COWEN
`
`AND COMPANY
`
`Equity Research
`Health Care
`
`Therapeutic Categories Outlook
`Comprehensive Study
`
`February 2015
`
`Alzheimer’s Disease
`Bone Diseases
`Cardiovascular
`Central Nervous System
`Dermatology
`Diabetes
`Epilepsy
`Gastrointestinal/Ulcer
`Hepatitis B Virus/Hepatitis C Virus
`Infectious Disease
`Multiple Sclerosis
`Obesity
`Oncology/Hematology
`Ophthalmology
`Orphan Diseases
`Pain Management
`Respiratory
`Rheumatology
`Sleep Disorders
`Urinary Incontinence
`Women’s Health
`
`Please see addendum of this report for important disclosures.
`
`www.cowen.com
`
`v
`
`
`
` @CowenResearch
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`
`
`+9% CGR 2014-19E
`
`Diabetes
`
`Diabetes: Many New Drugs, But Insulin Still Dominant
`
`Diabetes is characterized by the inability to produce insulin and/or a dysregulated
`response to insulin. Insulin is a hormone that regulates metabolic substrate utilization
`and promotes the cellular uptake and conversion of sugars and starches into
`biochemical energy. The cause of diabetes is not fully understood, although both
`genetics and environmental factors such as obesity and physical inactivity appear to
`play roles. If left unregulated, diabetes-related metabolic sequela can result in organ
`damage involving the nervous system, kidneys, eyes, immune system, and
`cardiovascular system.
`
`There are an estimated 382MM people worldwide with diabetes. There are 24MM
`children and adults in the United States, or 8% of the population, who have diabetes.
`An estimated 17.9MM (75%) have been diagnosed with the disease, approximately
`90% with type 2 diabetes. While the pathogenesis of type 2 diabetes is not completely
`understood, early stages are associated with euglycemia (normal blood sugar levels)
`maintained by compensatory hyperinsulinemia in the setting of peripheral insulin
`resistance (metabolic syndrome). Prolonged elevations in circulating insulin result in
`the down-regulation of insulin receptors in muscle and adipose, increasing insulin
`resistance and hyperglycemia, and further increasing insulin secretion (non-insulin
`dependent diabetes mellitus). In the latter stages of type 2 diabetes, the pancreas is
`no longer able to produce enough insulin to compensate for extreme insulin
`resistance and beta-cells begin to die. At this stage patients often require exogenous
`insulin to maintain appropriate blood glucose concentrations (insulin-dependent
`diabetes mellitus). Type 2 diabetics are currently managed with diet, exercise, oral
`antidiabetes agents, and insulin when necessary. The use of insulin is increasing
`among type 2 patients as oral agents fail to get patients to goal (HbA1c <7%) and
`insulin becomes easier to administer. It is estimated that by 2030 the worldwide type 2
`prevalence will have grown from 190MM to 330MM patients.
`
`Five to ten percent of diabetics have type 1 diabetes, which is a state of absolute
`insulin deficiency stemming from autoimmune destruction of the insulin-producing
`beta-cells in the pancreas. Patients with type 1 diabetes produce little or no insulin
`and are dependent on insulin injections for survival. A small percentage of diabetics
`who appear to have type 2 diabetes actually have a slowly progressing form of type 1
`diabetes and require insulin therapy.
`
`478
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`
`
`Diabetes Category Market Share By $ Sales
`
`Diabetes
`
` I
`
`2019P
`$54B
`
`Other
`9%
`
`GSK
`2%
`
`NVO
`32%
`
`JNJ
`3%
`
`AZN
`8%
`
`MRK
`13%
`
`LLY
`15%
`
`SNY
`17%
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`2014
`$36B
`
`Other
`11%
`
`JNJ
`1%
`
`Takeda
`2%
`
`BMY
`3%
`
`LLY
`13%
`
`NVO
`30%
`
`4i0
`
`MRK
`17%
`
`SNY
`23%
`
`Source: Cowen and Company
`
`
`
`MAJOR TRENDS & ISSUES
`
`In 2014, Novo Nordisk, Sanofi, Merck, and Eli Lilly led the $36B diabetes category. In
`2019, we forecast that these same four companies plus AstraZeneca (post its January
`2014 acquisition of Bristol’s diabetes assets) will dominate a projected $54B market,
`driven by growth of their insulins and GLP-1 products.
