Experimental design of cancer clinical trials

Question 1

Why is it good to give pateints choices?

Answer 1

Helps you determine treatment plan

Question 2

What would you want a clinical trial to do?

Answer 2

Prevent diseases

Cure infections

Cure cancer

Reverse diabetes

Treat high blood pressure

Etc…

Question 3

What are randomised trials?

Answer 3

Randomly selecting a group of people e.g. by pateitns, within patient, across groups/towns, cohorts, observational, sample collections

Question 4

What are N=1 randomised trials and how is it hard in cancer?

Answer 4

give many different therapies in different orders to different patients however this is hard in cancer trials as by the time you give person 1 the fourth treatment their cancer could be completely different due to the other 3 treatments.

Question 5

Why would you screen someone across groups and towns?

Answer 5

if someone knew their Neighbour was getting a beneficial test e.g. screening for breast cancer they might try and get a test done as well therefore making this less random and so screening by town is fairer for everyone.

Question 6

What are observational tests?

Answer 6

comparing across geography and time

Question 7

What would be a cohort study?

Answer 7

you follow more than one group of people/patients

Question 8

What are some cons to randomised disease trials e.g. negative controls?

Answer 8

You cant have any unless its a healthy person as you cannot give someone suffering from a disease no treatment

Question 9

Why are randomised trials good?

Answer 9

They reduce bias
The only difference between groups in the intervention

Question 10

How would you know if the result you got was any good?

Answer 10

minimize all other differences e.g. age, spread, type of cancer etc. However many unknowns will influence an outcome

Aim to have the treatment/ intervention being the only difference between the groups.

Statistics also help you answer complex questions

Question 11

How many false positives is acceptable in a clinical trial before they are thought to be biased?

Answer 11

Medicine considers less than 5% e.g. if you were to toss a coin in the air it would land on heads 5 times in a row.

Question 12

Should randomised trials have individuality?

Answer 12

No

Question 13

Breast cancer and lymph nodes example?

Answer 13

The number of lymph nodes next the breast cancer involved in the cancer correlates to survival rates.Eg.If there is no lymph node involvement survival rates are higher than if there is.

Question 14

How would lymph node involvement in breast cancer influence the trial?

Answer 14

In clinical trials this would mean you cannot treat someone with no lymph node involvement with antibody A and a group of people with lymph node involvement with antibody B and say that antibody A is a better treatment because they are different cancers with different survival chances.

Question 15

Herceptin cancer treatment example?

Answer 15

Is an expensive monoclonal antibody cancer drug which involved counting how many HER2 genes were mutated in the breast cancer. If it was less than 3 then survival rateswherehigher than those who have more than 3 genes mutated.

This gave rise to the treatmentherceptinwhich reduced the risk of dying from breast cancer but it only worked in HER2+ cancers which is only about 15% of all breast cancer

Question 16

How do you use translational treatments?

Answer 16

Collecting samples (blood, tissue) and linking them to what happens to the patients

May be within another trial

Prospective or retrospective

Question 17

What drugs are not good for all cancers and why is this example strange?

Answer 17

Cancers with more than 2 copies of chromosome 17 did better when given the drug anthracycline than those who have the normal number of chromosome 17 which is unusual (you’d think it would be the other way round).

Question 18

What are examples of qualitive research and how is this hard to measure?

Answer 18

quality of life for patients/relatives/carers/staff, cognitive function – stress, chemo brain, what bits of cognitive function matters, level of disability, food diariesetcas these are different for all people.

People will often lie about how bad side affects are as they want to live and the doctor might reduce their dose

Question 19

Why are overall cancer survival trials tricky?

Answer 19

Say someone is given a drug and lives 15 years, does this mean in those 15 years the drug shouldn’t be given to any one else?

Also, accidental deaths or deaths due to over illnesses could occur

Studies will be larger and more expensive in overall survival cancer trials.

Question 20

What would can trials measure (list)?

Answer 20

Overall survival (people could die from other things)
Time to progression/recurrence
Response rates (not all diseases are measurable), usually imaging
Toxicity– symptoms matter more to some people than others e.g. hair loss
Quality of life for patients (reported or experienced – people down play experiences if they think the doctor will change their dose because they want to live).
Biological changes in disease/patient

Question 21

What are survival surrogates?

Answer 21

Surrogate measurements are things that are linked to the disease that you can measure if you cannot measure the disease or thing itselfe.g:

• seat belts
• smoking cessation

Question 22

What would you study in incurable disease?

Answer 22

Patient specific things e.g. reduce treatments, increase quantity and quality of life

Question 23

What would you focus on in early disease?

Answer 23

Cure with no recurrance

Question 24

The different phases of cancer trials? Phase 0

Answer 24

Phase 0 – does the PK happen as expected, does the body respond how you are expecting it.

Question 25

The different phases of cancer trials? Phase 1

Answer 25

Phase 1 – demonstrate acceptable toxicity, in cancers these people will have no other drugs.

Question 26

The different phases of cancer trials? Phase 2

Answer 26

Phase 2 – is there evidence that the activity seen in test tubes is the same seen in humans.

Question 27

The different phases of cancer trials? Phase 3

Answer 27

Phase 3 – compare standard of care

Question 28

The different phases of cancer trials? Phase 4

Answer 28

Phase 4 – refine place in standard care –who gets it, should they get it.

Question 29

What questions would you need to think about when phase 0 trials happen?

Answer 29

Does it kill cancer in a test tube (although not everything grows in test tubes)

Dose it work in xenografts – labtumoursin lab animals (there is little similarities between animals and humans), often immunodeficient animals to get atumourgrowing and the role of the immune system in cancer is important.

Can we formulate it? - can you deliver it to someone.

Question 30

What would phase 1 involve in patients with advanced cancer?

Answer 30

Identification of maximally tolerated drug (bone marrow if the surrogate for cytotoxicity as if it can kill blood cells it can kill cancer cells)

New agents often target onocogenes

Question 31

Do you need to know the target in phase 2 unselected?

Answer 31

Yes

Question 32

What does phase 2 studies involve?

Answer 32

Usually heavily treated patients, pick sub-groups and randomise patients

Pre-operative (short term affects) can also be looked at

Question 33

What are subgroups of clinical trials?

Answer 33

Molecular defined subgroups - different outcomes, different treatments

Patient defined subgroups - risk of primary event (relapse, death), risks of abnormal phasrmacology

Other standard therapies - location

Question 34

What do phase 3 studies measure?

Answer 34

They are large multinational, multicentre studies which measure what happens in real life. They are powered to detect meaningful differences but its hard to deal with individuality and hard to identify those patients who wont benefit

Question 35

How are phase 3 trials analysed?

Answer 35

Usually analysed on an intent to treat basis (measure how new compares to old, so need to include patients who need this treatment)

However, you need to control the population, balance factors between the old control and new research arms

Question 36

Pros of phase 3 trials?

Answer 36

Defined patients in near-routine practice
Intensive monitoring of patients and outcomes
Protocol lead therapy
Treatment is the only difference between the groups

Question 37

Cons of phase 3 clinical trials?

Answer 37

Does not reflect all disease
Intensive biology difficult
Often uses older treatment
Not individualised