Trialome
Trialome is a body of knowledge (or data) generated by human clinical trials in medicine that goes beyond and extends existing clinical trial registration data from clinical trial registrieswith trial results data (published articles, raw data, summary data).[1][2][3] Studies show that only 46% of trials publish their results within 30 months from trial completion.[4] Clinical trial registries, such as ClinicalTrials.gov enable trial record managers to register a study in a register, link publications reporting trial results and to upload basic summary results. In USA andEU laws and regulations exist, such as Food and Drug Administration Amendments Act of 2007, that mandate trial registration and basic summary results submission for certain clinical trials. Public knowledge about all trials is important to prevent clinical trial sponsors from repeating again a failed trial.[5] Updating and creating the knowledge contained in trialome is facilitated by mandatory clinical trial registration policies and journal editor policies (e.g., manuscript requirements created by International Committee of Medical Journal Editors)
Contents
Clinical Trial Registries
Clinical trials registries play a key role in providing evidence that a trial exist. In some countries (e.g., USA), there are legislative mandates that certain clinical trials must be registered. The largest 5 registries are [2] (as of August 2012): ClinicalTrials.gov (130756 trials), EU register (18660 trials), ISRCTN (10853 trials), Japan registries network (JPRN, 10511 trials), Australia and New Zealand (ANZCTR, 6916 trials). The largest registry (ClinicalTrials.gov) contained more than 6 times more studies than the second largest registry.
Scientific publications with trial results
Significant trial result information is provided by trial investigators when they publish in a scientific journal about the conducted trial. However, a 2013 study of trials registered in ClinicalTrials.gov registry showed that only 27.8% trials had a linked result article.[3]
Due to a large number of registered and conducted trials and large number of publised trial result articles, it is not easy to link correct article to a correct trial. There are two types of trial-article links.[3]
Abstract link
One is created by the article authors when they write the article abstract and is often called abstract link. The International Committee of Medical Journal Editors requires an article to mention in article abstract the trial ID within the registry. About 23% of registered trials have linked result articles using the abstract link.[3]
Registry link
The second link is created at any time after article publication. A trial registry record manager updates the trial registry record with the article PubMed ID (PMID) that reports trial results. This type of link is used less often. About 7.3% of registered trials had linked result articles using the registry link.[3] Not all clinical trial registries enable trial registry record managers to provide a result article reference (e.g., ClinicalTrials.gov and ISRCTN and Australian New Zealand Clinical Trials Registry (ANZCTR) do support this functionality).
Posting trial results to public repositories
In USA, certain human clinical trials, due to legal mandate, have to post their results within 1 year after trial completion into ClinicalTrials.gov registry. Results of a clinical trial can be posted as basic summary results or detailed, individual patient data.
Summary trial results
Only some clinical trial registries enable uploading of trial results. For example, ClinicalTrials.gov and ISRCTN registry. A study using ClinicalTrials.gov showed that only 26.6% trials in the sample of 8907 trials provided basic summary results.[3] While summary results provide some insights into the trial, 59.1% of basic summary trial results submission consist of only the mandatory fields that do not include many important parameter for assessing the overall outcome of the trial.[3] Minimum legally required summary results consist of items such as number of participants starting and completing each trial arm, gender and age baseline characteristics, at least one outcome measure and a table of adverse events. ClinicalTrials.gov supports submission of more summary results items. Such submission can be considered extended results and those in addition to providing the legally required basic summary result items, also provide optional result items that are guided by standards for Good Clinical Practice and CONSORT statement and referred to as “Good reporting practice”. Optional items in participant flow report additional counts of patients “Lost to follow-up” or “Severe Non-Compliance to Protocol”. Other optional items in the outcomes measures module provide p-values of statistical significance of each individual primary or secondary outcome measure and information about the statistical tests used.
Individual patient results
In meta analyses, it is important to sometimes analyze individual patient data. Some trial sponsors (e.g., Roche and GSK) made a promise to make individual patient trial results available to qualified researchers. [6]
See also
- Clinical trials registry
- ClinicalTrials.gov
- Publication Bias
- Uniform Requirements for Manuscripts Submitted to Biomedical Journals
- List of omics topics in biology
References
- ^ Huser, V.; Cimino, J. J. (2012). "Precision and Negative Predictive Value of Links between ClinicalTrials.gov and PubMed". AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium 2012: 400–408. PMID 23304310.
- ^ a b Huser, V.; Cimino, J. J. (2013). "Evaluating adherence to the International Committee of Medical Journal Editors' policy of mandatory, timely clinical trial registration". Journal of the American Medical Informatics Association. doi:10.1136/amiajnl-2012-001501.
- ^ a b c d e f g Huser, V.; Cimino, J. J. (2013). "Linking ClinicalTrials.gov and PubMed to Track Results of Interventional Human Clinical Trials". In Sampson, Margaret. PLoS ONE 8 (7): e68409. doi:10.1371/journal.pone.0068409.
