How to Improve Access to New Medicine for Adolescents
Harris W. Dalrymple, PhD (Med), PhD (Law)
Harris W. Dalrymple, PhD (Med), PhD (Law)
Executive Director, Center for Pediatric Clinical Development

ACCELERATE recently proposed including patients over the age of 12 in clinical trials where the mechanism of action of the drug being studied and the characteristics of the target disease are potentially relevant to adolescents -- or when the disease is rarely present in the adolescent population. This makes perfect sense and is by no means new.

Since the beginning of the modern era of drug development regulation initiated in the 1960s after thalidomide resulted in an estimated 10,000 babies being born with malformations, strenuous efforts have been made to protect vulnerable populations from the perceived risks associated with new investigational drugs.

Over the years, children, women, pregnant women, ethnic minorities and a host of other definable groups have been classified as vulnerable, but with the exception of children, they are not vulnerable because they have impaired decision-making capacity. They have become vulnerable because in our determination not to expose those in these groups to “risk,” we have also excluded them from clinical trials. As a result, when new drugs are approved for use, the approval is based on an unrepresentatively narrow sample of the population. So, these groups, including children, are vulnerable to various disease conditions as a result of our relative lack of knowledge regarding the correct treatment of these conditions. Only in the last 20 to 25 years have we started to reverse that trend, and so far as children are concerned, we have adopted a “carrot and stick” approach, providing incentives (e.g., increased period of exclusivity) to manufacturers who assess the efficacy and safety of new drugs in children, and deterrents (e.g., refusal to grant marketing authorization for adult use) for those who do not. Clearly, that is not working as we had hoped.

...when new drugs are approved for use, the approval is based on an unrepresentatively narrow sample of the population.

Age is only one aspect of variability in drug response, and, like hepatic and renal functions, body weight and composition, it is a continuous variable. Yet we have positioned it as a categorical variable, like gender. Trial subjects are either “adult” or “child” with a fixed division at (in most countries) 18 years. No credible biological reason exists for this division. It is but another example of society deciding that individuals need to attain a certain “status” before being allowed to do something – such as driving a car or entering a contract. However, the safeguards for participants in clinical trials are already far greater than exist for those other activities, and far more people are killed every year by cars than are injured in clinical trials.

The reason for the age categorization appears to be tied to the age at which individuals are considered to be able to give consent in their own right. But even that varies by country; Austrian children may do so from the age of 14 years (15 in Finland, 16 in the UK). The administrative process appears to be impairing the medical objective – that of improving the availability of properly-evaluated new drugs to a population which may benefit from them.

The division of adult from child is reflected to an extent in the hospital infrastructure in many countries, but it is often less categorical than it is in the regulated world of clinical trials. In many countries, adults and children are treated in the same hospital, often in the same departments. “Children” are often transferred to “adult” clinics and facilities when they are aged 14 to 16, for a wide variety of sound medical, operational and sociological reasons. They are of physical similar size to adults and so can use the same equipment and devices. Their cognitive function and language is much closer to that of adults than to children only a few years younger. Most importantly, they can normally be dosed with the same drugs and on the same regimen as adults.

Adolescents meds03

Routes to a Solution

So, how do we correct this artificial, administrative division in clinical trials? We must surely start with the end in mind: Will the major agencies accept an application to approve a drug for patients aged, say, 12 and older? The FDA has already signaled it may be willing to do so, but that hardly seems sufficient to encourage a company to invest over $200M in a single development program, only for the agency to reject it and impose a requirement for one adequately-powered and adequately-controlled trial in patients aged 18 and over, and another in patients aged 12 to 17.

Last year, the EMA issued a reflection paper on the pharmaceutical development of medicines for use in the older population, which, sensibly, avoided proposing a categorical definition of “elderly,” instead recognizing that older people of the same chronological age can be quite different from one another. Recently, a publication appeared in the New England Journal of Medicine describing the efficacy of a drug in a particular tumor type in adults and children, demonstrating not only that such trials are possible but also appropriate when the disease target is sufficiently similar in adults and adolescents.

