Abstract
This cross-sectional study reviews New Drug Applications for new molecular entities and Biologics License Applications for new and biosimilar biological products submitted to the US Food and Drug Administration to assess how frequently applicants disclose applications in the media.
Current regulations prohibit the US Food and Drug Administration (FDA) from publicly disclosing the existence of pending New Drug Applications (NDAs) or Biologics License Applications (BLAs) unless previously publicly disclosed or acknowledged.1 Applicants may disclose applications in press releases, Securities and Exchange Commission (SEC) filings, or in other media; how frequently they do so is not known.
Methods
We conducted a cross-sectional study of all NDAs for new molecular entities and BLAs for new and biosimilar biological products submitted to the FDA between January 1, 2010, and December 31, 2016. Data were collected in July 2017. Because this was a cross-sectional study based on nonclinical data, we did not submit this study for human subjects ethics review.
Primary outcomes included disclosures of applications in press releases, SEC filings, and other media. For press releases, we searched applicants’ websites and websites where press releases are disseminated. For SEC disclosures, we searched a public SEC database for filings from firms publicly traded in the United States.2 If disclosures were not found, we conducted Google searches to identify any disclosures in other media (eg, investor materials or annual reports). We calculated the percentage of submissions disclosed and the median time to disclosure (days with interquartile range [IQR]).
We also searched Pharmaprojects, a subscription-based drug development database, and determined the percentage of submissions disclosed; such a listing was not considered to be public disclosure.3
Results
Of the 249 applications (188 NDAs and 61 BLAs), 222 (89.2%) were disclosed in at least 1 public medium (Figure), including in a press release for 196 applications (78.7%) (Table). Of the press releases, 196 (86.2%) were found on applicants’ websites. Of the applications, 157 (63.1%) were disclosed in SEC filings, including 136 (54.6%) in press releases and SEC filings. The overall rates of public disclosures increased from 87.5% in 2010 to 97.6% in 2016. Publicly traded companies were more likely to disclose applications in any medium compared with nonpublic companies (relative risk, 1.43; 95% CI, 1.09-1.90); private companies were less likely to disclose applications (relative risk, 0.70; 95% CI, 0.53-0.92) (Table). The median time to disclosure was 6 days (IQR, 1-63 days) for press releases, 55 days (IQR, 9-86 days) for SEC filings, and 6 days (IQR, 1-63 days) for any public media.
Figure. Public Disclosure of Application Submissions in Press Releases (PRs), Regulatory Filings From the US Securities and Exchange Commission (SEC), and Any Other Documents.
Data from a subscription-based database (secondary outcome) are not included.
aThe “other” documents were only searched when a PR or SEC filing was not found (n = 32). The following documents were found: a financial highlights document (n = 1), a notice of patent extension (n = 1), a stockholder update document (n = 1), and miscellaneous press articles (n = 2).
Table. General Characteristics of All Applications and Disclosuresa.
| Characteristic | All Applications, No. (%) (N = 249)b | Application Submissions Announced in Press Releases (n = 196) | Applications for Companies Publicly Traded in the United States, No. (%) (n = 221)b | Application Submissions Declared in US Securities and Exchange Commission Filings (n = 152)c | Application Submissions Disclosed in Any Medium (n = 222)d | |||
