Essential Bioinformatics and Biocomputing (LSM2104: Section I) Biological Databases and Bioinformatics Software Prof. Chen Yu Zong Tel: [phone] Email: [e-mail] http://xin.cz3.nus.edu.sg Room 07-24, level 7, SOC1, NUS January 2003

1
Advanced Bioinformatics Lecture 9 Drug resistant
cancerous mutation
ZHU FENG zhufeng_at_cqu.edu.cn http//idrb.cqu.edu.cn
/ Innovative Drug Research Centre in CQU
???????????????
2
Table of Content

1. Differential drug efficacy
2. Pharmacogenetics
3. Pharmacogenetic response
4. Drug resistance mutation
5. Prediction of drug resistance

2
3
Differential drug efficacy
Different patients
Same symptoms Same disease
Same drug Same dose
Different Effects
At a recommended prescribed dosage (1) a drug
is efficacious in most (2) not efficacious in
others (3) harmful in a few.
Lack of efficacy
Unexpected side-effects
3
4
People react differently to drugs One size does
not fit all
Patients with drug toxicity
Genotyping
Patients with non-response to drug therapy
Patient population with same disease phenotype
Toxic responders Non-responders Responders
Patients with normal response to drug therapy
4
5
Why does drug response vary?
Different patients
Same symptoms Same disease
Same drug Same dose
Different Effects
Genetic Differences
Possible Reasons Individual variation By chance
Ethnicity Age Pregnancy Genetic
factors Disease Drug interactions
SNP
5
6
Why does drug response vary? Genetic variation

• Primarily 2 types of genetic mutation events
  create all forms of variations
• Single base mutation which substitutes 1
  nucleotide
• Single nucleotide polymorphisms (SNPs)
• Insertion or deletion of 1 or more nucleotide(s)
• Tandem Repeat Polymorphisms
• Insertion/Deletion Polymorphisms
• Polymorphism A genetic variation that is
  observed at a frequency of gt1 in a population

6
7
Single nucleotide polymorphism (SNP)

• SNPs are single base pair positions in genomic
  DNA at which different sequence alternatives
  (alleles) exist wherein the least frequent allele
  has an abundance of 1 or greater.
• For example a SNP might change the DNA sequence
• from AAGCTTAC
• to ATGCTTAC
• SNPs are the most commonly occurring genetic
  differences.

7
8
Single nucleotide polymorphism (SNP)

• SNPs are very common in the human population.
• Between any two people, there is an average of
  one SNP every 1250 bases.
• Most of these have no phenotypic effect
• Venter et al. estimate that only lt1 of all human
  SNPs impact protein function (lots of in
  non-coding regions)
• Some are alleles of genes.

8
9
Tandem repeat polymorphisms

• Tandem repeats or variable number of tandem
  repeats (VNTR) are a very common class of
  polymorphism, consisting of variable length of
  sequence motifs that are repeated in tandem in a
  variable copy number.
• Based on the size of the tandem repeat units
• Venter et al. estimate that only lt1 of all human
  SNPs impact protein function (lots of in
  non-coding regions)
• Repeat unit 1-6 (dinucleotide repeat
  CACACACACACA)
• Minisatellites
• Repeat unit 14-100

9
10
Insertion/deletion polymorphisms

• Insertion/Deletion (INDEL) polymorphisms are
  quite common and widely distributed throughout
  the human genome.

10
11
Due to individual variation

• 20-40 of patients benefit from an approved drug
• 70-80 of drug candidates fail in clinical trials
• Many approved drugs removed from the market due
  to adverse drug effects
• The use of DNA sequence information to measure
  and predict the reaction of individuals to drugs.
• Personalized drugs
• Faster clinical trials
• Less drug side effects

Pharmacogenetics
11
12
Pharmacogenetics

• Study of inter-individual variation in DNA
  sequence related to drug absorption and
  disposition (Pharmacokinetics) and/or drug action
  (Pharmacodynamics) including polymorphic
  variation in genes that encode the functions of
  transporters, metabolizing enzymes, receptors and
  other proteins
• The study of how people respond differently to
  medicines due to their genetic inheritance is
  called pharmacogenetics
• Correlating heritable genetic variation to drug
  response
• An ultimate goal of pharmacogenetics is to
  understand how someone's genetic make-up
  determines, how well a medicine works in his or
  her body, as well as what side effects are likely
  to occur.
• Right medicine for the right patient

12
13
Pharmacogenetics vs. pharmacogenomics

• Pharmacogenetics Study of variability in drug
  response determined by single genes.
• Pharmacogenomics Study of variability in drug
  response determined by multiple genes within the
  genome.

