FDA, Research Organizations Seek to Bolster Drug Submissions

One sign of a healthy biomedical research enterprise is steady growth in new drug applications (NDAs) submitted to the Food and Drug Administration. There are signs of trouble, however, in the latest announcement of user fees for 2014 authorized by the Prescription Drug User Fee Act (PDUFA). FDA is increasing fees paid by pharma and biotech firms, as it has done steadily over the last 20 years. A key reason is that the agency calculates that it will receive only 116 submissions in 2014, slightly lower than the 122 applications filed in 2012. And because the agency has to collect a certain amount in fees each year, fewer applications means that each sponsor pays a little more.

Thus it will cost $2.2 million to submit an NDA or biologics license application (BLA) for agency review, according to a Federal Register notice published on August 2, 2013. A biosimilar application that carries clinical data will be just as pricey, while an efficacy supplement with clinical data to support a new indication or expanded labeling will cost $1 million. And although a $2 million application fee is relatively inconsequential for a large pharma company that spends billions on clinical trials and product development, it may be considerable for small firms with limited resources.

FDA sets its overall fee revenue target based on a "workload adjuster" that reflects inflation and the increased complexity of the drug review process. In addition to some 115-130 NDAs and BLAs filed each year, FDA receives nearly 7,000 commercial INDs, 140 efficacy supplements, and about 2,500 manufacturing supplements, according to the user-fee announcement. Sponsors will pay a total of $252 million in application fees, as well as equal amounts for FDA oversight of manufacturing establishments and marketed pharmaceutical products.

The drop in anticipated NDAs may reflect the squeeze on pharma investment in R&D and an ever-longer and more costly drug development process. FDA is on track to approve nearly as many new molecular entities (NMEs) as the near-record 45 in 2012, but those gains could halt in future years if applications diminish.

A more positive view of the approvals-and-innovative issue is to distinguish truly important NMEs—"first-in-class" and "advance-in-class" medicines that account for a growing proportion of new drugs—from "addition-to-class" drugs that are declining, say FDA analysts (http://bit.ly/14ZZ5OY). A rising number of advanced new therapies are coming to market, according to this new look at application approval rates, which counters concerns about total approvals remaining static.

Everyone would be a winner, moreover, if pharma and biotech companies filed more high quality applications. That would boost the approval rate despite fewer initial submissions.

Streamlining studies

One strategy for increasing the productivity of biopharma R&D is to develop a more robust clinical trial infrastructure that taps into health information more effectively, says Janet Woodcock, Director of the Center for Drug Evaluation and Research. She would like to see more ongoing clinical trials that utilize standardized methodologies to rapidly screen candidate compounds for signs of efficacy. "Let's not reinvent the wheel with every trial," she said at a July conference sponsored by the Brookings Institution. She noted that a "network of data sources" linked by common protocols and standards could greatly compress the clinical testing process.

Similarly, the National Institutes of Health aims to reduce the high failure rate of clinical trials through a target validation consortium, starting with pilots for Alzheimer's disease, diabetes, rheumatoid arthritis, and schizophrenia. The Patient-Centered Outcomes Research Institute (PCORI) also is taking steps to improve the conduct of clinical outcomes research through its National Patient-Centered Clinical Research Network. The program will consist of some 25 research networks formed by academia, research organizations, and patient groups, each able to tap into health data for over 1 million individuals, explained PCORI Chief Science Officer Bryan Luce at the Brookings conference. He acknowledged challenges, such as devising more efficient oversight by institutional review boards, but predicted that the program will create "a dynamically linked clinical research network" able to test hypotheses and conduct adaptive studies, Bayesian trials, and observational studies.

By providing data from real-world patients in usual care settings, the PCORI research network appears likely to support industry post-marketing studies, as opposed to randomized trials for investigational medical products. The program also could help identify and recruit patients for experimental studies, encourage use of patient-reported outcomes, indicate to sponsors why certain patients don't respond to treatments, and what endpoints are important. Key issues are how PCORI develops the governance and structure of the research networks, as well as who has access to the data.