`
` Insulin remains the cornerstone of diabetes treatment with an estimated 2014 WW
`market of $20B. Sales growth will be driven by increased penetration of insulin
`analogs resulting in an estimated market of $25B by 2019 (5% CAGR). Sanofi’s
`Lantus is the leading basal insulin but competitive basal insulins and biosimilar
`glargine may be available by 2015-16. In August 2014 the FDA issued a tentative
`approval of BI/LLY insulin glargine (Basaglar). The approval is subject to an
`automatic hold as a result of a litigation filed by Sanofi that claims patents
`infringement. Merck is also stepping up its biosimilar glargine efforts by running
`two Phase III non-inferiority studies versus Lantus (both type 1 and 2 diabetes) with
`estimated primary completions in March and May 2015. A number of novel long
`acting basal insulins are in development; Novo’s degludec (Tresiba) is rolling out in
`the E.U., Japan and may file in the U.S. H1:15 if DEVOTE (CV study) interim analysis
`is positive. Filing of Lilly’s novel basal insulin (Peglispro) has been delayed to after
`2016 as new clinical data is required to clear hepatotoxicity concerns. Sanofi’s
`Toujeo concentrated glargine formulation allows for smaller volumes of
`administration and awaits regulatory approval in Q1:15. Sanofi hopes to expand the
`Lantus market opportunity. Eli Lilly’s Humalog and Novo’s Novalog split the short-
`acting market; new competitors have had very little impact.
`
` Oral DPP-IV inhibitors reduce the breakdown of GLP-1 and have become successful
`as add-on therapies, usually with metformin. 2014 WW market is valued $8.8B and
`is estimated to grow to $12.6B by 2019 (CAGR 7%). Merck’s Januvia (sitagliptin) has
`performed well, but faces competition from AstraZeneca’s Onglyza (saxagliptin),
`BI/Lilly’s Tradjenta (linagliptin), and Takeda’s Nesina (alogliptin). Novartis’ Galvus
`(vildagliptin) has done well OUS. Takeda and Merck have once weekly oral
`compound (trelagliptin and omarigliptin) that were filed in Japan in March and
`November 2014.
`
` Our physician consultants view the oral DPP-IVs as less efficacious than GLP-1
`analogs given their inability to engage GLP-1 in a wide variety of tissues. Head-to-
`
`www.cowen.com
`
`479
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`head studies vs. GLP-1 have shown DPP-IVs to produce inferior lowering of HbA1c.
`Despite their limited efficacy, DPP-IV success has been driven by their ease of use.
`Onglyza’s SAVOR CV outcomes trial met its primary safety endpoint, but failed to
`improve CV outcomes; Januvia’s TECOS CV trial is expected to have data available
`for presentation at ADA 2015. Our physician experts give DPP-IV inhibitors only a
`10-15% chance of meeting CV endpoints.
`
` GLP-1 receptor agonist enhance pancreatic insulin secretion and reduce hepatic
`glucose production. 2014 WW market is valued $3B and is estimated to grow to $9B
`by 2019 (CAGR 24%). Sales growth of Astra’s GLP-1s Byetta and Bydureon remains
`sluggish. Safety concerns have emerged for the class including drug-induced
`pancreatitis and a possible link to pancreatic neoplasms. FDA and EMA have largely
`dismissed a causal relationship between GLP-1 agonist and pancreatic cancer,
`although safety data from CV outcome trials will be key. Scrip data suggest
`endocrinologists are putting more new patients on Novo’s Victoza, but that it has
`grown the GLP-1 market only modestly. Sanofi’s Lyxumia (lixisenatide), has been
`launched in the E.U. but submission was pulled in the U.S. until Phase III trial
`complete which is expected for Q3:15; its differentiation will be in combo with
`Lantus. Lilly’s Trulicity (dulaglutide), approved in late 2014 appears to have very
`good efficacy and acceptable safety. Glaxo’s Tanzeum/Eperzan (albiglutide) appears
`to offer no differentiating features.
`
` Latest entrants to the oral diabetes treatment portfolio are the SGLT-2s. 2014 WW
`market is valued $574MM and is estimated to grow to $3.6B by 2019 (CAGR 44%).
`JNJ’s Invokana, approved March 2013, has enjoyed first to market advantage. AZN’s
`Farxiga, which was approved in January 2014, provides competition. Lilly/BI’s
`Jardiance (empagliflozin) was approved in the U.S. and E.U. in late 2014.
`Merck/Pfizer’s ertugliflozin is in Phase III. In early 2015 BI/Lilly first-in-class SGLT-
`2/DPP-4 combination (Glyxambi) was approved by the FDA and AZN is expected to
`file a similar combination (saxagliptin+dapagliflozin) in early 2015.
