MannKind Corporation (MNKD) Q2 2013 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by. Welcome to the MannKind Corporation Second Quarter 2013 Conference Call. [Operator Instructions] As a reminder, this call is being recorded today, August 14, 2013. Joining us today from MannKind are Chairman and CEO, Alfred Mann; President and COO, Hakan Edstrom; Chief Financial Officer, Matthew Pfeffer; and Senior Vice President, Clinical Sciences, Robert Baughman. I would now like to turn the call over to Matthew Pfeffer, Chief Financial Officer of MannKind Corporation. Please go ahead.

Good afternoon, everybody, and thank you for participating in today's call. I will discuss very briefly our financial results for the second quarter of 2013 as reported last Friday and will then turn the call over to Hakan. Before we proceed further, please note that comments made during this call will include forward-looking statements within the meaning of federal securities laws. It is possible that actual results could differ from these stated expectations. For factors which could cause actual results to differ from expectations, please refer to the reports filed by the company with the Securities and Exchange Commission under the Securities and Exchange Act of 1934. This conference call contains time-sensitive information that is accurate only as of the date of this live broadcast, August 14, 2013. We undertake no obligation to revise or update any statements to reflect events or circumstances after the date of this call. Turning to the financials. The net loss applicable to common stockholders for the second quarter of 2013 was $46.1 million or $0.16 per share compared to a net loss applicable to common stockholders of $33.6 million -- or excuse me, $36.6 million or $0.23 per share for the second quarter of 2012. Primary factors resulting in this change were an increase in research and development expense resulting primarily from noncash compensation expenses, partly offset by a decrease in clinical trial-related costs resulting from the completion of the Affinity studies. Also affecting this outcome was a decrease in general and administrative expenses, due primarily to the non-recurrence of a litigation accrual made in the prior year. Finally, there was a large decrease in other income in the current quarter due to the non-recurrence of a noncash entry recording a gain on a forward purchase contract relating to Al's forward commitment purchase common shares in 2012. Cash and cash equivalents were $28.5 million at June 30, 2013, not including funding under the Deerfield $160 million debt facility announced July 1. In addition, we still anticipate receiving approximately $64.6 million from the exercise of remaining short-dated warrants expiring in this October. So far this year, we've received proceeds of $29.7 million from the warrant exercises, including $4.1 million of February 2012 warrants and $25.6 million of October 2012 warrants. Finally, $125.4 million remains available for future borrowings under the loan agreement with The Mann Group. Needless to say, we feel comfortable with our financial position, our ability to not only repay our December 2013 convertible debt obligations, but also to fund our operations comfortably into 2014. With that, I know that the recent financial results are not what most of you called in to hear about, so I'll now turn the call over to Hakan. Hakan?

Thank you, Matt. It's my pleasure to proceed immediately to the essential part of today's call, which, of course, are the results of FDA mandated Phase III trials for AFREZZA in type 1 and type 2 diabetes patients. Earlier today, we released preliminary results from these studies and reported that both studies met their primary endpoints, that the trial -- and type 1 trial achieved its other main objective of bridging the Generation 2 inhaler to the pulmonary safety data collected with the MedTone inhaler. With those words of introduction, let me now introduce our Senior Vice President of Clinical Science, Robert Baughman, to present the trial data to you. Robert, please proceed.

