analysis of alternatives - europa. analytical method transfer ... declaration. analysis of...

ANALYSIS OF ALTERNATIVES and SOCIO-ECONOMIC ANALYSIS

i

ANALYSIS OF ALTERNATIVES

and

SOCIO-ECONOMIC ANALYSIS

Non-confidential report

Legal name of applicant(s): Novartis Ringaskiddy Limited

IE Ringaskiddy, County Cork

Submitted by: Novartis Ringaskiddy Limited

IE Ringaskiddy, County Cork

Substance: Diglyme (Bis(2-methoxyethyl)ether),

CAS No 111-96-6

Use title: Use of diglyme as solvent in the manufacturing process of an intermediate for further conversion into a pharmaceutical compound used in medicinal products for treatment of respiratory diseases.

Use number: 1


Use number:1 Legal name of the applicant: Novartis Ringaskiddy Limited

ii

CONTENTS

LIST OF ABBREVIATIONS ............................................................................................................................ vi

Glossary............................................................................................................................................................... viii

DECLARATION ................................................................................................................................................. ix

1. SUMMARY .................................................................................................................................................. 1

2. AIMS AND SCOPE OF THE ANALYSIS .............................................................................................. 3

3. APPLIED FOR “USE” SCENARIO ......................................................................................................... 4

3.1. Analysis of substance function ........................................................................................................ 4 3.1.1. The substance..................................................................................................................................... 4 3.1.2. Purpose and benefits of diglyme ..................................................................................................... 4 3.1.3. The diglyme specific production process ...................................................................................... 5 3.2. Market and business trends including the use of the substance .................................................. 7 3.2.1. Annual tonnage .................................................................................................................................. 8 3.3. Remaining risk of the “applied for use” scenario ......................................................................... 8 3.4. Human health and environmental impacts of the applied for use scenario ............................... 8 3.4.1. Number of people exposed .............................................................................................................. 8 3.5. Monetised damage of human health and environmental impacts .............................................. 8

4. SELECTION OF THE “NON-USE” SCENARIO .................................................................................. 9

4.1. Efforts made to identify alternatives .............................................................................................. 9 4.1.1. Research and development .............................................................................................................. 9 4.1.2. Data searches ................................................................................................................................... 13 4.2. Assessment of shortlisted alternatives.......................................................................................... 15 4.2.1. Category 3 alternatives ................................................................................................................... 15 4.2.2. Category 2 alternatives: xylene, sulfolane, and 1,2,3,4-tetrahydronaphtalene ....................... 15 4.2.2.1. Substance ID, properties, and availability .................................................................. 16 4.2.2.2. Technical feasibility of n-butyl acetate ....................................................................... 17 4.2.2.3. Economic feasibility and economic impacts of n-butyl acetate .............................. 20 4.2.2.4. Availability of n-butyl acetate ...................................................................................... 20 4.2.2.5. Hazard and risk of n-butyl acetate ............................................................................... 20 4.2.2.6. Conclusions on n-butyl acetate..................................................................................... 21 4.2.3. Category 1 Alternative: n-butanol ................................................................................................ 22 4.2.3.1. Substance ID, properties, and availability .................................................................. 22 4.2.3.2. Technical feasibility of n-butanol ................................................................................ 23 4.2.3.3. Economic feasibility and economic impacts of n-butanol ........................................ 24 4.2.3.4. Availability of n-butanol ............................................................................................... 25 4.2.3.5. Hazard and risk of n-butanol ........................................................................................ 25 4.2.3.6. Conclusions on n-butanol .............................................................................................. 26 4.3. The most likely non-use scenario ................................................................................................. 26



iii

5. IMPACTS OF GRANTING AUTHORISATION ................................................................................ 28

5.1. Economic impacts ........................................................................................................................... 28 5.2. Human Health or Environmental Impact ..................................................................................... 29 5.3. Social impacts .................................................................................................................................. 29 5.4. Wider economic impacts ................................................................................................................ 30 5.5. Distributional impacts .................................................................................................................... 30 5.6. Uncertainty analysis ........................................................................................................................ 30

