Clinical trials results
This page is about the publication of clinical trial results. There is information about
Trial results can be a long time coming. To fully understand the risks and benefits of a new treatment, researchers have to know how well treatments work in the long term. This may take up to 10 years or more. Sometimes full results, or perhaps partial results (often called an interim analysis) are available in a year or so. But it is more usual for trial results to take 2 to 5 years longer.
When analysing trial results, doctors look at end points. An example of an end point may be response to treatment, which means whether a cancer has shrunk. But that doesn’t give the whole story. Researchers also need to look at how long the response lasts (the duration of response).
The aims and design of a trial may also affect how long results take. A trial that is recruiting 2,000 patients will often take longer to run than one recruiting 100. But this also varies depending on what the trial is looking at. A trial looking at treatment for a rare cancer can often take longer to recruit than a trial for one of the more common cancers, such as breast cancer. Generally, the larger the trial, the more reliable the results are likely to be when we get them.
If doctors are trying to prevent cancer developing in the first place, or looking for causes, it can take 10 or 20 years to get the full results. This is simply because they need to follow up the people who took part in the trial and see how many of them go on to develop cancer decades later.
Since 2009, we have been writing trial results in plain English and putting them on CancerHelp UK. On the trials search page, you can tick the results box to show trials that have results available.
We include both early (interim) results as well as results that show long term follow up. We hope to write up the results of all the trials listed on the database since it began in 2000. Some will have already been published in medical journals. Others may not have been published yet, but results might have been presented at a conference or made available to us by the researchers running the trial.
Unfortunately, some trials never produce results. If we know that it is unlikely results will ever be available, we say so. This might happen because a trial didn’t recruit enough people to be able to show anything new. Again, we will say if this is the case. But we don’t always know why a trial has not reported any results, or why the trial team has not made them available to us.
If you have taken part in a clinical trial and want to know the results, the first thing to do is ask your doctor. If the results are available, your doctor should be able to find them for you.
Everyone who takes part in a clinical trial has a patient information sheet (PIS) to read when they join the trial. Once they have read and understood it, they sign a consent form to agree to take part in the trial. The PIS usually says what will happen to the results of the trial. It is very unusual for the researchers to say that they will let each individual patient know the results. It is much more common to suggest that patients ask their doctor.
Clinical trial results are usually published in specialist medical journals. There are many different journals published around the world. They contain reports and results of both clinical trials and laboratory research. Some of the most well known medical journals include
- The Lancet
- The Lancet Oncology
- The British Medical Journal (BMJ)
- The British Journal of Cancer (BJC)
You can search for articles and read summaries (abstracts). But you can't usually see the full articles without a subscription to the journal. These papers are not written in plain English and often use many medical, scientific and statistical terms. So they can be difficult to understand if you are not used to reading this sort of information.
If you have access to the internet, you can search for trials results on a number of websites. If you don't have your own computer, internet access is available in many public libraries these days. The librarian will be happy to help you with your search. Examples of websites that list articles from medical journals include PubMed and Doctors Guide. These are aimed at doctors, researchers and other health professionals, and unfortunately are not always user friendly or easy to read.
Sometimes an organisation or drug company running a trial releases results to the public, especially if the results are very promising. They may send out a press release or put the information on their website.
An organisation called The Cochrane Collaboration carries out systematic reviews. These are overviews of all the research into a specific issue. They look at the published results of all the trials that have investigated a particular treatment in a particular situation. They pull all that information together and draw conclusions. They then write a Cochrane Review and this is published in The Cochrane Library. The Cochrane Database of Systematic Reviews is available free in some countries, including the UK. Reviews like this are very important. It is often not possible to draw firm conclusions from looking at the results of a single trial. But if several have shown the same result, you can be a bit more certain.
One criticism of the current system is that not all trials results are necessarily published. There are several reasons for that. The most common is that the treatment being tested wasn’t really any better. But we still need these results. If they are known, it can save money and time by stopping another researcher trying the same thing. It can also help to present balanced information about treatments to doctors. If one trial shows a treatment to be successful and another doesn’t, doctors know that they can’t be too certain about the likely success of that treatment until they know more.
Many trials now include a quality of life study. When treatments in a trial work equally well, it is important to find out if there is any difference in how people felt during the treatment. Even when a trial doesn’t show that one treatment is better than another, the quality of life study may show that one had fewer side effects. One treatment may be hospital based, while you can take the other at home. The quality of life study should show up which patients preferred. These are clearly still valuable results for the trial.
The research community are thinking about the best way to make sure all results are published in the future.
A few years ago, cancer research wasn’t very well organised in the UK. Researchers from different parts of the country started to work together and share information more and more. But there was no central record of all the clinical trials and laboratory research going on in the UK. So it was difficult to try and pull all the results together. But this is now changing.
In April 2001 the National Cancer Research Institute (NCRI) was formed to co-ordinate cancer research in the UK. The NCRI are currently working on a project called The Cancer Informatics Initiative. They want to bring together all the information we have on cancer research. As part of this they have developed a data sharing policy. They hope that everyone involved with publicly funded research will share their data in line with this policy. This will include results from both laboratory research and clinical trials.
In May 2004, the European Union brought in a Clinical Trials Directive. This covers many aspects of running clinical trials in Europe. Among other things, it aims to improve the recording and reporting of clinical trials.
In 2008 the National Cancer Intelligence Network (NCIN) was launched. NCIN is part of the National Cancer Research Institute and brings together cancer registries, doctors, health service researchers, the Office for National Statistics and others. These organisations now share information so that in the future we will have helpful data about how many people get cancer, how well treatments work, and the outcomes of treatment for different types and stages of cancer.
Increase text
Decrease text
Print this page