Thank you, Hakan. Let me begin by saying that the data should still be considered as preliminary as the final clinical study reports have not yet been finalized. No data beyond that contained in this morning's releases will be presented here. Since it is our intention to publish the results of these trials, any further data disclosure may be considered to be in excess of what is currently acceptable as prepublication disclosure. The 171 trial in subjects with type 1 diabetes was conducted to evaluate whether the reduction in A1C following AFREZZA administered using the Gen2 inhaler in combination with the basal insulin is non-inferior to that obtained with subcutaneous insulin aspart in combination with the basal insulin. In addition, a third treatment arm, AFREZZA using the MedTone inhaler plus basal insulin, was incorporated to enable the main safety objective, which was to compare the changes in lung function from baseline to the final treatment visit at week 24 between the AFREZZA treatment groups, and thus provide a safety bridge between the Gen2 and MedTone devices. The study design and analysis plan were finalized in collaboration with the U.S. FDA. The study included a 4-week basal insulin optimization period and, after randomization, a 12-week prandial insulin titration phase followed by a 12-week stable dosing phase and then a 4-week safety follow-up period, during which all patients were transferred to aspart plus pure basal insulin. In total, 518 subjects were randomized into 1 of the 3 treatment groups. There would be 174 in the Gen2 group, 174 into the MedTone group and 170 were randomized to the insulin aspart group. In brief, the study met its primary efficacy and safety objectives. Now let me highlight several findings. The A1C reductions were minus 0.21% in the AFREZZA-Gen2 group and minus 0.4% in the insulin aspart group. The upper bound of the 95% confidence interval of the between-group difference was less than the prespecified 0.4% margin as established by the FDA. Accordingly, non-inferiority of AFREZZA to insulin aspart was achieved. In addition to this effect on A1C levels, we observed a significant difference in fasting blood glucose concentrations between the 2 treatment arms. At the end of the treatment period, fasting blood glucose levels decreased in the AFREZZA-Gen2 group, whereas they increased in the insulin aspart group, resulting in a between-group difference of 35.5 milligrams per deciliter. We also observed a significant weight advantage for AFREZZA in that patients in the AFREZZA-Gen2 group lost weight over the treatment period, whereas the insulin aspart group gained weight during the same period. As I mentioned, the main safety objective of the study was to establish a bridge between Gen2 and the MedTone pulmonary safety data by demonstrating that the mean change in FEV1, the forced expiratory volume in 1 second, from baseline to week 24 was comparable between the Gen2 and the MedTone treatment groups. We achieved this objective, observing that these 2 values differed at week 24 by an insignificant 0.01 liters, and that's from a baseline value in excess of 3.4 liters. We also examined the difference in the mean change in FEV1 between AFREZZA-Gen2 and insulin aspart treatment groups, observing a small and not clinically meaningful difference of 0.03 liters. During the follow-up period, FEV1 values increased in the group that had used the Gen2 inhaler, but were transitioned to insulin aspart, returning to comparator values by week 28. We saw a clear hypoglycemia advantage for the AFREZZA-Gen2 group. The hypoglycemia event rates per subject-month in this group were significantly lower than in the insulin aspart group for mild, moderate and total hypoglycemic events, as well as for hypoglycemia with a glucose less than 36-milligram percent. The event rate for severe hypoglycemia was numerically lower for the AFREZZA-Gen2 group than in insulin aspart, but this difference did not achieve statistical significance. As well, for a given A1C level achieved at the end of the 24-week treatment period, total mild, moderate and severe hypoglycemia event rates were all lower in the AFREZZA-Gen2 group than in the insulin aspart group. Now let me point out a correction that we made to the event rates reported for severe hypoglycemia in our initial press release, in which we reported the events per subject-months. To be clear, we observed 8.05 events per 100 subject-months in the AFREZZA arm compared to 14.45 events per 100 subject-months in the insulin aspart arm. As reported in the press release, this difference was not statistically significant. Not all of our secondary endpoints were achieved. The proportion of subjects achieving A1C target levels of less than or equal to 7% and less than or equal to 6.5% during the 24-week treatment period was less for the AFREZZA-Gen2 group than in the insulin aspart group. The most common drug-related adverse event was cough, with more AFREZZA-treated subjects experiencing cough than insulin aspart treated subjects. Consistent with other AFREZZA clinical studies, the cough was predominantly dry, single or intermittent and usually occurred within 10 minutes of inhalation. The incidence of cough was highest during the first week of the treatment period and diminished quickly thereafter. In summary, the 171 trial met the primary efficacy endpoint of non-inferior reduction in A1C when comparing AFREZZA-Gen2 to insulin aspart, as well as the primary safety endpoint of showing a comparable mean change in FEV1 over the treatment period between the Gen2 and the MedTone treatment groups. Additionally, AFREZZA-Gen2 group had significantly less hypoglycemia, lower fasting blood glucose levels and a significant weight advantage compared to the insulin aspart group. Treatment with AFREZZA was generally well tolerated over the duration of the study. Turning now to the 175 trial. This trial was conducted to demonstrate that prandial AFREZZA Inhalation Powder is superior to inhaled placebo powder in reducing A1C levels when added to an oral anti-diabetic regimen in subjects with type 2 diabetes who are not adequately controlled on optimal or maximally tolerated doses of metformin or metformin plus one or more oral anti-diabetic agents. The study design and analysis plan were finalized in collaboration with the FDA. Following screening, subjects underwent diabetes education, followed by a 6-week oral therapy optimization period, then to randomization to either AFREZZA or placebo powder in addition to the subjects' current oral therapy. Each subject then underwent a 12-week prandial insulin titration phase, 12 weeks of stable dosing and a 4-week safety follow-up period, during which subjects returned to oral therapy alone. 353 subjects in total were randomized into 1 of the 2 treatment groups, 177 into the AFREZZA group and 176 into the comparator oral therapy group. Over the 24-week treatment period, mean A1C levels decreased by 0.82% in the AFREZZA group compared to 0.42% in the oral therapy group. The between-group difference in the mean change in A1C was statistically significant at the P less than 0.0001 level, thereby establishing superiority of AFREZZA when added to their existing oral agents over the comparator oral therapy treatment alone. Some of you may recall that in designing this study, we made certain assumptions about the statistical power necessary to detect the between-group A1C difference of 0.5% with a one-sided alpha less than 0.025. We used that statistical model to determine the number of subjects to be enrolled into each treatment group in order to perform the superiority analysis. The actual between-group A1C difference was 0.4%, but with a P value of less than 0.0001, there can be no doubt that the study was adequately powered to make an unequivocal superiority conclusion in keeping with the statistical analysis plan that was agreed to by the FDA. This study also met another efficacy endpoint. The AFREZZA group had a significantly greater percentage of patients reaching specified A1C targets of less than 7% and less than 6.5% than did the comparator oral therapy group. After 24 weeks of treatment, 37.7% of patients in the AFREZZA group achieved A1C levels below 7% compared to only 19% of patients in the comparator oral therapy group. Similarly, 15.9% of patients in the AFREZZA group achieved A1C levels below 6.5% compared to only 4.2% of the patients receiving only oral therapy. Other secondary efficacy endpoints were also achieved. During the treatment period, postprandial glucose excursions were reduced in the AFREZZA group compared to those in the comparator oral therapy group. At the follow-up visit, after the AFREZZA subjects had reverted back to oral therapy alone, the significant reduction in postprandial glucose excursions in the AFREZZA group was lost and the post-meal glucose excursions were indistinguishable between the 2 treatment groups. Additionally, over the treatment period, mean fasting blood glucose levels decreased more in this group than in the comparator oral therapy group, although that difference was not statistically significant. As we expected, significantly more patients experienced mild and moderate hypoglycemia in the AFREZZA group than in the comparator oral therapy group. However, there was no significant difference in the incidence of severe hypoglycemia, which was reported in 5.1% of the AFREZZA patients compared to 1.7% of the oral therapy patients. We did not see a weight advantage in the study. Patients in the AFREZZA group gained weight over the treatment period compared to a weight loss by patients in the comparator oral therapy. We also looked at pulmonary function measures in this study. In the AFREZZA group, the mean change in FEV1 over the treatment period declined by 0.09 liters more than the FEV1 decline in the comparator oral therapy group. This difference was not considered to be clinically significant. As in our other trials, FEV1 values in the AFREZZA group increased during the follow-up period returning to comparator values by week 28. In summary, the 175 trial met the primary efficacy endpoint of superiority in A1C reduction when comparing AFREZZA plus oral therapy to the oral therapy alone. Additionally, the AFREZZA group had a significantly greater percentage of patients reaching specified A1C targets of less than 7% and less than 6.5% and a greater reduction in postprandial glucose at week 24 than did the comparator oral therapy. Treatment with AFREZZA was generally well tolerated over the entire duration of the study. And with that, I'll return it to Hakan.

