicated and at least a further 29 novel susceptibility loci identified.
Disease prognosis is mostly defined in the early phase of the disease
Better Prognosis Poorer Prognosis• Female
• Younger (<40 years) age at onset
• Caucasian
• Optic neuritis or isolated sensory symptoms at initial presentation
• Monoregional onset
• Infrequent attacks within the first 5 years
• Long interval between first and second relapse
• No or low disability at 5 years
• Male
• Older (>40 years) age at onset
• Afro-American or non-white
• Motor, cerebellar, or sphincter symptoms at initial presentation
• Poor recovery from the first attachs
• Polyregional onset
• Frequent attacks within the first 5 years
• Short interval between first two attacks
• Short time to reach EDSS level of 4
• Disability at 5 years
Kantarci et al. Prognostic Factors in Multiple Sclerosis. In Handbook of Multiple Sclerosis (3rd edition). Cook SD editor. New York: Marcel Dekker. 2001. 449-463.
MRI prognostic factors
• MRI activity predicts relapses (Kappos 1999; Sormani 2003, 2007)
• MRI activity predicts brain atrophy (Paolillo 2004; Filippi 2004)
• MRI T2 lesion load predicts relapses and long term disability (O’Riordan 1998; Sormani 2003; Fisnisku 2008)
• Cortical lesions predict long term disability (Calabrese 2009; 2010)
• “MT MRI provides useful prognostic markers for the prediction of the long-term evolution of multiple sclerosis (Agosta 2006)
• Spinal cord atrophy predicts EDSS (Rocca 2011; Filippi 2011)
• fMRI correlates with cognitive dysfunction (Rocca 2010)
Pelayo, Montalban et al. Multiple Sclerosis 2010
CIS: n. abnormal EPs - time to EDSS 3.0
SEPsVEPsBAEPs
no MEPs
n: 247 (out of 335)
TeriTeripopo
Disease Modifying Drugs in the market or under evaluation Disease Modifying Drugs in the market or under evaluation by EMA:by EMA:
ALZALZiviv
FTY720FTY720po qdpo qd
LAQLAQpopo
BG12BG12
Wolinsky, modified
CLADCLADiviv
Targets of new MS treatments (Adapted from Linker 2008)
Teriflunomide
Comparative effects of MS therapies in disease activity, disease burden and safety
Relapses Disability Active lesions Brain atrophy
IFN/GA + +/- ++ -
Natalizumab +++ +++ +++ +
Alemtuzumab +++ +++ +++ +++
Fingolimod +++ ++ ++ +++
Teriflunomide ++ + ++ ?
Fumarate +++ ++ +++ +
Laquinimod + +++ + +++
Safety & tolerability
+++
+/-
+/-
+
++
++
+++
How to determine the risk-benefit profile of a drug?
• On the drug side, risks & benefits to be taken in account are:
• Efficacy, short and long-term
• Safety (trusting the drug)
• Tolerability
• Monitoring requirements
• Convenience
• Patient’s personal experience of previous therapies
• Doctor’s personal experience of previous therapies
• Cost of the drug
• (New MOA: FTY, cladribine is trying)
New drug must be evaluated in comparison with existing first-line therapies though the times are not the same, nor the trials’ protocols
IFN-GA
TeriflunomideLaquinimod
Efficacy
Burden of TherapyFactors affecting burden of therapy include convenience, monitoring, tolerability, and safety.
Alemtuzumab
Selecting Optimal Therapy in Relapsing MS: Potential Options?
BG12
Natalizumab
Fingolimod
Mitoxantrone
Individualised MS treatment
• When start treatment
• Which treatment start first
• Early detection of non responders
• Quick change of treatment in non responders
• Change of treatment strategy
Degeneration
Disability
time
EDSS 3
Inflammation
?
Response to treatment
Clinical onset
“Delaying treatment in MS:What is lost is not regained”
BR
B
R
B
R
Escalating therapy
1st line therapy
More aggressive approach
3rd line therapy
4th line therapy
5th line therapy
Mitoxantrone / Cyclophosphamide
CampathRituximab
BMT
Beta-Interferons / Glatiramer acetateLaquinimod/BG12/Teriflunomide
Natalizumab2nd line therapy
Fingolimod
Induction therapy
1st line therapy
2nd line therapy
3rd line therapy
4th line therapy
Natalizumab / IFN / GA/ Laquinimod/ Teriflunomide/BG12
Combination therapy
BMT
Mitoxantrone/Cyclophosphamide Fingolimod/ Campath/Rituximab
Preventive strategies
• Cancer risk screening procedures focused on oncogenic infections
• Baseline risk infection risk management
• Optimization of metabolic control
• Cardiovascular risk modification
• Vaccine preventable diseases
• ……………..
NATALIZUMABProgressive multifocal leukoencephalopathy
PML Risk Stratification in patients treated with Natalizumab
Evaluation of Anti-JCV AbsEvaluation of Anti-JCV Abs
Negative
≤0.10/1,000 (95%CI: 0-0.56)
Positive
0-24 months 1.50/1,000(95%CI: 0.83-2.50)
IS+
25-48 months
10.60/1,000(95%CI: 7.70-14.20)
Positive
0-24 months 0.53/1,000(95%CI: 0.33-0.81)
IS-
25-48 months 3.90/1,000(95%CI: 3.00-4.90)
Adapted from Sölberg-Sørensen P et al. Mult Scler 2012.
