Recently MannKind's (MNKD) Afrezza NDA decision was delayed for approximately four weeks with no reason given. The market responded positively, sending MNKD share prices up over 8% on the news before eventually settling back into its pre-announcement trading range. But what does the delay actually mean for MannKind? And what about the delay for Human Genome Sciences' (HGSI) Benlysta? Is there actually a trend between delays and the eventual outcome of the decision?
Some articles have hastily cobbled together some statistics about delays but unfortunately have fallen short of offering a real analysis. I therefore wanted to take a in-depth look at several NDA action date (aka PDUFA date) delays to dig up some data that might be actionable.
First let's collect the data. Here is a run-down of several NDA action date delays, along with the reasons for the delay and the ultimate decision by the FDA. I've tried to be as comprehensive as possible, but if I've missed something please mention it in the comments and we'll update our data. While the analysis is post-hoc and hardly statistically significant, it should be enough for us to establish some trends pertaining to the aforementioned delays.
Spectrum Pharmaceuticals (SPPI) - Levoleucovorin (calcium leucovorin, pure isomer) NDA for adjuvant for methotrexate
Original PDUFA date: Feb 10, 2008
1st Extension date: May 10, 2008
Reason: A REMS program was submitted and led to a three month extension (although not explicitly stated as the reason).
2nd Extension date: The FDA did not provide a specific date, but informed the company that it expects to issue the action letter shortly after the scheduled PDUFA date
Result: Approval - May 20, 2008
Daiichi Sankyo/Eli Lilly (LLY) - Prasugrel for acute coronary syndromes
Original PDUFA date: June 26, 2008
Extension date: Sept 26, 2008 (The FDA later missed this extended date in dramatic fashion and the drug ended up in front of a panel in Feb 2009. This is quite an extraordinary story and I'd suggest a little further reading if you're unfamiliar with it).
Reason: "...based on supplemental information provided during the review period."
Result: Approval - July 10, 2009
Salix Pharmaceuticals (SLXP) - METOZOLV (metoclopramide) NDA for gastroesophageal reflux disease
Original PDUFA date: Nov 30, 2008
Extension date: Feb 27, 2009
Reason: "...to provide time for a full review of the submission." No official reason given but this appears to be a standard three month extension based on the submission of additional data.
Result: CRL - Feb 26, 2009
United Therapeutics (UTHR) - Tyvaso (inhaled treprostinil) NDA for pulmonary hypertension
Original PDUFA date: Apr 30, 2009
Extension date: July 30, 2009
Reason: "The three-month extension was triggered by United Therapeutics' April 2009 submission to the FDA of the results of a human factors study, which was considered a major amendment to the NDA."
Result: Approval - July 30, 2009
Covidien (COV), Nuvo Research - Pennsaid Topical Solution for the treatment of osteoarthritis of the knee.
Original PDUFA date: Aug 5, 2009
Extension date: Nov 4, 2009
Reason: "During the review process, Nuvo provided the FDA with supplemental information, which the Agency determined to be a major amendment to the Pennsaid New Drug Application."
Result: Approval - Nov 5, 2009
Original PDUFA date: Nov 9, 2009
Extension date: Feb 9, 2010
Reason: "The FDA determined that a Risk Evaluation and Mitigation Strategy was necessary for GSK1838262. In response to FDA’s request, GSK submitted a proposed REMS. The FDA accepted this submission as a solicited major amendment to the GSK1838262 NDA."
Result: CRL - Feb 17, 2010
Note: The delay between the extension date and final decision was weather related.
Cadence Pharma (CADX) - IV Acetaminophen NDA for pain
Original PDUFA date: Nov 12, 2009
Extension date: Feb 12, 2010
Reason: "The FDA designated one of Cadence's submissions to the NDA, which contained additional clinical pharmacology data requested by the agency during the review process, as a major amendment."
Result: CRL - Feb 11, 2010
Sepracor, now known as Sunovion - Stedesa NDA for seizure
Original PDUFA date: Jan 30, 2010
Extension date: April 30, 2010
Reason: "In November 2009, at the request of the FDA, Sepracor submitted additional information about STEDESA to the agency."
