Memorandum

Abbott’s FreeStyle Libre (real-time) obtains Medicare reimbursement for intensive insulin users; available immediately with same pricing, criteria as Dexcom’s G5 – January 4, 2018

Executive Highlights

  • Abbott announced minutes ago that Medicare will cover FreeStyle Libre (real time), having met the code for therapeutic CGM. Abbott’s 10-day wear sensor is immediately available via mail-order to all eligible Medicare beneficiaries with diabetes on intensive insulin therapy. With this news, FreeStyle Libre joins Dexcom’s G5, which had shipped to 4,000+ beneficiaries as of 3Q17. We are very enthusiastic about more people on Medicare gaining access to CGM, just as is true for the diabetes population as a whole. We do believe that CGM should be available to also those on SFUs, given the association with hypoglycemia, and the potential for safer diabetes management, particularly for the elderly.
  • The path to Medicare reimbursement apparently took just over three months, given that FDA approval for FreeStyle Libre came in late September. Though we figured Abbott would move fast, benefiting from all Dexcom’s and various advocates’ work on this front, CMS made the decision far faster than we expected. Indeed, Dexcom certainly helped pave the way, as it took ~two years for G5 to obtain an insulin-dosing claim from FDA and finally secure Medicare reimbursement in March 2017 – and this was viewed as major progress a year ago as that coverage also came faster than expected (and we credit departing CMS exec Andy Slavitt for signing the Executive Order that made this happen).   
  • The FreeStyle Libre bundle (reader, sensors, strips) will be priced the same as the G5 – ~$250 per month (~$8 per day) – despite a lower retail price of ~$4-$5/day. As we understand it, this means CMS is paying more for FreeStyle Libre than private insurers or even cash-paying patients. Presumably, this will change moving forward, but are excited to see CMS’s moves to propel CGM (whether intentional or unintentional), given the pricing decimation in BGM – it does make it even more important to show outcomes, which shouldn’t be difficult. We also see this as a clear positive for Dexcom, as the products will now be on the same pricing footing in Medicare and can compete on features and CGM should gain even more attention. We continue to believe that for multiple cohorts, alarms are extremely important, and for many, they are less so, and both products are miles ahead of traditional BGM for most patients, and certainly all patients on intensive insulin.  
  • With this pricing, the Medicare CGM market will certainly eventually hit an estimated >$1 billion-plus in sales, assuming penetration reaches >25% on a potential estimated market of ~1 million intensive insulin users (this market should also probably be far larger). Will CMS hold this pricing as the market expands? The glucose monitoring investment from Medicare certainly went down with the BGM pricing reduction and this announcement signals more government interest in diabetes investment. Given the better outcomes, we’re very excited for patients.
  • Abbott’s FreeStyle Libre US site has been updated, noting, “Most patients pay no more than $75 per month on eligible FreeStyle Libre prescriptions at major retail pharmacies.” This copay level indicates strong private payer coverage, as we expected. Abbott also just said via email that FreeStyle Libre is “as widely reimbursed commercially under medical benefit as other CGM systems,” and its goal is further accessibility through the pharmacy benefit channel. Payer conversations are underway. Globally, FreeStyle Libre is now fully or partially reimbursed in 21 countries (up from 17 in September).

In a huge win for the diabetes community and CGM expansion, Abbott just announced that Medicare will cover FreeStyle Libre (real-time), having met the codes for therapeutic CGM. Medicare patients with type 1 and type 2 diabetes on intensive insulin therapy will have access to FreeStyle Libre immediately via mail-order from DME suppliers (Edgepark Medical Supplies, Byram Healthcare, Solara Medical Supplies, Edwards Health Care Services, Better Living Now, and Mini Pharmacy), provided they meet the same requirements as we saw for Dexcom’s G5. That is, if they:

  • Have diabetes;
  • Have been using a home BGM and performing at least four checks per day;
  • Are insulin treated with MDI or a pump;
  • Have an insulin treatment regimen that requires frequent adjustment on the basis of therapeutic CGM results (likely most people on MDI, given the value of CGM and how difficult insulin is to dose).

We are very surprised by how quickly Abbott secured Medicare coverage for FreeStyle Libre, as we had guessed within 12 months would have been a victory; instead, it took just over three months, given that FDA approval for FreeStyle Libre came in late September. Obviously FreeStyle Libre’s non-adjunctive US label was the critical linchpin, and Dexcom paved the way at CMS –therapeutic CGM was classified as “Durable Medical Equipment” (DME) one year ago, and G5 reimbursement followed two months later last March. The battle was hard-fought by Dexcom for ~two years, with lots of help from various diabetes advocacy communities, particularly JDRF, and Abbott is now a fast follower in securing coverage for FreeStyle Libre. One year ago, we would not have guessed that two CGMs would be covered and available to Medicare patients! There are still issues being worked out – for example, the “Share” isn’t available with Dexcom yet – this will be a matter of time as we understand it (Abbott also doesn’t have LibreLink for CMS patients nor for any US patients).