|---|---|---|---|---|---|---|---|---|
| No. (%) | Relative Risk (95% CI)e | No. (%) | Relative Risk (95% CI)e | No. (%) | Relative Risk (95% CI)e | |||
| Application type | ||||||||
| Biologic | 61 (24.5) | 48 (78.7) | 0.99 (0.85-1.16) | 57 (25.8) | 40 (70.2) | 1.03 (0.84-1.25) | 56 (91.8) | 1.04 (0.95-1.14) |
| New drug | 188 (75.5) | 148 (78.7) | 1.00 (0.76-1.62) | 164 (74.2) | 112 (68.3) | 0.97 (0.80-1.19) | 166 (88.3) | 0.91 (0.88-1.05) |
| Drug class typef | ||||||||
| First in class | 93 (37.3) | 76 (81.7) | 1.06 (0.93-1.21) | 84 (38.0) | 61 (72.6) | 1.09 (0.92-1.31) | 84 (90.3) | 1.02 (0.94-1.11) |
| Not first in class | 156 (62.7) | 120 (76.9) | 0.94 (0.83-1.07) | 137 (62.0) | 91 (66.4) | 0.91 (0.77-1.09) | 138 (88.5) | 0.98 (0.90-1.07) |
| Application review typeg | ||||||||
| Rolling | 66 (26.5) | 55 (83.3) | 1.08 (0.95-1.24) | 62 (28.1) | 44 (71.0) | 1.04 (0.86-1.27) | 62 (93.9) | 1.07 (0.99-1.17) |
| Priority | 129 (51.8) | 110 (85.3) | 1.18 (1.04-1.35) | 114 (51.6) | 78 (68.4) | 0.98 (0.82-1.18) | 117 (90.7)h | 1.03 (0.94-1.12) |
| Standard | 118 (47.4) | 85 (72.0) | 0.84 (0.74-0.97) | 105 (47.5) | 73 (69.5) | 1.02 (0.85-1.21) | 104 (88.1)h | 0.97 (0.89-1.06) |
| Application status | ||||||||
| Approved | 199 (79.9) | 165 (82.9) | 1.34 (1.07-1.68) | 179 (81.0) | 122 (68.2) | 0.95 (0.77-1.18) | 183 (92,0) | 1.18 (1.01-1.37) |
| Pending | 11 (4.4) | 8 (72.7) | 0.92 (0.64-1.33) | 8 (3.6) | 8 (100) | 1.48 (1.35-1.62) | 9 (81.8) | 0.91 (0.69-1.21) |
| Complete response letteri | 18 (7.2) | 12 (66.7) | 0.84 (0.60-1.17) | 16 (7.2) | 13 (81.3) | 1.20 (0.93-1.54) | 16 (88.9) | 1.00 (0.84-1.18) |
| Refuse to file | 8 (3.2) | 4 (50.0) | 0.63 (0.31-1.26) | 6 (2.7) | 2 (33.3) | 0.48 (0.15-1.49) | 4 (50.0) | 0.55 (0.28-1.11) |
| Withdrawn | 13 (5.2) | 7 (53.8) | 0.67 (0.40-1.12) | 12 (5.4) | 7 (58.3) | 0.84 (0.52-1.37) | 10 (76.9) | 0.86 (0.63-1.16) |
| Applicant | ||||||||
| Publicly traded in United States (based anywhere) | 221 (88.8) | 183 (82.8) | 1.78 (1.19-2.67) | 221 (100) | 152 (68.8) | NA | 204 (92.3) | 1.43 (1.09-1.90) |
| Publicly traded and based in United States | 179 (71.9) | 147 (82.1) | 1.17 (0.99-1.39) | 179 (81.0) | 149 (83.2) | 11.70 (3.90-34.74) | 168 (93.9) | 1.22 (1.07-1.39) |
| Private | 28 (11.2) | 13 (46.4) | 0.56 (0.37-0.84) | 0 | NA | NA | 18 (64.3) | 0.70 (0.53-0.92) |
| Applicant size | ||||||||
| Small (≤1250 employees) | 78 (31.3) | 66 (84.6) | 1.11 (0.98-1.26) | 67 (30.3) | 54 (80.6) | 1.27 (1.07-1.50) | 71 (91.0) | 1.03 (0.94-1.13) |
| Large (>1250 employees) | 171 (68.7) | 130 (76.0) | 0.90 (0.79-1.02) | 154 (69.7) | 98 (63.6) | 0.80 (0.67-0.93) | 151 (88.3) | 0.97 (0.89-1.06) |
| Expedited programs or special statusj | ||||||||
| Accelerated | 33 (13.3) | 28 (84.8) | 1.09 (0.93-1.28) | 31 (14.0) | 25 (80.6) | 1.21 (0.99-1.47) | 31 (93.9) | 1.06 (0.96-1.17) |
| Breakthrough | 39 (15.7) | 35 (89.7) | 1.17 (1.03-1.33) | 36 (16.3) | 29 (80.6) | 1.21 (1.00-1.47) | 37 (94.9) | 1.08 (0.99-1.18) |
| Fast track | 94 (37.8) | 76 (80.9) | 1.04 (0.92-1.19) | 83 (37.6) | 59 (71.1) | 1.05 (0.88-1.26) | 88 (93.6) | 1.08 (1.00-1.18) |
| Orphan drug | 105 (42.2) | 81 (77.1) | 0.97 (0.85-1.10) | 92 (41.6) | 62 (67.4) | 0.97 (0.81-1.16) | 92 (87.6) | 0.97 (0.89-1.06) |
| Qualified infectious disease product | 7 (2.8) | 7 (100) | 1.28 (1.20-1.37) | 5 (2.3) | 4 (80.0) | 1.17 (0.75-1.83) | 7 (100) | 1.13 (1.08-1.18) |
Abbreviations: IQR, interquartile range; NA, not applicable.