6. CONCLUSIONS ........................................................................................................................................ 31

6.1. Comparison of the benefits and risk ............................................................................................. 31 6.2. Information for the length of the review period .......................................................................... 32

7. SUBSTITUTION EFFORT TAKEN BY THE APPLICANT ............................................................ 33

7.1. Summary .......................................................................................................................................... 33 7.2. Factors affecting substitution ........................................................................................................ 35 7.2.1. International supply chain of indacaterol containing products ................................................. 35 7.2.2. Identification of alternatives and process change ....................................................................... 36 7.2.3. Technical and organisational ......................................................................................................... 37 7.2.4. Regulatory Challenges ................................................................................................................... 38 7.3. List of actions and timetable with milestones ............................................................................. 41 7.3.1. Initiation and implementation of substitution process ............................................................... 41 7.3.1.1. DD activities ................................................................................................................... 42 7.3.1.2. CPD activities ................................................................................................................. 43 7.3.2. Regulatory tasks .............................................................................................................................. 45 7.3.2.1. Analytical method transfer ............................................................................................ 45 7.3.2.2. Change Control and Change Request activities ......................................................... 46 7.3.2.3. Other tasks ....................................................................................................................... 46 7.3.2.4. Stability testing ............................................................................................................... 47 7.3.2.5. Core Documentation preparation and submission ..................................................... 47 7.3.2.6. Core Regulatory Documentation review period ........................................................ 49 7.3.3. Overall time-line of diglyme substitution .................................................................................... 53 7.4. Monitoring of the implementation of the substitution process ................................................. 55 7.5. Conclusions ...................................................................................................................................... 56

8. REFERENCES ........................................................................................................................................... 57



iv

TABLES

Table 1: Impacts for the applied for use and the non-use scenario for a review period of 7 years. 2

Table 2: Substance of this AoA. 4

Table 3: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 5

Table 4: List of tested diglyme alternatives 14

Table 5: Overview on yield and area results for category 3 alternatives. 15

Table 6: Overview on feasibility of category 2 alternatives compared to the benchmark of diglyme. 16

Table 7: Substance IDs and properties of n-butyl acetate (European Chemicals Agency: www.echa.europa.eu) 17

Table 8: Scoring system evaluation of diglyme and n-butyl acetate in comparison. 18

Table 9: Isolated yields and impurities for n-butyl acetate compared to the benchmark of diglyme. 19

Table 10: Classification and labelling of n-butyl acetate (European Chemicals Agency: www.echa.europa.eu). 20

Table 11: Substance IDs and properties of n-butanol (European Chemicals Agency: www.echa.europa.eu) 22

Table 12: Scoring system evaluation of diglyme and n-butanol in comparison. 23

Table 13: Isolated yields and impurities for n-butanol compared to diglyme. 24

Table 14: Classification and labelling of n-butanol (European Chemicals Agency: www.echa.europa.eu). 25

Table 15. Minimum sales revenue in Euro. 28

Table 16: Uncertainties on economic impacts. 30

Table 17: Comparison of impacts for the applied for use and the non-use scenario. 31

Table 18: Impacts for the applied for use and the non-use scenario for a review period of 7 years. 32

Table 19: XXXXXXXXXXXXXXXXXXXXXX 41












Table 31: Tasks and time-frames from regulatory submission to the competent health authority until approval 51



v

FIGURES

Figure 1: Results from the SciFinder search for the keyword “green solvents” in Journals from dated 08/01/2013. Novartis, 2013. 9

Figure 2: NPV of 7 years of value added foregone due to production shutdown. 29

Figure 3: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 35

Figure 4: Summary and timeframe until global approval process is accomplished 52

Figure 5: Overall timeline of diglyme substitution 54

Figure 6: XXXXXXXXXXXXXXXXXXXXXXXXX 55



vi

LIST OF ABBREVIATIONS

ACS GCIPR American Chemical Society Green Chemistry Institute Pharmaceutical Roundtable

