J&J 1Q16 – Invokana sales up 17% YOY, down 13% sequentially to $325 million; Record-low global and US LifeScan/Animas sales; No commentary on Vibe, Calibra, or AP – April 19, 2016

J&J provided its 1Q16 update this morning in a call led by CFO Dominic Caruso and VP of Investor Relations Ms. Louise Mehrotra. Below we include our top ten highlights from the call, followed by Q&A.

Janssen Highlights

1. Sales of SGLT-2 inhibitor Invokana (canagliflozin) and Invokamet (canagliflozin/metformin) rose 17% year over year (YOY) but declined 13% sequentially to $325 million in 1Q16. TRx for Invokana in the US type 2 diabetes market (excluding insulin and metformin) was 6.1% in 1Q16, down from 6.5% in 4Q15.

2. The EMA initiated a new safety review of Invokana/Invokamet last week after an increase in amputations (mostly affecting toes) was observed in the ongoing CANVAS CVOT. We were quite surprised to hear this though note that it is a very sick group of patients in this trial.

3. While it was not mentioned during the call, full DKA results from the phase 2 trial of Invokana in type 1 diabetes were published in the latest issue of Diabetes Care. The incidence of ketone-related adverse events at 18 weeks was 5.1% (6/117) with the 100 mg dose of Invokana and 9.4% (11/117) with the 300 mg dose. We believe this is a very manageable problem but that education among most patients is low.

4. There was no mention of Janssen’s numerous other ongoing efforts in diabetes and obesity; recent news includes the announcement of three new collaborations related to type 1 diabetes under the Disease Interception Accelerator program.

LifeScan/Animas Highlights

5. Global LifeScan/Animas revenue totaled $429 million in 1Q16, down 11% as reported and 8% operationally YOY. This marked the lowest quarterly revenue ever in our J&J Animas/LifeScan model (which goes back to 2003), the fifth consecutive quarter in which global sales have come in under $500 million, and 15 of the past 16 quarters with global sales declines. The field continues to be battered, with Medicare dropping prices again recently.

6. US Diabetes Care sales totaled $180 million in 1Q16, dropping 15% YOY against a challenging comparison (sales grew 10% YOY in 1Q15). This was the lowest US revenue ever recorded in our J&J model and the second-ever quarterly revenue to dip below $200 million (only sales of $198 million in 1Q14 came in under this milestone). US sales fell 11% sequentially in 1Q15.

7. International Diabetes Care revenue totaled $249 million in 1Q16, down 9% as reported and 3% operationally YOY against an easy comparison (down 15% as reported and up 1% operationally in 1Q15). We suspect currency headwinds played a small role in the decline. Sales fell 10% sequentially.

8. J&J’s supplementary materials highlighted the “strength” of Animas Vibe sales. We thought this was notable for a couple reasons: (i) Dexcom’s G5 launched in 3Q15-4Q15 and saw major uptake, representing a theoretical headwind for J&J (patients have to decide between data on the pump with the G4-integrated Vibe vs. data on the phone with G5); and (ii) the Vibe is no longer the only Dexcom-integrated pump on the market, as the Tandem t:slim G4 has been available since September 2015.

9. The accompanying slide deck highlighted J&J/LifeScan’s new collaboration with WellDoc to advance mobile care for patients with type 2 diabetes. Announced in early March, the partnership will integrate WellDoc’s FDA-approved BlueStar software for type 2 diabetes with LifeScan’s recently launched OneTouch Verio Flex Bluetooth-enabled BGM and Reveal app.

10. J&J did not comment on its Calibra Finesse bolus-only insulin delivery patch device (the company said in 2Q15 that it would launch this year); hypoglycemia-hyperglycemia minimizer (HHM) pivotal study (in the planning phase); or (iii) the Bluetooth-enabled OneTouch Verio Flex BGM launched in March.

Honorable Mention: Pooled global revenue for J&J and Roche (who also reported this morning) reached ~$875 million, falling 14% relative to pooled revenue in 1Q15 (~$1.0 billion). Typically, J&J reports a week before the other BGM companies. We bring you our full analysis of the pooled results as a bonus (also included in our Roche coverage).

