Genome, transcriptome and proteome features of malignant cells as a source of combinatorial biomarkers for clinical translation

1
Genome, transcriptome and proteome features of
malignant cells as a source of combinatorial
biomarkers for clinical translation

• Ancha Baranova
• George Mason University, Fairfax, VA

2
Key word to understanding of cancer phenomenon
PROGRESSION
From bad to worse then to full catastrophe
3
LUNG CARCINOMA PROGRESSION
e.g. carcinogen from the smoke
e.g. Smoke
4
How long it usually takes DECADES
It takes 17 years for a large benign tumor to
evolve into an advanced cancer but lt2 years for
cells within that cancer to acquire the ability
to metastasize
Sequencing studies show that virtually all of the
mutations necessary for metastasis are already
present in all of the cells of the antecedent
carcinoma.
17 years
10.1073/pnas.0712345105 Jones et al., 2008
5
Major implication for translational research
1) need for early biomarkers2) need for
evaluation of how far away the tumor was from
sprouting metastasis ? basis for post-surgical
treatment and care
TRUTH Majority of human tumors sit in our bodies
undiagnosed for decades
6
Problem in biomarker research all low-hanging
fruits already collected
PSA, CEA, AFP.
7
Single biomarkers have problems with
differentiating grey area diseases, i.e.
inflammatory conditions
Solution Biomarker panels
PSA LeveL
Controls
Prostate cancer
Other cancers
Benign genitourinary diseases
BPH
Barak et al., 1989
8
Now from where these biomarker for biomarker
panels are usually coming ?
A fishing expedition Differences often
reflect Inflammation, fibrosis and other
common properties
Image courtesy of Purdue University, Dr. W. Andy
Tao
9
Two-pronged approach proposed
DISTANCE ANALYSIS Use entire pattern of gene
expression to evaluate of how far away the
excised tumor was from sprouting metastasis
TUMOR vs. a MIX of various normal cells Perform
transcriptome (in silico) and proteome (using
antigen screening) subtractions
Most GENERAL approach
Most TARGETED approach
10
Will uncover tumor biomarkers that cannot be
masked by antigen production in normal tissue
Two-pronged approach proposed
TUMOR vs. a MIX of various normal cells Perform
transcriptome (in silico) and proteome (using
antigen screening) subtractions
Most TARGETED approach
11
Most Targeted Approach Consortium
Blokhin Russian Oncological Scientific Center,
Moscow (sample collection and processing) Caerus
Discovery LLC, Manassas, VA (capture of protein
biomarker as differentially expressed antigens)
Institute of Chemical Biology and Fundamental
Medicine SB RAS, Novosibirsk (development of
tumor protein biomarkers as targets for
antibody-guided drug delivery) Biomedical
Center, StPeterburg (differential display of
tumor RNAs) Research Center for Medical
Genetics RAMS (non-coding Tumor RNAs) George
Mason University, Fairfax, USA (biomarker
validation, bioinformatics support and general
coordination)
Most TARGETED approach
12
Two-pronged approach proposed
DISTANCE ANALYSIS Use entire pattern of gene
expression to evaluate of how far away the
excised tumor was from sprouting metastasis
Most GENERAL approach
13
Simple words explanation forusing entire
transcriptome as biomarker

• Who is that? -- Instant answer (same person,
  Darwin)

Biomarker approach X Distance from the end of
nose to the upper lip Y Diameter of the eye
orbit Z Number of hairs in the beard
14
Two-pronged approach proposed
DISTANCE ANALYSIS Use entire pattern of gene
expression to evaluate of how far away the
excised tumor was from sprouting metastasis
Blokhin Russian Oncological Scientific Center,
Moscow (sample collection and processing) Institu
te of Chemical Biology and Fundamental Medicine
SB RAS, Novosibirsk (next gene sequencing and
microarray) Vavilov Institute of General
Genetics (next gene sequencing and microarray)
Research Center for Medical Genetics
RAMS (distance analysis of samples based on
summed evaluation of non-coding RNAs) University
of Arkansas for Medical Sciences (Little Rock,
AK) (distance analysis of whole genome
patterns) George Mason University, Fairfax,
USA (development of attractor descriptors,
bioinformatics support and general coordination)
Most GENERAL approach
15
VISION for the FUTURE

• New generation of tumor markers specific to the
  malignancy but not to the any normal cell type
  will allow
• True early diagnostics of dormant tumors
• 2) More specific antibody-guided delivery of
  therapeutic agents to the tumors

16
VISION for the FUTURE
Comparison of prognoses for This tumor and That
tumor
17
Same strategy can be realized using NextGen Seq

• Very possible
• low sequence coverage of transcripts will be
  enough
• cost-efficient cut-off needs to be determined

18
VISION for the FUTURE
REFERENCE LAB
Profiles 100s normal samples for prostate
Establishes its own, equipment-specific NORMAL
SPACE
For every single tumor sample, the distance from
normal tissue space is measured
19
How bad is my tumor?
Answer exact distance from Ideal norm
Your tumor
20
(No Transcript)