12 - Drug development and clinical trials

Question 1

How long does a drug typically take to develop?

Answer 1

10 years from discovery through to market

Question 2

Discovery vs development?

Answer 2

Discover = the molecule is synthesised
Involved test tubes, cell cultures
Development involves clinical trials and getting it ready for general use in the market

Question 3

How much does it cost to develop a drug

Answer 3

3 billion per drug that is successfully developed

Question 4

What portion of drugs that are tested in humans reach the market

Answer 4

Only 1/10

Question 5

Why is patency important?

Answer 5

In order to make money to cover the costs of the drug, developers rely on the patency of 18 YEARS from DISCOVERY (is then open to generic competition

Question 6

In which phases has there been an increase in cost during drug development?

Answer 6

Phases 2 and 3
Means these phases are becoming LESS EFFICIENT
Discovery and preclinical development is about the same cost

Question 7

What are the 3 top revenue earning drugs?

Answer 7

IMMUNOLOGY (things that end in ab are antibodies)

also infectious disease and oncology

Question 8

What is a blockbuster drug?

Answer 8

makes more than 1000 million/1 billion a year in revenue

Question 9

What are the 4 phases of drug development?

Answer 9

0 (PREDICTION of what effective doses would be for humans using non-human species. Helps to determine reasonable starting doses for phase 1)

1 (tolerability - PK and how well the drug is tolerated)

2 (effectiveness - does the drug work)

3 (safety - looking at toxicity)

4 (post marketing - send out to market and see how people actually use them)

Question 10

Biomarker?

Answer 10

Is a readily measurable indicator of response

is not a clinical outcome - is used to see if the drug seems to be having an effect that is related to the dose/conc

Question 11

Surrogate?

Answer 11

A biomarker that is known to be associated with the outcome (is a good predictor of the outcome) can be used as a surrogate endpoint i.e. viral load in HIV > mortality

Question 12

Outcome

Answer 12

How the patient feels/functions/survives

Question 13

What are the 2 reasons that clinical trials are done?

Answer 13

Learn and confirm

1. Learn - exploration of the unknown and development of hypothesis
2. Confirm - develop confidence and TEST a hypothesis

Question 14

Phase 0?

Answer 14

• non-clinical phase
• data from non-human animals gives indication of:
• probable mechanism of action
• what likely effective concentrations are
• major routes of elimination
• oral absorption properties

Question 15

Phase 1?

Answer 15

Tolerability

• start with very small doses as based on phase 0
• slow increase and watch what happens
• stop when adverse effects are noticed (max tolerable dose!)

Question 16

Difference between tolerability and tolerance?

Answer 16

Tolerability - the max dose you can tolerate before experiencing adverse effects
Tolerance - describes when you take a medicine repeatedly and you get tolerance and you do not react with as big of an effect

Question 17

What do you learn from phase 1?

Answer 17

• single and multiple dose PK
• adverse effect PD?

20-30 people

Phase 1 is about learning and tolerability

Question 18

Phase 2?

Answer 18

Effectiveness (confirm the drug works)
- Phase 2A
> proof of concept 
> yes/no decision point 
- Phase 2B
> learn dose-response curve (most actually do conc-response curves)
> learn effective doses
> learn target concentration 

• 100 people

Question 19

Phase 3?

Answer 19

Safety
- learn the adverse effects in a target population
- confirm the effective dose (need to consider method effectiveness i.e. did they take their pills?)
- learn the pharmacodynamics of surrogate/outcome
- PK and PD of predictable vairables i.e. what are the things that are predictive of individual differences for improving dose i.e. sex, age, other drugs
- about 1000 patients
( only find out about the more common adverse effects)

Question 20

Phase 4?

Answer 20

Post-marketing
- confirm effective dose
- confirm common adverse effects
- LEARN UNcommon adverse effects
- learn use effectiveness
> use in real clinical practice i.e. don’t like taste
- learn pharmacoECONOMICS (is based on the use effectiveness)

Goes form being in well controlled conditions to 100 000 in first few months. Risky time.

Question 21

4 examples of herbal alternative medicines?