Janssen Highlights

1. Sales of SGLT-2 inhibitor Invokana (canagliflozin) and Invokamet (canagliflozin/metformin) rose 17% year over year (YOY) but declined 13% sequentially to $325 million in 1Q16. This is the first time Invokana has seen a sequential decline since its launch, though growth had been tapering off over the past few quarters. As in past quarters, the vast majority of revenue (91%) came from the US, where sales rose 12% YOY but declined 15% sequentially to $297 million. Ex-US sales more than doubled YOY and rose 17% sequentially (from a much lower base) to $28 million. Management continued to characterize Invokana as a strong driver of growth during the call and emphasized that the drug retains 80% preferred access in the combined commercial and Medicare Part D segments. The company attributed the sequential decline to extra shipping days, incremental rebates and savings program redemptions, and increased competition for formulary positioning. Management did not elaborate further on Invokana’s challenges when pressed during Q&A, instead emphasizing the strong YOY growth. We wonder how much of the increased formulary competition stemmed from the positive EMPA-REG OUTCOME results for Lilly/BI’s Jardiance (empagliflozin) – we hope to learn more during Lilly’s 1Q16 update on April 26. The impact of EMPA-REG OUTCOME on Invokana’s performance is one of our main unanswered questions for Janssen in 2016, as we could imagine the results helping Invokana by expanding the SGLT-2 inhibitor class or hurting it by leading more providers and payers to favor Jardiance.

• TRx for Invokana in the US type 2 diabetes market (excluding insulin and metformin) was 6.1% in 1Q16, down from 6.5% in 4Q15. TRx among US endocrinologists fell to 11.5% from 12.8% in 4Q15, while TRx among US primary care physicians fell from 5.8% to 5.5%. Again, we are very curious to learn from Lilly whether these losses were due primarily to gains for Jardiance post-EMPA-REG OUTCOME or whether other factors were responsible. Formulary competition was likely a major factor here as well, as Janssen has often cited strong access as the main explanation for Invokana’s market share gains in the past.

Figure 1: Invokana Sales (1Q14-1Q16)

2. The EMA initiated a new safety review of Invokana/Invokamet last week after an increase in amputations (mostly affecting toes) was observed in the ongoing CANVAS CVOT. The agency’s Pharmacovigilance Risk Assessment Committee has requested more information from Janssen to assess whether Invokana increases the risk of lower limb amputations; it also plans to request information from other SGLT-2 inhibitor manufacturers. The EMA notice stated that the incidence of lower limb amputation in CANVAS is currently 7/1,000 patient-years with the 100 mg dose of Invokana and 5/1,000 patient-years with the 300 mg dose vs. 3/1,000 patient-years with placebo after an average of 4.5 years of follow-up. There is currently no significant difference in amputations between groups in the CANVAS-R renal outcomes trial (incidence of 7/1,000 patient-years with Invokana and 5/1,000 patient-years with placebo). During Q&A, J&J management emphasized that the 12 previous clinical trials of Invokana have not found any signal of increased amputation risk; notably, the Independent Data Monitoring Committee for CANVAS has recommended that the trial should continue. To our knowledge, there is no mechanistic or clinical evidence to suggest that SGLT-2 inhibitors would increase amputation risk, so while we will be keeping a close eye on the investigation, we expect that this will not prove to be a major concern. The amputation rates in EMPA-REG OUTCOME were not reported in the published paper, and we assume this means there were no concerning signals. 

