FDA counted its first-cycle complete responses, and the conclusion is addressed to sponsors

Over the past year a number of outside analyses have argued that FDA is issuing more complete responses on the first review cycle. On 16 September CDER answered with its own numbers: First-Cycle Complete Response Rates for CDER NDA and BLA Applications: A Longitudinal Analysis, FY 2008–2024. It uses the agency's internal tracking system rather than public data, partly because, as the paper says, "some applications in the CR dataset are not yet publicly available."

Most of the coverage has focused on whether FDA is right that the trend is down. For a team preparing an NDA or BLA, the more practical parts of the paper are the BLA figures and the conclusion, which says plainly what the agency expects sponsors to do with the advice they already receive.

What the numbers say

FDA defines the first-cycle complete response rate for a fiscal year as the share of original NDAs and BLAs filed that year that received a complete response at the end of their first review cycle. Applications refused to file or withdrawn before filing are excluded. FY2025 is left out because 31 applications from that year were still within their review goal when the data were pulled in June.

On that basis, the rate fell "from 47% CR rate for FY2008 cohort to approximately 27% in FY2024," with an average of 33.7% over the period. The paper describes a step down around FY2011 and a plateau near 30% since. It acknowledges a rise from 22% for the FY2018 cohort to 37% for FY2020, and treats it as a temporary perturbation, possibly linked to the limits on inspections during the pandemic.

Two other figures matter for planning. The median first-cycle approval rate across the annual FY2008–2024 cohorts was about 64%. The median current approval rate across the same cohorts, using data as of 1 June 2026, was about 77%. Approvals on a second or later cycle account for the higher current rate. And efficacy supplements have become far less likely to receive a complete response, falling from about 30% in FY2008 to 7% in FY2024 while their volume nearly doubled.

New BLAs are the exception

The paper is direct about the one category moving the wrong way: "Of note is the increase in CR rates of CDER's original BLAs since FY2019. This increase is more pronounced for new BLAs (Original BLAs containing new active ingredients)." It also notes that BLA filings have grown, from an average of 9 a year under PDUFA IV to 22 under PDUFA VI, and recommends continued monitoring "particularly for BLAs."

The paper does not say why. Its footnotes point to a 2024 FDA and Duke-Margolis meeting on "the root causes of CRs related to quality and facility issues" for biologics, and to the CMC Development and Readiness Pilot as an example of FDA trying to surface problems early. That is consistent with what the published rejection letters show, which we went through in July: manufacturing, facility and product-quality deficiencies appear far more often than efficacy (FDA says CMC readiness is the bottleneck). The FDA analysis itself does not establish what caused the recent rise in BLA complete responses, and the published letters are a selected sample of products that were eventually approved. Still, for a biotech preparing its first BLA, quality and facility readiness deserve explicit planning alongside the clinical evidence.

The paragraph addressed to sponsors

The conclusion sets out a division of responsibility:

"FDA is accountable for providing transparent statutory and regulatory expectations. Industry, in turn, is responsible for leveraging FDA guidance, pre-NDA and pre-BLA meeting feedback, and publicly available information to construct high-quality submissions."

Read as a writer, that sentence names three inputs to an application, and each one has a document trail.

Guidance. An application written over two or three years can be drafted against a guidance that has since been revised, finalised or withdrawn. Someone has to know which sections relied on which version, and check them before submission. We wrote about how often that happened in a single summer in Five sources moved in five weeks.

Meeting feedback. Pre-NDA and pre-BLA minutes often contain the agency's most specific expectations: an analysis it wants to see, a dataset it expects in a particular form, a question on a manufacturing site. Those agreements are made by the regulatory lead, but they have to be carried out by whoever writes the clinical summary, the integrated safety analysis or Module 3, sometimes months later. A practical check before filing is to list every request and agreement in the minutes and point to the section of the application where each one is answered.

Publicly available information. Since FDA began publishing complete response letters, the public record includes the deficiencies the agency has actually cited in comparable programmes (FDA reversed its CRL pause). FDA's own paper now counts that record among the things sponsors are expected to use.

What the gap costs

The difference between first-cycle and eventual approval is measured in time. A 2026 study of publicly available complete response letters for novel therapies ultimately approved between 2020 and 2024 found median times of 0.60 years from receipt of the letter to resubmission and 1.28 years from receipt of the letter to approval (Therapeutic Innovation & Regulatory Science). For a small company, most of that period is spent on a programme that is not yet earning anything.

The paper closes by suggesting that "there is perhaps value in a more in-depth analysis of complete responses" to see what FDA, industry or both could do differently. Until that analysis exists, the sentence about guidance, meeting feedback and public information is the clearest statement of what the agency expects, and each of the three can be checked against the draft before it is filed rather than after the letter arrives.