What is a special protocol assessment worth, anyway?

After bending over backwards to counter criticism and approve more drugs, the FDA might have drawn a line in the sand. The scathing panel opinions on Aveo Oncology and Delcath Systems’ respective cancer projects two weeks ago were unambiguous and, despite each having apparently met the terms of its special protocol assessment, chances of approval of either are now hopelessly low.

At least three other companies are similarly pinning their hopes on success in single oncology studies that, under an SPA, use improvement in progression-free survival as their primary endpoint (see table below). Aveo and Delcath’s crushing setbacks could be making it difficult for these groups’ management teams to sleep easy.

It is striking that at the May 2 panel meeting both Aveo and Delcath were sharply criticised for submitting data that they had previously been warned was likely insufficient (Adcom double play erases Aveo and Delcath, May 3, 2013). Nevertheless, each company would argue that all it was doing was complying with an SPA, having met the terms set out within.

Under the 1997 FDA Modernization Act an SPA represents an understanding with a company that the agency backs a pivotal study’s design, and indicates that a positive result in such a trial can form the basis of regulatory approval. Biotech sponsors routinely boast that an SPA is “binding”.

The endpoint conundrum

The trouble starts when a relatively weak endpoint is agreed on – such as progression-free survival (PFS) rather than the normally accepted oncology gold standard of overall survival.

This in itself is not necessarily a problem; in some cancers PFS is the best that any drug has shown, while certain study designs make it unrealistic to measure overall survival. But, as Aveo and Delcath found, if a negative effect on overall survival is seen the FDA will not ignore it, and even the hint of a worsened outcome will likely prove terminal.

A quick perusal of EvaluatePharma reveals three other companies pinning their hopes on the success of a study that the FDA has signed off under an SPA and which has improvement in PFS as the primary endpoint. One of these – Pharmacyclics – has two such studies of its Bruton’s tyrosine kinase inhibitor ibrutinib under way, the first of which should provide investors with the next test of the FDA’s thinking after it reads out next year.

The recently floated Puma Biotechnology has initiated a trial of neratinib in Her2-positive breast cancer, although this will not read out for some time, while a study of CytRx’s aldoxorubicin in soft tissue sarcoma has yet to begin. All will measure overall survival as a secondary endpoint, and the companies involved will no doubt hope that this does not throw a spanner in the works.

Pharmacyclics had earlier carried out a phase III trial under an SPA with Xcytrin, but this failed to demonstrate an effect on PFS, and Celsion’s Thermodox suffered a similar fate. Meanwhile, Ariad/Merck & Co’s ridaforolimus did show a statistically significant effect on PFS in a phase III soft tissue sarcoma trial carried out under an SPA – with a numerical overall survival benefit – but the agency still refused to approve it without further studies.

On the other hand, Exelixis agreed a PFS-based phase III study under an SPA for cabozantinib, and a positive result was enough for approval, although there was also a strong numerical improvement in overall survival.

Design flaws

Still, perhaps we should not be too quick to draw conclusions from Aveo's tivozanib and Delcath's Melblez; the bull case is that Aveo was slammed for major design flaws, particularly allowing sorafenib recipients to switch to tivozanib but not the other way around. Delcath would have faced questions over severe toxicity irrespective of pooled data showing a non-significant increased risk of death.

And panel members did reiterate that – in principle, at least – PFS remained an acceptable endpoint in kidney cancer studies. But the May 2 panel meeting might have been a watershed moment, and it is clear that a significant PFS benefit is potentially good enough only in the absence of a detrimental effect on overall survival.

Even a non-significant numerical deterioration in overall survival – or a signal in the pooled trial database – should set off investors’ alarm bells. Cynics will now be sure that an SPA is binding on only one party.

To contact the writer of this story email Jacob Plieth in London at [email protected] or follow @JacobEPVantage on Twitter