Memorandum

Roche 1Q18 – Global Diabetes Care sales up 7% YOY, driven by 14% North American growth; Solo Patch Pump Listed under key 2018 launches; Lucentis Flat YOY – April 27, 2018

Executive Highlights

  • Global Diabetes Care revenue of 478 million CHF (~$490 million) rose 7% YOY, Roche Diabetes’ strongest YOY growth quarter in over ten years. This came against a moderate comparison to 1% growth in 1Q17. However, management was quick to point out the “very slow growth” in 1Q17 and base of sales, projecting for YOY growth <5% for the remainder of the year. On a sequential basis, sales declined 5% from a base of 501 million CHF (~$508 million). Revenue was driven by strong sales in the US and Asia Pacific, with North American revenue rising 14% YOY. International revenue of 422 million CHF (~$433 million) increased 6% YOY, marking the largest OUS YOY growth recorded in our model since 1Q10. Performance reported for both geographies came on moderate comparisons to 1% and 2% 1Q17 growth, respectively. The slide deck noted impressive 24% YOY growth in US BGM sales – presumably signaling uptake for Accu-Chek Guide.

  • Pooled 1Q18 global sales for Roche, Abbott, and J&J totaled $1.25 billion, up 10% YOY on a moderate comparison to 2% growth in 1Q17. Revenue was clearly driven by a strong Q1 performance from Abbott’s FreeStyle Libre – when Abbott is excluded from pool sales, total Roche and J&J revenue of $829 million declines 2% YOY on an easy comparison to a 4% decline in 1Q17. Our estimates suggest that isolated BGM sales from the Big Three grew ~3% YOY to ~$1.0 billion.

  • While the slide deck listed the Solo Patch Pump as a key 2018 EU launch, there was no mention of the device during the call. We did notice that Roche posted a participant-reported outcomes study for the Solo (n=180 adults with type 1) on ClinicalTrials.gov in late March. The multi-site study, currently listed as “not yet recruiting” investigates treatment satisfaction for the Solo vs. MDI. Treatment satisfaction comparing the Solo with the Omnipod will be evaluated as a secondary outcome – a bold move on Roche’s part. The study is slated for completion in July 2019.

  • There was no mention of Senseonics, which could receive FDA approval any day now following a very successful Advisory Panel meeting. Similarly, there was no mention of mySugr, which had ~1.2-1.4 million users as of ATTD and has begun rolling out insights based on hyperglycemia/hypoglycemia pattern recognition.

  • Lucentis (anti-VEGF therapy) sales of 393 million CHF (~$403 million) remained flat YOY on a very tough comparison to 10% growth in 1Q17. Sales increased 37% sequentially, albeit from a record low base of 288 million CHF (~$291 million) in 4Q17. Readout for the phase 2 Lucentis port delivery study LADDER (n=220) is expected in 2H18.

    Roche CEO Dr. Severin Schwan led the company’s 1Q18 financial update yesterday morning. Global Diabetes Care revenue rose a promising 7% YOY – perhaps an indication that the business has bottomed out and is now on the ascent. There were very few diabetes mentions during the call, but we were glad to see Roche’s Solo Patch Pump remain listed under key 2018 EU launches in the slide deck. After years of silence, it’s reassuring to see an indication that Roche is now sticking with its timeline for Solo – we hope it actually launches. Below, we bring you top financial, pipeline, and therapy highlights from the call, as well as pooled “Big Three” (Abbott, J&J, Roche) analysis and selected Q&A.

    Table of Contents 

    Financial Highlights

    1. Global Diabetes Care Revenue Up 7% YOY on Moderate Comparison; <5% Growth Expected for Remainder of ’18; Growth in All Regions

    Global Diabetes Care revenue of 478 million CHF (~$490 million) rose a promising 7% YOY against a moderate comparison to 1% growth in 1Q17. Sales decreased 5% sequentially from 4Q17. On a YOY basis, 1Q18 represents the greatest quarterly growth in more than ten years (+7% in 4Q07). Per the investor update, although “market conditions continue to be challenging” – presumably in reference to CMS’ competitive bidding and other pricing and competitive pressures in BGM – YOY growth was observed in all regions. In fact, Diabetes Care contributed 16% of Roche’s Diagnostics division revenue. During prepared remarks, Roche Diagnostics CEO Mr. Roland Diggelmann highlighted strong sales particularly in the US and Asia Pacific. Mr. Diggelmann also noted that growth was driven by new product launches – the pared down Accu-Chek Instant bG system was listed as a key 2017 launch in the 4Q17 slide deck and the Accu-Chek Instant S launched in India in March. However, Mr. Diggelmann was careful to draw attention to “very slow growth” in 1Q17 and the resulting YOY comparison, warning listeners to not expect similar trends for the remainder of 2018 and instead prepare for growth <5% YOY.