`
` Our scatter plot shows that, through 2019, we expect the diabetes therapies of Novo
`Nordisk, Sanofi, Eli Lilly, AstraZeneca and Merck will contribute significantly to
`sales.
`
`480
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`n
`
`•
`
`NVO
`
`Diabetes
`
`•
`
`MRK
`
`•
`
`LLY
`
`SNY
`
`Diabetes Scatter Plot
`
`125%
`
`•
`
`AZN
`
`_
`••
`
`NVS
`
`JNJ
`
`••
`•
`
`Takeda
`GSK
`
`BMY
`
`% Of Company 2014-19 Sales Growth From Category
`
`100%
`
`75%
`
`50%
`
`25%
`
`0%
`
`-25%
`
`$0.0
`
`$2.0
`
`$4.0
`
`$6.0
`
`$8.0
`
`$10.0
`
`$12.0
`
`$14.0
`
`$16.0
`
`$18.0
`
`2019 Sales Contribution By Company To Category ($ In B)
`
`
`Source: Company data; Cowen and Company
`
`
`
`
`
`www.cowen.com
`
`481
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`■
`
`Estimated Worldwide Market For Diabetes/Metabolic Drugs By Class ($MM)
`
`Drug Class
`Insulins
`DPP-IV Inhibitors
`
`2014
`Market
`$20,015
`8,773
`
`% Total
`56%
`25%
`
`2019P
`Market
`$25,178
`12,557
`
`% Total
`46%
`23%
`
`$
`14-19
`CGR
`5%
`7%
`
`NRx
`‘87-14
`CGR Comments
`-1% - LLY, NVO and SNY dominate
`NM - MRK's Januvia, NVS' Galvus, AZN's Onglyza
`BI/LLY's Trajenta
`
`GLP-1 Analogs
`
`2,992
`
`8%
`
`8,939
`
`16%
`
`24%
`
`SGLT2 Inhibitors
`
`574
`
`2%
`
`3,585
`
`7%
`
`44%
`
`Sulfonylureas
`Glitazones
`Other Oral Agents
`Total Market
`
`571
`349
`2,345
`$35,620
`
`2%
`1%
`7%
`100%
`
`473
`195
`3,290
`$54,217
`
`1%
`0%
`6%
`100%
`
`-4%
`-11%
`7%
`9%
`
`NM - AZN's Byetta/Bydureon, Novo's Victoza, SN
`Lyxumia; LLY's Trulicity, GSK's Tanzeum
`
`NM - AZN's Forxiga; JNJ's Invokana; LLY/BI's
`Jardiance
`
`NM - Various therapies
`NM - Takeda's Actos
`9% - Glucophage/metformin
`8% - Driven by GLP-1 agonists, DPP-IV inhibito
`SGLT2 inhibitors
`
`
`
`
`
`Source: Cowen and Company estimates
`
`These agents are used as mono or combination therapies depending on HbA1c levels
`and resistance to previous therapies. AACE guidelines (below) suggest mono therapy if
`HbA1c is around 7.5% and no previous therapies, dual therapy if HbA1c>7.5% and
`resistant to monotherapy, and triple therapy if resistant to dual therapy and
`HbA1c>9.0%.
`
`482
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`American Association of Clinical Endocrinologist (AACE) diabetes treatment guidelines
`
`LIFESTYLE MODIFICATION
`
`(Including Medically Assisted Weight Loss)
`
`ONOTHERA PY
`
`127 ALP.. RA
`Qo Ai
`(5 AG.,
`
`TZO
`
`II Alc >6.5%
`in 3 %%ha add
`and dump
`ANA 1.rapyl
`
`O
`
`ENTRY Ai c a 7.5%
`
`DUAL THERAPY
`
`GLAI RA fl/il
`
`WAN.
`
`TIP
`
`Baal
`
`MET
`
`cca.....in
`
`in'nl
`
`alb
`A
`,,
`
``ale
`
`• Dula of medications INA/area suggetted hiAorthy of uNge
`• ' Based %on phase ldlnral vials data
`
`NO SYMPTOMS
`
`S MP 0
`
`DUAL
`THERAPY
`
`DR
`
`TRIPLE
`THERAPY
`
`SULIN
`OTHER
`AGENTS
`
`ADD OR INTENSIFY INSULIN
`
`LEGERE,
`
`- ?..grbirtp=
`
`A
`
`TRIPLE THERAPY
`
`NA., AA o f
`
`LTD
`
`' 501.2
`
`Spalinstdit
`
`DPP% e)
`
`MET
`IA Nor
`
`ANN
`
`enmanptneM e
`AG.,5
`
`SU/GLIN
`
`not at gaol in :I.
