Potential Red Flags as Chelsea Faces CRDAC Vote Second Time Around

FDA Advisory Committee meetings are synonymous with drama; the expectation for run-up trades, the release of briefing documents and of course the actual proceedings and panel vote.  Chelsea Therapeutics’ Northera has drummed up much enthusiasm from bulls, optimistic for a positive outcome second time around at the Cardiovascular and Renal Drugs Advisory Committee (CRDAC) meeting scheduled for January 14^th, 2014. 

As a reminder, Chelsea is developing Northera (droxidopa) for the treatment of symptomatic neurogenic orthostatic hypotension (nOH) in patients with underlying primary autonomic failure (Parkinson’s disease, multiple system atrophy, and pure autonomic failure), dopamine-β-hydroxylase deficiency, and non-diabetic autonomic neuropathy.  Patients with nOH experience significant drops in systolic blood pressure upon standing up. In healthy individuals, the autonomic nervous system detects changes in posture and ensures that blood pressure is maintained when standing from sitting or lying.  As nOH patients have primary autonomic failure, they are unable to regulate blood pressure and can experience severe symptoms such as dizziness, light-headedness, fainting, changes in vision and experience can falls.  Currently there are no FDA approved drugs to treat these symptoms.

Chelsea originally submitted a new drug application (NDA) for Northera in September 2011 on the basis of data from combined safety and efficacy data from Chelsea's two completed Phase III efficacy studies in nOH (Studies 301 and 302), two long-term open-label extension studies, a dedicated thorough QTc study, and a 24-hour ambulatory blood pressure monitoring safety study. Having conducted the pivotal study under Special Protocol Assessment, and having received Orphan and Fast-track drug designations, the company requested priority review.  FDA recognizing the unmet patient need granted priority review to the sponsor and scheduled review by CRDAC in February 2012. Some of the key issues that came up in the original review included the following:

·        In Study 301 droxidopa conferred at least one week of symptomatic benefit on the Orthostatic Hypotension Symptom Assessment (OHSA) Item 1 with a robust p value of < 0.001. However, Study 302 failed to meet its’ end points; there was no symptomatic benefit and no increase in standing blood pressure with Northera treatment versus placebo. There were some supportive data from a smaller study.

·         The absolute magnitude of symptomatic benefit was questioned as to whether a mean change of 1.3 on a patient reported outcome instrument was clinically meaningful, especially when this effect size was not maintained beyond 2 weeks of therapy.  nOH is a chronic condition and durability of effect was questioned repeatedly by FDA reviewers

·         Inadequate or perhaps unclear safety characterization given the unexplained deaths and risk of rare serious adverse events

The CRDAC meeting was a very split meeting despite the final 7:4 vote in favor to approve Northera.  The committee agreed that for a subset of patients there appeared to be a significant symptomatic benefit, but shared concerns that durability of effect needed to be demonstrated beyond 2 weeks; ideally in studies 4 - 12 weeks long.  In terms of safety, the committee did not appear overly concerned with safety profile but did note that a lot of nOH patients can exhibit supine hypertension at baseline and the effects of treatment on this group of patients had not been characterized. Despite the positive vote, FDA issued a complete response letter to the sponsor requesting Chelsea submit data from an additional study to support efficacy demonstrated in Study 301, along with the recommendation that such a study be designed to demonstrate durability of effect over a 2- to 3-month period.

Fortuitously in 2010 Chelsea had initiated another study in nOH patients with Parkinson’s disease in order to assess the impact on Orthostatic Hypertension Questionnaire (OHQ) score improvements at 8 weeks; Study 306. Unfortunately at the first interim futility analysis of this trial which occurred in early 2011, Chelsea reported that the study was unblinded. At the end of the initial eight-week double-blind treatment period, patients taking Northera reported a mean improvement in OHQ composite score of 2.3 units (+/- 2.52) from their mean baseline OHQ composite score of 6.0, vs. patients randomized to placebo who reported a mean OHQ composite score of 3.5 (+/- 2.49), based upon a 2.1 unit improvement from baseline OHQ composite score of 5.6.

Potential Red flag #1: Another blinded study suggested Northera was no more active than placebo at 8 weeks and that the drug had no durability of response

Chelsea undertook some data mining and noticed a positive reduction in the number of falls compared with placebo so in 2011, subsequently decided to change the primary end point for Study 306B to assess the impact of Northera on reducing falls.

Fast-forward to 2012, having just received a CRL for Northera and in need of a second confirmatory study as quickly as possible, Chelsea proposed  a change in the primary endpoint of Study 306B from measuring a reduction in falls to the orthostatic hypotension symptom assessment (OHSA) item #1 score (dizziness, lightheadedness, feeling faint or "feeling like you might black out") at visit 5 (two weeks post-titration) and also proposed an increase in the number of patients targeted for enrollment from approximately 160 to 200. However, in its response, FDA advised Chelsea that, based on the theoretical potential for certain patients from Study 306B to have been unblinded in conjunction with the reporting of 306A data, the FDA could not be confident that this information did not influence an amendment of the statistical analytic plan, and therefore believes, as presently planned, Study 306B is unlikely to provide sufficient confirmatory evidence to support a Northera NDA. The FDA did not provide feedback or express any concerns regarding Chelsea's proposal to assess efficacy two-weeks post titration using OHSA item 1 (dizziness).