    Figure 1. Global, North America, International Quarterly Sales (1Q11-1Q18)

    • BGM continues to account for the lion’s share of Diabetes Care revenue (~93%), as depicted in the figure below, and remarkably grew faster than pumps in 1Q18. BGM rose 5% YOY, while Roche’s “Other” products (mostly pumps) rose 2% YOY. Although it’s heartening to see positive growth from both sectors, this performance comes against easy to moderate comparisons: in 1Q17 BGM rose just 2% YOY and “Other” fell 7% YOY. As Roche continues to shift more towards an ecosystem approach, we wonder how much longer it will separate sales out like this.

    Figure 2. 1Q18 BGM and “Other” (Mostly Pumps) Sales

    2. North American Revenue up 14% YOY on Moderate Comparison; US BGM Climbs 24% YOY; 89% Sequential Increase from Record Low

    North America Diabetes Care sales of 56 million CHF (~$57 million) rose a very strong 14% YOY against a moderate comparison to 2% growth in 1Q17. This performance marks the largest YOY North America growth recorded in our model in more than ten years (4Q07)! Revenue was driven by impressive 24% YOY growth in US BGM (per the slide deck) – perhaps the US launch of the Accu-Chek Guide Meter and SimplePay Savings Program in May is beginning to pay off. We also wonder if the mySugr Bundle is helping here – no metrics have ever been shared on uptake of the unlimited strips, Accu-Chek Guide BGM, and coaching. As a reminder, Roche discontinued new pump sales in the US as of the beginning of 2017 – further evidence that the observed growth was concentrated in BGM. Sales increased a whopping 89% sequentially, albeit from a record low base of just 30 million CHF (~$30 million) in 4Q17.

    3. International Sales Rise 6% YOY on Moderate Comparison; 10% Sequential Decline

    International revenue of 422 million CHF (~$433 million) increased 6% YOY against a moderate comparison to 1% growth in 1Q17. Sales decreased 10% sequentially from a high base of 471 million CHF (~$477 million), aligning with Roche’s typical seasonal trend of 1Q and 3Q sequential declines interspersed with 2Q and 4Q growth (dating back to 2007). Still, this performance marks the largest YOY international increase recorded in our model since 7% YOY growth in 1Q10. Based on Mr. Diggelman’s remarks, we imagine this growth was spearheaded by expansion of the Accu-Chek Instant bG and introduction of the Accu-Chek Instant S in India. We continue to believe that deeper penetration in developing markets with simple meters has loads of potential, both from an outcomes-improving standpoint and from a business standpoint.

    Pooled Financial Highlights

    1. Pooled Global Sales for “Big Three” BGM Rise 10% YOY to $1.3 Billion, Driven by Strong Abbott Performance

    Pooled global sales for Roche, Abbott, and J&J (the “Big Three” public BGM companies) totaled $1.25 billion in 1Q18, up 10% YOY on a moderate comparison to 2% growth in 1Q17. Pooled revenue decreased 5% sequentially from a base of $1.31 billion in 4Q17, breaking the three-quarter streak of positive sequential growth. Still, this performance marks three consecutive quarters of YOY growth for these three companies – largely reflecting Abbott’s FreeStyle Libre. To this end, Abbott absolutely drove pooled Q1 growth, reporting global increases of 44% YOY (+33% operationally) in 1Q18. These increases were clearly due to continued strong uptake of FreeStyle Libre (we estimate that Libre sales contributed ~50% of Abbott’s total Diabetes Care revenue in the quarter), not BGM. Roche contributed modestly to growth, rising 7% YOY. J&J on the other hand pulled down pooled revenue, with sales declining 15% YOY (-20% operationally) to a record low of $339 million in 1Q18 – of course, J&J no longer has the (however small) benefit of Animas revenue since the unit announced its closing in October. Excluding Abbott sales, pooled J&J and Roche revenue of $829 million declined 2% YOY on an easy comparison to a 4% decline in 1Q17. And estimating BGM revenue alone (excluding revenue from FreeStyle Libre and Roche pumps), sales climbed ~3% in 1Q18 (to ~$1.0 billion)