The FreeStyle Libre Medicare bundle will include the reader (three-year life), 10-day sensors, and strips for the reader’s built-in FreeStyle BGM. We’re not sure how many strips Abbott will provide, as FreeStyle Libre doesn’t require calibration, but it does have a lengthy 12-hour warmup. (Dexcom’s G5 Medicare bundle includes two strips per day = 60/month.)

FreeStyle Libre will be reimbursed at the same pricing as Dexcom’s G5. This comes in accordance with a 16-page CMS Ruling letter released last January with pricing for therapeutic CGMs: A durable receiver is reimbursed at $236-$277, with a monthly price of $248.38 for sensors/transmitters ($2,981/year). This translates to ~$8/day per user pricing, nearly double what Abbott is currently receiving in the US retail pharmacy channel (~$4-$5/day for cash-paying users). This pricing certainly signifies a business win (through higher margins) for Abbott near-term though anything can happen longer term. With Medicare out-of-pocket costs identical between the two devices, the products can compete on features.

With this pricing, we expect the Medicare CGM market could easily eventually exceed an estimated >$1 billion in sales, assuming penetration reaches >25% on an estimated market of ~1.3 million intensive insulin users (sensitivity analysis below). We also believe penetration should be far higher, depending on education. Will CMS hold this pricing as the CGM market expands? We assume that for CMS –and for Abbott – it was easiest and fastest to put FreeStyle Libre into the existing reimbursement structure for Dexcom’s G5. Over time, it’s conceivable that CMS will reconsider this structure, especially because it is currently paying more for FreeStyle Libre than private insurers or even cash-paying patients in the pharmacy channel. On the other hand, if Abbott and Dexcom can show value, that level of pricing may be well justified – particularly compared to diabetes drugs and the cost of hospitalizations for severe hypoglycemia. We certainly think outcomes that have been shown warrant current pricing.

Ultimately, it is very positive to see CMS’s effort to cover CGM in type 1 and type 2 diabetes, unlike how it decimated pricing in BGM. We applaud the federal government for making this valuable investment in public health. Below, we explore potential Medicare CGM market dynamics, share a major FreeStyle Libre private insurance update, note CMS’s controversial prohibition of remote monitoring with therapeutic CGM (and what it means for LibreLink), discuss the therapeutic CGM landscape, and share our questions.

 

How big could the Medicare CGM market become? 1%-25% penetration in intensive insulin users = $40 million to $1 billion/yr

  • We look forward to watching how quickly G5 and FreeStyle Libre expand into the Medicare population eligible for CGM coverage. We estimate this market could easily be more than one million patients and potentially far more. Dexcom estimated Medicare-aged patients are ~20% of the overall type 1 market, translating to ~300,000 patients (assuming 1.5 million US type 1s). Meanwhile, there are ~12 million total people with diabetes on Medicare (per CMS data), and assuming 95% of them have type 2 – with an estimated one in ten qualifying as intensive insulin users (this is speculation on our part) – that translates to roughly 1.3 million people eligible for CGM coverage, assuming all are doing four fingersticks per day and can meet the other criteria. As people with diabetes stay older longer with improved management, these numbers will only grow. We also strongly believe that those on SFUs should qualify for CGM, given how associated SFUs are with hypoglycemia.
  • See below for our estimates on potential market size at various levels of penetration. For context, as of our 1H17 Roundup, we estimated the global CGM market would total ~$1.5 billion in 2017; the Medicare market alone could reach this level once CGM penetration hits ~25% in intensive insulin users – that of course assume pricing remains at the current level, which is a big “if.”

CGM Penetration and Annual Sales Based on
Estimated ~1.3 million Medicare Intensive Insulin Users