Potential Red flag #2: The primary end point for Study 306 has been changed twice and the sponsor is excluding the results from part A selectively because they were negative.  FDA is not usually one for glossing over “details” when it comes to end points or intent-to-treat vs. modified intent-to-treat outcomes. It is quite possible that FDA could look at both populations together as excluding Study 306A biases the results in favor of Northera

In December 2012, the company reported the results of study 306B as positive: The results showed that treatment with Northera provided clinically meaningful and statistically significant improvements compared to placebo in dizziness/lightheadedness at week 1 (1.0 unit change; p=0.018), the primary endpoint. Study results also demonstrated a statistically significant increase in standing systolic blood pressure (SBP) at week 1 (5.6 mmHg; p=0.032), a key secondary endpoint of the study. At time points beyond week 1, dizziness/lightheadedness and standing blood pressure predominantly favored Northera-treated patients, although the results were not statistically significant. 

Potential Red flag #3: Unless you were paying close attention, you may have missed that Chelsea just moved the primary end point a third time, from 2 weeks post-titration (visit 5), to week 1 assessment. They did this because the study failed to meet significance at the pre-specified primary end point at week 2.

Potential Red flag #4: This was an 8-week study that failed to show Northera durability of response beyond week 1.  In other words, not a single study including either study 306A or 306B have shown durability of response beyond week 1.

Following more regulatory meetings, FDA issued guidance to the sponsor in February 2013 suggesting that "data strongly demonstrating a short-term clinical benefit (e.g., improvement in symptoms or ability to function) of droxidopa in patients with nOH would be adequate for approval, with a possible requirement to verify durable clinical benefit post-approval." It further noted that any decision regarding the outcome of an FDA review, to be performed by the DCRP will be based on the strength of Study 306B and its ability to provide substantial evidence of effectiveness to support approval. Buoyed by this change of feedback, Chelsea refiled the response to the CRL in July and were issued a PDUFA goal for February 2014.

Potential Red flag #5: “strongly demonstrating short-term benefit” “strength of Study 306B”.  It is still a matter of clinical debate whether a change of 1 on the subjective OHQ composite “strongly” demonstrates benefit, or how robust Study 306B actually is given that it failed to meet the pre-specified end point at week 2 (or beyond).  The reviewers may also question why the primary end point for this trial changed 3 times

Potential Red flag #6: Adding to the safety database may uncover new safety signals. FDA is not known for its ability to selectively focus on positive clinical data at the exclusion of negative results.  Rest assured FDA reviewers will be picking through the entire Study 306 database parts A and B and will be pooling these with the existing data tables to look closely at safety.  It is quite possible that new safety signals could be identified given the larger patient pool

The company feels they can demonstrate durability of response in a post-market setting. Infact patient enrollment has just commenced in a 450-patient study of Northera in nOH –Study 401. Unfortunately, Chelsea is heading into the next panel with a failed 8-week study (which failed in part A and B) and not one single blinded study conducted to date has demonstrated durability of response beyond 1 week. Given that the study population and end point measures for the confirmatory trial are consistent with those studied to date, why should the outcome of any future study be different?  I highly doubt this trial will demonstrate any durable effect beyond 1 week.

Potential Red flag #7: If the sponsor felt more confident about 12-week durability perhaps they would have selected this as the primary end point for their large “post-marketing” confirmatory trial. You would think?  Instead they have opted for a week 1 assessment as the primary end point in order to boost the odds of a “positive” outcome.

Bulls are buoyed by the prospect of addressing a large unmet patient need, estimating >100,000 patients that may be eligible for therapy coupled with the economic benefits of premium orphan drug pricing.  Some analysts are projecting potential peak sales in the hundreds of millions.

Potential Red flag #8:  How could FDA craft and enforce a label for a drug with unproven symptomatic benefits beyond 1 week of therapy which would be used in a chronic disease setting?  The practicalities of providing physicians guidance as to how to identify patients who will benefit from therapy and how to discontinue treatment in nonresponders should not be underestimated, especially as this drug could be used widely off-label. It’s not as if physicians are routinely administering OHQ’s in a real world setting and it would be extremely unusual for FDA to issue a label with disclaimer along the lines of “symptomatic benefit beyond one week of therapy has not been demonstrated in clinical trials”…. The FDA that has to safeguard public health by ensuring patients are not inappropriately exposed to therapy when benefits do not outweigh safety risks, so the path forward for labeling may not be clear cut.

Potential Red flag #9: How can premium pricing be negotiated for a drug that confers 1 week of symptomatic benefit?   From a payer perspective, I have a hard time understanding how the sponsor will be able to negotiate a premium price for a drug that makes a subset of patients “feel” less dizzy by a score of 1 on a subjective questionnaire for one week. Conclusive pharmacoeconomic data are lacking, and the drug does not significantly reduce the rate of falls, hospitalizations for falls, nor does it reduce mortality.  Orphan status or not, based on the clinical data and modest subjective efficacy benefit it’s hard to make any compelling case that 8-12 weeks of therapy at a premium price are warranted.

While some traders may use the argument that FDA would not guide Chelsea to file if they were going to reject them again, in my opinion, a second review does not always guarantee approval. Traders may be prudent to wait for the FDA briefing documents to be released before betting on Northera’s outcome second time around.

Disclosure:  I have no position in the company mentioned in this article