AstraZeneca 2Q14 – “Excellent” US Farxiga launch progress; Onglyza falls 1%; Bydureon momentum builds – July 31, 2014

AstraZeneca (AZ) provided its 2Q14 financial update this morning in a call led by CEO Mr. Pascal Soriot. The quarter 2Q14 was the first full quarter following AZ’s acquisition of BMS’ diabetes portfolio in February. Due to complexities in the way BMS and AZ presented diabetes portfolio revenue in 1Q14, meaningful sequential comparisons are not available for most of the 2Q14 results. Below, we include our top ten highlights from the call, followed by selections from Q&A.

1. According to management, the US Farxiga (dapagliflozin) launch continues to be the most successful oral diabetes drug launch since Januvia; Farxiga’s US new-to-brand prescription share among SGLT-2 inhibitors is at 40%.

2. Forxiga is performing well in its German re-launch following successful re-arbitration on pricing with government authorities (one chapter in the long G-BA/IQWiG saga).

3. Global Onglyza (saxagliptin) sales fell 1% to $238 million, driven by a 14% decline in the US.

4. Bydureon (exenatide once weekly) sales grew 70% from 2Q13 to $112 million in 2Q14; Byetta (exenatide twice daily) sales fell 15% in 2Q14 to $88 million.

5. The Bydureon dual-chambered pen should be launched in the US in 2H14, and it recently received a positive CHMP opinion in Europe (earlier than the previously expected 4Q14) and has been submitted in Japan.

6. AZ does not believe that GSK’s aggressive pricing strategy for Tanzeum (albiglutide) will force prices down across the class.  

7. During Q&A, CEO Mr. Pascal Soriot noted that AZ sees great potential in the parallel use of GLP-1 agonists and SGLT-2 inhibitors, and suggested that AZ hopes to explore this area further.

8. The saxaglipin/dapagliflozin (“saxa/dapa”) fixed-dose combination remains on track for regulatory submission in the US and EU by the end of the year.

9. During BMS’ 2Q14 update we gained insight into the logistics and agreements that will ensure that the handoff of BMS’ diabetes portfolio will progress smoothly.

10. Roxadustat, AZ’s therapy for chronic kidney disease (and associated anemia), began two phase 3 pivotal trials.

Top Ten Highlights

1. Consistent with early observations from 1Q14, management characterized the US launch of the SGLT-2 inhibitor Farxiga (dapagliflozin) as the most successful oral diabetes drug launch since Merck’s DPP-4 inhibitor Januvia (sitagliptin) – although the presentation did not include specific sales figures (as we used to receive from BMS), management shared that Farxiga has achieved a 40% new-to-brand prescriptions share (NBRx) in the US among SGLT-2 inhibitors. In an IMS Health chart included in the presentation slides, Farxiga’s new prescription (NRx) volume in the US post-launch tracked slightly above the penetration curve for J&J’s Invokana following its launch. Another chart showing the monthly NRx volume of the entire SGLT-2 inhibitor class had a clear positive inflection point corresponding with the date of Farxiga’s launch in late February.