13
14
The study of variations in genes that determine
an individuals response to drug therapy.
Pharmacogenetics
Common variation in DNA sequence (i.e. in gt1 of
population)
Genetic Polymorphism SNPs INDEL VNTRs
Potential Target Genes are those that
encode Drug-metabolizing enzymes Transporters Dru
g targets
14
15
Determinants of drug efficacy and toxicity

• Patients response to drug may depend on factors
  that can vary according to the alleles that an
  individual carries, including

• Pharmacokinetic factors
• Absorption
• Distribution
• Metabolism
• Elimination
• Pharmacodynamic factors
• Target proteins
• Downstream messengers

15
16
Examples

• EM phenotype Extensive metabolizer IM
  phenotype intermediate metabolizer PM
  phenotype poor metabolizer UM phenotype
  ultrarapid metabolizers

16
17

• Individual variations in drug response are
  frequently associated with three groups of
  protein
• ADME-associated proteins proteins responsible
  for the absorption, distribution, metabolism and
  excretion (ADME) of drugs
• Therapeutic targets proteins that can be
  modified by an external stimulus (drug
  molecules).
• ADR related proteins drug adverse reaction
  related proteins
• The factors in variations of drug responses
• Sequence polymorphism
• Transcriptional processing of proteins altered
  methylations of genes, differential splicing of
  mRNAS
• Post-transcriptional processing of proteins
  differences in protein folding, glycosylation,
  turnover and trafficking.

17
18
Medicines are not safe or effective in all
patients
18
19
Medicines are not safe or effective in all
patients
Drug Group Efficacy Incomplete/Absent
SSRI 10-25
Beta blockers 15-25
Statins 30-70
Beta2 agonists 40-70

when considered in further detail, we can see
that efficacy of some of our major drug classes
vary from 10-70 incomplete efficacy.
19
20
The needs of prediction of pharmacogenetic
response to drugs

• Pharmacogenetic prediction and mechanistic
  elucidation of individual variations of drug
  responses is important for facilitating the
  design of personalized drugs and optimum dosages.
• For most drugs, not all of the ADME-associated
  proteins responsible for metabolism and
  disposition of pharmaceutical agents are known.

20
21
The feasibility of prediction of pharmacogenetic
response to drugs

• A number of studies have explored the possibility
  of using polymorphisms as indicators of specific
  drug responses.
• Computational methods have been developed for
  analyzing complex genetic, expression and
  environmental data to analyze the association
  between drug response and the profiles of
  polymorphism, expression and environmental
  factors and to derive pharmacogenetic predictors
  of drug response
• A number of Freely accessible internet resources

21
22
The approach of prediction of pharmacogenetic
response to drugs

• Reported polymorphisms of ADME-associated
  proteins
• By a comprehensive search of the abstracts of
  Medline database

22
23
The approach of prediction of pharmacogenetic
response to drugs

• ADME-associated proteins linked to reported drug
  response variations
• Also by a comprehensive search of the abstracts
  of Medline database

23
24
The approach of prediction of pharmacogenetic
response to drugs

• Rule-based prediction of drug responses from the
  polymorphisms of ADME-associated proteins

the analysis of clinical samples of the variation
of drug responses
Used as indicators for predicting individual
variations of drug response

the results of genetic analysis of the
participating patients
24
25
The approach of prediction of pharmacogenetic
response to drugs

• Similar to the Simple rules-based method for
  using HIV-1 genotype to predict antiretroviral
  drug susceptibility (HIV drug resistant genotype
  interpretation systems)
• Comparative Evaluation of Three Computerized
  Algorithms for Prediction of Antiretroviral
  Susceptibility from HIV Type 1 Genotype. J
  Antimicrob Chemother 53, 356-360 (2004).