3. While it was not mentioned during the call, full DKA results from the phase 2 trial of Invokana in type 1 diabetes were published in the latest issue of Diabetes Care. Results from the trial, first presented at EASD 2015, demonstrated reductions in A1c, body weight, and insulin doses but a dose-dependent increased risk of ketoacidosis with Invokana vs. placebo. The incidence of ketone-related adverse events at 18 weeks was 5.1% (6/117) with the 100 mg dose of Invokana and 9.4% (11/117) with the 300 mg dose; there were no events in the placebo group. The incidence of serious DKA events was 4.3% (5/117) with the 100 mg dose of Invokana and 6.0% (7/117) with the 300 mg dose. All serious events occurred following precipitating factors such as infection, reduced carbohydrate intake, pump failure, or other interruption/reduction of insulin therapy. In the paper, the authors (led by Dr. Anne Peters [USC, Los Angeles, CA], who has been a leading voice on the SGLT-2 inhibitor/DKA issue since it first emerged) noted that because very little was known about the risk of DKA with SGLT-2 inhibitors when the study began, the protocol did not include any specific monitoring guidelines or mitigation strategies. They recommended several such strategies to mitigate the risk in future studies of SGLT-2 inhibitors in type 1 diabetes, including routine ketone monitoring, lower doses of the SGLT-2 inhibitor, interruption of treatment during illness/surgery, discontinuation of treatment if ketones are elevated, and more detailed algorithms for adjusting insulin doses and carbohydrate intake. While we expect the regulatory bar for SGLT-2 inhibitors in type 1 diabetes to be fairly high, we hope that Janssen will continue to pursue the indication and that phase 3 studies can provide more clarity on how to reduce the risk of DKA.

• We recently saw data at ENDO demonstrating a ~10% incidence of DKA with AZ’s Farxiga (dapagliflozin) in type 1 diabetes. The study took 30 patients on insulin + liraglutide at baseline and randomized them to add Farxiga or placebo. The two DKA cases occurred immediately after a dose increase from 5 mg to 10 mg dapagliflozin, with low final insulin doses (≤0.5 U/kg) identified as the primary risk predictor. More broadly, there were statistically significant increases in plasma glucagon, free fatty acids, and ketones across the Farxiga group despite the counteracting effect of the GLP-1 agonist. AZ is currently investigating Farxiga in two phase 3 trials (DEPICT 1 and DEPICT 2) in type 1 diabetes, and we hope these studies will help provide a more complete picture of the class’ risk/benefit profile in this population. Lilly/BI are also investigating Jardiance in type 1 diabetes in the phase 3 EASE-2 and EASE-3 trials.

4. While not mentioned during the call, Janssen has a number of ongoing efforts in diabetes and obesity:

• Type 1 diabetes: In addition to a possible type 1 diabetes indication for Invokana, Janssen has a significant presence in the type 1 prevention/cure field. The innovative Disease Interception Accelerator program recently announced three new collaborations related to type 1 diabetes and a new project focused on identifying biomarkers that predict gestational diabetes risk. Janssen also has a hand in multiple efforts to develop cell replacement therapies for type 1 diabetes, through subsidiary BetaLogics and an option to license/acquire ViaCyte’s VC-01 cell replacement therapy.
• Type 2 diabetes: Janssen has taken several steps to refill its type 2 diabetes/metabolic disease pipeline in recent months. It licensed Hanmi’s phase 1 GLP-1/glucagon dual agonist last November and announced a collaboration with Intrexon in December to develop oral biologics for type 2 diabetes, obesity, and other metabolic disorders.
• Obesity/prediabetes: J&J has indicated on several recent occasions that it considers obesity to be an area with great potential, including for Invokana. The company completed a phase 2 trial of Invokana in combination with phentermine in June 2015, and another trial investigating Invokana’s effects on body weight and metabolism in patients with type 2 diabetes and a BMI ≥25 kg/m² is expected to complete in 2017. We were also intrigued to hear Janssen mention the possibility of a prediabetes indication for Invokana in its Pharmaceutical Business Review last year but have not heard any updates since then.

LifeScan/Animas Highlights

5. Global LifeScan/Animas revenue totaled $429 million in 1Q16, down 11% as reported and 8% operationally YOY. The quarterly performance in actual revenue represents the lowest quarterly revenue all-time in our J&J LifeScan/Animas model (which goes back to 2003) and the fifth consecutive quarter in which global sales have come in under $500 million – a pattern we have never before seen in our model. Further, 1Q16 now marks 15 of the past 16 quarters with global sales declines (3Q14 was the lone exception and that was an easy comparison). Of course, part of this is currency driven in the past, though the relatively small 2% negative impact this quarter means the underlying business is still having a rough time. This results on an easy YOY comparison to a 6% sales decline in 1Q15, though a tougher comparison to 4% operational growth a year ago.

