Personalized Medicine

1
Personalized Medicine

• C. Kent Osborne
• Dan L. Duncan Cancer Center
• Lester and Sue Smith Breast Center
• Baylor College of Medicine, Houston, TX

2
Baylor College of Medicine

• Dan L. Duncan Cancer Center is the 3rd NCI
  designated Cancer Center in Texas
• 38M in NCI funding (top 1/3 of all CC) 100M
  total in cancer related funding
• Number 2 in Texas in cancer-related grant funding
• Number 1 medical school in Texas in NIH grant
  funding
• Number 2 in the country in research expenditures
  in biological sciences
• Human Genome Sequencing Center
• Theme is Personalized Medicine

3
Patient Example

• 57 yo healthy man
• No psychiatric history
• Admitted for suicide attempt
• Recent history of bad cold with cough
• Took 1 tablespoon of Robitussin DM qid the day
  before and day of admission

4
Example (Cont)

• Situation did not make sense
• Screened for a variant of a liver enzyme that
  degrades dextromethorphan (DM)
• Found to have a normal variant called 2D64 (slow
  metabolizer)
• Thus, high dose of DM plus slow degrading in
  liver led to build up in blood, mental confusion
  and suicide attempt

5
Example 2

• 45 yo woman with breast cancer diagnosed 2 years
  ago and on tamoxifen
• Tolerating tamoxifen very well
• Also on blood pressure meds and Prozac for mild
  depression
• Found to have a recurrence of her breast cancer

6
Example 2 (cont)

• Tamoxifen is metabolized by the same liver enzyme
  to its active metabolite
• Prozac and several other drugs block this enzyme
• Also 10 of women have the slow metabolizer
  enzyme variant like the prior example
• Thus, tamoxifen may have been ineffective in this
  woman because of the Prozac or because she might
  have had the slow metabolizer variant

7
Example 3

• 68 yo woman presents with a 4 cm breast cancer
  with tumor spread to 1 lymph node under the arm
• The tumor was ER, meaning estrogen driven
• Treated by lumpectomy and radiation
• She has a 40 chance of having seeds spread from
  the breast to other parts of the body

8
Example 3 (cont)

• In the past all patients like this would have
  received chemotherapy and tamoxifen
• But many patients did not need the chemo because
  it was not going to work in them
• A piece of the tumor was sent for the Oncotype DX
  assay which measures the activity of 21 of the
  25,000 genes in all cells

9
Example 3 (cont)

• Test tells us what the chances of having seeds
  are in this patient
• Also tells us whether she would receive benefit
  from chemo or whether tamoxifen alone is best
• The score in the patient was only 16 which is
  low despite the large tumor and spread to the
  lymph nodes
• Her risk of recurrence is only about 10 and
  chemo will be ineffective

10
Recurrence Score as a Continuous Predictor
Paik et al, SABCS 2003
Paik et al, SABCS 2003
11
Example 3 (cont)

• Thus, NO CHEMO in this patient

12
Personalized Medicine

• Use the genetic makeup of the patient and that of
  the tumor (diseased tissue) to determine optimal
  treatment for that patient.
• Provide sensitive and compassionate care for the
  individual patient.

13
Genetic Changes
DNA (25K genes)
RNA
Protein
14
Types of Genetic Changes

• Single change in one of the 4 molecules that make
  up the genetic code (A,T, G, C)
• ATGCGGTACCCAATGGA
• ATGCGGTACCAAATGGA
• Present in all cells at birth changes the
  protein very little so it has slightly different
  effects (variation in the activity of an enzyme
  such as 2D6)
• A normal variant. Subtle differences among
  people. May also cause susceptibility to disease
  such as cancer BRCA1 and 2

15
Genetic Changes (cont)

• Change in the gene after birth only in a specific
  tissue, not in all cells
• These changes are called mutations
• May cause disease like cancer if occur in genes
  controlling cell growth, etc
• Usually need several mutations for a cancer to
  form

16
New York Times Cover, Science Section December
25, 2008
17
Genetic Changes and Cancer

• Hundreds of the 25,000 genes can contribute to
  cancer if they are mutated
• For a cancer to form many genetic changes are
  required
• Therefore, there are many possible combinations
  of changes that can cause cancer

18
Genetic Changes and Cancer

• The changes that cause a cancer in an individual
  patient is called the molecular signature or
  genetic fingerprint
• The signature dictates how aggressive it will be
  and what treatment will work best
• We can now measure the genes in a tumor and
  classify the patient on the basis of its gene
  signature into groups

19
Implications

• Determine the particular gene fingerprint in each
  tumor.
• Learn whether the tumor can spread and what genes
  are controlling it.
• Individualize treatment according to the
  fingerprint.

20
Identify the Tumor FingerprintGene Profiling

• Impossible to measure all 25,000 genes one at a
  time
• New techniques to measure all genes
  simultaneously
• Gene arrays

21
5
3
Approximately 1.2M features on each chip
22
(No Transcript)
23
Research Theme

• Identify and characterize the gene alterations
  causing cancer development and progression and
  translate that information to improved
  prevention, diagnosis and treatment of the
  disease.

24
Gene Expression
Residual Tumor Size
Chang, J. C., E. C. Wooten, et al. (2003). "Gene
expression profiling for the prediction of
therapeutic response to docetaxel in patients
with breast cancer." Lancet 362(9381) 362-9.
25
What Does the Future Hold?

• Predict susceptibility to certain diseases based
  on the persons genotype
• Determine which medicines should or should not be
  used in a specific person
• Specify treatment based on the genes driving the
  tumor or disease
• Gene therapy to correct the abnormality

26
Gene alterations normal and disease

• Tip of the iceberg cancer genes, normal
  variants contributing to cancer, variants causing
  different response to meds, susceptibility genes

Unknown
27
Breast Cancer Key Events in History
Mortality
WOC
1990
1980
1970
2008
SN bx
HER2
Systemic disease
Surgery Hormone therapy
AIs
ER, PR tamoxifen
Adj. therapy Lumpectomy Screening
Prevention
Personalized medicine
BRCA 1,2
Targeted Rx
28
(No Transcript)
29
The Breast Cancer Problem

• Very common 1 in 8 newborn females
• Western life style prolonged estrogen exposure
• Hereditary gene mutations (BRCA1, BRCA2) uncommon

30
Breast Cancer (cont)

• 2/3 of cases above age 50
• Most common cause of death from disease in 25 to
  50 year olds
• African Americans have lower risk but higher
  mortality
• No two breast cancers are exactly alike
• They have their own genetic fingerprint

31
How Does Breast Cancer Evolve?
Mutations in Genes
Normal Duct Cells
Precancerous cells
In Situ Cancer
Invasive Cancer
E
E
E
E
E
E