European Journal of Cancer
Issue 6, 2010

By Helen Saul

PODIUM – An insight into the changing features of drug approvals – with Prof. Hubert Leufkens

Prof. Hubert Leufkens, Utrecht University, The Netherlands

Professor Hubert Leufkens (Utrecht University, the Netherlands) is chairman of the Dutch Medicines Evaluation Board and a co-opted member of the EC’s Committee for Medicinal Products for Human Use (CHMP). He is a co-author of a recent EJC paper ‘Therapeutic indications in oncology: Emerging features and regulatory dynamics’ (EJC 2010; 46(3): 471-475)

Does your paper imply a widespread change in the procedure leading to registration?

It is well known that efforts are being made to reduce the time needed for clinical development and for regulatory review. Our findings confirm this trend, which is positive if it implies quicker access to treatment and reduced regulatory delays. A key question for regulators is to assess how mature a clinical package should be at the time of the drug application. Regulatory authorities have to take decisions based on the best available evidence and, when data supporting the approval are not robust, restrictions to the indications are a means of identifying the patient population that may most benefit from the use of the drug, instead of depriving potential patients of an effective treatment. Obviously, when new knowledge becomes available after the approval, regulatory authorities may extend the indications in order to include a larger patient population. What’s needed is a sustainable life-cycle approach to the new therapeutic molecule including risk management plans for identifying and minimising possible drug induced harm after marketing approval.  

Are you therefore in favour of the trend towards restricting and later broadening indications?

Restrictions can allow the patient population to be tailored to those who will benefit most from the treatment, strictly based on the evidence provided by registration trials which mostly involve a highly selected population. On the other hand, regulators have to define indications that are applicable to the “real” patient population. If the population is too selected, it may be difficult to find patients in general practice who can actually benefit from treatment. Paradoxically, we may be faced with a new drug that is an orphan from the patients’ perspective.

This also means that the more the indication is restricted, the higher the risk of off-label drug use. In this sense, when regulators adopt heavy restrictions on a new drug which represents a unique therapeutic option, the result could be that the regulatory authority itself fuels off-label use. On the other hand, granting an authorisation without sufficient evidence could harm patients.

How can the needs of special populations such as children and the elderly be safeguarded?

The issue of special populations in clinical trials is not new and has been debated at length by the scientific and the regulatory community. Several actions have already been implemented at the EU level, such as the new paediatric regulation aimed at further stimulating paediatric research during the pre-marketing phase.

In previous work, we suggested that further registration trials could be avoided when consolidated robust evidence is available in a specific population such as children (Eur J Clin Pharmacol. 2009 Feb;65(2):209-16).

In what ways does the situation in oncology differ from that in other medical specialties?

There are many advances in oncology – in molecular biology, biomarkers, diagnostics, and so on – and many unmet medical needs. However, the translation of all these scientific developments into clinical practice often lacks the supportive clinical data needed for regulatory approval.

Why do most anticancer drugs have only a single indication? Is this a cause for concern?

Our analysis has shown that, contrary to common belief, most anticancer drugs only hold a single indication. This could be because once the drug is on the market, companies may not be interested in extensions if the drug has found its way on to standard treatment protocols. Generic competition may make further investments in drug development simply not very attractive from an industry perspective. This tends to lead to widespread off-label prescribing.

Another possibility is that, when we compare the wording of indications for new and older oncology drugs, we see a shift towards more detailed and precise patient targeting for the newer products. The indications for the older products were simply phrased more broadly, allowing a wider window for prescribers. 

And yet the relatively recent first-in-class drugs sunitinib, cetuximab and bevacizumab have multiple indications?

Such recently approved drugs may already have multiple indications because of multiple clinical developments, but it also reflects the fact that many modern oncology drugs are mechanism-based. This provides opportunities for more than one clinical application, which means that a compound can be tested in different oncology areas virtually at the same time. This has positive implications in terms of public health, provided that robust evidence is produced.

Were you surprised by any of your findings?

We expected older compounds to have more indications, as just discussed. Another issue was that the time needed to add a new indication for a drug has decreased dramatically over the years, down from 81 months in 1996 to 6 months in 2006. This may show how companies have speeded up the clinical development of drugs. Hopefully our findings will provoke useful discussion among the different stakeholders involved in the process, and lead to fruitful alignments between drug developers and the clinic.