Unblinded trials

There is one final opportunity to revisit the proposed methods of statistical analysis prior to the breaking of the blind, or in an unblinded trial, before the statistics group have seen study data. This so-called blind review usually takes place around the time of database lock and the following lists some of the aspects of analysis that would generally be considered ... 

In two separate open, randomized, controlled but unblinded trials, 549 patients were divided into five treatment groups. In the first study there were 375 patients in three groups ... 

In a comparative, randomized, unblinded trial for primary therapy of invasive aspergillosis, 144 patients received either intravenous voriconazole (6 mg/kg bd on... 

Acute Aspirin Therapy for AF-associated Stroke A combined analysis of the 1ST and CAST trials indicated a 21% RRR (95% Cl —5 to 41) in the frequency of early recurrent stroke associated with acute aspirin therapy compared to placebo in patients with AF. No difference in early mortality or sICH was found. This finding was largely driven by the relatively large (about 25% RRR) benefit observed in the unblinded 1ST, compared to the smaller benefit (5% RRR) observed in the double-blinded CAST. 

Swartzman, L. C. 8c Burkell, J. (1998). Expectations and the placebo effect in clinical drug trials why we should not turn a blind eye to unblinding, and other cautionary notes. Clin. Pharmacol. Ther., 64, 1-7. 

The phase-three results were unblinded in March 2003 and overall, for all types of metastatic brain cancer, were reported as nonpositive, showing the drug to have efficacy but not at the statistical level set at the beginning of the trial by Alios Therapeutics Inc. for a successful phase-three trial. The subset analysis for the different types of metastatic cancer to brain, however, showed that breast cancer patients had a 100% increase in survival. The 500 plus patient trial also demonstrated RSR 13 to be extremely safe, confirming our early design feature of using a known antilipidemic drug, clofibrate, as a substructure. An NDA for RSR 13 with trade name efaproxaril was reviewed by the PDA on a fast track review for use in treatment of metastases of breast cancer to brain. 

Are there clear instructions for reporting of adverse events and serious adverse events (SAEs) There should be full instructions for the reporting of SAEs (including addresses and fax numbers), with time limits. It should be clear that these rules also apply to SAEs that occur in subjects who have finished the study. All SAEs that come to the knowledge of the trialist should be reported unless the protocol provides guidance or time limit when the authors can justify that the occurrence of the SAE could not be related to the treatment received in the clinical trial. In a blinded study, there should be clear instructions on when and by whom the code for a particular study subject should be unblinded in an emergency. 

Non-randomized concurrent clinical trials initially assign participants to a control or test group in a non-random fashion. These trials are run concurrently, but are unblinded. This introduces a danger that the control and intervention groups are not strictly comparable. 

It is good practice to pre-specify in the protocol, or certainly in the statistical analysis plan, the statistical method to be used for analysis for each of the endpoints within the confirmatory part of the trial. This avoids the potential for bias at the analysis stage, which could arise if a method were chosen, for example, which maximised the treatment difference. As a consequence changing the method of analysis following unblinding of the study in an unplanned way, even if there seem sound statistical reasons for doing so, is problematic. Such a switch could only be supported if there was a clear algorithm contained within the statistical analysis plan which specified the rules for the switch. An example of this would be as follows ... 

There will also be at least one other statistician involved closely with the activities of the DMC and this is the statistician who supplies data tables to the committee for their deliberations. This statistician should also not be otherwise involved in the trial as they will potentially be supplying unblinded information to the DMC and attending their meetings. In the way that these things tend to be organised these days this individual may be part of a CRO that is providing this service (and potentially other services) to the sponsor. See Pocock (2004) and the FDA guideline for further discussion on this and related points. The DMC should also receive details of individual SAEs, usually in the form of narratives and these will often be supplied directly from the sponsor. These patients can be unblinded by the independent statistician if this has not already been done. 

What can be allowed in relation to adapting the trial design based on unblinded data ... 

Although the primary goal of the analysis of a clinical trial should be to answer the questions posed by its main objectives, new questions based on the observed data may well emerge during the unblinded analysis. Additional and perhaps complex statistical analysis may be the consequence. This additional work should be strictly distinguished in the report from work which was planned in the protocol. ... 

Investigator s qualifications and agreements Communication with IRB/IEC Compliance with protocol Randomization procedures and unblinding Informed consent of trial sutbjects Safety reporting... 

Before a trial starts, a charter needs to be written and agreed upon by the trial sponsor and the committee. This charter describes the structure and operation of the committee and specifies its activities and responsibilities. The DMC should have access to fully unblinded data, with actual treatments and not just codes available for its review. Except in certain limited circumstances, trial integrity is best protected when interim data comparing treatment groups are seen only by the DMC members and statisticians preparing the interim report (Ellenberg et al., 2003, see also O Neill, 2006). 

In addition to manufacturing both the test drug and the comparator drug, these materials need to be blinded. A trial may be called a double-blind study when neither the subjects nor any of the people concerned with their evaluation and care know which treatment the subjects are receiving. While double-blind trials predominate, in some cases a single-blind trial is conducted if it is not possible to blind the subjects (or sometimes the investigators). When trials are conducted in which everyone knows what treatment a subject is receiving, the trial is said to be unblinded. 

Ondansetron has also been investigated for the treatment of fatigue associated with primary biliary cirrhosis but results have been disappointing. A randomised, controlled crossover trial (n = 54) examined the effect of ondansetron 4 mg three times a day versus placebo for a four-week period, before being crossed over for a further four-week period. The study concluded that the use of ondansetron did not offer a clinically significant reduction in fatigue compared to placebo [20]. However, the results of the study may have been affected as patients were effectively unblinded during the second phase of the... 

Kava is currently promoted for relief of anxiety, stress, and insomnia. Stress may be prolonged and difficult to cope with and affected individuals may suffer from insomnia. Kava has been promoted as an axiolytic agent with little risk for dependence or adverse reactions. An unblinded, comparative, crossover trial of kava (120 mg) and valerian (600 mg) was conducted, each agent administered for 6 weeks with a 2-week wash-out period between. This was followed with administration of a combination of the two compounds. Both stress and insomnia were measured regarding social, personal, and life... 

---