Okay, this is Matt. Before Hakan starts to take over the call again, I want to apologize to a large number of people on this call. We've been getting some reports that there was a bit of a snafu in connecting some segment of the lines of this call and that a number of people didn't come in until 10 or 12 minutes in. I apologize for that. I'm not sure exactly what caused it. But given we weren't certain exactly when everybody came in or who was affected, we elected not to redo the call. But mostly, you missed my discussion of the financial results, which I'm sure most of you aren't that concerned about. And in any event, the entire replay of the call will be available within about 90 minutes after we finish this call. To the extent you have something in particular you want to hear or you want to go back and listen to the beginning, we invite you to do that. And then, of course, you'll have an opportunity to ask questions when we all finish our prepared remarks. With that, I'll turn it over to Hakan.

Thank you. Well, as you can tell from the data, we are very pleased with the outcome of these trials. And we will now expeditiously proceed to finalize the amended new drug application for a planned submission early in the fourth quarter. But let me now briefly address the status of partnership efforts for AFREZZA. As we stated before, we have been facilitating the initial diligence efforts of various parties over the past several months, while other potential parties have indicated that they would initiate or resume diligence activities upon the availability of Phase III data. Well, now that the data is available, we are kicking off a formal partnering process that is intended to allow multiple potential partners to evaluate the AFREZZA opportunities simultaneously. We have engaged with Greenhill & Co. to run this process and to advise us with respect to partnering proposals that are put forward by potential partners. Our goal is to secure a partner that has the right capabilities, the right approach to diabetes therapy to make AFREZZA the commercial success we know that it can be. And with those comments, let me turn the call over to Al. Al?