From a medical and biological perspective, we must consider age as a continuous variable. If agencies persist in separating those aged 12 to 17 (depending on the country of conduct, some of these will legally be adults already) and assess efficacy and safety dedicated trials in that group, then why not do the same those aged, say, 20 to 29, 30 to 29, 40 to 49, etc.? Depending on the strength of the data, the drug label could then indicate the drug is approved for use in those aged 20 to 29, 40 to 49 and 60 to 69, but not those aged under 20, 30 to 39 or 50 to 59. Medically, this would clearly be nonsense, but what of the genuinely elderly – our centenarians? The agencies have not suggested that separate development programs are required for that group and seem to be shying away from doing so. Are we really suggesting that the response of a 17-year old is so different to that of an 18-year old that separate trials are required, but the response of an 18-year old is so similar to that of a 103-year old that they can be included in the same study? Many years ago, upper age limits were common in clinical trials, but the Agencies have pushed for inclusion of subjects of “relevant” ages in clinical development programs, although the elderly still appear to be under-represented. So, as a society, we appear to face a similar issue in the elderly as we do in the young regarding trial participation – but we are not adopting a common approach to solutions.

Adolescents meds01

The next step is to establish where adolescents are actually treated. If a highly demarcated system exists in a country, state or region, with patients aged under 18 being treated by a pediatric specialist at a separate facility, then that’s fine; place the clinical trial with that facility, assuming all of the other requirements are met. If the separation of child from adult is more “fluid,” that’s fine too; place the clinical trial with sites which treat adolescents alongside adults and ensure that the staff understand the trial is open to all.

Studies suggest that consumer comprehension may be compromised if content exceeds a 7th- or 8th-grade reading level, which is the average American reading level identified by the United States Department of Health and Human Services. If documents adhered to that practice, then the need for radically different documents for those aged 12 to 17 could be eliminated. Simple grammatical changes would be all that would be required to ensure that relevant legal requirements were met which safeguarded those considered to lack legal competence on the basis of age, for example replacing “you” by “your parents/guardians.”

Ideally, information sheets and assent/consent documents would be structured differently, with Part A explaining what the trial entails, the procedures, duration of involvement, the subjects’ rights regarding assent/consent/dissent withdrawal, data privacy and protection, and other matters over which the adolescent may exercise full control, and Part B explaining insurance and compensation for injury, access to data and rights regarding biological samples and analysis, over which, due to the legal issues relating to age, the parents/guardians would more likely exercise control.

So far as the assenting/consenting process itself is concerned, physicians treating adults are accustomed to dealing with patients with widely varying capacity to understand and to assimilate written information in documents, and so should have the experience to explain the requirements of a trial to parents where required. At centers treating both adolescents and adults, the physicians will be accustomed to speaking with both groups, and so should be comfortable explaining a clinical trial.

The contractual position may indeed be more complex, particularly where the contract is not only with the institution or facility, but also with the investigator(s). However, many trials already require multiple contracts with sites, e.g., for ancillary services such as pharmacy, cardiology, radiology, and so a separate contract for a “pediatric specialist” is conceptually similar and should not be seen as an obstacle.

A less obvious obstacle may relate to insurance. Some insurers charge higher costs for trials in children, partly on the basis that the costs associated with life-long management of an iatrogenic injury are likely to be greater than for an adult. However, given the strengthening requirements for pediatric development in the form of PIPs and PSPs, these costs will need to be borne at some point.


The operational challenges of including those under 18 years of age in clinical trials are well-understood and none are impossible to overcome, other than the availability of patients in the age range. Including 12- to 17-year olds in clinical trials, except where good evidence exists not to do so, can be facilitated by a thorough understanding of the legal and administrative processes applicable to such subjects, particularly in the area of assent/consent, and by understanding where patients of that age are usually treated, recognizing international and within-country differences will exist. The trade bodies for the pharmaceutical and biotech industries need to engage with the regulatory agencies to establish that data generated in such trials will be acceptable, and the label they will generate.

Explore related content and topics