• Sequentially, revenue fell 11% against a low base, as sales reached just $480 million in 4Q15. 1Q16 marked the fourth quarter since the beginning of 2015 that quarter-to-quarter sequential results have been negative.
• CFO Mr. Dominic Caruso mentioned “Diabetes Care” (LifeScan/Animas) only very briefly in his prepared remarks, noting that the segment has experienced some “challenges” due to pricing pressures and competition with newly launched pumps. The latter must refer to the Dexcom-integrated Tandem t:slim G4 that launched in September in the US, and perhaps the MiniMed 640G that is seeing strong uptake in Europe. This is the first time in recent memory that management has directly highlighted competition in the pump market. Supplementary materials did allude to the “strength” of Vibe sales, suggesting that this side of the business is still growing, though perhaps not as quickly as the 32% operational growth for Vibe seen after it launched in early 2015. The competition is no surprise based on other quarterly updates: Tandem saw a “tremendous” response to the t:slim G4 in 4Q15, Insulet had a strong 4Q15 with record high OmniPod revenue, and Medtronic’s MiniMed Connect drove record-high US sales in 4Q15.  

Figure 2: Global, Us, International Quarterly Sales (1Q12 – 1Q16)

6. US Diabetes Care sales totaled just $180 million in 1Q16, dropping 15% YOY against a modest comparison (sales grew 10% YOY in 1Q15). As expected, management did not comment extensively on the US business except to acknowledge pricing pressures and competition from competitive pumps (see above). The quarterly performance marks the lowest US revenue ever recorded in our J&J model and the second-ever quarterly revenue to dip below $200 million (only sales of $198 million in 1Q14 came in under this milestone). US LifeScan/Animas sales peaked at $362 million in 4Q07, more than double this quarter’s sales – that’s a striking testament to how devastating the impacts of competitive bidding have been. Sequentially, sales fell 11% against a low base ($202 million in 4Q15). The Animas Vibe did secure pediatric approval in 4Q15 down to age two. This is apparently driving Vibe growth internationally and we wonder if that can buoy the US business in future quarters. In pediatrics, MiniMed Connect is pretty compelling for remote monitoring, while Insulet has the tubeless advantage. The Vibe cannot be used for remote monitoring, as the pump does not contain any Bluetooth or smartphone connectivity. While it’s possible to use a Share receiver with the Vibe, it requires separate calibration and negates the value of integration.

• While not mentioned during the call, CMS recently announced results from the second round of its competitive bidding program, establishing even lower payment amounts for mail order blood glucose strips – $8.32 per 50-count box ($0.17 per strip), down 20% from the current allowable payment of $10.41 ($0.21 per strip). The new prices are slated to go into effect on July 1, 2016 and we’re not sure how this will impact J&J’s Diabetes Care business, which continues to have a rough time. For the latest on competitive bidding, see our detailed report here.

7. International Diabetes Care revenue totaled $249 million in 1Q16, down 9% as reported and 3% operationally YOY against a relatively easy comparison (down 15% as reported and up 1% operationally in 1Q15). This performance marks the lowest revenue since 3Q06 (which totaled $230 million) and the seventh consecutive quarter of negative reported growth. At the same time, the operational performance gives us some hope that declines are driven in large part by currency as opposed to significant underlying weakness in the business’ fundamentals – indeed, the OUS business has grown operationally in three of the past five quarters. Sequentially, sales fell 10% against a low base (international sales totaled $278 million in 4Q15).

• J&J’s slide deck attributed international performance to Diabetes Care “softness” in Europe partially offset by strength of the Animas Vibe. The latter was buoyed by the recent approval and launch in pediatrics. J&J has not recently brought a new product to the OUS market, which could explain some of the declining sales numbers.  

8. J&J’s supplementary materials called out the “strength” of Animas Vibe sales. We thought this was notable for a couple reasons: (i) Dexcom’s G5 launched in 3Q15-4Q15 and saw major uptake, representing a theoretical headwind for J&J (patients have to decide between data on the pump with the G4-integrated Vibe vs. data on the phone with G5); and (ii) the Vibe is no longer the only Dexcom-integrated pump on the market, as the Tandem t:slim G4 has been available in the US since September 2015. We’ll be keeping an eye on competition between the two moving forward, especially in the pediatric age group. As a reminder, Animas received FDA approval of the pediatric indication (ages 2+) for the Vibe integrated with Dexcom G4 in. Tandem countered by filing a 510(k) to expand the pediatric indication of the t:slim down to patients 6+ years old (it’s currently for 12+ year olds) and clearance is expected in 2Q16. It does sound like a significant number of pediatric patients are already using the t:slim off-label – reps recently shared that the under-12 population is currently among Tandem’s most profitable demographics, and we wonder whether Tandem will steal patients from J&J (especially younger patients and those new to pumps) with the cool factor of the touchscreen interface. Medtronic’s MiniMed Connect also launched in the fall and sales have far exceeded management’s expectations – this is also a strong competitor on the pediatric front.