• A bar chart comparing diabetes portfolio sales between 2Q13 and 2Q14 allows us to speculatively estimate what Farxiga sales might currently be. The two bars contain sections for Onglyza, Byetta, Bydureon, and “Other.” The Other category was $63 million in 2Q14, and appeared to be roughly $10 million in 2Q13. The Other category contains Symlin (pramlintide) and Forxiga/Farxiga. Assuming that quarterly Symlin sales have held roughly constant from 2012 ($25-$30 million) and that European Forxiga sales in 2Q14 were $10-$15 million (see below), this means that 2Q14 Farxiga sales were roughly $20-$30 million.
• Other companies with SGLT-2 inhibitors include: J&J (Invokana, approved in the US and EU), Lilly/BI’s Jardiance (empagliflozin, approved in the EU in May and under review in the US), Pfizer/Merck (ertugliflozin, phase 3), Astellas/Kotobuki (ipragliflozin, approved in Japan), Taisho (luseogliflozin, approved in Japan), Islet Pharma/BHV Pharma/Kissei’s remogliflozin etabonate (phase 2 –recently reported results), Theracos (EGT0001442, phase 2), Lexicon (SGLT-1/SGLT-2 dual inhibitor LX4211, phase 3-ready upon partnering), and Novartis (SGLT-1/SGLT-2 dual inhibitor LIK066, phase 2).
• AZ announced earlier this year that it will begin a phase 3 trial investigating Farxiga in type 1 diabetes patients later this year. J&J recently registered a phase 2 trial testing Invokana (canagliflozin) in type 1 diabetes patients, and Lexicon is likely to move into phase 3 soon in type 1 diabetes for its SGLT-1/2 dual inhibitor LX4211.

2. Consistent with 1Q14, AZ did not break out dapagliflozin sales in Europe, where the product is marketed under the brand name Forxiga. From back when BMS reported Forxiga revenues, we know that sales in Europe climbed fairly linearly from $3 million in 1Q13 to $8 million in 4Q13. This was fairly modest growth for the first SGLT-2 inhibitor approved in Europe, but the reimbursement environment in Europe has become a significant obstacle throughout diabetes care.

• As a reminder, in December, AZ announced its decision to withdraw Forxiga from the German market after failing to reach an agreement on pricing with the government. However, during the company’s 1Q14 update in April, management enthusiastically shared that the drug was being re-launched in Germany following successful re-arbitration on pricing. Management shared during Q&A that the re-launch has been going well in Germany – this momentum stands to continue, as German authorities recently handed down a negative cost-effectiveness decision on J&J’s Invokana.    

3. Compared to combined BMS/AZ sales in 2Q13, AZ’s global 2Q14 sales of the DPP-4 inhibitor franchise Onglyza (saxagliptin) fell 1% year-over-year (YOY) to $238 million. The franchise’s performance was strikingly different inside vs. outside the US: domestic sales fell 14% to $144 million while ex-US sales rose 30% to $94 million. There were a number of contributors to the poor domestic performance: the press release noted that total prescriptions for Onglyza in the US fell by 3% relative to 2Q13 and that average realized selling prices were lower in 2Q14. Interestingly, the reduction in Onglyza’s share of the DPP-4 inhibitor market (down 0.4% points, to 15.3%) was smaller than the reduction in script volume, indicating that the overall DPP-4 inhibitor class shrunk slightly in 2Q14.

• For comparison, in 2Q14, sales of Merck’s Januvia (sitagliptin) rose 2% to $1.6 billion, while Novartis’ Galvus (vildagliptin; marketed ex-US only) grew 13% to $328 million, and Lilly’s Tradjenta (linagliptin) grew 65% to $90 million.
• Although the slowdown in Onglyza’s growth was disappointing for the company, it is not wholly surprising: during AZ’s 1Q14 update, management fairly explicitly stated that the company planned to focus its attention Bydureon and Farxiga/Forxiga. As a result of this decision, management noted, it would be fair to expect a “flattish picture” for Onglyza moving forward. The strategic decision to move focus away from Onglyza was certainly due in part to the exciting newer drug classes in the company’s portfolio, but there are also a number of reasons why the DPP-4 inhibitor class (except for fixed-dose combinations with other classes) might have trouble escaping its current flattish period (see below).
• Onglyza’s 2Q14 performance was likely affected by the class-wide slowdown for DPP-4 inhibitors. Causes of this slowdown include: (i) the growing focus on cost-effectiveness by payers; (ii) the increased price competition due to the entry of more competitors; (iii) the introduction of SGLT-2 inhibitors; (iv) the saturation of patient transfers from TZDs to other oral agents (albeit, due in part to the decreasing number of patients still on TZDs); and (v) the reverberations of the incretin-pancreatitis/pancreatic cancer scare peaking in 2013. The class slowdown seems to be particularly pronounced in the US.
• In addition to the class-wide slowdown, Onglyza could potentially be suffering from clinicians’ perception of the 27% increase in hospitalization for heart failure seen in the drug’s cardiovascular outcomes study SAVOR. In February, we learned that the FDA requested heart failure data from SAVOR, and in the months since then we have seen growing support behind the theory that there might be a class-wide DPP-4 inhibitor effect on heart failure. We have also heard suggestions from conference speakers that although the signal seen in SAVOR could have been due to chance, it might be safest to avoid prescribing Onglyza in patients at high risk for heart failure. In a class with so many more similarities than differences, even small safety signals that emerge in clinical testing can have a significant impact on clinician uptake. Most KOLs, however, seem to feel that any “real” risk of heart failure would likely be quite moderate given that the patients in SAVOR who experienced an excess of heart failure did not experience any higher rates of MACE or death.