AoA Analysis of Alternatives

API Active Pharmaceutical Ingredients

CHF Swiss Franc (currency)

COPD Chronic obstructive pulmonary disease

CPD Commercial production division

CR Change request

CSR Chemical Safety Report

CTD Common technical document

DD Development division

DMC Development and Manufacturing Committee

DNEL Derived no effect level

DoE Design of Experiments

DP Drug product

DS Drug Substance

EEA European Economic Area

EMA European Medicals Agency

EU European Union

FDA Food and Drug Agency

(c)GMP (Current) Good Manufacturing Practices

GTI Genotoxic impurities

HSE Health, Safety and Environment

ICH International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use

LABA Long-acting beta-2 agonist

MAH Marketing Authorization Holder



vii

NDA New Drug Application

NPV Net Present Value

NRL Novartis Ringaskiddy Ireland Limited

PharmOps Pharma Operations

PMDA Pharmaceuticals and Medical Devices Agency

QA Quality Assessment

QRA Quality Risk Assessment

REACH Registration, Evaluation, Authorisation and Restriction of Chemicals

Reg CMC Regulatory Chemistry Manufacturing Control

RoW Rest of World

SP Substitution process

SVHC Substance of Very High Concern

USD United States of America currency (US-Dollar)

VOC Volatile Organic Compound

ZRA Zurich Risk Assessment



viii

Glossary

API

APIs are substances or combination of substances used in a finished pharmaceutical product (FPP), intended to furnish pharmacological activity. The API might be the substance that has direct effect in restoring, correcting or modifying physiological functions in human beings. API are often referred to as drug substance (DS).

CR Manufacturing Process changes requiring control are generally documented as CR. In this document the applicant for the change proposes the type of grade-evaluation of the change and specifies the time frames.

Drug product (DP)

Drug product is defined as a finished dosage form (e.g. tablet, capsule, solution), that contains an API, generally, but not necessarily, in association with one or more other ingredients.

EMA European Union agency responsible for the protection of public and animal health through the scientific evaluation and supervision of medicines.

FDA The FDA is an agency within the U.S. Department of Health and Human Services with the core functions: Medical Products and Tobacco, Foods and Veterinary Medicine, Global Regulatory Operations and Policy, and Operations.

(c)GMP

A practices that conforms to the respective guidelines for manufacturing, testing, and quality assurance to ensure that a drug product is safe for human use. Many countries have legislated that pharmaceutical and medical device manufacturers follow GMP procedures and create their own GMP guidelines that correspond with their legislation.

ICH

The International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) is a project that brings together the regulatory authorities of Europe, Japan and the United States and experts from the pharmaceutical industry in the three regions to discuss scientific and technical aspects of pharmaceutical product registration.

NDA The vehicle in the United States through which drug sponsors formally propose that the Food and Drug Administration (FDA) approve a new pharmaceutical for sale and marketing.

PDMA Japanese regulatory agency for pharmaceuticals and medical devices, working together with Ministry of Health, Labour and Welfare. The aim is to protect the public health by assuring safety, efficacy and quality of the respective products.

Process Development Laboratory (PD lab)

Process development laboratory at Novartis manufacturing site where the adapted manufacturing process is performed at small scale prior to performing the large scale manufacturing process.

QRA

The QRA identifies and quantifies the operational parameters and associated ranges of the process with respect to the quality of the product. The QRA document and the underlying experimental results are auditable. Chemical production uses the QRA document as a basis for the validation of a production process.