9. The accompanying slide deck highlighted J&J/LifeScan’s collaboration with WellDoc as an “important development” – we were glad to see this mentioned even though the partnership didn’t come up in prepared remarks or Q&A. Announced in early March, the partnership will integrate WellDoc’s FDA-approved BlueStar software for type 2 diabetes with LifeScan’s recently launched OneTouch Verio Flex Bluetooth-enabled BGM and Reveal app. J&J also made an investment (size not specified), completing WellDoc’s Series B financing round at a total of $29.5 million. While the integration is still under discussion, we assume the first generation will use Flex’s built-in Bluetooth and paired app to automatically populate blood glucose fields in BlueStar, saving patients manual entry. J&J has a lot to gain from this partnership: (i) augmenting its Reveal app and Flex BGM with the clinically proven BlueStar software – instead of just displaying a number, BlueStar gives patients real-time feedback and education; (ii) setting itself apart from competitors selling just another meter (or just another connected meter); (iii) giving providers BlueStar’s powerful tools, including clinical decision support; (iv) getting WellDoc’s expertise on software, which is not a strength for the hardware-focused LifeScan; and (v) a clear path to reinvigorating the pipeline at a time of continued struggle in BGM. As this partnership evolves, we wonder how it might impact J&J’s US sales – the BlueStar integration is a key differentiator in our view, though it remains to be seen how patients, payers, and providers will perceive its benefits.

• Roche also announced a partnership yesterday to enhance the Bluetooth-enabled Accu-Chek Connect meter with mySugr’s Logbook app (the world’s most popular diabetes app, with 600,000+ users). The integration has already launched in Germany and Austria and is coming to the US soon. While the focus of the two partnerships is different, we see them in the same ballpark of enhancing BGMs and making data more actionable and meaningful. We’ll compare them once J&J and WellDoc launch their integration, which will presumably happen in phases over the next year.

10. There was no commentary on the pipeline, and we were particularly disappointed not to hear about the company’s automated insulin delivery product or Finesse bolus-only patch device. We learned in early March that Animas is planning a large-scale, multi-center pivotal study of its hypoglycemia-hyperglycemia minimizer (HHM) with the FDA – the first major timing update ever on the company’s automated insulin delivery product with Dexcom. At this point, there have been no updates in terms of trial size, timing, or type of device, and there is still no sign of the study on clinicaltrials.gov. For more on automated insulin delivery, see our competitive landscape here.

• J&J was unfortunately silent on the Calibra Finesse bolus-only insulin delivery patch device. Guidance on Finesse in 2Q15 called for a launch in 2016 and we hope that has not changed, though the absence of news is not confidence inspiring. According to clinicaltrials.gov, J&J is still “currently recruiting” for a 24-week clinical trial (n=312) of the device that will randomize type 2 diabetes patients not achieving glycemic targets (A1c 7.5-10%) to either Finesse or the Novo Nordisk FlexPen to initiate bolus insulin therapy. The primary endpoint is A1c at 24 weeks, with secondary endpoints including time-in-range and treatment satisfaction. Primary completion is slated for December 2016.
  