`month. orate.,
`to or mtensily
`oorolin Theron
`
`O
`
`PROGRESSION
`
`OF DI SEASES
`
`
`Source: AACE, Reprinted with permission from American Association of Clinical Endocrinologists. Garber AJ, Abrahamson, MJ, Barzilay
`JI, et al. AACE Comprehensive Diabetes Management Algorithm. Endocr Pract. 2013;19:327-336
`
`CopyoRbto nu Ma
`
`mcs
`
`Various ongoing trials are studying the long-term CV effects of using insulin analogs,
`DPP-4 inhibitors, GLP-1 agonist and SGLT-2 inhibitors. Tresiba’s DEVOTE (interim
`look), Januvia’s TECOS and Jardiance’s EMPA-OUTCOMES are all expected to readout
`in 2015.
`
`Long-term CV studies and estimated data readouts
`
`Company
`
`Drug
`
`Class
`
`Study name
`
`n
`
`7,644
`
`Data readout
`
`H1:15 (interim)
`
`NVO
`
`MRK
`
`Tresiba
`
`Januvia
`
`Insulin
`
`DPP-4
`
`DEVOTE
`
`TECOS
`
`BI/LLY
`
`Jardiance
`
`SGLT-2
`
`EMPA OUTCOME
`
`NVO
`
`JNJ
`
`Victoza
`
`GLP-1
`
`Invokana
`
`SGLT-2
`
`DPP-4
`
`GLP-1
`
`SGLT-2
`
`GLP-1
`
`BI/LLY
`
`Tradjenta
`
`AZN
`
`AZN
`
`LLY
`
`Bydureon
`
`Farxiga
`
`Trulicity
`
`Source: Company data
`
`
`
`LEADER
`
`CANVAS
`
`CARMELINA
`
`EXSCEL
`
`DECLARE
`
`REWIND
`
`14,000
`
`H1:15 (ADA:15)
`
`7,000
`
`9,340
`
`4,365
`
`8,300
`
`14,000
`
`17,150
`
`9,622
`
`Q3:15
`
`H1:16
`
`H2:17
`
`H1:18
`
`H1:18
`
`H1:19
`
`H1:19
`
`
`
`DETAILED DISCUSSION
`
`Injectable Insulin: The Cornerstone Of Therapy
`
`There are an estimated 11-12MM type 1 diabetics worldwide (approximately 1.6MM in
`the U.S.), and most are on insulin therapy. Among U.S. residents ages 65 and older,
`26.9% had diabetes in 2010 with 13-14MM total diabetics in the U.S. According to
`CDC statistics, 14% of all U.S diabetics are on insulin alone and another 13% are on a
`combination of insulin and oral therapy. Patients with type 2 diabetes require higher
`doses of insulin than patients with type 1 diabetes because of their resistance to its
`
`www.cowen.com
`
`483
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`action. Given the enormous population of type 2 diabetics worldwide (90MM+), the
`market for insulin is large and should continue to grow, even with the availability of
`newer agents that may delay or reduce the need for insulin. This trend will be driven
`by increasing standards of living in the developing world combined with life style
`trends and increasing obesity in developed nations.
`
`There are several types of insulins available defined by their length of action: rapid-
`acting (Humalog, Novolog, and Apidra), short-acting (Humulin, Novolin, and
`Velosulin), intermediate-acting (NPH and Lente), and the long-acting (Ultralente,
`Lantus, and Levemir). Some forms are available as premixed combinations. The insulin
`market changed substantially following the introduction of Sanofi’s Lantus, a once-
`daily basal insulin analog, which has become the basal insulin of choice for type 1
`diabetics and the dominant add-on therapy for type 2 patients failing oral anti-diabetic
`agents.
`
`484
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Comparison Of FDA Approved Insulins
`
`Diabetes
`
`■
`
`Type of Insulin &
`Brand Names
`Rapid-Acting
`Humalog or lispro
`Novolog or aspart
`Apidra or glulisine
`
`Short-Acting
`Regular (R) humulin or
` novolin
`Velosulin (for use in the
` insulin pump)
`
`Intermediate-Acting
`NPH (N); NovolinN,
` Humalin N
`
`Lente (L)
`
`Long-Acting
`
`Ultralente (U)
`
`Company
`
`Onset
`
`Peak
`
`Duration
`
`Role in Blood Sugar Management
`
`Lilly
`NovoNordisk
`Sanofi
`
`15-30 min.