CGM Penetration

1%

5%

10%

25%

50%

Estimated Number of Medicare CGM Users

~13,000

~65,000

~130,000

~325,000

~650,000

Annual Revenue Based on ~$250/month pricing

~$40 million

~$200 million

~$400 million

~$1 billion

~$2 billion

  • While there is plenty of runway for both CGM products to be successful in the Medicare population, there are notable differences between G5 and FreeStyle Libre. How will the Medicare market segment? G5’s alarms make it the obvious choice for people with impaired hypoglycemia awareness or those experiencing hypoglycemia regularly who want the “safety net” of continuous communication CGM – this is likely even more common in the 65+ years crowd than in younger age groups, and will be a strong advantage for Dexcom until next-gen Libre is out. Dexcom has the big advantage of a two-hour warmup, while FreeStyle Libre won’t show data until 12-hours after insertion. (Unless a Medicare patient separately buys a FreeStyle Libre reader and overlaps wear times.) Abbott has the advantage on factory calibration, smaller on-body footprint, fully disposable design, longer labelled wear time, and easier education for HCPs – all will be strong in the Medicare population. Notably, these differences may well shrink pretty quickly: Dexcom’s slimmer G6 is currently under FDA review for both once/daily- and zero-calibration, while the Verily gen one sensor should follow in short order with a fully disposable design. Abbott has its second-gen CGM with continuous communication (enabling alarms) under development for use in Bigfoot’s ecosystem. It is unclear if Abbott’s second-gen Libre sensor will also be launched as a standalone device, though we’d guess it will (this is speculation).
  • Like the non-Medicare market, we expect FreeStyle Libre should raise category awareness in a very meaningful way, ultimately benefitting both organizations as well as CMS itself. For now, it’s very early days: as of early November, Dexcom had shipped 4,000+ Medicare bundles, and expected to complete an additional “close to 20,000” shipments by the end of 2017. Per the analysis above, this would imply just under 2% penetration of the (estimated) intensive insulin market, assuming our estimates are reasonable. The potential for both Dexcom and Abbott to change the standard of care for such a vast and growing population is big!

Libre Private (US) and OUS Insurance Update

  • In a brand-new update on Abbott’s US website, we’ve learned that “most commercially insured patients pay no more than $75 per month” (<$2.50/day) for FreeStyle Libre at major retail pharmacies. This is very strong copay pricing for those with private insurance out of the gates. Via email, the company also told us that Libre is “as widely reimbursed commercially under medical benefit as other CGM systems” and that they intend to further expand accessibility through the pharmacy benefit channel. We’re not sure how coverage varies by plan (likely widely at this early stage), but this certainly puts FreeStyle Libre very much on par with Dexcom’s private payer reimbursement for many patients – and ahead for those with high coinsurance for DME supplies. We’ll be fascinated to see if it encourages Dexcom to move to the pharmacy faster, especially since the 10-14-day wear G6 will give the company more room to budge on pricing.   
  • Notably, the website also shows, and Abbott confirmed, that there is no prior authorization required for FreeStyle Libre in the US. This is obviously a huge win as it lifts the administrative hassle and time necessitated by PAs, and could grease the wheels to promote much higher prescription rates.
  • Globally, FreeStyle Libre is now fully or partially reimbursed in 21 countries (up from 17 in September). One of those is obviously the US, but this means that three additional markets have begun covering Libre in the past four months. We assume none of them are as large as UK, Japan, or France – all of which got a dedicated press release – but the continued momentum is a positive sign and we expect this to be reflected in user base growth over 2018.

Remote Monitoring Status

  • As a reminder, to ensure CMS reimbursement compliance, Dexcom had to turn off app communication for Medicare G5 transmitters, disabling remote monitoring and data sharing. Dexcom CEO Mr. Kevin Sayer has previously noted multiple irons in the fire to include remote monitoring for Medicare beneficiaries – there has not been a timing update on this, though perhaps we’ll hear one at JPM or in the 4Q17 call. If Dexcom can figure this out near-term, it will be a sizeable Medicare product advantage over Abbott. (It is extremely frustrating for patients and providers that this problem exists – we do expect it to change but ideally it would never have been an issue).
    • The status of FreeStyle Libre’s Android reader app, LibreLink, is currently unknown in the US for both the Medicare and non-Medicare populations – it was not included in the initial FDA submission, and Abbott has not commented on submission or launch timing. This means that patients have to use the handheld “reader” to scan for data and cannot share readings with caregivers via LibreLinkUp. We imagine it won’t be a too long (we are defining this broadly, however) before we see LibreLink and LibreLinkUp approved and implemented in the commercial pathway, though Abbott is obviously far behind Dexcom on this front in the US.  Notably, Dexcom has direct-to-Apple-Watch communication under FDA review, in addition to smartphone apps for iOS and Android already out.
    • As noted, this Medicare coverage criteria remains a source of enormous frustration for many users. Perhaps combined pressure from Dexcom, Abbott, and the patient community – one recently told us that signing up for therapeutic CGM through Medicare is “like signing a suicide note” – will be sufficient to result in change.

Therapeutic CGM landscape – What about Senseonics and Medtronic?