Yield Amount of product synthesised in a chemical reaction.



ix

DECLARATION



1

1. SUMMARY

Use of diglyme at Novartis Ringaskiddy Limited

Diglyme (Bis(2-methoxyethyl)ether; CAS No 111-96-6) is a well-established organic solvent used by Novartis Ringaskiddy Limited in one particular pharmaceutical synthesis: the production process of indacaterol. Indacaterol is a long-acting ß2 adrenergic agonist (LABA). Indacaterol is an active ingredient in a number of Novartis medicinal products. All products are approved for the treatment of COPD.

Diglyme is used at one Novartis production facility in the EU in one production synthesis step for the batch wise production of indacaterol. The synthesis is performed within enclosed equipment in accordance with Good Manufacturing Practices (GMP). Diglyme and other solvents are introduced into the reactors by trained personnel using appropriate protective equipment and via transfer systems designed to minimize unintentional release. The annual tonnage band for the use of diglyme in API production is 1-10 tonnes per year.

Significant effort has been undertaken by Novartis since 2013 according to its ‘Green Solvent Policy’ with the goal to identify and replace solvents with a high hazard potential, including diglyme. Novartis and its affiliates performed a detailed search for alternatives to diglyme in the production process of indacaterol. Based on Novartis’ existing expertise, 16 standard solvents were identified as potential alternatives to diglyme. In a multi-step process all alternatives were tested and the ones that revealed significantly reduced yield or failed other key requirements, were not further investigated (category 3). Substances that provided acceptable yield and good selectivity in the reaction mixture were further investigated (category 1 and 2). Category 2 alternatives are potential candidates concerning the impurity profile and yield but have constraints regarding their ecological innocuousness. Category 1 alternatives are suitable to substitute diglyme and were tested for the parameters, impurity profile, yield and ecological innocuousness. The outcome of this process was that n-butyl acetate was identified as a suitable alternative for diglyme.

The present analysis of alternatives (AoA) demonstrates that the substitution from diglyme to n-butyl acetate in the synthesis of indacaterol is technically feasible. Novartis already initiated the solvent replacement process towards using n-butyl acetate instead of diglyme. For details on this process, the reader is referred to the substitution process (SP) as explained in detail in chapter 7.

Expected time-scale of diglyme substitution and length of the review period

When a medicinal product is registered and marketed, it must be ensured that it continues meeting the requirements for quality, safety and efficacy in accordance with regulations of the different health authorities of the corresponding markets worldwide. Therefore, changes to pharmaceutical manufacturing processes require manifold organisational, technical and regulatory tasks over a long time-period of several years, with respect to compliance with the extensive and diverse regulatory requirements worldwide associated with such changes. Altogether technical and regulatory tasks have to verify and document, under GMP conditions, that the change has no influence on the product in terms of quality, efficacy and safety.



2

Taking the technical readiness level and the extensive regulatory evaluation process into account, it can be concluded that the substitution of diglyme cannot be fully implemented before the sunset date, especially because of the timeframes for validation of the adapted manufacturing process and for world-wide submission, review and approval by health authorities over which Novartis has no influence. Since the sunset date for diglyme is in August 2017, a review period of 7 years until 2024 is necessary to validate the adapted manufacturing process and to obtain the necessary approvals from health authorities worldwide.

Further justification for this review period is outlined in chapter 7 (SP) and SEA. The SEA clearly demonstrates the benefits of permitting the continued use of diglyme, especially in circumstances where Novartis intends to substitute diglyme right after the approval of the regulatory bodies is granted.

Human health and environmental impacts of the applied for use scenario

The chemical safety report (CSR) provides a detailed assessment of exposure relating to the “applied for use scenario” and demonstrates that there are no adverse human health or environmental impacts. Since it is demonstrated that worker exposure for diglyme is clearly below the Derived no effect level (DNEL) and no release into the environment is measurable, the risks to human health – expressed in monetary terms – is zero.