Biomarkers
• MRI
• EPs
• OCT
• Body fluid
Geurts et al, Radiology 2005
Double inversion recovery (DIR) - cortex
MS: spinal MR imaging
Brain atrophy – early and profound
De Stefano, Neurology 2010
R=0.6p<0.0001 (Spearmann rank correlation)
Global basal EP* score vs EDSS*VEP, MEP, SEP, BAEP
glo
bal
bas
al E
P s
core
EDSS
N=84
Leocani et al, JNNP 2006
Trip et al. 2005
OCT in chronic ONRNFL thickness correlates with VEP amplitude (axonal loss)
unaffected affected
body fluid biomarkers in MS
Markers for a disrupted blood-brain barrier (BBB): s-ICAM, s-
VCAM (IFN- tx), MMP-9/TIMP-1
Markers for demyelination: myelin breakdown products (?)
Markers for axonal damage: NSE, NFL, CK-BB, not clear
Markers for gliosis: S-100, GFAP, not clear
Markers for remyelination: NCAM, CNTF, not clear
Definition of treatment response
Time
MR
I A
ctiv
ity
New Enhancing Lesions
New T2 lesions have a finite time when they enhance, but timing the scan frequency to catch all of these is near impossible in practice
Monitoring Disease Activity Using MRI: A Recipe for Misinterpretation
Criteria for classification of complete response
No brain active lesions
+ No spinal cord active lesions
+ No relapses
+ No disease progression
+ No brain atrophy increase
+ No spinal cord atrophy increase
+ No increase of NAWM abnormalities
+ No EPs changessensitivity specificity
++
--
--
++
Pozzilli et al., Neurol Sci 2005
MRI MARKERS IN MS RRMS / Identification of individual responders
242 RRMS, 4.3 years IFN treatment
Patients with vs. without one-year MRI active scan (i.e., at least 1 EL)
Patients with vs. without one-year new T2 lesions (i.e., at least 1 new T2 L)
Predictivity of the response to DMTs
• Clinical and demographic
• MRI
– The drug effects on MRI activity predicts the effects on disease activity (Rio 2008; Rio 2009;Durelli 2008; Sormani 2009) and on brain atrophy (Filippi 2004)
• Laboratory
• Pharmacogenomic
• At treatment onset
• Early during treatment
Predictive value of biomarkers in MS TREATMENT
Treatment effects according to Activity/Dissemination of the Disease at First
Event in BENEFIT Parameter indicating
more dissemination or activity at onset
Treatmenteffect*
Parameter indicating less dissemination or activity
at onset
Treatmenteffect*
Multifocal 37% Monofocal 55%
> 9 T2 lesions 43% < 9 T2 lesions 60%
> 1 Gd+ lesion 38% No Gd+ lesion 57%
Treatment effects were significant in all subgroups by log rank tests
*Risk reduction according to unadjusted hazard ratios
EPs global score at treatment onset and treatment response
4,3 4
10,7
0
2
4
6
8
10
12
full-responders partial-responders non-responders
valo
re m
edio
sco
re g
lob
ale
bas
ale
EP
36
30
30
Mann Withney test, * p<0.001
*
*
Biomarkers predictive of the reponse to IFN
• NABs• Overexpression of genes of the NFKb network• increase in the frequency and suppressive activity
of CD4+CD25+Foxp3+ regulatory T cells • baseline overexpression of type I IFN-responsive
genes in nonresponders to IFN-β• High baseline levels of Il-17 in non responders• …………….
Poor predictive value in individual patients
Genome Wide Association Study (GWAS) ForPredicting Response To GA: Results
We selected SNPs with a P-Value < 10-4 31 SNPs
P-value = 10-4
Chr: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
The Genetic Model
• As every SNP predicted response only partially, we used the 31 significant SNPs from the GWAS to develop a genetic predictive model:
the best and most parsimonious model includes only 6 SNPs
Super Responders
Super Non-Responders Total
Genetic prediction of response**
Super Responders 45 6 51
Super Non-Responders
6 55 61
Total 51* 61 112
* One of the 52 subjects was removed because of missing genotype for this model** Patients were genetically classified responders if probability of super-response is ≥ 0.5
Sensitivity 87.9%Specificity 92.7%Positive predictive value 90.6%Negative predictive value 90.5%Correctly classified 90.5%
• At treatment onset
• Early during treatment
Predictive value of biomarkers in MS TREATMENT
MRI MRIDMT
12 mMS
Predicts:
Relapses
Disability
MRI activity: at least 2 new T2 / gad lesions
AND
or
Río et al. Mult Scler 2010
36 m
Relationship between active lesions on MRI and treatment response
Rio 2009
Take home message
• Prognostic information are available to orient treatment choice (better value at disease onset!)
• At present we have moderate possibility to predict which patients will respond to the various disease modifying agents in advance of therapy onset
• Pharmacogenomic information is still immature to allow its translation to clinical practice
• In treatment monitoring, MRI is of help in predicting therapeutic response
M. Comola
Neurorehabilitation
L. Leocani
Neurophysiology V. Martinelli
Neurology
M. Filippi
Neuroimaging
M. Falautano
Neuropsychology
G. Martino
Nuroimmnology
I. Sangion M. Galdabini