`10-20 min.
`20-30 min.
`
`30-90 min
`40-50 min.
`30-90 min.
`
`3-5 hours
`3-5 hours
`1-2½ hours
`
`Rapid-acting insulin covers insulin needs for
`meals eaten at the same time as the injection.
`This type of insulin is used with longer-acting
`insulin.
`
`30 min. -1 hour
`
`2-5 hours
`
`5-8 hours
`
`30 min.-1 hour
`
`2-3 hours
`
`2-3 hours
`
`Short-acting insulin covers insulin needs for
`meals eaten within 30-60 minutes
`
`NovoNordisk,
`Lilly
`
`1-2 hours
`
`4-12 hours
`
`18-24 hours
`
`1-2½ hours
`
`3-10 hours
`
`18-24 hours
`
`Intermediate-acting insulin covers insulin
`needs for about half the day or overnight. This
`type of insulin is often combined with rapid- or
`short-acting insulin.
`
`30 min.-3 hours
`
`10-20 hours
`
`20-36 hours
`
`Lantus (glargine)
`
`Sanofi
`
`1-1½ hour
`
`No peak time;
`insulin is delivered
`at a steady level
`
`20-24 hours
`
`Long-acting insulin covers insulin needs for
`about 1 full day. This type of insulin is often
`combined, when needed, with rapid- or short-
`acting insulin.
`
`Levemir (detemir)
`
`NovoNordisk
`
`1-2 hours
`
`6-8 hours
`
`Up to 24 hours
`
`Pre-Mixed*
`14-24 hours
`2-4 hours
`30 min.
`Lilly
`Humulin 70/30
`Up to 24 hours
`2-12 hours
`30 min.
`NovoNordisk
`Novolin 70/30
`Up to 24 hours
`1-4 hours
`10-20 min.
`NovoNordisk
`Novolog 70/30
`18-24 hours
`2-5 hours
`30 min.
`Lilly
`Humulin 50/50
`16-20 hours
`30 min.-2½ hours
`15 min.
`Lilly
`Humalog mix 75/25
`*Premixed insulins are a combination of specific proportions of intermediate-acting and short-acting insulin in one bottle or insulin pen (the numbers
` following the brand name indicate the percentage of each type of insulin).
`
`These products are generally taken twice a day
`before mealtime.
`
`
`
`Source: Adapted from WebMD; Cowen and Company
`
`
`
`
`
`www.cowen.com
`
`485
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`US INSULIN MARKET
`
`1--. ----1-1--F
`
`• A"-•-•-•-•-•-a-A
`
`Dec-14
`
`Sep-14
`
`Jun-14
`
`Mar-14
`
`Dec-13
`
`Sep-13
`
`Jun-13
`
`Mar-13
`
`Dec-12
`
`Sep-12
`
`Jun-12
`
`Mar-12
`
`Dec-11
`
`40.0%
`
`35.0%
`
`30.0%
`
`25.0%
`
`20.0%
`
`15.0%
`
`10.0%
`
`5.0%
`
`0.0%
`
`Market Share
`
`Humulin Group
`
`Humalog Group
`--M--
`
`--A--
`
`Novolin Group
`
`Lantus/Solostar
`--%--
`
`-if--
`
`Apidra/Solostar
`
`-III-
`
`Levemir
`
`-_,-
`
`Novolog Group
`
`Human Insulin
`
`Source: IMS Health
`
`Lantus Still Dominant Basal Insulin, But Sanofi Preparing For The Future
`
`Lantus (insulin glargine) is a once-daily, long-acting basal insulin injection used for
`type 1 diabetes and type 2 diabetes. Lantus offers lower hypoglycemia risk and
`improved metabolic control over 24-hours basal insulin coverage, while limiting peak
`concentrations. The efficacy benefits provided by Lantus (dosed via a once-daily
`subcutaneous injection) in combination with mealtime fast-acting insulin injections
`raised the benchmark for glucose control in type 1 diabetics. Lantus is the leading
`branded insulin in the U.S. and in many major European markets. The SoloStar
`prefilled disposable pen, launched in 2007, has been a major growth driver for the
`Lantus franchise. Lantus is the add-on treatment of choice in patients with type 2
`diabetes failing oral agents. An estimated 80% of current U.S. Lantus patients are type
`2 diabetics. Lantus cannot be mixed with short-acting insulin, but this has proven to
`be only a minor issue.