Comparison of benefits and risks

The combined assessment of impacts was performed to estimate the overall costs and benefits. The uncertainty analysis (see section 5.6) demonstrates that the effects of uncertainties on the overall result of the SEA are low. The impacts for the two scenarios (use and non-use) are described in chapter 6.1. It could be clearly demonstrated that the benefits of a granted authorisation outweigh the risks. The following Table 1 summarises the quantitative impacts based on a review period of 7 years. Health and environmental impacts have to be considered to be zero over the whole review period, therefore even low socio-economic impacts clearly outweigh the risks.

Table 1: Impacts for the applied for use and the non-use scenario for a review period of 7 years.

Type of impact € million

Value added Xxxx) (45-75)

Health and environmental impacts 0

Net benefits of a granted authorisation Xxxx (45-75)

In conclusion, Novartis identified a replacement substance which is currently being implemented into the pharmaceutical manufacturing process taking extensive technical, organisational and regulatory requirements into account. A review period of 7 years is needed to successfully finalize these tasks and get the necessary global approvals. The SEA underlines that for the requested review period the socio-economic impacts clearly outweigh the risks.



3

2. AIMS AND SCOPE OF THE ANALYSIS Diglyme (Bis(2-methoxyethyl)ether; CAS No 111-96-6) is a well-established organic solvent used in one particular pharmaceutical synthesis: the production process of indacaterol. Indacaterol, a LABA, is the active pharmaceutical ingredient (API) contained in medicinal products registered by Novartis under different trade names for the treatment of COPD in more than 110 countries worldwide.

Diglyme was added to REACH Annex XIV in August 2014 and therefore should be substituted by alternative solvents. In this AoA it will be demonstrated that a detailed search for replacement solvents was performed by Novartis in recent years trying to identify alternatives for diglyme in the production process of indacaterol. The most promising solvents were evaluated for their technical and economic suitability against the ECHA authorization criteria.

It will be shown that beside the technical evaluation of alternative solvents, other factors, such as the industrial scale up, the reevaluation of the whole synthesis and most importantly regulatory requirements, must be taken into account when setting up the substitution strategy for diglyme. This is especially the case as the solvent is used for production of an API, where highest requirements must be fulfilled to ensure health and safety of patients.

The substitution strategy (including current and projected research and development effort) is described in detail in chapter 7.

This SEA demonstrates that benefits of a granted authorisation by far outweigh the risk for health and the environment by continuing the use of diglyme. Economic, social, wider economic as well as distributional impacts are considered in this analysis. Additionally, market and business trends for a highly competitive market are described. A qualitative uncertainty analysis is provided to determine the influence of uncertainty factors on the overall result.



4

3. APPLIED FOR “USE” SCENARIO

3.1. Analysis of substance function

3.1.1. The substance

The following substance is subject to this AoA.

Table 2: Substance of this AoA.

# Substance Intrinsic property(ies)1 Latest application date2 Sunset date3

1 Diglyme (Bis(2-methoxyethyl)ether, DEGDME) EC No: 203-924-4 CAS No: 111-96-6

Toxic for reproduction (category 1B)

21 February 2016 22 August 2017

1 Referred to in Article 57 of Regulation (EC) No. 1907/2006 ² Date referred to in Article 58(1)(c)(ii) of Regulation (EC) No. 1907/2006 3 Date referred to in Article 58(1)(c)(i) of Regulation (EC) No. 1907/2006

Diglyme is categorized as a Substance of Very High Concern (SVHC) and is listed on Annex XIV of Regulation (EC) No 1907/2006. Adverse effects are evaluated in detail in the CSR.

3.1.2. Purpose and benefits of diglyme

Diglyme is a versatile substance with dipolar aprotic properties which is used in a variety of applications. The most important applications are:

- Solvent for synthesis of many chemicals and pharmaceuticals, - In extraction, distillation, and purification processes, - In the production of plastic and rubber products including magnetic polystyrene beads, - In the production of binding agents, - In sealed batteries as solvent of electrolytes, - In polytetrafluoroethylene etchant solutions, and - Filling and packaging for Scientific Research and Development.

