CT SOP version no 1.0.

Main changes from LM SOP no.2.05 version no. 3.1. Adapted to the wording of the clinical trial regulation no 536/2014.

The sponsor has the overall responsibility for the preparation and review of overall procedures for randomisation, blinding and unblinding in clinical drug trials.

The sponsor’s responsibilities shall be described in the quality system of the sponsor institution. Tasks are delegated according to SOP Roles and Responsibilities in clinical trials implemented in the institution.

The sponsor may transfer any or all of the sponsor's trial-related duties and functions to a third-party vendor such as a Contract Research Organisation (CRO), but the ultimate responsibility for the quality and integrity of the trial data always resides with the sponsor. The sponsor should ensure oversight of any trial-related duties and functions carried out on its behalf. Transfer of duties shall be specified in a written agreement

The project leader should implement this SOP in studies involving randomisation. 

WHO

The project leader must plan and describe in detail procedures for randomisation, blinding and unblinding and the methods used to ensure hidden allocation. This should be described in the protocol or in separate study-specific procedures.

The principal investigators/investigators and others involved in the study should be informed of blinding and unblinding routines. It must be documented that information is given and to whom (for example, in the start-up meeting).

Randomisation

Randomisation ensures random distribution of subjects to treatment groups. Randomisation is an instrument to reduce or eliminate selection bias. The randomisation process consists of two parts that should be described accurately in the protocol:

1. Specification of the random allocation of patients to treatment groups.
   • The allocation can be carried out continuously using an electronic randomisation solution or using a pre-generated allocation list. The allocation procedure should include a description of the selected method, including information about any stratification. A method that includes varying block sizes is recommended for block randomisation. Regardless of the solution chosen, the task of creating a randomisation procedure/list should be done by someone not affiliated with the study. Specific data such as block size and programming code should be kept out of reach of study personnel until database lock, while information about allocation ratio and stratification groups should be verifiable before start-up. All solutions should be tested before study start to ensure that the results of the testing are within the requirements of specification.
2. Patient randomisation.
   • The protocol should detail the process from the specified allocation procedure to the allocated patient. As a rule, this involves the following:
     1. How eligible patients who have given informed consent are assigned to the treatment group according to the allocation procedure. This may involve, for example, envelopes or login to a website or data capture solutions.
     2. How to hide the allocation at least until the patient is allocated for treatment. The principle is that treating or assessing study personnel as well as the patient do not know the sequence before the allocation. This can, for example, involve the use of opaque envelopes or Data Entry Application (DEA).
     3. How the allocation time and allocation are recorded. In blinded studies, only the randomisation number is recorded.
     4. How to ensure that the patient receives the treatment that has been allocated.
     5. Procedures for retaining the blinding through the allocation.

Blinding

Blinding is a procedure in which one or more of the parties involved in the study are kept from the treatment allocation in order not to be affected by this information. Blinding is an instrument for reducing or eliminating systematic biases as a result of knowing treatment allocation. It is particularly important to ensure that everyone who performs different measurements or assesses response/endpoint is blinded to treatment allocation, especially in studies with subjective outcome measures.

Always consider who should be blinded to treatment allocation. In drug trials, the principal investigator/investigator and healthcare professional who are in contact with the subject and the subject should be blinded. In addition, consider blinding of other study personnel, study nurses, monitor, statistician and others who could potentially influence measurement results in the study by knowing the treatment allocation. One must decide which people need to be blinded/unblinded and take measures to ensure this in the planning and implementation of the study.

To reduce the risk of accidental unblinding of blinded personnel, the number of people who know what the patient is getting should be limited to a minimum. Strategies to reduce the risk of accidental blinding should be described in protocol or in a separate procedure. Examples of strategies include restricting access to the medicine room/closet where the study drugs are stored, dealing with glass and ampoules in such a way that those who are to be blinded cannot see what the patient has been given, restrict access to documents related to the medicines (e.g. medical accounting logs and randomisation confirmations), as well as use of an unblinded monitor to monitor drug management.

Who is blinded to treatment allocation and what measures are taken to maintain blinding throughout the study should be described in detail in the study protocol.

When planning a blinded study, one must clarify whether this is practically achievable. One way to achieve blinding in clinical drug trials is to compare two drugs. Often the comparator used is without an active substance, i.e. placebo. To ensure blinding, the alternative treatments (active substance and placebo) should be completely identical (similar in appearance, taste, consistency, etc.). If it is impossible to make identical comparative drugs, the use of "double dummy" is an option. If the blinding cannot be taken care of by the drug alone, other measures must be taken to ensure that the blinding in the study is maintained, e.g. in that those who administer the treatment are unblinded and those who measure outcomes are blinded.

It is recommended to consult statisticians or other professionals and/or pharmacies that can assist with quality assurance of the randomisation process and blinding in the study.

In the event of an adverse event (AE) or adverse reaction (AR), it may be necessary for the investigator to know quickly what treatment the subject is receiving. In such cases, it may be appropriate to unblind (break the code) to determine what treatment the subject is receiving. Unblinding should be carried out according to pre-described routines.

Also see Safety reporting - SOP 1.4.b

Routines for unblinding/breaking codes in the individual study

Planning phase

Unblinding routines must be documented in the protocol or in another document. This may involve details of:

• Names and phone numbers of people who are able to unblind.
• What requirements must be met for unblinding, e.g. that the choice of further treatment depends on the allocated treatment.
• Timeframe for unblinding.
• How the people who can unblind have access to the allocation list.

Conduct phase

Necessary information about unblinding procedures should be documented in the patient medical record, e.g. under “kritisk informasjon”, or otherwise to ensure that the information is available in case of immediate assistance.

When a code is broken, you must ensure:

• It is documented in the patient record if relevant for the health care of the patient.
• It is documented in the study documentation as described in the specific study, e.g. directly on the envelope or in the electronic system. The date and the reason for the unblinding are documented.
• The sponsor is informed immediately.
• The unblinding is carried out in such a way that access to the code is limited. Only the code for the study participant(s) in question should be revealed, and the blinding for other subjects in the study must be maintained.

The main rule according to the intention-to-treat principle is that the subject continues in the study after breaking the code.

Close-out phase

Unblinding of the study should only be done once the study has ended, the database is locked, and the results are to be considered. This unblinding should be documented.

The monitor will document opened/unopened codes in the monitoring report from the closing visit.