`
`Lantus is marketed in more than 70 countries, including Japan. Sanofi pays Novo an
`undisclosed royalty on Lantus sales due to cross-licensing of patent rights. Lantus is
`covered by a composition patent expiring in February 2015. Sanofi has implemented
`many efforts over the years to protect its insulin franchise including bundling diabetes
`care products into packages for certain emerging markets; competing on price when
`necessary; and use with SoloStar (prefilled disposable insulin pen). More recently,
`Sanofi has filed its nex-gen version of Lantus, Toujeo (U-300), for which regulatory
`decisions are expected in the U.S for Q1:15 and in E.U. and Japan for Q2:15.
`Development of lixisenatide (GLP-1) both alone and in combination with Lantus (with
`intellectual property protecting the single agent and combination, possible filing by
`year end 2015), and legal action to delay biosimilar competition (see next paragraph).
`Lantus sales were €6.34B in 2014 and we estimate sales of €6.72B in 2015, €6.28 in
`2016 and €4.82B in 2019.
`
`
`
`
`
`486
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`Sanofi Files Suits Against Lilly, Likely Protecting Lantus Until Mid-2016 In The U.S.;
`E.U. Competition On Tap For 2015
`
`Lilly announced on December 20, 2013 that it had filed a 505(b)(2) application seeking
`approval of its insulin glargine biosimilar (Basaglar). On January 30, 2014, Sanofi filed
`suit against Lilly, prompting a 30-month stay of litigation, based on formulation
`patents that extend well beyond the February 2015 composition patent. The 505(b)(2)
`filing allows a variant of the innovator product to be approved, but allows the
`challenger to use innovator data to support the filing. In July 2014, Sanofi filed an
`additional lawsuit accusing Lilly of infringement on seven patents related to Lantus
`and the SoloStar pen. The trial is scheduled for September 2015.
`In September 2014, BI/Lilly’s insulin glargine product (Abasria) was approved in the
`E.U. It could launch in mid-2015 post the 2/15 patent expiry.
`
`U.S. Patent Situation
`
`Lantus patents expiring in 2023 and 2024 cover a range of formulation parameters,
`including pH (1-6.8), inclusion of a preservative, and addition of stability agents such
`as polysorbate and polyhydric alcohol. Whether or not Lilly has formulated around
`these parameters is a key unknown, but it is likely Sanofi selected them for specific
`reasons. For instance, it would appear that an acidic pH is key to the Lantus
`formulation and Sanofi appears to have covered the entire range.
`
`Sanofi filed the litigation in Delaware, where there are four active judges with
`extensive experience in pharmaceutical issues. Historically stays have gone the entire
`30 months, and it is likely that the Lantus stay will as well. The only exception might be
`if there is a clear situation of non-infringement, e.g., a completely different
`formulation, but even in that case, none of these four judges is likely to make a quick
`decision. In about 60% of pharmaceutical patent cases before this court, the innovator
`has prevailed, although most of these are small molecule cases. The success rate is
`probably lower when only 505(b)(2) cases are considered. Sanofi is likely to assert a
`variety of claims early on, and they will be narrowed later. Ultimately the broadest
`claims on which validity can be argued will be brought to trial.
`
`Lantus Failed To Demonstrate CV Benefit In ORIGIN, But Cancer Concerns Alleviated
`
`Sanofi presented results from Lantus’ CV outcomes trial (ORIGIN) at ADA 2012. In
`ORIGIN, patients treated with Lantus did not show a statistically significant difference
`in the incidence of CV events compared to standard of care. However, Lantus delayed
`the progression from pre-diabetes to diabetes and a cancer signal was not observed.
`The lack of a CV benefit for Lantus, while a disappointment, is unlikely to change the
`outlook for Lantus given that ORIGIN was a landmark trial with broad implications for
`the insulin class.
`
`ORIGIN was a six-year trial to evaluate the impact of Lantus on CV outcomes in 12,500
`patients with pre-diabetes or early type II diabetes with high CV risk. 6,264 patients
`were initiated on Lantus therapy and titrated to achieve fasting normoglycemia. The
`co-primary endpoints for ORIGIN were the composite of CV death, non-fatal MI, or
`non-fatal stroke (co-primary #1); and the composite of CV death, non-fatal MI non-
`fatal stroke, revascularization, or hospitalization for heart failure (co-primary #2).
`
`Normalizing blood glucose with Lantus did not improve CV outcomes when compared
`to standard of care (co-primary #1: HR=1.02, p=0.63; co-primary #2: HR=1.04,
`p=0.27). Our physician consultants believe that normalizing blood glucose with insulin
`is important to maintaining endothelial health and preventing vascular complications.