Thank you, Hakan, and good afternoon, ladies and gentlemen. We, at MannKind, are pleased to share with you the significant results of the Affinity study. As you know, the protocols of these trials were carefully crafted after discussions with the FDA and were specifically designed to address the questions raised by the FDA in their complete response letter. We are very pleased that we have met the primary endpoints of both these studies and have additionally shown important advantages in the secondary objectives. We are planning several Phase IIIb and IV studies in order to explore ways in which to further differentiate AFREZZA, but that is a topic for another time. We appreciate your patience and understanding for the postponement of our quarterly conference call to enable us to include a discussion of these Affinity trials. Before I discuss the trials, I want to acknowledge and praise the MannKind team that conducted these sophisticated studies, completed them on time, locked the immense amount of data very quickly and analyzed the key information in order to enable us to share preliminary results with you today. I congratulate our MannKind employees and our CROs, who successfully completed this work on virtually the exact schedule laid out almost a year ago when we neared the end of enrollment. Now let me add some comments and perspective on the Affinity results. There were 2 primary objectives for the Affinity 1 trial conducted in type 1 diabetes. The primary efficacy point was to compare the A1C lowering effect of AFREZZA to that of an injected rapid-acting analog, in this case, NovoLog, in both cases, along with Lantus as the basal insulin. We met the endpoint of demonstrating non-inferiority to have confirmed that AFREZZA can reduce A1C levels at least as well as the best-selling prandial insulin. But that is where the similarities between AFREZZA and NovoLog end. In Study 171, we saw clear advantages among these differences in the effect of the prandial insulin on fasting blood glucose levels. We saw a significant decrease in the AFREZZA arm compared to an increase in the NovoLog arm. Moreover, this improvement in fasting glucose levels in the AFREZZA group disappeared completely during the follow-up period when these patients switched back to NovoLog, indicating that this lowering effect was attributable to AFREZZA. More importantly, these improvements in glucose measurements were associated with a significantly lower risk of hypoglycemia. As you heard from Bob, we saw a lower risk of mild, moderate and severe hypos, as well as hypos with the blood glucose levels observed to be less than 36 milligrams per deciliter. When we compared event rates by week as patients progressed through the titration algorithm, we saw that the AFREZZA group consistently experienced fewer hypos than did the NovoLog group. We also looked at the event rates for hypos on the basis of the ending A1C level at week 24, and again, the AFREZZA group reported fewer hypoglycemic events than did the comparator group. Another area in which AFREZZA distinguished itself from NovoLog was weight. NovoLog patients gained weight during the treatment period, while AFREZZA patients remained at baseline levels or even lost weight. So study A1C is really a safety story that tells type 1 patients and presumably insulin-dependent type 2 patients that AFREZZA can reduce their A1C levels at least as well as NovoLog, but with the additional benefits of lower fasting blood glucose levels, a lower risk of hypoglycemia and no weight gain. That is a compelling story. We've been telling it for a while, but that story becomes even more compelling now that we have demonstrated these advantages with the Dreamboat inhaler. Now let me turn to the second objective of the Affinity 1 trial, which was the comparative performance of AFREZZA with the Dreamboat device to that with the MedTone inhaler. This comparison was intended to provide a bridge to the pulmonary safety data from our earlier studies. In Study 171, we observed that the pulmonary function results with the 2 devices were virtually identical, so we regard the bridge as having been established. Just to put that in perspective, with this 6-month trial, we are now able to support the safety of the Dreamboat inhaler by drawing all of the safety data from our earlier development program using the MedTone device, including not only a 2-year pulmonary safety study involving 2,035 subjects, but also the thousands of chest CT scans and tens of thousands of pulmonary function tests that were administered in other Phase II and Phase III clinical studies. The consistent finding across all these studies is that the use of AFREZZA is associated with a clinically insignificant decrease in lung function that appears at the onset of therapy, it does not progress during therapy and it resolves fully upon cessation of the therapy. All in all, it is clear to me that Affinity 1 has validated AFREZZA as a better prandial insulin for use in basal/bolus therapy for type 1 and surely also for insulin-dependent type 2 patients. But if you've been following my commentary over the past few years, you know that I'm not really surprised by this outcome. Now let me talk about Affinity 2. In many ways, this was a very difficult study to successfully conduct, so I'm extremely pleased that we got such a clear and statistically significant result in insulin-naive type 2 patients, who, unlike insulin-dependent patients, can more significantly improve their glycemic control by being more attentive into diet and exercise and lifestyle factors. Our protocol went to considerable lengths to control these factors by providing extensive diabetes education to all the patients prior to the run-in phase, including topics such as nutritional counseling; incorporating physical activity into lifestyle; using medications effectively, monitoring blood glucose and using measurement to improve glycemic control; preventing complications through risk reduction behavior; and goal setting for daily living. During this period, we also provided glucose meters to all the subjects and instructed them on the proper technique for obtaining self-monitored blood glucose measurement. As might be expected, the patients in Affinity 2 responded to this education and monitoring with improved A1C level. Even during the running period before a single patient was randomized, A1C levels declined in all patients. Nevertheless, we are able to demonstrate that adding AFREZZA to an oral treatment regimen causes a statistically significant further decrease in A1C level. I won't speculate how this signing would translate to a real world setting in type 2 patients that are not nearly as attentive to diet, exercise and glucose monitoring, but in the controlled setting of this clinical trial, AFREZZA passed a difficult test with flying colors. Not only did AFREZZA meet the standard of superiority in reducing A1C levels, but we saw a clear difference in a number of patients that reached the ADA, the American Diabetes Association, and the American Association of Clinical Endocrinologists' goals for A1C levels. Twice as many AFREZZA patients reached the ADA goal of 7% or less than did the oral group, and almost 4x as many AFREZZA patients reached the AACE goal of 6.5% or less than did patients on the comparator group. Still another clear difference between the 2 treatment arms was in the 7-point blood glucose profile collected over the course of the study. These profiles show glucose measurements at 7 different time points throughout the day: before breakfast, 90 minutes after breakfast, before lunch, 90 minutes after lunch, before dinner, 90 minutes after dinner and at bedtime. At bench line, these profiles had the standard sawtooth-like appearance typically seen in diabetic patients whose glucose levels swing wildly up and after each meal and then down again as their medication attempts to control postprandial excursion. What we saw in the AFREZZA group was that the 7-point profile became much flatter and much lower overall as these patients gained control over their postprandial glucose excursions. These significant improvements with AFREZZA were observed early in the treatment phase and became more pronounced the longer the patients remained on AFREZZA. Tellingly, at the follow-up visit after all patients had returned to a regimen of oral therapy alone, the 7-point profiles for all the patients once again displayed the large peaks and valleys in their glucose levels throughout the day. As we expected, we saw more mild and moderate hypoglycemic events in the AFREZZA group in the Affinity 2 trial. After all, this was a trial that added a potent glucose lowering agent insulin to a background regimen of oral medications, some of which can also cause hypos. However, it was reassuring that severe hypos did not occur in a significantly greater number of patients or at a significantly higher rate than in the comparator group. Moreover, no patient discontinued the trial because of hypoglycemic events. Put another way, it would appear that the benefits we saw in this trial of early-stage type 2 in terms of glycemic control were not outweighed by an increased risk of hypoglycemia. We continue to see the ultrafast kinetics and dynamics of AFREZZA as providing far more physiologic prandial insulin therapy that is so conveniently and easily delivered. Patients have widely expressed enthusiasm for the non-injection modality. In all of our clinical trials involving a total of more than 6,700 subjects, about 4,000 of which were administered with AFREZZA, we've seen nothing to raise concern for safety with AFREZZA. In summary, the Affinity trials have demonstrated significant advantages, suggesting the opportunity for improved glucose control throughout the entire spectrum of diabetes. I am excited and proud of what our team has achieved. We attribute many of the benefits AFREZZA has demonstrated in the Affinity trial to faster onset and a shorter duration of action, which portends to create a new class of ultrarapid-acting insulin, and AFREZZA will be the first product in this new category. I continue to believe that AFREZZA will likely become a valuable anti-glycemic therapy throughout the vast early-stage type 2 market, as well as becoming a premier prandial insulin in basal/bolus therapy for type 1 and late type 2 diabetes. And now, let's open the call to questions. Operator?

