New Guidances Try to Keep Up with Progress

In both oncology and cardiology, new EU rules correct some errors but leave others overlooked.

Progress marches on-and with it the regulatory framework.

A new guideline should emerge from the European Medicines Agency by the end of this year to revise European Union guidelines on anticancer medicine evaluation in man.¹ The revision will remedy an obsolete focus mainly on conventional cytotoxic compounds.

What is striking about this is that the last revision is only a year old, and the initial guideline itself dates only from 2001. Regulations, like innovations, are made of fragile material. Science and technology now move forward so far in such a short time that a new approach is needed in designing clinical trials for many new treatments.

Toxicity, for instance, has for long been accepted as a surrogate marker for efficacy in early drug development. Most licensed anticancer medicines have had a narrow therapeutic index, so the mechanisms underlying toxicity are traditionally postulated to be the same as those promoting antitumor activity.

But new compounds targeting the activity of tumor growth factors, intra-cellular signalling, or structures in the tumor environment demand new approaches other than basic toxicity.

At the same time, some substances only work in combination with other compounds. And for substances targeting specific structures, such as constitutively activated enzymes, drug development focused on conventional "pathological-anatomical" indications might not always be regarded as the most appropriate.

And new methodological questions are posed by the assessment of progression-free survival in open label studies, pivotal studies without a randomized reference, or the external validity of confirmatory studies.

For all these reasons, alternative designs of clinical studies are needed, says the concept paper on which the EMA is basing its work. New designs are to be suggested for exploratory studies making use of functional imaging or tissue biopsies (including molecular pharmacology) in order to define the pharmacodynamics of new compounds. And new mechanisms are needed for proof-of-concept studies when response rate is considered inappropriate or in case combined use of agents is needed to obtain anti-tumor activity.

The methodological issues to be addressed will include the use of historical controls in exploratory trials and choice of endpoints. Thought will be given to the role of external review and sensitivity analyses to reduce possible bias in open label confirmatory studies based on progression-free survival; and to possible bias comparing "cytostatic" and "cytotoxic" compounds with respect to progression-free survival. Other areas to be reviewed include external validity of confirmatory studies and special patient groups; rare tumors; appropriate comparators in confirmatory studies when there are evidence-based comparators and where there is not; and licensing based on studies without a randomized reference.

The next stage will be a late autumn discussion by the Agency's Efficacy Working Party and its Oncology Scientific Advisory Group.

Lipid-disorder drugs: New guidance
The agency has moved faster on treatments for lipid disorders-another area where science and technology are moving the frontiers forward for clinical trials and treatment. It has just adopted new guidance on the evaluation of drugs for this indication,² which will come into effect at the start of 2005.

Current mechanisms of disease categorization are too lax, the guidance says. The common classification of lipid disorders according to the prevailing laboratory abnormality "does not accurately represent the different genetic and metabolic defects, or clinical syndromes." Blood lipid levels may be affected by other clinical conditions such as diabetes mellitus, thyroid disorders, or nephrotic syndrome, so underlying conditions should be treated, and lipid levels should be reassessed once the disease has been controlled.

The guideline defines the primary goal of treating lipid disorders as prevention of cardiovascular morbidity and mortality associated with lipid levels-although it recognizes that in rare cases of very high triglyceride levels, the initial aim is prevention of acute pancreatitis.

The advances have been dramatic in this field, the guideline acknowledges. For instance, most HMG-CoA reductase inhibitors have been shown to reduce cardiovascular events (including stroke) and overall mortality in patients at high cardiovascular risk, irrespective of their cholesterol levels. Some data also suggest that fibrates have been shown to reduce the rate of coronary events both in patients with mixed hyperlipidemia and in men with coronary heart disease with only low levels of HDL cholesterol without hypercholesterolemia.

Nevertheless, in the regulator's role of doubting Thomas, the agency points out that "positive effects on mortality and morbidity can only be evaluated properly in large-scale and long-term clinical trials, in patients with lipid disorders and/or high cardiovascular risk." And until this data is available, it should be specifically mentioned in the product characteristics that beneficial effects on mortality and morbidity are unknown.