25
26
Basic idea of using HIV-1 genotype to predict
antiretroviral drug susceptibility
Phenotype resistant drug 1, drug 2, drug 3
HIV-1 genotype 1
Phenotype susceptible drug a, drug b, drug c
Phenotype resistant drug 2, drug 3, drug a
HIV-1 genotype 2
Phenotype susceptible drug b, drug c
Phenotype resistant drug 1, drug 3
HIV-1 genotype 3
Phenotype susceptible drug 2, drug a
Phenotype resistant

Phenotype susceptible
26
27
The approach of prediction of pharmacogenetic
response to drugs

• Examples of the ADME-associated proteins having a
  known pharmacogenetic polymorphism and a
  sufficiently accurate rule for predicting
  responses to a specific drug or drug group
  reported in the literature.

27
28
Limitation of Simple rules based methods

• Low predicting accuracies of simple rules based
  methods 50100 (comparable to those of 8197
  for predicting HIV drug resistance mutations from
  the HIV resistant genotype interpretation
  systems)
• Variation of response to some drugs associated
  with complex interaction of polymorphisms in
  multiple proteins
• Simple rules
• Limited predicting capacity for prediction of
  drug responses
• The basis for developing more sophisticated
  interpretation systems like those of the HIV
  resistant genotype interpretation system

28
29
Other methods

• Computational methods for analysis and prediction
  of pharmacogenetics of drug responses from the
  polymorphisms of ADME-associated proteins
• Examples recently explored for pharmacogenetic
  prediction of drug responses
• Discriminant analysis (DA) Chiang et al., 2003
• Unconditional logistic regression Yu et al.,
  2000
• Random regression model Zanardi et al., 2001
• Logistic regression, 2004 Zheng et al., 2004b
• Artificial neural networks (ANN) Chiang et al.,
  2003 Serretti et al., 2004
• Maximum likelihood context model from haplotype
  structure provided by hapmap Lin et al., 2005

29
30
Examples

• Statistical analysis and statistical learning
  methods used for pharmacogenetic prediction of
  drug responses

30
31
What is the drug resistance?

• Organisms are said to be drug-resistant when
  drugs meant to neutralize them have reduced
  effect or even no effect.
• Main cause of drug fail during the treatment of
  infectious disease , cancers (chemotherapy)
• Main cause of the drug resistance
• Mutation in drug-interacting disease proteins
  (genetic resistance)
• Development of alternative disease related pathway

31
32
Example of drug resistance mutations

• HIV-1
• Protease mutations (could be quickly developed)
• Integrase mutations

Henderson L. and Arthur L. 2005. NIH AIDS
Research and Reference Reagent Program
32
33
The needs for drug resistance mutations prediction

• The molecular analysis of drug resistance
  mechanisms
• Design new agents to against resistant strains
• Guide the clinical regimen to fight with disease

33
34
Methods for mechanistic study and prediction of
resistance mutations

• Structure-based approaches
• molecular modeling approach
• evolutionary simulation model
• neural network model
• Sequence-based approaches
• Statistical learning methods
• Neural networks (NN) (classification,
  association, regression)
• Support vector machines (SVM) )(classification,
  regression)
• Decision tree (DT)
• Simple rules (HIVdb, HIValg, ARS, and VGI etc)

34
35
Methods for mechanistic study and prediction of
resistance mutations

• Simple rules

Phenotypic
Drugs
Protein Mutations
Genotypic
35
36
Methods for mechanistic study and prediction of
resistance mutations

• Simple rules

susceptible potential low-level
resistance low-level resistance Intermediate
resistance high-level resistance
36
37
Methods for mechanistic study and prediction of
resistance mutations

• Simple rules

37
38
Projects QA!
Biological pathway simulation
2. Computer-aided anti-cancer drug design
3. Disease-causing mutation on drug target
Any questions? Thank you!
38