EMA Releases New Advice on Human Trials for Cancer Drugs

April 1, 2016
Philip Ward

Philip Ward is ACT's European editor, phone +44 1244 538583, philipward1@btconnect.com

Applied Clinical Trials

Volume 25, Issue 4

The agency is concerned about the lack of non-clinical models with good predictive properties in the oncology space.

The European Medicines Agency (EMA) has published draft guidance on the evaluation of anti-cancer medicines in humans.

The 39-page document, which was issued on March 15 and is open for feedback until September 15, covers all stages of clinical development and addresses the development of treatments for malignancies, including drug resistance modifiers or normal tissue protective compounds.

The agency is concerned that although many anti-cancer compounds are being developed, companies have only obtained a marketing authorization for a relatively small percentage of them due to poor activity or evidence of a detrimental safety profile. Until non-clinical models with good predictive properties have been defined, this situation is unlikely to change, and the absence of such models represents the greatest hurdle for efficient drug development in the near future, EMA noted.

A central aim of the document is to promote the development of molecule-specific preclinical models to assess and predict anticipated activity as well as safety, which tends to be a standard approach used by developers of targeted molecules. The validation and predictive reliability of these models is a complex, time-consuming, and specialist process that requires tumor immunologists.

EMA has sought to classify compounds according to reasonable designs of exploratory studies. This applies to cytotoxic compounds where the toxicity and overall response rate (ORR) are thought to be suitable markers of activity in dose-finding studies, compared with non-cytotoxic compounds where ORR and/or toxicity may not serve this purpose.

Importantly, intra-patient dose escalation in Phase 1 can allow more effective drug levels to be reached, provided no toxicity is seen in two dosing cycles, state the authors of the document. This might help smaller biotech companies, for example.

“The requirements of the characterization of the safety profile have changed with the emergence of molecularly targeted agents, immunomodulating drugs, and other non-cytotoxic agents. These types of agents may have other types of toxicity and are often dosed differently to conventional chemotherapy,” they wrote. “The dose-finding process and concepts such as dose-limiting toxicity may, therefore, need to be addressed differently than for standard cytotoxic agents.”

Moreover, cumulative incidences by toxicity grade are not sufficient to characterize the toxicity profile. The impact of an adverse drug reaction on the benefit-risk balance may, for instance, differ importantly depending on how the incidence, prevalence, and severity change with time on treatment, and on the possibility to alleviate the reaction by dose reduction.

Survival clues

EMA is urging companies to submit overall survival data compatible with a trend towards favorable outcome to capture potential negative effects on the activity of next-line therapies and treatment-related deaths. This approach is likely to have consequences on interim analyses, other than for futility, and cross-over, which should be undertaken only when available survival data provide the information needed for a proper evaluation of benefits and risks, explained the authors.

As well as defining the appropriate doses and schedules of a cancer drug, the EMA emphasize’s the importance of identifying a target population with optimized benefits and risks in the section about exploratory studies. Advice is also supplied about studies for combinations of drugs with minimal activity, as well as combinations of conventional cytotoxics.

No precise definition is given for either “trend towards favorable effects on survival” or “reasonably excluding negative effects on overall survival,” but the authors explain that if a major increase in toxicity is foreseeable, it is recommended that confirmatory studies are undertaken with the aim of showing overall survival benefit. They acknowledge that improved safety without loss in efficacy may constitute tangible aims and the design of non-inferiority efficacy studies.

The safety focus of the document has added relevance in the light of the recent events of Zydelig, the PI3K inhibitor made by Gilead Sciences. The drug is being investigated by EMA after reports of serious side effects-including multiple deaths-among patients in several studies testing Zydelig in newly diagnosed leukemia and lymphoma. Gilead halted those trials in March. 

Earlier in March, EMA published long-awaited guidance on how to comply with its policy on publication of clinical data. The agency is moving toward the operational implementation of its proactive publication policy, which has launched a new era of transparency, said Noël Wathion, the agency’s deputy executive director.

The guidance will ensure that companies are aware of what is expected of them and are ready for the publication of these critical data, he added. EMA wants to work with companies that are concerned by the first wave of publication (i.e., those for which the decision-making process has been finalized since the policy entered into force). EMA is organizing a webinar in the second quarter of 2016 to allow companies to ask any outstanding practical questions. This webinar will be a live broadcast and will be available for future reference on the EMA website.

download issueDownload Issue : Applied Clinical Trials-04-01-2016