[Operator Instructions] Our first question comes from Simos Simeonidis from Cowen and Company.

My first question has to do with the changes in weight that we saw in the trials with AFREZZA. In the type 1 trial, we saw a delta, almost 3 pounds weight loss difference. In the type 2, we saw a 3.5-pound difference of weight gain between the arms. Do you have an explanation why that may be?

Yes. Let me suggest that we did a great deal of diabetes education in the 171 trial, in the type 1s, we believe, is because we are managing their glucose more effectively. So the weight loss there was actually related to the drug product. In 175, we believe that what we see there is the impact of heavy diabetes education. As you know, with these type 2s, early in their disease, many of them had not used glucose meters before, they went to frequent visits to the clinical site. So we believe that, that's the outcome. It was probably really due to lifestyle modification.

So you're saying that -- I'm sorry, I don't follow. So they put on weight because of lifestyle modification?

No. What I'm saying is that the difference is what is significant. The amount of weight added in the 175 study was very small, just over a pound. Where we saw the difference was there was a weight loss in those that are on the controlled treatment. And we believe that was due to the fact that there were not in as good a control and that we had also participated in increasing their knowledge of their lifestyle as diabetics.

Simos, keep in mind that the cause of weight gain in insulin therapy is because people are eating snacks to avoid hypoglycemia and the liver is supporting out glucose as well to avoid the hypoglycemic issues. We don't have those issues in -- with AFREZZA, so that our view is -- my view is that AFREZZA is really weight neutral.

Okay. I was wondering if you have the numbers for Affinity 1 in terms of the percentage of patients that were achieving A1Cs below 7% or 6.5%?

No, I do not have those number presently. What I can tell you is that in the 171 trial, those reaching 7% and 6.5% did favor the aspart group. But I do not have the numbers with me, I'm sorry.