  • There is a chance that Senseonics might also secure a non-adjunctive claim, which would pave the way for possible Medicare coverage. Senseonics’ implantable Eversense CGM system has been submitted to FDA with an adjunctive label (it also requires two fingersticks calibrations per day), and CEO Mr. Tim Goodnow has said that he doesn’t want to put the primary approval at risk by going for a non-adjunctive claim. Management hopes for an “early 2018” approval following an Advisory Committee meeting expected this quarter; the company has not yet shared if this meeting will occur. If the first-round approval is for adjunctive use, we assume a secondary submission could obtain a non-adjunctive indication, given Eversense’s very strong accuracy in its US pivotal study – an MARD of 8.8% over 90 days. Presumably Senseonics would need to provide a durable receiver, as the current product is app-only (at least that we know of).
  • We’re guessing (but aren’t positive) that Medtronic’s Guardian Connect standalone mobile CGM will remain adjunctive initially. The sensor is currently under review with a US launch expected by April. While the 670G modulates basal insulin delivery based solely on the Guardian Sensor 3’s reading, the 670G/Guardian Sensor 3 label is explicit in stating that a confirmatory fingerstick is always to be used when bolusing. We’d guess the standalone mobile version will have a similar indication. The sensor has accuracy in the range of Dexcom and Libre (MARD of ~10%), but three to four calibrations are recommended daily.

Close Concerns’ Questions

Medicare Reimbursement-Specific Questions

Q: How much pent up demand is there for the Libre in the Medicare population, and how quickly will FreeStyle Libre Medicare shipments ramp up? It took Dexcom some time to get off the ground due to administrative hassle (e.g., BGM bundling, prescribers have to dig up records of four fingersticks/day), and Dexcom had only shipped a few “test cases” by early August following the March reimbursement victory. However, as of 3Q17, Dexcom has shipped the G5 to over 4,000 Medicare beneficiaries. Will Abbott encounter a similar delay? 

Q: How will Medicare beneficiaries and their providers weigh the comparative benefits and drawbacks of Dexcom’s G5 and Abbott’s FreeStyle Libre? Who will see more traction in the Medicare population in 2018, accounting for Dexcom’s head start?

Q: How much pricing risk is there for CGM in the Medicare populations? For now, it’s potentially eyebrow-raising that the government is paying Abbott more than private insurers and cash-paying patients – will that continue, especially if adoption expands dramatically? In a perfect world, the companies would both be paid for their sensors based on the outcomes they delivered. Will CMS move in this direction within five years?

Q: Will Abbott conduct a real-world study for the Medicare population? Real-world outcomes for the Libre outside of the US have been exceptionally strong, and demonstrating Libre’s value in the Medicare population could drive CGM penetration and reimbursement status.

Q: How will the pathways to LibreLink differ for the commercial and Medicare populations? Has LibreLink been submitted to FDA? When will approval come? When it does, will Medicare beneficiaries with Android phones be allowed to use it? How will Abbott’s app and data efforts compare to Dexcom and Medtronic? What is Abbott’s Apple iOS and Apple Watch strategy for FreeStyle Libre?

Other Questions

Q: How strictly will CMS continue to assess eligibility for therapeutic CGM? Dexcom has reported hassle associated with the process – might the government eventually relax protocol? Will Medicare be overwhelmed with demand and claims?

Q: Will Senseonics’s Eversense receive FDA approval as a non-adjunctive CGM on the primary submission? Will CMS reimburse the implantable CGM, or will Senseonics have additional hoops to jump through (e.g., providing a receiver)?

Q: Will elevated visibility of FreeStyle Libre impact FreeStyle Libre Pro professional CGM sales in a positive way? We imagine it will, given the provider enthusiasm for Pro and the sizeable fraction of Medicare patients who are not intensive insulin users. Professional CGM is reimbursed for Medicare, though we’re not sure how much administrative hassle it requires. Abbott’s FreeStyle Libre Pro has a very strong product advantage over Dexcom’s G4 Professional CGM and Medtronic’s iPro2.

Q: How quickly will Abbott obtain a pediatric indication from FDA? 

Q: What can be expected from advocacy groups? We expect advocacy groups to begin to request CGM use for additional groups beyond those on intensive insulin management. Those on SFUs is a very logical request, given the high rates of hypoglcyemia associated with SFU use. In turn, there may be lobby for more patients to gain access to glycemic-dependent medications like DPP-4 and SGLT-2 inhibitors as well as GLP-1. What studies would be required? How far out is this?

Q: What other changes in patient cohorts may happen? Presumably, with so many patients with diabetes (nearly 50%) not in adequate control, many more would benefit from intensive insulin therapy. We expect we may see more patients move to this with the benefit of far easier and better monitoring. This would be a positive for both companies (probably more Abbott given greater ease of use currently) as well as possibly insulin manufacturers. 

 

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