Only the first mentioned use is subject to this application: “solvent for the synthesis of Active Pharmaceutical Ingredients (API) at one industrial site”.

The key functionalities of diglyme are based on the following desirable physicochemical properties:

- Boiling point: 162 °C - Vapour pressure at 20 °C: 3 mmHg - Excellent chemical stability even at high temperatures (e.g. above 100 °C). Thus it is ideal for

reactions at high temperatures. It also shows a high stability under basic conditions and a moderate stability under acidic conditions.

- Highly selective ligand with improved reactivity. - Highly solubilizing substance in which the intermediate ingredients and APIs are able to go

into solution, even though they mostly display a poor or limited solubility in classical organic solvents.

- Chemical inertness towards reaction compounds: one of the base compounds used in the process reacts with a number of more electrophilic solvents.



5

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx. Diglyme is used in only one of sixteen working steps in the comprehensive and complex production of indacaterol. This diglyme-specific production process is illustrated in the following.

3.1.3. The diglyme specific production process

Diglyme is used at one production facility based in the EU in one process chain, in batch wise production. The manufacture of indacaterol is performed within enclosed equipment in accordance with Good Manufacturing Practices (GMP). Diglyme and other solvents are introduced into the reactors via transfer systems designed to minimize release, by trained personnel using appropriate protective equipment.

Diglyme is handled in an enclosed system while charging it into the head tank and dosing it into the reaction vessel. The vessels and transfer lines are all hard piped and flanged with the exception of a manifold flexible hose which connects head tank to the reaction module. The product is discharged to closed flexible containment bags. These modules are also hard piped with exception of the manifolds flexible hoses and connection to mobile filter dryer. The integrity of the equipment is regularly maintained and verified e.g. by preventative inspections and leak testing.

Substances involved xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

Table 3: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Abbrev. Full name Process role

xxx xxxxxxxxxxxxxxx xxxxxxxxxxxxx

xxx xxxxxxxxxxxxxxx xxxxxxxxxxxxx

xxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxx

xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxx

x xxxxx Xxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx xxxxxxx

x xxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxx

X xxxxxxxxxxxxxxxxx Xxxxxxxxxxxxxxxxxxx

Xxxxxxx Xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Xxxxxxxxxxxxxxxxxxx

Xxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Xxxxxxxx



6

Abbrev. Full name Process role

xxx Xxxxxxxxxxxxxx xxxxxxx

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Process description xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Using diglyme as solvent for this reaction is critical to control genotoxic impurities which need to be kept to a level of 100 ppm for the final product (drug substance, DS). Therefore, the alternative solvent must not have adverse quality impacts on the final DS. This requires verification by a research and development campaign prior to going into production with an alternative solvent.

Regulatory Requirements Change / Approval Process Bringing pharmaceutical substances from conception to the market is a complex and stepwise process which can last decades. Most importantly, it must be ensured that the medicinal product meets the requirements for quality, safety and efficacy. Therefore, pharmaceutical substances require an approval from regulatory bodies to ensure the conformity of the product with all relevant quality, safety and efficacy regulations.

In order to follow for example GMP, manufacturers must comply with extensive quality and documentation requirements. Additionally, production conditions which are established must be reproducible. The latter includes suitable rooms, certified facilities, trained personnel and documentation. These requirements are not only valid for the first time a medicinal product is synthesized but for the whole life-cycle of a medicinal product or pharmaceutical process. In order to make a change to an existing manufacturing process (such as the replacement of a solvent in a synthesis step) it has to be shown that the amended procedure still meets the requirements in terms of



7

quality, safety and efficacy. Replacement of diglyme is considered a regulatory relevant change and a regulatory submission will need to be made to all relevant health authorities where the medicinal product containing the API is approved. To fulfil this need, an integrated team approach is necessary for the revalidation of the synthesis. This includes expertise from a variety of disciplines (e.g., process engineering, industrial pharmacy, analytical chemistry, microbiology, statistics, manufacturing, and quality assurance).