Okay. And, I guess, the last question is more of a general one. Has the FDA told you that if you meet the primary endpoints of these trials, you would get approval regardless of what else is in the package in terms of the other trials or what happens with the secondary endpoint, is that, that cut and dry? Or do you not have that level of clarity from the agency?

Yes, what I can share with you is after 30 years in the business, I've learned I cannot speak for the FDA. What I can tell you is that they asked us -- for them to consider approval for the product, we had to address questions that came out of the complete response letters. We have addressed their key issues in those complete response letters. So we believe that has opened the door for an effective review by the agency.

Our next question comes from Ian Somaiya from Piper Jaffray.

It's Matthew on for Ian. So just a couple, and congratulations on successfully meeting the primary endpoints. First, in the type 1 trial, can you provide us -- the press release gives A1C reductions for the second-generation device, as well as for the NovoLog group. But can you give us a comparison of how either the first-generation and second-generation devices performed relative to each other, or tell us how the first-generation device performed relative to NovoLog?

What I can do, Ian, is just share with you that the MedTone device performed comparable to the TI Gen2 data, so that we do not see an appreciable difference between those 2.

Okay. And, I guess, turning to the cough, you provided some color on it. And I was just wondering if you could maybe expand a little bit on the -- what was sort of the reason as it relates to cough for dropping out.

Yes, if you remember, the dropping out due to cough was an insignificant component of the 5% who left the study in those receiving aspart, especially in the type 2s, whereas only 1.9% of the control dropped. What I can tell you is that if you go back and look, I mean, cough essentially was a nonevent in the 171 study in type 1s because the comparator -- I mean, I think they had 4 reported cases total. So we saw a significant incidence of cough compared to insulin aspart in type 1. The cough incidence in type 2, where we're comparing AFREZZA-Gen2 to the placebo, was essentially comparable between the 2.

Our next question comes from Cory Kasimov from JPMorgan. Matthew J. Lowe - JP Morgan Chase & Co, Research Division: It's actually Matt Lowe in for Cory today. First question I have is, did the FDA explicitly sign off on all aspects of these studies? I guess, I'm thinking specifically about the size of Affinity 1 and with the 3 arms having 170 patients in each arm. Was that something the FDA explicitly signed off on? And then secondly, regarding the dropout rate, I guess, how did you define a dropout in the studies? And could the FDA potentially use a different method for defining a dropout?

Okay, the first question was regarding the size of the studies, and all of the projections for the number of patients required in each treatment arm was vetted with the agency. They did review that as they reviewed the protocols as well. In the second study, the dropouts are those who withdraw from the trial due to either adverse events or other reasons. Again, just withdrawing your consent is sufficient, so in some cases, we do not know all of the reasons for the drop-outs. What I can tell you is that the drop-outs that occurred, occurred early in the trial, especially in the AFREZZA and the MedTone -- I'm sorry, AFREZZA-Gen2 and the AFREZZA-MedTone groups. And it was essentially related to the fact that it was a new route of delivery for them and it was also becoming -- the trials were becoming a burden to their lifestyle.

Our next question comes from Jason Butler from JMP Securities.

First, a couple on the type 1 study. Can you talk about the use of basal insulin in the study, whether or not the dosing of the basal insulin was balanced between the 2 -- the 3 arms, I guess?

Yes, what I can tell you is that greater than 70% of the patients were taking insulin glargine and the remaining patients were divided rather equally between NPH insulin and detemir. At the end of the trial, we evaluated the total basal insulin dose, and it was slightly lower in the insulin aspart group and it was comparable between the 2 AFREZZA treatment groups.

Okay. And then can you talk about the number of prandial insulin units? Was there anything there to suggest that -- or to support that both the control and experimental insulins were being titrated to the same level?

Well, what I can tell you is that over the course of the titration period, that was the first 12 weeks post-randomization, that both the MedTone and Gen2 treatment groups had a significant increase in their dose of AFREZZA. So they were being effectively titrated, and that also being mandated by the titration management committee, which reviewed glucose values over various periods throughout the titration.

Okay, great. Just on MedTone and the change in FEV1. Can you comment, I'm sorry if I missed this, whether the change in FEV1 in this study was similar to what you'd seen with the MedTone device in previous studies?

Yes, that's correct. It was actually quite comparable to all of our previous type 1 studies.

Okay, great. Last question, you mentioned that the number of patients reaching goal was somewhat more on the control on NovoLog than it was on AFREZZA. Was there any suggestion of the 3 statistical significance in any groups?

Yes, yes, it was statistically different. It was a greater number of subjects reaching those levels in the aspart group than in the TI Gen2 group.

Okay, great. And I'm sorry, just one last question, I guess a broad question maybe for Al. You achieved your goal here of lowering fasting glucose, which I know that's something you've been trying to do for a while in these kinds of studies. Are you surprised by the magnitude of the A1C reduction in the type 1 study given that you did get much lower fasting levels?