`However, significant heterogeneity across the ORIGIN population may have masked
`
`www.cowen.com
`
`487
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`I
`
`secondary effects from independent risk factors and contributed to the negative
`outcome of this study.
`
`Sanofi Seeks To Expand Lantus Opportunity With Toujeo
`
`Sanofi’s new glargine formulation, Toujeo, is designed to provide a unique PK/PD
`profile and to allow for a lower injection volume. Sanofi expects Toujeo to target both
`high dose insulin users with type II diabetes and users administering more typical
`doses. A formulation that allows for insulin granules to be packed more densely
`should allow for greater amounts of insulin to be delivered in smaller volumes. Patents
`describe the new Lantus formulation as being more concentrated than the current
`formulation (300mg/mL vs. 100mg/mL). Given the increased glargine concentration in
`the new Lantus, a greater amount of Lantus precipitates at neutral pH than in the
`existing formulation and provides a unique PK profile. This is exemplified by more flat
`glucose infusion rates compared to Lantus, suggesting >24hrs duration of action for
`Toujeo.
`
`The patent on Toujeo was filed in 2011 and examination by the USPTO is ongoing. The
`patent covers a wide range of glargine concentrations. Our legal consultants believe
`that the dissolution of the new Lantus formulation at neutral pHs is what would be
`expected by increasing the glargine concentration and does not believe the new
`Lantus patent application is likely to be granted. A patent was granted in the EU in
`July 2014 with an expiry in 2031. Toujeo has also been granted patents in Australia,
`Japan and other smaller markets.
`
`Management indicated that it expects growth for basal insulin to be driven by
`switches in emerging markets (from pre-mixed insulin to basal) as well as the
`continued growth in new-to-insulin patients elsewhere. Toujeo was filed in the U.S.
`and E.U. in April 2014 and in Japan in July 2014. The FDA accepted the filing on July 8,
`2014 and the EMA in May 2014. Sanofi expects regulatory decisions for U.S and E.U.
`in H1:15. We estimate Toujeo sales of €180MM in 2015, €500MM in 2016, and
`€1,400MM in 2019.
`
`New Insulin Glargine Formulation Reduces Depot Surface Area By Half
`
`Lantus®
`
`
`
`Source: Company data
`
`Pooled Meta-Analysis Demonstrates Reduced Hypoglycemia
`
`Data presented at ADA 2014 showed Toujeo to be as effective at HbA1c control as
`Lantus (and with slightly less weight gain and a lower incidence of hypoglycemia). A
`pooled meta-analysis of three Phase III EDITION trials in T2DM presented at ADA
`2014 showed a decrease in nocturnal hypoglycemia of 31% (as measured by event
`
`488
`
`www.cowen.com
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`rate per patient year across the 6 month study period) and a 14% decrease in
`hypoglycemia at any time. The reduction in nocturnal hypoglycemia was observed
`during the initial titration phase (first 8 weeks, 31% reduction) as well as after the
`titration phase (week 9 to week 24, 20% reduction). Sanofi believes that the
`hypoglycemic reduction post titration is particularly meaningful and reflects the Toujeo
`smoother PK/PD profile compared to Lantus. Sanofi believes that patient concerns
`over hypoglycemia are a key contributor to insulin non-compliance. However, our
`physician experts have been more cautious on Toujeo, citing concerns that patient
`doing errors of the more concentrated product could magnify hypoglycemic reactions.
`
`Latest Data For Toujeo Solid But Not Spectacular
`
`In 2013 and 2014, Sanofi released complete data for the Phase III EDITION trials
`comparing Toujeo to Lantus. The EDITION studies enrolled both type 1 and type 2
`diabetes patients at different stages of the disease. In type 2 trials, EDITION III
`enrolled insulin naïve patients only on oral anti diabetic drugs (OAD); EDITION II and
`EDITION JP2 enrolled patients on OADs and basal insulin; and EDITION I enrolled
`patients on OADs, basal insulin and mealtime insulin. In type 1 trials, EDITION 4 and
`EDITION JP1 enrolled patients on basal and mealtime insulin.
`
`Results from EDITION III (n=878) were presented at ADA 2014 and showed
`comparable improvement in glycemic control (HbA1c -1.42% vs -1.46), a slightly
`lower weight gain for Toujeo (+0.50 Kg vs +0.71 Kg) and a comparable nocturnal
`hypoglycemia rate post titration phase (RR 0.89, p=0.454).
