SNP Associations with Event Free Survival
in Myeloma from Two Phase III Clinical Trials
using the Bank On A Cure SNP Chip
International Myeloma Foundation
Brian Van Ness
Genetic influence on disease progression and response
} Geneticvariations
among
individuals
Genetic
deregulation in
· Drug metabolism
tumor cells
· Bone microenvironment
· Immune responses
· Genetic variations
· DNA repair
lead to differences in
disease progression
and response
Genetic variation in the population
DNA sequence
Biological trait
(Genotype)
(Phenotype)
Environment
ATTCGCATGGACC (3 billion)
C
A
SNPs
Single Nucleotide Polymorphisms }Disease risk
Outcome / response
~ 1/1000 bp
BOAC CHIP Functional Groups of SNPs
3404 SNPs
Other functional
groups:
Drug metabolism
Cell proliferation
Cell interactions
Cell signaling
Gene regulation
Bone metabolism
DNA Repair
Immune response
BOAC target SNP associations
Myeloma vs unaffected (risk)
Survival
Toxicities
Bone disease
Age of onset
Study Samples run on BOAC SNP Chip
Trial
Treatment
N
(Controls)
118
(Survival)
E9486
VBMCP +/- IFN/CP
<1yr
26
>3yr
26
S9321
VAD + VBMCP
<1yr
45
>3yr
46
EA100
Dex vs Thal-dex
56
Total Therapy 2 VAD (Thal) / Tnsplt
301
(UK Lab - Gareth Morgan)
MRC IX
CTD vs CVAD or MP
161
Hovon
TAD vs VAD
46
TOTAL
825
Top 2 networks of most significant SNP associations with
survival in 2 myeloma phase III trials
DNA Repair
Immune response
SNP association with extreme phenotype - survival
Chemotherapy trials: ECOG E9486 and ECOG/SWOG S9321
CSTB
Conglomerate model, recursive partitioning
from top 20 univariate p-value rank:
Cyp4F
GSTA
< 1yr vs >3yr
BMP2
Event free Survival for Clinical data for both studies combined
Clinical Data (All Data)
1.0
Actual
Long
PROC
Short
logrank p-val<0.001
0.8
Event free survival
0.6
Classification prediction
Event free survival using classification from both studies combined
Both Classification based on SNPs (All Data)
E9486 = 87%
0.4
1.0
S9321 = 67%
Long
5 SNP fit
0.2
Short
logrank p-val<0.0001
0.8
[Combined = 74%]
0.0
0.6
0
50
100
150
Months from Enrollment
0.4
0.2
0.0
0
50
100
150
Months from Enrollment
Conclusions:
· Developed high throughput, gene targeted SNP panel
· SNP Chip design targets networks and pathways to allow hypothesis
driven analysis of cellular functions associated with disease outcomes
(risk, complications, response, survival, toxicities)
· Analytical approaches emphasize groups of genes associated with
biologic functions that define each individual.
· Variations in DNA repair, drug metabolism, and immune response
may impact patient response, toxicities , and survival
*
· We should consider combining the genetic influence of the tumor
cells and the genetic variation of the patient to better direct therapies
Study
*
led by Dr. Gareth Morgan, co-director Bank On A Cure (London)