Research updates
Phase III clinical trials   page 7
Go back a page
Go forward a page
6. The practicalities Link to the Medical Research Council web site
Trials can be very expensive to run. Prevention and screening trials are generally much more expensive than treatment trials (because they require more people to assess the intervention as generally there will be proportionately few endpoint events). For example the major trial of ovarian cancer screening in post menopausal women announced in March 2000 involves 200,000 women from 12 collaborating sites in the UK and will take about ten years to complete. The cost will be about £24 million pounds.
Cover of a protocol document
Figure 7. The protocol (document) for a trial. The purpose of the trial is to compare which of two chemotherapy regimens would be the better in the treatment of a recurrent malignant glioma (a type of brain tumour). Click on the links below to see pdf files of:
Timeline graphic with roll overs
Who is involved? What do they do?
Reporting back
The results of trials for new treatments are reported to a number of different groups in different ways. These include:
  • the scientific community
  • doctors
  • the general public.

The methods of reporting include presentations by the research team (usually at scientific conferences), peer review articles in journals, interviews and news reports on the TV and in mainstream newspapers.

The public find stories about health interesting; newspaper sub-editors like ‘cure’ or ‘scare’ in headlines - it is their job to produce an article that captures the public's imagination. New treatments are very regularly reported in the media but the reports can be a nuisance - raising false hopes when a treatment is only at the stage where it has been shown to have activity in a laboratory experiment. In some cases a ‘trial’ is described but it turns out to be only Phase I or II. Nevertheless it is often reported as though the new treatment is round the corner for everyone.

The next time you see a report of a miracle treatment, try to consider and discuss what stage you think it is at. You could research it further by looking it up in a journal such as The Lancet.

Go back a page Go to the top of the page Go forward a page
Question 6
A Phase III clinical trial has many similarities to the investigations you carry out during your course. The stages above are similar to the stages you go through when you plan an investigation. How do they correspond?
The stages roll over the diagram to find out more about each stage.
1. The proposal. Trials of new drugs (before licensing) are usually conducted by the company developing the drug. Indeed, they may be responsible for all aspects of the trial. In this case, the process for stages 1 and 2 will not necessarily be as described here.
A multi-disciplinary team (including at least one doctor who is active in the field, Trial Managers and statisticians who are experienced in clinical trials) meet together to draw up an outline proposal for the trial. Nowadays the team usually includes a patient representative. Their first step is to identify the important clinical question which they think a trial can answer.

The statistician will address the key question ‘How many patients might we need?’ but practical matters such as the availability of patients and resources must be considered too. Sample size calculations are based on expectations of how patients will fare on the control arm (based on previous trial data or historical data) and what differences may be expected between the research arm and the standard arm in terms of safety as well as efficacy (based on data from Phase II trials).

They then set out how the trial will be designed and organised.

2. Peer review and funding. The proposal is submitted for peer review and funding. If funding is granted, the trial proceeds to stage 3.
The proposal is examined by experts in the disease area and on clinical trials (peer review).

Funding may come from the Medical Research Council, from medical charities (e.g. Cancer Research UK, British Heart Foundation) or from the Department of Health.

Peer review may result in changes being made to the proposal (including the design of the trial).

Funding is granted or not. The decision will depend both on the quality of the proposal and the availability of funds. There are sometimes not enough finances to fund all trials that deserve funding.

3. Development of full protocol. Trials are run according to set procedures; the design and performance of each experimental procedure is clearly formulated in the protocol.
The multidisciplinary team described in Stage 1.

(In a drug trial one or more drug companies may also be involved and may supply the drug and prepare the placebo.)

They draw up a full description of how the investigation will be carried out. This is called a protocol. The protocol includes:

Background, rationale, eligibility, assessments and when they are to be undertaken, treatments, adverse events and what to do if these occur, statistics, Patient Information Sheet, Patient Consent form, case report forms (for collecting data) You can see examples of the forms using the links under Figure 7

4. Ethical approval. The protocol is now submitted to the ethics committees (see page 2) for local and national ethical review and approval.
Multicentre Research Ethics Committee (MREC). Multicentre because the proposal must be approved by MREC if it is to be run at more than one site. There are 12 MRECS in the UK.

Local Research Ethics Committee (LREC) for each of the sites involved (even if the proposal already has MREC approval). LREC approval covers any site within the (Strategic) Health Authority in which the LRECs are based. There may be more than one LREC in a (Strategic) Health Authority.