All percentages are row based unless otherwise noted. All statistical tests had a P < .05 significance level. Data from a subscription-based database (secondary outcome) are not included.
Percentages are column based.
Calculations based only on applications belonging to firms that were publicly traded in the United States (n = 221), of which 152 had a Securities and Exchange Commission filing.
Includes disclosure of an application in any source (press releases, securities regulatory filings, or other medium). Includes 5 disclosures to the Securities and Exchange Commission made by publicly traded partners of private companies who submitted the applications to US Food and Drug Administration.
Relative risk calculations were conducted using the complements of the categorical (row) and disclosure (column) variables.
First-in-class status at application submission; some products may be first in class at submission but not at approval if another product of the class gets approved first.
Number of applications for priority or standard review are 247; 2 applications were withdrawn before the US Food and Drug Administration assigned review status.
The sum is 221 (not 222) because of exclusion of 2 applications that were withdrawn before US Food and Drug Administration assigned review status.
A letter documenting deficiencies in the application that preclude marketing.
One application may use multiple expedited development programs or special statuses; priority review, an expedited review program, is listed under review type.
The Pharmaprojects database listed 246 of the 249 study drugs (98.8%). We found a reference to FDA submission for only 162 applications (65.9%), including 11 not disclosed in public media. For 43 of the 162 applications, the listing date preceded the applications’ receipt by the FDA (median, 59 days before date of FDA receipt; IQR, 13-195 days).
Discussion
We found that information that the FDA treats as confidential with regard to applications for NDAs and BLAs was in most cases already available to the public and that the disclosure rate in 2016 was 97.6%. Most disclosures took place in press releases, SEC filings, or both and occurred within 1 week of application submission.
A 2010 FDA report recommended that these applications be publicly disclosed at filing.1 Implementing such a recommendation would mark a significant departure from longstanding FDA practice that may necessitate changes to statutes or regulations.1 Other transparency proposals in the 2010 report, such as the recommendation to disclose the FDA’s complete response letters, which list deficiencies in applications that preclude approval, would reveal that an NDA or BLA had been filed and thus necessarily include eliminating the prohibition against disclosing the existence of unapproved applications. We also found that subscription-based databases would not obviate the need for FDA disclosure because they are not public databases, and the database that we reviewed contained incomplete and, in some instances, inaccurate data.
The limitations of our study include underestimated disclosure of older applications because materials may no longer be Internet accessible and more recent applications in some disclosure media (eg, annual reports) may not yet be disclosed. We also did not review foreign securities filings or other subscription-based databases that may provide different information. Finally, findings cannot be generalized to other types of FDA applications, which may have different disclosure rates.
References
- 1.US Food and Drug Administration FDA Transparency Initiative: Draft Proposals for Public Comment Regarding Disclosure Policies of the U.S. Food And Drug Administration. Washington, DC: US Dept of Health and Human Services; 2010. [Google Scholar]
- 2.US Securities and Exchange Commission (SEC) and Electronic Data Gathering, Analysis, and Retrieval System (EDGAR). 2017. http://www.sec.gov/edgar/searchedgar/companysearch.html. Accessed July 1, 2017.
- 3.Informa PLC. Pharmaprojects: Complete drug development database. 2017. https://pharmaintelligence.informa.com/products-and-services/data-and-analysis/pharmaprojects. Accessed July 1, 2017.