Some small concessions are made to the realities of developing new treatments. A relative reduction in LDL cholesterol is acceptable in patients with primary hypercholesterolemia as a valid surrogate endpoint, provided that no claims are made regarding morbidity and mortality. But it sharply reminds that "ideally a new lipid-modifying agent is expected to demonstrate an effect on the prevention of cardiovascular morbidity and mortality."

And reduction in triglyceride levels and/or increase in HDL-cholesterol might also be considered as a relevant component of the primary endpoint for particular target populations. "However," it insists, "an isolated effect on these parameters is in principle not expected to be the sole basis for the demonstration of the efficacy of a new lipid-modifying agent, but should be seen in conjunction with the effect on non-HDL cholesterol and the underlying mechanism as well."

It is also cautious on assessment in terms of effects on vascular damage. Although target organ damage of heart, brain, kidneys, and blood vessels is "presumably and plausibly" associated with morbidity and mortality, "the prognostic value of these drug effects with regard to morbidity and mortality remains to be established," it says. The agency is particularly skeptical over changes in intimal media thickness and plaque stability.

The best it can offer is the prospect that such studies may become more useful in the future. "These endpoints can be particularly valuable for the scientific community as they may bring information on how the medications act and can have clinical protective effects. As more becomes known, such studies may substitute clinical studies using hard clinical endpoints in selected subgroups."

Planning for a mortality study factors in all-cause mortality and/or cardiovascular mortality, as adjudicated by a blinded, independent committee. If cardiovascular mortality is chosen as a primary or co-primary endpoint, effects on noncardiovascular mortality should also be considered. And the evaluation of cardiovascular morbidity should especially take into account signs and symptoms of organ damage (such as myocardial infarction or stroke) and their therapeutic management. An unqualified veto is also placed on placebo-controlled trials, since there are now effective treatments available.

Selection of subjects
Evaluating the effects of a new agent for treatment of lipid disorders will generally require a study on the type of lipid disorders for which the drug is intended. Studies for the evaluation of efficacy or safety of a new lipid-modifying agent are mainly performed in subjects with primary hypercholesterolemia and mixed hyperlipidemia with moderately to very highly elevated cholesterol levels. Attention should be paid to effects of gender, race, and age. Subjects with clinical and/or other manifestations of atherosclerosis and/or type 2 diabetes mellitus should be represented in adequate numbers to allow statistical (sub)group evaluation.

The guideline urges no radical departures in strategy design for studies of the first administration of products for lipid disorders to man: they should not differ essentially from those dealing with other cardiovascular products. Following initial screening, a dietary lead-in period is obligatory before randomization in the study. Inclusion criteria and the reliability of the methods used should be justified, taking into account such factors as the target population and assay accuracy.

The progress in science, technology, and clinical trial design makes possible new therapies-but also means continual production of new rules and guidelines.

Meanwhile, the chief European Union regulators are going to be changed at the end of October (see sidebar below). The senior figures in the European Commission, the civil service of the EU, are being replaced: out goes Commission President Romano Prodi and nearly all his team. In comes Jos Manuel Barroso, until now the prime minister of Portugal, and 24 colleagues from across the enlarged EU.

References
1.CHMP/EWP/l068/04: Recommendation on the need for revision of the CPMP note for guidance on evaluation of anticancer medicinal products in man; http://www.emea.eu.int/pdfs/human/ewp/106804en.pdf.

2. Note for guidance on clinical investigation of medicinal products in the treatment of lipid disorders, CHMP, July 2004; http://www.emea.eu.int/pdfs/human/ewp/302003en.pdf.

SIDEBAR: Key figures in the new European CommissionGnter Verheugen from Germany, responsible for industry affairs-including enterprise policy, competitiveness, and the European Agency for the Evaluation of Medicinal Products
Stavros Dimas from Greece, responsible for environment-including questions over genetically modified products
Janez Potocnik from Slovenia, responsible for science and research, including the EU's joint research centre and the EU's multibillion euro research budget
Markos Kyprianou from Cyprus, responsible for health and consumers protection, and in charge of the new European Centre for Disease Prevention and Control, which is shortly to open up in Stockholm
Charlie McCreevy from Ireland, responsible for questions relating to access to the 25-state single market that the EU now represents
Peter Mandelson from the United Kingdom, responsible for trade, including cooperation (and dispute resolution) with the United States and Japan