    • Abbott has firmly surpassed J&J in diabetes revenue (after pulling ahead for the first time in 4Q17) and is inching ever closer to Roche – just $79 million in quarterly revenue separated the two in 1Q18. Given that Abbott CEO Mr. Miles White anticipates 2018 FreeStyle Libre sales of at least $1 billion in 2018, and Roche expects diabetes care growth <5% for the remainder of 2018, this margin will only become narrower. At this rate, Abbott’s global sales may exceed Roche’s in 2018 or perhaps 2019. Although Roche certainly has some irons in the fire with Senseonics, mySugr, automated insulin delivery, and its growing digital ecosystem, translating these initiatives into actual topline revenue may take some time. Meanwhile, FreeStyle Libre is gaining serious ground with robust OUS momentum, an encouraging US launch, and pediatric indication filing with the FDA expected in 2018. J&J’s future is far more tenuous – the Platinum Equity ~$2.1 billion offer to acquire LifeScan is still on the table and the company remains silent on Calibra (manufacturer of the OneTouch Via patch bolus delivery device).

    • We’d note that pooled “Big Three BGM” growth trends are becoming more challenging, given differences between the three companies. Our estimates suggest that FreeStyle Libre is getting close to surpassing Abbott’s BGM business (perhaps as soon as Q2). BGM still dominates Roche and J&J for now. Roche is expanding its focus to add AID and digital health, but these are not likely generating meaningful revenue now. J&J’s future is uncertain, though we think Ascensia is probably the best case study of what it may do under Platinum.

    Figure 3. Worldwide Pooled Roche, Abbott, and J&J Quarterly Sales (1Q12-1Q18)

    Device Pipeline Highlights

    1. Solo Patch Pump Listed Under Key 2018 Launches; New PRO Solo Study (n=180 type 1s) Comparing Solo to MDI and Omnipod on CT.gov

    Although there was no mention of Roche’s Solo Patch Pump during the call, it was once again listed in the slide deck under key 2018 launches. Per the slide, Solo is slated for a EU launch, although it’s likely to eventually expand globally – Roche representatives at ATTD anticipated a US Solo launch at least two years following the complete EU launch. Solo, described as “a small and tubeless insulin delivery device operated through a remote control, which includes a BGM” was acquired from Medingo back in 2010. At ATTD, we learned from Roche representatives that a launch in certain EU pilot countries is expected by the end of 2018 – one representative anticipated an initial launch as early as this Summer or Fall. At the time, the Solo PDM was undergoing its last clinical trial, so it’s uncertain whether Roche will be able to hit this timeline. In late March, Roche posted a multi-site, randomized controlled participant-reported outcomes study for the Solo (n=180 adults with type 1) on CT.gov, currently listed as “not yet recruiting”. Pump-naïve participants will be randomized to either remain on MDI, initiate pump therapy with Solo, or initiate pump therapy with the Omnipod for 26 weeks. Subsequently, all participants will switch to Solo for an additional 13 weeks. Although the primary outcome will be treatment satisfaction for the Solo vs. MDI, treatment satisfaction comparing the Solo with the Omnipod will also be evaluated as a secondary outcome. It’s a pretty bold move to include a direct comparison to a competitor as established and well-received as Insulet, so we applaud Roche’s confidence and transparency. It’s also reassuring to see an early commitment to evaluating the user experience and (hopefully) incorporating feedback before an initial launch. On the downside, we’re a little disappointed to see this study limited to type 1s – we think getting user feedback from type 2s would be equally as, if not more, valuable, considering the size of the potential type 2 market that could benefit from a slim patch pump. Then again, there is a logic to starting with type 1s, who are more likely to be cut from the early adopter mold (even if they are pump-naïve). The launch of Medtronic’s MiniMed Pro-set with BD FlowSmart technology – paused due to a higher-than-anticipated rate of complaints – serves as a recent reminder of how important human factors and training are. This study is slated to end in July 2019, but we assume management has decided not to wait for a readout prior launch.

    *Red box added for emphasis

    • See below for a picture of the device from 2010, although the form factor and specs may have meaningfully changed since that time – we assume the handheld has been radically redesigned with a more modern look and feel.

