PRA: pediatric trials require specialized endpoints
Clinical trial teams have a wealth of guidance on acceptable endpoints for adults to draw upon in their work. However, when it comes to pediatric endpoints, the resources may be less plentiful, and simply using the same as for adults could lead to less than reliable results.
Outsourcing Pharma (OSP) spoke with Jo Dewhurst, Director of the Center for Pediatrics Clinical Development with PRA Health Sciences, about why determining appropriate endpoints for pediatric trials, and how to accomplish the job.
OSP: What are some of the challenges associated with endpoints for pediatric patients?
JD: We should not assume that adult study endpoints are always appropriate for pediatric studies, but at the same time we should not assume they must be different. The disease, the patient population, the patient age, and the needs of the patients should be considered.
It is important to assess practical considerations as well as scientific considerations when selecting endpoints for pediatric trials to ensure they are achievable and appropriate for the targeted age range. Another key factor to consider is to determine the reliability and validity of endpoints in the pediatric population.
OSP: What are some of the questions and variables that are unique to pediatric trial participants?
JD: There are a number of differences between adults and children, such as organ maturation and developmental stage. These can impact how they metabolize different drugs resulting in different PK profiles.
There can also be differences in the pediatric form of some conditions, with the conditions appearing outwardly the same in adults and children but perhaps having a different disease mechanism or be secondary to another condition, for example childhood hypertension. For these reasons, the primary endpoint used for adult trials may or may not be appropriate for the pediatric trial.
OSP: What are some ways in which endpoints come into play regarding drug approval? What are some of the considerations involved?
JD: For a new drug to be approved, the relevant regulatory authorities must be provided with evidence that the new drug has a clinically meaningful effect. This is done by providing data on selected endpoints and so requires consideration of which relevant disease related symptoms are anticipated to be affected by the treatment.
There are also evidentiary considerations such as whether the methods of assessment are well defined and reliable and does the endpoint represent something meaningful to the patient regarding how such endpoint data will be collected and whether this is a validated method is also a consideration. This may mean that the sponsor needs to consider alternative endpoints specific to pediatrics to fulfill the regulatory requirements.
Often endpoints used in the adult population are not validated for the pediatric population. Sponsors should consider early in development which endpoints they intend to use and start discussions with the relevant regulatory authorities to determine the evidence required to support the use of the proposed endpoint and how such evidence can be generated; this may include the use of novel exploratory endpoints in pediatrics in the early stages of development to help gather the necessary evidence to support later use as primary endpoints for pivotal studies.
OSP: How can trial teams effectively select appropriate endpoints for pediatric clinical trials?
JD: When designing a pediatric clinical trial it is important to select appropriate endpoints based on the disease and endpoint validity.
For example, nonalcoholic fatty liver disease (NAFLD) is a disease in children that may progress to non-alcoholic steatohepatitis (NASH), cirrhosis, or advanced fibrosis. A large number of clinical trials is being conducted in adults with NASH and the FDA draft guidance on development of treatments for NASH in adults is extensive, but is less clear for pediatric NAFLD. Pediatric NAFLD has unique aspects in its histology and natural history, meaning different endpoints to those used for adult NASH studies may be appropriate.
However, it is also important to consider children’s needs and the challenges that may be associated with the endpoint and, ideally, to include pediatric patients and their parents/caregivers in endpoint development. An example of considering the patient perspective arose in determining age appropriate measures of growth velocity for pediatric growth disorders. Such an endpoint is objective but uses non-invasive measurements routinely used in pediatric clinical care.
However, it is also heterogeneous, both by condition and perspective. When asked by the FDA for feedback, patients and parents voiced a concern that improvement in height was a goal of treatment for pediatric growth disorders and, due to the limited window of opportunity for meaningful treatment, prolonged placebo exposure was a key concern.
The FDA Pediatric and Endocrinologic and Metabolic Drugs Advisory Committee meeting on May 11, 2018 discussed potentially clinically meaningful study endpoints in drug development for achondroplasia (ACH). They agreed that average growth velocity (AGV) was an acceptable endpoint, being easy to measure and reproducible, but it was felt that data on final adult height should be obtained to validate long-term clinical efficacy.
Endpoints related to improvement in complications of the condition and improvements in quality of life measured through validated PRO’s can be evaluated as secondary endpoints and may need to be age specific.
OSP: What are the dangers of not coming up with appropriate/fitting endpoints in a pediatric trial?