The ethical review committees have to be independent of the investigator, the sponsor or any undue influence.

The MRECs and LRECs review the protocol. If necessary they may suggest amendments, for example to the Patient Information Sheet.

If the committees consider that the trial is ethically acceptable the protocol is approved.

5a) Promotion. Steps are taken to ensure that people are aware of the trial.
The Principal Investigator (PI) who is the lead medical doctor, the Trial Manager, and all the clinical staff (doctors and nurses etc.). The Principal Investigator and Trial Manager contact doctors about the trial. They may hold a meeting which often includes people who haven’t worked with them before. It might take place at the Clinical Trials Unit. Further details will be forwarded to those who are interested, including the MREC approval letter, case report forms (CRFs) and other necessary documents. The Trial Manager may also visit the sites.
The participating consultants then submit the plans for the trial to the Local Research Ethics Committee (LREC) for approval.
5b) The trial is added to trial registers
These databases keep a record of ongoing trials so that the community knows what is going on and does not duplicate research. Examples are current controlled trials ( and the national research register.
6. Databases. Databases are set up to help with the running of the trial.
The Trial Manager, Principal Investigator, statistician and computer programmer. Set up databases which have three functions:
  • to randomise patients
  • to store data collected on case report forms
  • to manage and chase queries or outstanding data.
7. Trial begins: recruiting patients. The trial is opened and recruitment to the trial starts.
Medical doctors, nurses and patients or healthy volunteers (e.g. in a screening trial) Doctors and nurses approach patients and ask them to participate. They explain the trial and give patients the Patient Information Sheet. The patients have a chance to ask further questions. To maintain confidentiality, details of the patients who consent are phoned or faxed to a secure location without explicitly revealing the patient’s identity. The patients are randomised and entered on a computer-database. They now receive the allocated trial treatment.
8. Recruitment on-going. The trial continues to recruit patients - amongst the many things that are happening in this period.
Principal Investigator, nurses, physicians, possibly surgeons, the Trial Manager, Statistician. The doctors or nurses post or email the (clinical) case report forms (CRFs) to the Trial Manager who ensures that the information is entered onto the database. The forms are generally quite short (one or two sheets of A4). Any unexpected entries, unusual results or missing data are checked.

The Trial Manager may visit some or all of the trial sites to assess progress, and sometimes to check that some of the key information on the CRFs corresponds to the patient's hospital notes for all or a sample of patients. This is called 'Source Data Verification'.

The statistician and Trial Manager prepare separate reports for the Trial Management Group and the Trial Steering Committee and the Research Ethics Committees.

Trial Management Group: includes the Principal Investigator, some physicians or surgeons, the Trial Manager, statistician and sometimes others. The Trial Management Group is responsible for the day to day running of the trial.
Trial Steering Committee: an independent body which includes a majority of members who are not involved in running the trial. The Trial Steering Committee ultimately considers options and makes any major decisions about the trial recommended by the other committees.
Data Monitoring Committee (DMC): an independent body whose members are not involved in the trial; it usually has at least 3 members (including one or more clinicians expert in the disease area and one or more statisticians with experience in trials). The Data Monitoring Committee: the members should be the only people to see the results separated by treatment group during the trial. They are independent and look at the trial from the point of view of trial patients, future patients and society in general. It is their responsibility to prevent patients being exposed to any excess risks by recommending the trial stops early if the safety or efficacy results are sufficiently convincing.
9. Trial stops recruiting. The trial stops recruiting when the target number of people is reached (in the case of a cancer trial this is typically after 2 to 5 years). A trial is occasionally stopped before this on the recommendation of the Data Monitoring Committee. This would happen if a treatment were shown to be effective or detrimental.
About 300 to 2000 patients will generally have been recruited and randomised. Follow up of patients continues after recruitment stops until enough endpoint events have been reported. Again this model is based on cancer trials and does not apply to all trials. Some kinds of trial may have a fixed time point when follow up stops.
10. Writing up. After peer review, the report will be published as a paper in a medical journal. It is important that all trials are written up and published regardless of whether the results are positive, negative or inconclusive. Some will be picked up as news stories and will reach a wider audience.
The clinical investigators and the statisticians and, to a lesser extent, the Trial Manager. The Data Monitoring Committee are involved in an advisory capacity. They analyse the results and write a report. This will include their interpretation of the results and the implications for the care of future patients.