4. Bydureon (exenatide once weekly) sales continued their positive momentum, growing 70% YOY as reported to $112 million in 2Q14. In the US, sales grew from $57 million in $95 million (67%), while OUS sales grew from $9 million to $17 million (90%). Management highlighted that the product’s total US prescription (TRx) share within diabetes has grown 0.3% from March 2014 to the present. It is exciting to see Bydureon sales pick up speed following a slightly anemic launch, and as we learned later in the call, a series of line extensions could help keep that momentum going (see below). The franchise is likely benefitting (at least in terms of volume) from Express Scripts’ decision to replace Novo Nordisk’s Victoza (liraglutide) on its 2014 formulary with Bydureon and Byetta.

• Byetta (exenatide twice daily) sales continued their downward trend, falling 15% YOY to $88 million. The decline was most prominent in the US, where sales fell 28% from $74 million in 2Q13 to $53 million in 2Q14. Big-picture, however, the decline in Byetta revenue is more than matched by Bydureon’s upswing, and in cases where patients have access to both, we see the convenience of once-weekly administration as a compelling factor favoring Bydureon. Byetta has the cost advantage, and therefore might be the more attainable choice for patients without insurance.

5. The Bydureon dual-chambered pen, which encapsulates the entire reconstitution process within one device, recently received a positive CHMP opinion in Europe, paving the way for a likely approval in one to three months. Management noted that this decision arrived somewhat sooner than expected. The dual-chambered pen has also been filed for regulatory approval in Japan. In the US, the dual-chambered pen is on track for a launch some time before the end of the year – the device was approved by the FDA in March. Management noted during Q&A that AZ sees the device factor as a key determinant of the success of GLP-1 agonists. The dual-chambered is a big step up from the old tray product, but the reconstitution process still requires over 15 minutes.

• Consistent with previous guidance, management plans to submit a Bydureon suspension formulation in the US and EU in 2015 (which would enable an autoinjector device). We learned during the company’s 1Q14 update that a once-monthly Bydureon suspension was unfortunately dropped from the company pipeline. Ease of delivery for GLP-1 agonists is becoming one of the most important differentiating features for this class – if Bydureon gains an autoinjector device, it would join Novo Nordisk’s Victoza, Sanofi’s Lyxumia, and Lilly’s dulaglutide (still under review) as GLP-1 agonists that come in “ready-to-use” devices. GSK’s Tanzeum (albiglutide) currently requires patient-end reconstitution.
• During Q&A, management noted that enrollment for the EXSCEL CVOT for Bydureon recently hit its 75% enrollment target slightly ahead of schedule. The study’s estimated primary completion date is in late 2018.

