Trial and error: pivotal trial design is key to medical-device success
07 April 2015
Author: Bill Peatman, CIRTEC Medical Systems
‘Pivotal’ clinical trials with human subjects are typically the last step in the medical-device approval process. As the US Food and Drug Administration (FDA) states in its design considerations document for clinical trials, “Evidence from one or more pivotal clinical studies generally serves as the primary basis for the determination of reasonable assurance of safety and effectiveness of the medical device of a pre-market approval application (PMA) and the FDA’s overall benefit-risk determination.”
A successful clinical trial for a new medical device is by no means guaranteed. Device performance on a statistically significant number of human subjects can introduce variations in performance not seen in earlier work. A successful trial depends a great deal on trial design. Poor design can lead to the failure of a trial for an otherwise effective device, and can doom devices and their sponsor companies. Well-designed trials, on the other hand, can help ensure that the device is successful.
Recent high-profile devices that have failed at the trial phase include the Medtronic Symplicity renal denervation device. The device promised to reduce hypertension by ablating nerves connected to the kidney. While previous trials had shown significant benefits, the US pivotal SYMPLICITY HTN-3 trial results did not meet the endpoints defined as successful outcomes. The results not only impacted Medtronic, but cast doubt on other products in the category.
What went wrong? A secondary analysis of the trial found that nearly 39 per cent of patients treated in the trial had been subject to medication changes during the study. Each patient was supposed to be on the same maximally tolerated dose before enrollment in the study. The analysis found, however, that there were changes in the dosage and in the type of medications taken by these patients. Also, some patients received only a single ablation, while others received up to 26. The analysis showed that a greater number of ablations correlated with a reduction in blood pressure.
Best practice for clinical trial design
What can medical device developers do to ensure these kinds of mistakes do not happen? Trial design is critical. Following best practices for trial design will help avoid costly, sometimes fatal, trial errors.
- Choose your endpoint carefully: Clinical trials are all about whether the device has the specific impact that the device maker claims. For example, the trial of another high-profile device failure, the Enteromedics Maestro vagus nerve stimulator for weight control, found that the device worked 8.5 per cent better than its control group. However, the threshold for success was set at 10 per cent, and the device was not approved. The device worked, but not as well as device makers claimed it would. (In early 2015, an FDA advisory panel ultimately approved the device after considering additional data, concluding that the benefits of the device outweighed the risks for certain patients.)
- Select the right patient population: In 2012, the FDA updated guidance on choosing patients for clinical trials, stating, “Clinical trials are not designed to demonstrate the effectiveness of a treatment in a random sample of the general population.” The guidance document recommends selecting patients whose conditions do not change frequently, who are likely to comply with treatment, who are likely to have a similar rate of disease progression, and who have some likelihood of response to the treatment. Trial patients should also represent the demographics of the actual patient group who would use the commercialised device, including the age and gender of the patients enrolled.
- Enforcing patient compliance: Trials must be designed to ensure patient compliance. This can be difficult, especially when trial time frames are long and patient interactions with physicians infrequent. As seen in the SYMPLICITY HTN-3 trial, post-trial analysis found patients took uneven dosages of medication during the trial. Anticipating obstacles to compliance and possible miscues is critical to trial success. A better-designed trial may have interviewed participants more frequently and discovered these variations in time to make necessary changes in the trial population and controls.
- Selecting a control: Selecting the right control arm of a medical device trial is more complex than for, say, a new drug where the control group is often given a placebo. Using a sham (placebo) device on a control group raises safety and ethical questions, especially with invasive devices such as implantable products. One option is to use subjects already using an existing treatment (active intervention), but the treatment for the trial will likely be administered by some of the best clinicians in their field, leading to better results for the control group than what would happen in a more real world scenario. Other control options include control groups receiving no treatment (no intervention); patients serving as self-controls, where the subject of the trial also serves as the control (subject as control). Placebo controls are particularly difficult with medical devices because it is usually quite obvious when a subject is not using a medical device.
- Timing: Clinical trials must be completed in the specified time frame. One issue that can delay trials is recruiting subjects. Depending on the complexity of the trial, the demographics of the subjects and the prevalence of the condition to be treated, recruiting qualified subjects can be a challenge. It is important to take this into account when designing the trial schedules.
Medical device trials – conclusion
Demonstrating that a new medical device is safe and effective for use in humans can be extremely challenging. Clinical trials are fraught with risk, and there are many variables that must be managed to ensure success. It all starts with trial design. Errors in trial design can prove fatal to a device being approved. The approval process for medical devices is not especially standardised, and regulators frequently set requirements for approval on a case-by-case basis.
Engaging an experienced manufacturing partner that has brought similar products to market is one way to de-risk the device development effort, and allow device makers to work on trial design with the confidence that the device will operate as designed.
CIRTEC Medical Systems provides design, development and manufacturing support to the medical-device industry with a particular focus on active implantable devices and peripherals as well as minimally invasive systems. It supports the full product development cycle, from requirements definition through commercial manufacturing. CIRTEC is ISO 13485 certified.http://www.engineersjournal.ie/2015/04/07/medical-device-trials-design/http://www.engineersjournal.ie/wp-content/uploads/2015/03/clinicaltrial-1024x623.jpghttp://www.engineersjournal.ie/wp-content/uploads/2015/03/clinicaltrial-300x300.jpgBiomedical devices,research