    • Following the JDRF-/Helmsley Charitable Trust-funded meeting earlier this week, we’re still buzzing from the excitement surrounding interoperable automated insulin delivery components. Roche established itself at the forefront of this movement from the industry side, becoming the first company to publicly disclose participation in JDRF’s initiative to accelerate “Open Protocol” AID systems at ATTD, meaning that it will provide seamless, secure, interoperable connectivity with other devices and smartphone apps (e.g., Bluetooth). It seems likely Roche will also move to automating delivery with Solo over time. Hopefully, it will also pursue a De Novo “integrated pump” (“iPump”) classification, similar to the FDA’s recently created integrated CGM (iCGM) pathway with special controls, for Solo or a different Accu-Chek device. To be clear, the hypothetical iPump (i.e., an insulin delivery device preemptively cleared for interoperability) has not yet been created, but it was a hot button topic and FDA’s Dr. Courtney Lias seemed very open to it at the interoperability meeting. The consensus from regulatory, industry, DIY, academia, and other representatives in the room was that an iPump could be possible, but several questions remain (namely related to what special controls should be met). At this point, it seems like the best way forward would be for one interested pump company to take the regulatory plunge, as Dexcom did in requesting an iCGM designation. Be sure to read our expanded full report on the meeting for further details. Given Roche’s commitment to open protocols and active presence at the meeting, we would think that it would jump at the opportunity to develop an iPump first.

    • If high European demand for Insulet’s Omnipod is any indication, Solo could be a big product for Roche, especially if manufacturing can scale and Roche delivers on simple training and use. We wonder how much investment the company will put in hardware vs. software development moving forward, and how it will divide resources between its two pump platforms vs. BGM vs. CGM (Senseonics) vs. digital health (mySugr). Assuming Solo does indeed launch, Roche would be the only company to boast both a tubed and tubeless pump in-house, establishing quite the compelling portfolio for future AID systems down the line.       

    • As a reminder, Solo is equipped with a disposable 200-unit insulin reservoir (the only disposable piece, which is changed every ~three days), a reusable cannula “cradle” infusion set for insertion, a 90-day pump “base” (which stores pump settings) with adhesive that should be changed every three days, and a touchscreen remote. Solo is truly tubeless, distinguishing it from Cellnovo and Kaleido, which have short infusion sets, and is fairly low-profile on the body. However, it does have more on-body components than the Omnipod and is not quite all-in-one integrated. Eventually, per Roche representatives at ATTD, Solo will be controlled via an app directly on users’ smartphones – a huge win for Roche, as SOOIL’s Dana pump is the only option currently on the market with this feature, although Ypsomed reportedly is close behind.

    • The type 2 pump market is extremely underpenetrated and could be a promising avenue for Roche to explore with Solo. Only a handful of potential competitors exist in the US, and virtually none are present in the EU at the moment. BD’s Swatch type 2 patch pump and Insulet/Lilly’s U500 Omnipod (both with connectivity) are expected to launch by “September 2019” (BD) and in “2019” (Insulet) in the US. Meanwhile, Valeritas announced earlier this month that V-Go SIM (Simple Insulin Management), its durable accessory for the V-Go patch insulin delivery device, will launch in 1H19 in the US, providing one-way Bluetooth communication and Glooko integration. Other potential competitors including Cequr’s PAQ and J&J’s bolus-only OneTouch Via (neither of which bears connectivity) have yet to launch. There could be very high demand for the slim-looking Solo pump in the type 2 sector. Might Roche develop a different handheld for type 2, or wait until they can provide a mobile display/control app, before targeting this population?

    2. No Senseonics Mention; Roche Sponsoring Eversense XL RCT Investigating A1c and Time <54 mg/dl

    Though Roche has previously invested in Senseonics and distributes Eversense overseas, there was no mention of the CGM manufacturer during the call. After an FDA Advisory Committee voted unanimously in favor of the Eversense 90-day implantable CGM in March, we wouldn’t be surprised to hear word of approval any day now. As of Senseonics’ 4Q17 call, the company expects to be commercially ready for a US launch by mid-2018 (an ADA launch?), citing expected pricing of ~$10 per-day per-user and positive payer discussions thus far. Roche’s distribution territory is currently contained to Europe, the Middle East, and Africa, excluding Scandinavia, Finland, and Israel, and Senseonics has said it will go direct in the US with its own sales force. We continue to wonder if Senseonics is a Roche acquisition target in the future. Roche has certainly demonstrated a willingness to integrate external components (mySugr) and with a likely FDA approval, Senseonics is more attractive than ever. We’d also note that based on our read of FDA’s special controls for its new 510(k) pathway for iCGM, Senseonics is the only other CGM currently on the market besides the Dexcom G6 that appears to meet the high accuracy bar. Given these exciting prospects, Senseonics’ value could rise – with a market cap of $438 million, Roche could also remain a strong distribution partner.