6. During Q&A, management was prompted to discuss the recent news that GSK will be launching its once-weekly GLP-1 agonist Tanzeum (albiglutide) in the US at a heavy price discount relative to existing players. Consistent with what we heard from Lilly management during their 2Q14 update, AZ management believes that the product’s pricing stems from a potentially weaker clinical profile, although CEO Mr. Pascal Soriot did acknowledge that the GLP-1 agonist field stands to become increasingly competitive in the future. Notably, he also said that “other competitors will come in that will have a bigger impact” than Tanzeum (possibly referring to Lilly’s dulaglutide, which is the only once-weekly GLP-1 agonist to have proven non-inferiority vs. Novo Nordisk’s market-leading Victoza [liraglutide]). Although he did not directly suggest that prices for Bydureon will be coming down any time soon, Mr. Soriot did note that any reduction in price could lead to substantial volume gains for the class as a whole, especially in emerging markets.

7. Management emphasized twice during Q&A that GLP-1 agonists have great potential as co-therapy with SGLT-2 inhibitors. CEO Mr. Pascal Soriot specifically cited recent findings that SGLT-2 inhibitors’ efficacy is blunted by a counterregulatory glucagon response that increases hepatic glucose production. A GLP-1 agonist (or, to a slightly lesser extent, a DPP-4 inhibitor) could blunt this glucagon effect. Mr. Soriot suggested that AZ hopes to explore this combination further. It is exciting to hear AZ so enthused about the potential of combination therapy in diabetes (see our section on the saxa/dapa DPP-4 inhibitor/SGLT-2 inhibitor fixed dose combo below). We will be interested to see if AZ explores the combination through formal clinical trials, or if it thinks the mechanistic data on the two classes’ effects on glucagon will be enough to convince clinicians of the combination’s power to prescribe each component separately.

8. On the topic of incretin/non-incretin combinations, AZ confirmed its plan for a regulatory submission for its phase 3 saxagliptin/dapagliflozin fixed-dose combination (saxa/dapa) by the end of the year. We saw the results of the first phase 3 study on saxa/dapa presented at ADA: the combination achieved an A1c reduction of 1.5% from a baseline of 9.0% relative to reductions of 1.2% in the dapagliflozin arm and 0.9% in the saxagliptin arm. While not quite additive, and certainly not synergistic (as the glucagon hypothesis explained above might have suggested), these are quite impressive A1c-lowering results for an oral medication and come with minimal risk of hypoglycemia and some weight loss. Lilly/BI submitted a fixed-dose combination of their SGLT-2 inhibitor empagliflozin and their DPP-4 inhibitor linagliptin in the US in 1Q14, and plan to submit the combination in Europe in 2015. SGLT-2 inhibitor/DPP-4 inhibitor FDCs would likely not be able to match the efficacy of a GLP-1 agonist/SGLT-2 inhibitor combination (or, for that matter, a GLP-1 agonist/basal insulin combination), but SGLT-2/DPP-4 inhibitor FDCs have the advantage of oral administration. As a result, they are likely the most patient-friendly (and primary care-friendly) combination on the horizon. Pfizer/Merck have plans to develop a FDC of their partnered phase 3 SGLT-2 inhibitor ertugliflozin and Merck’s market-leading DPP-4 inhibitor Januvia (sitagliptin).

9. During BMS’ 2Q14 update, we learned some additional logistical details on how the transfer of BMS’ diabetes business to AZ will progress. BMS will continue providing the active ingredient for Farxiga and Onglyza (~$30 million quarterly), and will also provide a set of transitional services (~$30 million in 2Q13, expected to taper off towards the end of the year). For details on the royalty payment scheme, see our original report on the AZ-BMS portfolio acquisition.

10. Roxadustat, AZ’s therapy for chronic kidney disease (CKD) and end-stage renal disease (ESRD) and associated anemia, entered phase 3 in 2014. Filing in the US is planned for 2018, while a filing in China is expected in 2016. Management spoke about roxadustat in great detail during prepared remarks, outlining two planned phase 3 trials: both OLYMPUS (in non-dialysis-dependent CKD patients with anemia) and ROCKIES (in dialysis-dependent CKD patients with anemia) are expected to read out in 1Q17, and to use MACE as a primary endpoint. 

