More and more frequently, treatment switching is occurring in clinical trials, particularly in the area of oncology. Not surprisingly, patients are demanding to be switched (most commonly) from the control treatment arm of a trial to the experimental treatment or intervention arm to ensure what they perceive to be their best chance of cancer control if not cancer recovery. Human research ethics committees are also requesting the switch, or early crossover. Even the industry sees switching as necessary, particularly to help with recruitment to clinical trials, which can be challenging at the best of times.
However switching does cause issues for many of the decision makers involved in the development, regulatory approval and reimbursement of medicines.
For regulators, treatment switching confounds the clinical effect of novel therapies. For payers (health technology assessment agencies), treatment switching confounds cost-effectiveness results for economic evaluations that rely on intention to treat (ITT) analyses. Even for clinical trialists (investigators), switching can cause a conundrum – the tension between treating current patients and treating future patients. That is, needing to treat the patient in front of them giving them the best chance of fighting their cancer versus the likely consequence of confounding trial results to the extent that regulatory and funding approval may be delayed or even prevented, so access for future patients is impacted.
So how to best address the issue?
- The solution will need to be designed on a case-by-case (trial-by-trial) basis. There cannot be a simple solution given the differing requirements of the stakeholders involved and the complexity of managing the needs of each. Clinical trials must be ‘fit for purpose’.
- Switching will need to be well defined upfront in protocols. Protocols will need to stipulate when switching is allowed, on what basis switching will be allowed and who decides when switching is to occur.
- Justification on the statistical adjustment method(s) and assumptions utilised to analyse data from clinical trials where switching has occurred will also be required.
- Appropriate study design with the involvement of key stakeholders will be critical, including the increasing importance of incorporating health technology assessments into, and generating robust Q of L data from, Randomised Controlled Trials.
Although oncology clinical trials are the current focus, switching will not remain the exclusive domain of this therapeutic area. It will rapidly expand to other therapeutic areas which is why agreement, or at least consensus, must be reached sooner rather than later.