Well, I'd rather Bob answer that question. But I'm not really -- I wouldn't say I'm disappointed.

Can you be specific, which trial are you asking about?

The type 1 study.

Yes. No, that response in the type 1 trial was very, very consistent with what's been published for the last 5, 6 years as people have gone through the DPP IV inhibitors and other oral agents. So the reduction in A1C that we saw in the aspart group was -- I mean, we did a very good job of titrating aspart, as well as we did a good job of titrating the -- in TI Gen2. So we think those numbers fall right in line with very effective titration of both products. So we're quite comfortable with the results.

Our next question comes from Steve Byrne from Bank of America.

I just want to follow up a little bit on Jason's line of questioning here. So if the fasting glucose levels were lower in the AFREZZA arm in the type 1 study, and I think you had said earlier that the prandial insulin use was actually a little higher in the AFREZZA arm, wouldn't both of those suggest that you could achieve a lower A1C level, and yet the A1C level was higher in the AFREZZA arm? Is there -- is the 1 thing that's different here is that the post -- in the post-prandial time period, you actually had more hypos in the NovoLog arm, and thus, you had effectively lower blood glucose during that time period? Is that what drove the A1C levels lower?

Yes, let me see if I can bring us back here, you covered everything including hypos there. What we did see was we did see at 24 weeks a significant reduction in the fasting plasma glucoses in those subjects receiving TI Gen2. There's no question, it was quite remarkable. But greater than 35-milligram percent difference between the 2 groups. Over the course of the day, when we looked at the profiles, the 7-point glucose profiles that Al spelled out for you, exactly pre- and post-meal, all the way to bedtime, what we saw was that later in the day, there were lower glucose values in the insulin aspart group. We had pushed that dose very well. They were slightly higher in the evening, however, the next morning, it would be lower. So the fasting plasma glucoses were actually lower in the morning. But throughout the day, the TI Gen2 group increased slightly, whereas the aspart group did not.

Okay. And then a question about the type 2 study. The fact that you had some more -- you had more hypos in the AFREZZA arm, I was wondering if those were in patients that actually had the lowest A1C levels. Is it possible that, for some of those patients, they might have been overly medicated?

Well, I mean, again, the hypo incidence was greater, that is the number of subjects who reported a hypoglycemic event was greater. But when we really looked at the ones that concerned us were the severe hypos that wasn't at all correlated to their A1C level. So I don't have in front of me the correlation of all hypos to A1C, but I can tell you that the ones that concerned us, the severe hypos, was not correlated to A1C.

Our next question comes from Michael Higgins [ph] from Highland Research [ph].

Just a follow-up on Steve's line of questioning and the hypoglycemic events. I understand you're early on here, but are you able to detect any difference in the mild, moderate or in the severe according to the type of oral anti-diabetic drug that the patient was on?

That's a very good question, and I do not have that data. We have not -- it's data that we will have to generate, but we do not have that here. And as I'd mentioned earlier, I mean, there are some data that you're asking for that I won't disclose because I don't want to exceed what I think would be too great a prepublication disclosure. So what I can tell you is that's data that we will be able to have, but I do not have it here today.

Okay. And so just to reiterate what Steve said -- or what you had said when Steve asked was, in the mild and moderate group, doesn't seem to be any correlation on HbA1c level with hypoglycemic event, but it isn't severe. How about baseline characteristics, anything you can pull away from that piece of information?

No, nothing that stands out.

Okay. In terms of the dozen or so cardiovascular events, or actually less than that, just the serious cardiovascular events, can you give us any indication as to what those were, and were they related to the study drug?

Yes. We had 5 cardiovascular events, of which none were attributable to the study drug. And they were balanced between the 2 groups, 2 and 3, 2 in the TI Gen2 group and 3 in the comparator.

Our next question comes from Graig Suvannavejh from MLV & Co. Graig C. Suvannavejh - MLV & Co LLC, Research Division: I just had a few questions, if I could. My first, I just would love if you could just confirm for us that the data that you did generate from these 2 trials you feel is completely adequate to address the complete response letters that were issued -- or the latest complete response letter that was issued by FDA, specifically as they relate to -- anything that related to the device?

Related to the device? Graig C. Suvannavejh - MLV & Co LLC, Research Division: Yes.

Yes. As you would imagine, we've had a number of correspondences back-and-forth to the agency to ensure that we understood where they wanted information from the complete response letters. And we are comfortable that the information that we've generated on the device will satisfy the reviewing division. Graig C. Suvannavejh - MLV & Co LLC, Research Division: Okay, great. My second question just has to do with as you look to now do your kickoff through your partnering discussions and that process there. I know it's early days, but any sense of, now that you have the data in hand, how that will, one way or another, dictate of sorts how you go about partnering discussions? And I guess, generally speaking, just overall, if you could just remind us what you're ideally looking for in a partner.