Answer 21

digoxin (fox glove)
morphine
aspirin 
quinine
> all come from plant sources

Question 22

Why aren’t herbal medicines medcines?

Answer 22

Not approved by Medsafe to be safe and effective.
Cannot make claims.
They may be effective but there are no real testing to show they are safe and effective

There is no money in it - patent protection is unlikely

Question 23

Example of St Johns Wort drug interaction

Answer 23

St Johns Wort + cyclosporin > cardiac transplant rejection

Question 24

What alternative medicines commonly interact with HF medications?

Answer 24

• ST Johns
• grapefruit juice
• green tea

Question 25

What are the ABCS of clinical trial design?

Answer 25

A - Assignment B - Blinding C - Comparison S - Sequence

Question 26

2 ways to Assign in clinical trials?

Answer 26

1. First come first served 2. Randomized > used to determine which subjects get which treatment

Question 27

First come first served assignment?

Answer 27

Often introduce bias or confounding. I.e. first group given treatment during one season. There is a loss of blinding.

Question 28

Randomized?

Answer 28

Needs to be balanced to ensure similar people in each group. If different subgroups might have different responses i.e. have had a previous stroke then it is common to stratify the randomization sequence.

Question 29

Types of blinding?

Answer 29

Open - unblinded Single blind - investigator knows subject doesn't i.e. dr giving surgery Double blind - both are not aware (most common) > double dummy: can tell what they get from formulation i.e. pill and inhaler Triple blind - no one knows. Error. i.e. randomisation sequence is lost or misinterpretated. No one knows what treatment was given

Question 30

Why is blinding used?

Answer 30

To reduce bias due to expectation of treatment

Question 31

3 kinds of comparisons as the control group

Answer 31

1. Active treatment comparison 2. Placebo 3. Standard treatment

Question 32

Active treatment comparison

Answer 32

1. Dose control (dif doses) 2. Concentration control (helps to reduce the influence of random between individuals in PKs) 3. Biomarker control (reflects the effect of the drug and so can be used and altered to control the intensity of treatment)

Question 33

Why use a control group?

Answer 33

To account for factors that might influence the outcome that are NOT experimentally assigned

Question 34

Placebo

Answer 34

Inactive substance. Best. If there is genuine uncertainty about the effect of the active treatment then it is considered ethical to randomise the placebo.

Question 35

Standard treatment?

Answer 35

i. e. known to be effective and is normally always used (most common in clinical trials) - non-inferiority - add-on

Question 36

Sequence?

Answer 36

The sequence of treatments can influence what is learnt from a trial and the kind of bias that can arise - parallel - cross over - titration > forced or flexible

Question 37

Parallel sequence?

Answer 37

Most robust with the least amount of assumptions needed to be made for analysis. There are different treatments assigned to different groups of subjects. Good design to answer 'does the drug work?' BUT gives unclear answers to learning questions that ask about the shape of the dose-response relationship of what dose is needed for a particular effect to be achieved

Question 38

Crossover?

Answer 38

Uses 2 or more treatments for each individual. Compare a person to themselves. More efficient. Allows individual dose response curves to be observed > CONS - carryover treatment effects (if long half life)

Question 39

Titration?

Answer 39

Special kind of crossover design. Forced: Giving a fixed sequence of doses to each subject to learn about dose response relationship (10mL > 20mL) Flexible: Start with a low dose and if subject responds then the dose is kept constant. Dose is only increased if the desired response isn't reached

Question 40

Analysis perspective?

Answer 40

There are perspectives to be considered when analysing a clinical trial. Intention to treat analysis As treated

Question 41

Intention to treat?

Answer 41

- only considers the treatment assignment - does NOT take into consideration whether or not the patient ACTUALLY took the treatment - means that the size of the treatment effect will be UNDER estimated if some subjects DONT take the active treatment or vv. - useful in making pharmacoECONOMIC decisions where the cost of the drug has to be paid whether or not the drug was actually taken i.e. assumes use effectiveness (BIAS)

Question 42

As treated analysis?

Answer 42

Method effectiveness - takes into account info about what the subject actually took - less likely to have underestimation bias - more suited to making scientific decisions