JD: Well-defined, interpretable, and reliable efficacy endpoints that are directly applicable to the disease or condition are critical to the success of a trial. If these endpoints are not appropriate or not validated this can result in the clinical trial failing to meet the endpoint and the pediatric label not being granted, despite the drug apparently working.
One such example is the six-minute walk test that is standardly used in adults with pulmonary arterial hypertension (PAH). When designing the STARTS -1 pivotal pediatric PAH trial for sildenafil there was concern from many specialists that pediatric patients would struggle to comply with the directions for the six-minute walk test.
They may get bored or disinterested by factors unrelated to their PAH, or may walk at a variable pace, both of which could impact the reliability of the test. Instead peak oxygen consumption (VO2 maximal exercise test) was used as the primary efficacy endpoint to look at the patients ability to perform aerobic work. The trial gave the opportunity to evaluate peak VO2 as an endpoint with regard to its correlation with other clinical endpoints relevant to PAH, with peak VO2 showing good reliability and suggesting it is a suitable and robust measure to evaluate a drug treatment for PAH in children.
OSP: Could you please share some data about pediatric-adult endpoint comparisons?
JD: An analysis of 234 pediatric pivotal efficacy studies submitted to the FDA between September 2007 and July 2016 for which there were corresponding adult trials in the same indication showed that 141 of these trials had the same adult and pediatric endpoints, with 122 (86.5%) succeeding to meet their primary endpoints.
However, of the 93 trials where the adult and pediatric endpoints differed, the difference being the end point itself and/or the time of the measurement, only 57 (61.3%) succeeded. It was also noted that trials including both adult and pediatric patients succeeded more frequently (89.5% v 67.6%). This may be because the disease is better understood, is more similar in adults and pediatrics, and/or that there is more prior experience with the endpoints.
As we can see the same endpoints for the adult and pediatric population can be appropriate but must be considered carefully.
OSP: Could you talk about endpoints and their role in trial failure? Please feel free to give examples, either of trials that failed or succeeded.
JD: When a trial fails it may not be because of the endpoint selected. It may be because the endpoint used is not validated for the purpose for which it is being used in the pediatric trial, or that the drug does not work in pediatric patients. However, it may also be that the endpoint selected was not suitable or practical for the patient population, such as the six-minute walk test mentioned previously for PAH.
OSP: What are some challenges associated with using PROs in pediatric studies?
JD: PRO data must be collected from rigorously developed and reliable patient questionnaires covering topics such as patient symptoms and the burden of the symptoms on their day-to-day life. Many PROs used in adult studies have been assessed in adolescents, sometimes in a slightly modified form, and can be used effectively. However, validation in younger subjects is often more difficult due to language/vocabulary development and attention span.
For this reason the following is generally considered appropriate for PROs:
- Under the age of 5 care giver reporting is required as there is no clear evidence of reliability or validity of child self-reported measures
- Aged 5-7 years it may be possible for the child to self-report but reliability and validity is often questionable
- Aged 8-11 years the reliability and validity of child self-reporting improves
- Aged 12-18 years child self-reporting is preferable
OSP: Could you please explain digital endpoints? How do they work and what are some benefits/challenges?
JD: Digital endpoints are the most recent type of endpoint and are captured using a sensor. Such sensors may be worn, remote (such as voice sensors), or ingested. Such continuous or frequent monitoring does offer the prospect of more accurate and precise assessments; the benefit may be reduced trial participation burden, such as reduced site attendance, and a better understanding of how new therapies affect and hopefully improve daily life.
However, like any endpoint, digital endpoints require robust evidence to support their use. To develop such evidence, particularly in the pediatric population, will require significant resources and, ideally, collaboration. The Digital Medicine Society has issued a library of digital endpoints collected in clinical trials giving information on digital measures already being included in industry-sponsored trials, providing a strong signal of where digital endpoints may be most valuable in drug development and where there is already a growing body of evidence of feasibility and performance.
Digital endpoints currently are often exploratory endpoints but there is a move to promote digital endpoints as primary and secondary endpoints as more evidence to support their use becomes available.
The FDA MyStudies mobile application is being used to support the Childhood Arthritis and Rheumatology Research Alliance Limit Juvenile Idiopathic Arthritis Trial, in which it will be used to collect the uveitis endpoint and includes questions regarding administration of the medication. It is also being used to support the Crohn’s and Colitis Foundation Inflammatory Bowel Disease (IDB) Plexus Platform to collect patient reported outcomes. Such initiatives will help to support the use of digital endpoints to support pediatric clinical trials.