Well, as I said, we have formed a relationship with Greenhill & Co. in regards to advising us on partnering discussions. And we are putting basically together a formal process, where we are going back to potential partners that have been in contact with us before that either proceeded to do the due diligence or some of them decided to wait or resume due diligence activities upon the availability of Phase III data. So that, basically, starts as of today, and we will go back and resume those activities. Following that, certainly, what we've said before is that our preference is still a global partner that will help us to launch the product and position the product on a global basis. We may look to participate in the, say, market on the product in the U.S. market, where we could have some type of a co-promotion, at least the right to co-promotion for the price in the U.S. And again, together with Greenhill & Co., and we will then, say, look at who we believe will be the right partner in terms of experience, in terms of their investment in the opportunity. And that will really determine the process going forward and the timing for any type of agreement. Graig C. Suvannavejh - MLV & Co LLC, Research Division: I just had another, just follow-up question real fast. On the call, I believe, you mentioned that you were going to embark back on the clinical side on perhaps some Phase IIIb or some other studies. Can you just give us some color on what kinds of studies you're looking to do on a go-forward basis?

We're still just planning those, and it's too early to give you any details. But what -- these trials were -- that we just reported, the Affinity trials, were constructed very carefully to meet the requirements of the FDA to set forth in the CRL and in accordance with our discussions with the agency, whereas we want to do some trials that will really further differentiate us in a significant way in glucose control in both primary and secondary endpoints. So we can do that by -- again, these trials were directed to the approval process. And now, we want to target some marketing questions.

And quite importantly also, this is one of the topics that we would discuss with potential partners and even hopefully have the potential partners pay for. Graig C. Suvannavejh - MLV & Co LLC, Research Division: Okay. And if I could just sneak one more question, and this is really more for Matt. I know no one really cares about the numbers per se. Just piggy-backing on the future trials, how should we be thinking about R&D spend on a go-forward basis now that Affinity 1 and 2 are pretty much wrapped up?

Well, obviously, you've already seen there's been a slight decrease, but I'm not going to project a lot because we have a lot of things coming in to take the place of the Affinity trials. So I think we need to be pretty conservative there. Remember, we are planning to ramp up for commercialization, that's going to require some expansion in our Denver facility, and you've seen some of that trickling in, in the form of capital spending already, and you'll probably see more. So I -- you know I tend to be somewhat conservative, but I'd say, all in all, it may be shifting a little bit from R to D, but -- or just preparation for commercial launch. But on balance, it's not going to change all that much.

Our next question comes from Anthony Esposito from Imperial Capital.

Did this release of the data trigger the second tranche of the financing agreement that was signed in July with Deerfield?

I'm glad you asked that question. It's a fairly formal process with our friends at Deerfield. We're embarking upon that now, but ultimately, they will have to make that decision. But our belief is that yes, it will.

Our last question comes from Keith Markey from Griffin Securities.

Two questions. I was just wondering if you could tell us a little bit about the range of HbA1c levels in the patients who were enrolled in the trials? And were there any real differences at all between the 2 groups or 3 groups depending upon the trial that you were talking about?

Which trial are you asking about?

Either or both. Whichever is easiest. Why don't we focus on type 2, the Affinity 2.

So what we had with the Affinity 2 trial was that when the patients came in to the study, they averaged about 8.6% or 8.7% A1C. After the run-in phase, remember, they go through a 6-week optimization phase, we saw a decline occurring already in the patients before they were even randomized. At the point of randomization, they had comparable A1C levels.

Okay, great. And then related to the phase -- the type 2 trial, sometime back, you did a very well-controlled, very small trial in which you titrated patients, not based upon the amount of insulin given, but based upon the amount of food they were consuming, and came up with really very, what seemed to be incredibly, great success in -- of weighting hypoglycemic events in -- regardless of how much they ate or -- and I was just wondering, is there anything in particular that you feel might be contributing to the hypo events that you did see in the type 2 that are perhaps beyond -- are beyond the control that you could provide in the clinical setting that you used?

Yes. What I can tell you is that we have a host of data and we'll do a considerable post-talk evaluation. What we reported here were essentially our prespecified endpoints. The question you're asking is something that we'll be able to address once we get into the post-talk analysis, looking at meals, frequency of meals, how many were skipped, all of that is in our database. But as you can imagine, with more than 55,000 meals per group, we have a lot of data to look at. So I do not have that at this time.

We have no further questions at this time. I will now turn it back to Mr. Mann for closing remarks.

Thank you, all, for joining us for this conference call today. And we look forward to updating you again at our next quarterly call that's somewhere in October. Thank you.

Thank you, ladies and gentlemen. This concludes today's conference. The playback will be available in 90 minutes at (888) 843-7419, using passcode 34087263 followed by the pound sign. Thank you for participating. You may now disconnect.