• Tempanor (AZD1722) remains in phase 2 for ESRD/CKD with type 2 diabetes.

Honorable Mention: During Q&A, management briefly suggested that oral GLP-1 agonists have a long way to go before reaching the clinic. The company believes that the candidates currently in development have a number of unanswered questions on bioavailability. Novo Nordisk is currently the biggest player working to develop an oral GLP-1 agonist.

Questions and Answers

Q: You mentioned you had a very successful US launch for Farxiga, but you haven’t given any sales numbers.

A: We for the time being will not provide sales numbers. We decided not to do that, and I’m sure you’ve noticed that we’re not the only ones not doing it. The only thing I can tell you is that the launch is going very well in the US. It’s going very well in Germany too, by the way, and also in Japan, although it is still pretty early there. In the US we have captured 40% of new-to-brand prescriptions, so we are pretty happy with the performance to date.

Q: Your 2023 guidance for the exenatide franchise seem to point to $2.5 billion in sales. We have seen some new long-acting GLP-1 analogs enter the market at a pretty huge discount, it appears. Others will joint the market over the next couple of years, and we will see combination with basal insulin and perhaps oral GLP-1 as well. Has any of the recent news about GSK’s pricing changed your view of this franchise going forward?

A: As a general statement, we expect it to be more competitive, both in terms of promotional activity and access. That being said, we have good access with the product, and we have a well-established team, we’ve integrated the BMS team in, and we have good targeting. I think on top of that, we have the next stage in our device strategy, which would be launching in the second half of the year, which is the dual chamber pen, which is very much on track. I think that ultimately when we look at the competitive situation, we recognize the importance of device, which is why we are launching that pen. We’re confident in the clinical profile of Bydureon, which has shown consistent A1c reductions in the range of 1.3-1.9% very durably over six years of data. The parameters of efficacy and convenience of device will remain core factors. In terms of oral GLP-1s, I think they are farther out. There are a number of questions about feasibility and the amount of volume of drug.

A: Regarding the GSK product and the pricing strategy, we don’t expect this product to really have a substantial impact because the pricing probably reflects the clinical profile of the agent. Certainly other competitors will come in that will have a bigger impact. Our strategy here relies on the device, and the development of volume for the class. The GLP-1 agonist class is one that has an important place in diabetes. If prices decline, you can expect growth in volume, especially in emerging markets. It is very competitive, without a doubt, but we also see good complementarity with Forxiga. We think this will be a very competitive combination, and we will leverage the portfolio as much as we can.

Q: What are the prospects for and interest in a combination of a GLP-1 agonist with an SGLT-2 inhibitor? We’ve been seeing some interesting case reports around that.

A: There is a very strong logic to combine a GLP-1 agonist and Farxiga; it is clearly a combination we will be exploring and a case where the two products can support each other. As you know, the potential problem with SGLT-2 inhibitors as a class is that they increase liver glucose production, but with DPP-4 inhibitors and even more so GLP-1 agonists you should see substantial clinical benefits. We will certainly be exploring that one.

Q: How are events accruing in the outcomes study for Bydureon, and when do you anticipate that the study will read out?

A: The EXSCEL trial is enrolling very well. We just hit our 75% enrollment target a little bit early, actually, and if I recall correctly I don’t think the data will read out until 2018.

Q: Could you clarify the amount of royalties you paid to BMS for their shared products and where it is reported in your accounts?

A: The accounting treatment for the BMS rights acquisition is a business combination, so we capitalize the payment at fair market value and then we amortize them. Regarding the royalties, I can’t give you the number of royalties separately, but I can tell you the number that we have paid as royalties for the first half, which was $58 million, paid from the acquisition in February through June.

-- by Manu Venkat and Kelly Close