Result: CRL - May 3, 2010
MannKind (MNKD) - Afrezza NDA for diabetes
Original PDUFA date: Jan 16, 2010
Extension date: Not given
Reason: "The FDA explained that it has not yet completed its inspection of the insulin manufacturing facilities of N.V. Organon, a third-party supplier to MannKind."
Result: CRL - Mar 15, 2010
Note: While the extension appears to be almost three months to the day. This does not appear to be a standard additional data extension due to the reason given by MannKind (basically, I'm assuming they aren't lying).
OSI Pharma (OSIP)/Genentech - Tarceva (erlotinib) sNDA for advanced non-small cell lung cancer
Questcor (QCOR) - Acthar gel sNDA for Infantile Spasms
Original PDUFA date: June 11, 2010
Extension date: Sept 11, 2010
Reason: "The FDA extended the PDUFA date in order to review information regarding labeling and potential post-approval commitments that they solicited from Questcor." Additional data triggered a three month extension.
Result: Approval - Oct 15, 2010
Warner Chilcott (WCRX) - Actonel (risedronate sodium) extended-release for osteoporosis
Original PDUFA date: July 24, 2010
Extension date: Oct 24, 2010
Reason: "...to allow sufficient time to review additional information solicited by the FDA and previously provided by the Company."
Result: Approval - Oct 11, 2010
Cumberland Pharma (CPIX) - Acetadote sNDA for non-acetaminophen acute liver failure
AstraZeneca (AZN) - Ticagrelor (BRILINTA) NDA for Acute Coronary Syndromes
Before analyzing the data, an important dichotomy must first be made: many delays were three month extensions, often triggered by submission of additional data and the others appear to be due to the FDA simply needing more time. There should be much different results from either category based on this important difference. Keep in mind that MannKind's current delay was not triggered by additional data and the delay for Benlysta was triggered by additional data (I'm assuming here HGSI gave no reason for extension but reasons for a three month extension are almost always additional data).
First, let's look at the "standard" delays - the delays most pertinent to HGSI. In this case, there were 13 total extensions that appear to be due to the submission of additional data. Of these extensions, seven resulted in approval and six resulted in a complete response letter (CRL). The approval rate for those NDA decisions, then, was 54%. This falls approximately in line with a 66% first cycle approval rate given by the FDA (pdf) for 2006.
Perhaps more interesting, though, was the fact EVERY company that got a standard three month extension without providing a reason for said extension got a CRL. Three of three in this case. It may just be a correlation between, "Good managements tend to keep their shareholders informed," and "Good management tends to have a better approval percentage," but I can't help but wonder if these companies withheld reasons that could have adversely affected their share prices. Whatever the case, it'll be interesting to see if that holds up on decisions in the future.
HGSI, it should be noted, provided a reason for their three-month extension.
For MannKind, on the other hand, we have significantly less data. In total, there is only record of three delays that were not standard three month extensions, four if you count Prasugrel's lengthy unannounced second delay. In these cases, three of the four extensions resulted in approval - a 75% approval rate. The lone CRL, in fact, came from MannKind itself. And if MannKind's extension was considered a standard additional data submission, contrary to my assumptions, then the approval rate climbs to 100%.
Due to the low number of delays similar to those of MannKind's current delay, it's hard to pull anything concrete from the data provided. In a more qualitative sense, however, it would appear that of all the delays given, the delays of Adolor's Entereg and Spectrum's Levoleucovorin are most similar to MannKind's current delay. If you want to go that route, the delay is most certainly a good one; both of these delays resulted in approval.
It's up to you how to interpret the above data, but I hope it provided a better insight into what delays might mean for approval of both Afrezza and Benlysta. My own conclusions, based as much on logical reasoning as it is on the data, are:
- Three month delays due to submission of additional data mean virtually nothing in terms of change in approval odds.
- Companies that receive three month extensions without providing a reason may be worthy of additional scrutiny.
- Shorter delays by the FDA that are not based on additional data submission appear to be a good thing.
The market, then, may have been responding properly when MannKind's share price popped by over 8% on the delay news. The delay of HGSI's Benlysta, on the other hand, is unlikely to mean much of anything in terms of eventual approval odds.
Good luck to all of those carrying a position through MannKind's FDA decision on "approximately" January 24.