    • Per Senseonics’ 4Q17 call, user feedback has been positive in the EU, with ~2,000 users reported in 4Q17 (+135% sequentially) and strong user retention (77% decided to re-insert following the first sensor). The 180-day wear Eversense XL is currently available in the UK and Sweden, with plans to convert the entire existing Eversense user base to XL by 3Q18. Excitingly, we also learned during Senseonics’ 4Q17 call that the company has been in discussions with the FDA concerning a US trial for the XL, slated to begin enrollment this summer.

    • Roche is sponsoring a two-arm, randomized controlled study of Eversense XL (n=324 adults with type 1 and type 2 diabetes) investigating A1c and time spent in hypoglycemia (<54 mg/dl). Cohort 1 will include insulin-using patients with A1c >8% randomized to receive either the Eversense XL or the usual SMBG or flash glucose monitoring (FreeStyle Libre) device used by participants. Cohort 2 will include only those with type 1 diabetes who spend on average >1.5 hours with sensor glucose <70 mg/dl for at least 28 days. Cohort 2 participants will be randomized in the same fashion as cohort 1. The multi-site study will be conducted in France and is currently listed on CT.gov as not yet recruiting. The estimated study completion date is December 2018.

    3. No Mention of mySugr or Decision Support Efforts; ~1.2-1.4 Million mySugr Users as of February, Half T2!

    There was no mention on the call of mySugr or in the supplementary materials, which Roche acquired for 65 million euros (~$80 million) in June. During a Roche symposium at ATTD, we learned that mySugr now has ~1.2-1.4 million registered users, half of which are type 2! Exactly how the company defines a “registered user,” has never been divulged. We also weren’t sure why multiple user base numbers were shared at ATTD. We also heard at ATTD and the recent Bay Area Diabetes Summit that mySugr has begun incorporating contextual insights into its app, providing real-time feedback on hyperglycemia/hypoglycemia patterns and potential ways to address them. In line with Roche’s commitment to providing value-based, device-agnostic, integrated diabetes care, the company also shared a slide at ATTD outlining how it intends to market the mySugr bundle, which is already reimbursed by major German payers, and US payers are reportedly expected to come online in 1H18 per JPM. The slide depicted Medtronic, Dexcom, Abbott, and Roche devices along with insights for population management. Roche is certainly all-in with mySugr – at ATTD Roche Global Head of Diabetes Care Mr. Marcel Gmünder called the app “our main interface and basis for interaction with patients” and “an engaging and intuitive tool,” while Mr. Jürgens added that it is “the leading management platform” and “basically a little clinic in the hands of the patient.” We continue to be optimistic regarding this reinvigorated Roche, although major news updates or a major success are still elusive at this stage.

    • At ATTD, Roche VP Mr. Tim Jürgens introduced the professional platform, which boasts features including: (i) pattern detection; (ii) “meaningful” graphs (data download for different devices; Ambulatory Glucose Profile); (iii) connectivity (one single hub for all devices); (iv) communication (more intensive monitoring of patients between face-to-face visits); and (v) population management (patients prioritized based on critical situation; real-time tracking of test strip consumption). Mr. Jürgens said at the time that this platform was to launch later in 2018, but we haven’t heard an update since. This strikes us as a competitor to Glooko – what will Roche add?

    4. No Updates on Senseonics/Roche/TypeZero iDCL Pivotal Trial, Set to Begin in 2H18 in EU

    There were no updates on the Senseonics/Roche/TypeZero long-term closed loop system iDCL pivotal trial. Last we heard (Senseonics 4Q17), the EU pivotal trial is now expected to begin in 2H18, slightly delayed from DTM (1Q18). We weren’t able to find the trial on ClinicalTrials.gov. With Prof. Hans de Vries moving to Dr. Lutz Heinemann’s Profil Institute, we wonder if he will remain the PI on the study, or if someone else will take over the helm.

