Based on years of poor R&D productivity, the pharma industry is rethinking clinical development. Novartis CEO Joseph Jimenez is rethinking the traditional blockbuster. Tomas Philipson and former FDA Commissioner Andrew von Eschenbach are rethinking phase III trials. A third aspect of clinical development also requires rethinking but I’ll leave that for the end of this post.
Jimenez sees the new blockbuster as a therapeutic agent that targets a biological pathway rather than a single indication in a large population. By targeting a pathway, the new therapeutic agent can treat multiple health conditions and receive multiple niche approvals that collectively generate more than $1B. As John Carroll notes, basing development strategy on targeting biological pathways is not new – the strategy is natural for biotechs like Genentech. The novelty is having a pharma giant with about $57B a year in revenues see targeting biological pathways as the basis for a blockbuster-by-accretion strategy and the key to future growth.
Reduce Reliance on Phase III Trials – or Eliminate Them?
Like Jimenez, Philipson and von Eschenbach have targeted therapeutic agents in mind when they call for reducing reliance on phase III trials or even eliminating them. They would like to allow marketing new drugs after “promising early stage research in targeted patients, with appropriate post-marketing studies required.” Labeling would state that approval was based on “promising but provisional research.” The marketplace of payers and patients would be the ultimate judge of new products. Is this idea radical? Not really. Alistair Wood advocated earlier approvals conditional on phase IV trials years ago ((N Engl J Med. 2006 Aug 10;355(6):618-23)). In a 2011 opinion piece in Science, Intel’s Andy Grove proposed leaving phase I safety testing under FDA’s purview, but not efficacy testing. “E-trials” would replace later phase trials, with “qualified physicians” dispensing experimental medicines and response data entering a database for analysis. Conditional approvals find compassionate use today.
Derek Lowe expresses reservations about reducing reliance on phase III trials. For example, he asks, “What would keep the real bottom-of-the-barrel types from pumping out high-priced placebos for demanding diseases like Alzheimer’s, which compounds would fly through safety studies and reap big profits until they (slowly) were proved ineffective?” The potential for gaming the system would depend on the rules for progressing the drug. So would the scope of such a policy and the scale of potential benefits. Stringent rules for bypassing phase III would limit the impact because few drugs would skip phase III. With less stringent rules, there would still be many drugs that merit continued development but not skipping phase III.
The underlying issue is that the current structure calling for large studies with an overall benefit (p<0.05) was fine for traditional blockbusters but doesn’t work for small populations. Think of this not only in terms of orphan drugs (<200,000 pts/year) but getting down to N-of-1 scale – personalized medicine. Doing a study of a group like epidermolysis bullosa, with an estimated 30,000 patients scattered across the US is extremely difficult under our current model, increasing costs and delaying (or denying) useful therapies.
Conquering the Inefficiencies We Can’t Bypass
Much as we need to rethink the blockbuster and phase III, we can’t bypass the core of pharma’s productivity problem – the way we run clinical trials. To begin with, generating revenues from multiple indications is central to the blockbuster-by-accretion strategy. It follows that such collective blockbusters will require some clinical testing in each indication. Multiple indications mean more trials, not less. Furthermore, nothing in our experience of clinical research suggests that early phase results will usually be so promising that we can justify skipping phase III. That means we’ll still need to run phase III trials more efficiently. Finally, earlier phase studies will remain on all scenarios. These trials are less wasteful not because the industry runs them better than late-phase trials but only because they involve fewer subjects. So developing most drugs more efficiently will require running trials more efficiently across all phases.
Sometimes we will be able to bypass inefficiencies through measures such as skipping phase III but we can only bypass so much. The only way to solve pharma’s productivity problem is to confront and conquer inefficiency in program planning, study design, and, above all, in trial management and operations. We can’t win the war against inefficiency without fighting and winning the daily operational battles that dominate our budgets and timelines. But the outlook is bright: those with access to innovations such as adaptive monitoring and adaptive enrollment, are winning more of these battles every day.