  • What is taking Finesse so long to come to market? Manufacturing? Cost and payer discussions? Go-to market strategy? Device updates (e.g., prefilled)? Absence of clinical outcomes data? Packaging and labeling? What can J&J learn from Valeritas’ V-Go?
  • As a reminder, the Finesse is a three-day, very slim profile (2 inches long, 1 inch wide, and 0.25 thick), bolus-only delivery device. J&J acquired the device from Calibra Medical in July 2012 and it secured FDA clearance way back in 2010 for type 1 and type 2 diabetes. The clearance was updated in 2012. We’re not sure if J&J has made any modifications that require an additional submission – the FDA database does not list any new clearances.
• We were a bit surprised that neither prepared remarks nor Q&A mentioned the Bluetooth-enabled OneTouch Verio Flex BGM that launched in the US in 1Q16. The Verio BGM pairs with the updated OneTouch Reveal app and is priced at just $19.99. Aside from Bluetooth connectivity, the Flex’s key feature is an arrow that points to a color and adds context to glucose readouts – blue for hypoglycemia, green for in-range, red for hyperglycemia. J&J has clearly prioritized low-cost, so the screen is not color. The Flex replaces LifeScan’s former Bluetooth-connected OneTouch Verio Sync BGM, which has now been discontinued. For more on the OneTouch Verio Flex BGM, please see our detailed coverage.

Table 1: LifeScan/Animas Pipeline Summary

Honorable Mention: Pooled Global Sales

• J&J is typically the first of the BGM companies to report, but – for the first time in recent memory – Roche reported its quarterly financials in parallel with J&J. We bring you our pooled analysis below.
  
  • Pooled global revenue for J&J and Roche (who also reported this morning) reached ~$875 million, falling 14% relative to pooled revenue in 1Q15 (~$1.0 billion). This comes against an easy comparison as combined revenue declined 9% a year ago. While some of the weakness comes from currency headwinds, it cannot explain the entire drop since the US market is driving the declines – see below.
  • Pooled declines in 1Q16 were driven in large part by US weakness. Combined sales of ~$228 million fell a striking ~27% YOY against a relatively easy comparison to pooled sales in 1Q15 (sales grew 2%). Sales fell 16% sequentially, pretty standard seasonality effects. The performance marks an all-time low for J&J and Roche pooled quarterly US sales by significant margin – the previous low was $277 million in 3Q15.
  • Pooled international sales were not much better as combined sales of $647 million fell 8% in 1Q16 against an easy comparison to pooled revenue in 1Q15 (down 13%). As noted above, depressed revenues as a result of foreign exchange have characterized the business for several quarters now, though will the business return to growth once the currency effects attenuate? Pooled sales fell 14% sequentially.

Figure 3: Pooled J&J and Roche Quarterly Sales (1Q12– 1Q16)

Questions and Answers

Q: Could you comment a little bit more about the diabetes business? You had mentioned there are challenges in the market. Could you comment on Invokana and the additional competition that’s in the market?

A: To clarify, I was talking about challenges in the market with respect to our consumer medical device diabetes business, the LifeScan business. And what I meant by that was that we see continued pricing pressure in that market and we also see some competition with respect to new pumps being launched. We’ve just launched a new pump, obviously, and we’re also seeing good uptake with our current pump. So the dynamics I’m talking about as being challenging are in the medical device piece of the business. Overall Invokana continues to do well. Its uptake is strong quarter over quarter and it remains a growth driver for us going forward.

Q: Invokana is an important growth driver for the company, but sales did come in a bit lighter than we had anticipated despite the growing excitement about this class of drugs. Can you comment generally on what drug categories you’re seeing with widening gross to net discounts? Are you seeing more gross to net discounts or widening discounts specifically in diabetes?

A: I wouldn’t say that there’s any one particular area in the business where we see a widening of gross to net or rebating that’s any different than what we’ve seen before in the business. So rather than comment on any one particular area, I would say, generally speaking, there’s nothing that sticks out to us as a widening trend.

Q: There was an announcement last week that the EMA is looking into an increase in amputations seen in the CANVAS study. Do you have any comments on that? Is that something the FDA is also looking into?

A: Before the CANVAS study, there have been 12 previous studies with Invokana where we saw no indication of an increase in lower limb amputations, particularly the toe, so this seems to be focused only on the patients in the CANVAS study. It’s important to note that yes, the European regulatory body is taking a look at it, but already the independent monitoring committee that reviews the clinical trial concluded that the trial should continue after reviewing all the data. So that’s what I can tell you about that and obviously, we’ll look to advise physicians appropriately of the results once we have more understanding of them.

-- by Ava Runge, Emily Regier, Varun Iyengar, Adam Brown, and Kelly Close