`
`Results from EDITION II (n=811) were presented at ADA 2013 and subsequently
`published in Diabetes Care. The data showed comparable improvement in glycemic
`control (HbA1c -0.78% for Toujeo and -0.57% for Lantus), a lower weight gain for
`Toujeo (+0.10 Kg vs +0.66 Kg) and a better nocturnal hypoglycemia rate post titration
`phase for Toujeo (21.6% vs 27.9%, p=0.038).
`
`Results from EDITION I (n=807) were also presented at ADA 2013 and subsequently
`published in Diabetes Care. The data showed a comparable reduction in HbA1c (-
`0.88% vs -0.86%), comparable weight gain (0.9 kg for both) and a better nocturnal
`hypoglycemia rate post titration phase for Toujeo (37.4% vs 46.5%). While a
`statistically significant reduction in nocturnal hypoglycemia is positive, our clinical
`consultants believe the absolute number of severe hypoglycemic events is low on
`Lantus and that reductions afforded by novel formulations are of limited clinical utility.
`
`In both EDITION IV (type 1 diabetes; n=549) and EDITION JP1 (type 1 diabetics in
`Japan; n=243), Toujeo demonstrated non-inferiority to Lantus in HbA1c reduction at 6
`months. In EDITION IV, rates of any-time hypoglycemia were similar between Toujeo
`and Lantus. However, Toujeo demonstrated a statistically significant 31% relative
`reduction in nocturnal hypoglycemia vs Lantus (7.8% vs 11.2%). Adverse events for
`Toujeo and Lantus were reported as similar in all studies. Sanofi believes that the
`potential market for Toujeo is 6 million people considering the 5 million currently on
`basal insulin and the 1 million that initiates basal insulin every year.
`
`Novo’s Levemir Making Only Modest Gains On Lantus
`
`Novo Nordisk’s Levemir (detemir), launched in the U.S. in March 2006. Levemir has
`improved its market share modestly over time (16.4% NRx share in November 2014,
`+60% Y/Y) although has failed to make significant progress where it competes with
`Lantus. This is despite weight-gain data that may be differentiating. However, in a
`
`www.cowen.com
`
`489
`
`Mylan Ex.1088
`Mylan v. Sanofi - IPR2018-01676
`
`

`

`Cowen and Company
`
`Conferences
`
`Therapeutic Categories Outlook
`February 2015
`
`
`Diabetes
`
`direct comparison to Levemir, Lantus was shown to have activity levels more than four
`times greater than Levemir during the period from 12 to 24 hours after administration.
`The same study showed a marked and highly significant difference in terms of
`duration of action: Lantus showed near 24-hour coverage whereas Levemir had a
`duration of action of only 17.5 hours.
`
`Following the Lantus Diebetologia publications, Novo stated that all its insulin analogs
`have been tested for IGF-1 receptor binding in the early research phase and only
`insulins with a binding ratio between the insulin and IGF-1 receptors similar to, or
`better than, that of human insulin were accepted for further development. Studies on
`receptor binding have shown that Levemir in comparison to human insulin has a
`relative affinity to the IGF-1 receptor, which is equal to or slightly lower than the
`insulin receptor. Novo points out that Levemir distinguishes itself from Lantus, which
`has been shown to have increased affinity for the IGF-1 receptor compared to human
`insulin. Post the ORIGIN data, this argument is much less likely to hold weight
`amongst the endocrinology community.
`
`Novo’s Tresiba Marketed In E.U. And Japan; Interim Data For U.S.-Required CV Study
`Expected Early 2015; Filing Timing Uncertain
`
`Degludec (Tresiba) is Novo’s ultra-long-acting basal insulin analogue that is dosed
`once daily. Degludec’s 40 hour duration of action provides a prolonged steady state,
`reducing peaking concentrations, and hypothetically reducing the risk of
`hypoglycemia. Tresiba was approved in December 2012 in Japan (where
`reimbursement is similar to Lantus) and in January 2013 in the E.U. Tresiba has been
`launched in 15 countries through June 2014. In December 2014 Tresiba received a
`positive CHMP opinion for the added indication of treatment of children 1-17 years old
`with diabetes.
`
`Novo also developed Ryzodeg (combination of degludec +insulin aspart in a 70/30
`mix) which can be dosed twice a day to allow for dosing at main mealtime. Ryzodeg is
`approved in the E.U. and Japan. In September 2014, Novo presented data from a
`Phase IIIb study at EASD which compared Ryzodeg (twice daily) to basal insulin (once
`daily) plus 2-4 injections of insulin aspart. At 26 weeks, Ryzodeg