    5. No-Frills Accu-Chek Instant S Launches in India in March; No Mention of Accu-Chek Guide on Call

    Although there was no mention during the call, Roche launched the Accu-Chek Instant S, a low-cost, simplified BGM designed for lower-resource settings, in India in March. Instant S has also been available in Pakistan since at least last September. While the apparently identical Accu-Chek Instant bG, which was listed as a key 2017 launch in the EU in the 4Q17 slide deck, is equipped with Bluetooth, the Instant S is not. It will be interesting to see how Roche distinguishes the two devices, as Accu-Chek Instant has been positioned similarly as a no-frills meter. We’re still not sure what the US plans are for the Instant bG, but in 3Q17, Roche told us the initial countries for rollout are Greece, Romania, and South Africa. There was also no mention of the Accu-Chek Guide, which launched in the US in May (and has expanded to at least five other countries). In the past, Mr. Diggelmann has expressed confidence in the Guide to “protect and safeguard this franchise,” and reminded attendees at JPM that strips are “the majority of the business.” We are big fans of the Guide’s accurate, low-cost strips and Simple Pay program – it could be driving much of the US growth noted above.

    Drug Pipeline Highlights

    1. Lucentis Sales Remain Flat YOY on Tough Comparison to 10% Growth; 37% Sequential Increase; FDA Approves 0.3 mg Prefilled Syringe

    Lucentis sales of 393 million CHF (~$403 million) remained flat YOY on a very tough comparison to 10% growth in 1Q17. Sales increased 37% sequentially, albeit from a record low base of 288 million CHF (~$291 million) in 4Q17. Roche Diagnostics Chief Operating Officer Mr. Daniel O’Day attributed the strong YOY growth to the successful launch of the first prefilled syringe, noting that growth for Lucentis is expected this year, although some quarterly fluctuation is anticipated as was previously observed in 2017. Lucentis was once again included on a slide listing the top 20 products for 1Q18 Pharma Division sales. A slide highlighting major granted and pending 2018 approvals noted that the 0.3 mg prefilled syringe for diabetes-related retinopathy (filed with the FDA in December) achieved US approval in March. Other highlights include first-in-class launches in mCNV (myopic choroidal neovascularization, a complication of near-sightedness) and diabetic retinopathy without DME (launches ongoing) plus the launch of the first prefilled syringe for wAMD (age-related macular degeneration) and macular edema after retinal vein occlusion.

    Figure 4. Global Q1 Lucentis Sales (1Q14-1Q18)


    2. Phase II LADDER Readout Expected in 2H18

    During prepared remarks, Mr. O’Day announced that Phase II data from the LADDER study (phase 2 Lucentis port delivery study; n=220) is expected for readout in 2H18. The study was included in the slide deck under a list of approved products seeking additional indications, with the status of “recruitment completed.” CT.gov still lists an estimated primary completion date of September 2018 and a final completion date of March 2019 for LADDER. Lucentis is also involved in three studies investigating the anti-VEGF/Ang2 bispecific antibody with dual mechanism of action (RG7716), all of which are indicated to have completed recruitment.

    3. Two Studies Investigating Roche’s FGF21 Candidate; No Further Updates

    The slide deck noted two studies investigating Roche’s FGF21 candidate RG7992: One Phase Ia study in obesity (n=79 adults with overweight/obesity and no type 2 diabetes, but suspected insulin resistance) was completed in March 2017, while another (Phase Ib) is still recruiting in type 2 diabetes (n=140) and is expected to complete in June. It’s unclear whether Roche will pursue a diabetes indication, an obesity indication, or both for this candidate (RG7992). Given that Lilly and Pfizer both discontinued FGF21 analogs in development for type 2 diabetes due to lackluster glucose-lowering results, we’re inclined to have more confidence in the weight loss prospects of this therapeutic target for now. Novo Nordisk also has an FGF21 analog in phase 1 for obesity.

    Selected Questions and Answers

    Q: In terms of the margin given, you've had a better quote on diabetes and indeed you seem to think that might be sustainable. Are you confident this will be the turning year for the diagnostics division margins?

    Mr. Roland Diggelmann (CEO, Roche Diagnostics): The first quarter [for Diabetes Care] was actually strong. It was a plus 5%. I did mention though that it came on the back of a very slow first quarter, particularly in the US, in 2017. So, you should not expect those gross sales to continue through the year. Diabetes Care continues to be a difficult market in terms of the pricing and particularly in the U.S. and we see large swings from quarter to quarters based on that.

     

    -- by Maeve Serino, Brian Levine, Adam Brown, and Kelly Close