The company must notify the respective health authorities world-wide about the proposed change of the formerly approved manufacturing process as it must be shown that this change does not have an influence on the quality of the final medicinal product. Applicants must therefore assess possible effects carefully to ensure that the API still meets requirements for quality, safety and efficacy as justified during approval as new drug entity.

In principle, this procedure applies for health authorities worldwide. Changes within a manufacturing process of an API are generally considered significant as described above. Data requirements and timelines for the authorities´ evaluation process vary drastically between countries. Consequently, final implementation of these changes can take up to 5 years after preparation of the core submission documents, until the last of the global approvals is obtained, as described in more detail in the SP. In summary, in order to replace diglyme as a solvent within the synthesis of an API such as indacaterol, an extensive approval process for requalification must be passed before the drug can be manufactured with the amended process.

3.2. Market and business trends including the use of the substance

The API indacaterol is used for the Novartis brands globally known under different tradenames. The medicinal products were launched in 2009 and 2013, both for the treatment of COPD. COPD, a common preventable and treatable disease, is characterised by persistent airflow limitation that is usually progressive and associated with an enhanced chronic inflammatory response in the airways and the lung to noxious particles or gases. Exacerbations and comorbidities contribute to the overall severity in individual patients (Global Initiative for Chronic Obstructive Lung Disease, 2011). To date over 1.5 million patients with COPD have been treated, or are currently being managed, with the innovative bronchodilator therapies containing indacaterol. Another product which also contains indacaterol as active ingredient but at lower dose strength was approved in the US in 2011. XXX xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx. Currently COPD is the 5th leading cause of death worldwide and is expected to become the 3rd by 2020. Novartis is known as an innovative market leader in this field, nevertheless facing a very competitive market situation as many new original and generic compounds enter the market.

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.



8

3.2.1. Annual tonnage

The annual tonnage band for the use of diglyme in API production is 1-10 tonne per year.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

3.3. Remaining risk of the “applied for use” scenario

As stated before, diglyme is used as a solvent in the manufacturing of an API. As the solvent does not end up in the final API, there is no risk to human health at consumer level. The production usually takes place once a year for a limited period of weeks, with a maximum of 24-40 potentially exposed workers. A contained system is applied and there is no open handling, also not during the charging of diglyme.

Results of the exposure measurement report indicate that measurements during the last production cycle (campaign August to October 2014) are below the method detection limit or the current derived no effect level (DNEL). In 2015, RAC decided on a reference DNEL which was published by ECHA in July (RAC/33/2015/08 rev 1). For the assessment, this reference DNEL from RAC has been used and adequate control of the process is demonstrated.

3.4. Human health and environmental impacts of the applied for use scenario

The CSR provides a detailed assessment of exposure relating to the “applied for use scenario” and demonstrates that there are no adverse human health or environmental impacts. Specifically, detailed exposure modelling and workplace measurements demonstrate that worker exposure is below the respective DNELs for diglyme (see also section 3.3). There is no public exposure to diglyme from this use. All waste streams and filters are sent to offsite incineration, therefore environmental impacts can be considered as negligible.

3.4.1. Number of people exposed

24-40 workers are potentially exposed to diglyme in a batch production process which lasted XXXXX in 2014.

3.5. Monetised damage of human health and environmental impacts

Since it is demonstrated that worker exposure for diglyme is clearly below the DNEL, and there is no measurable release into the environment, the risks to human health, expressed in monetary terms, is zero.



9

4. SELECTION OF THE “NON-USE” SCENARIO

4.1. Efforts made to identify alternatives

4.1.1. Research and development

Solvents Significant effort has been undertaken to research and develop alternatives for solvents with a high hazard potential. In general a universal drop-in replacement for a solvent does not exist. There are a variety of solvent guides (described and referenced below) which all aim to assess the environmental, health and safety hazard potential based on varying criteria. Many of these guides propose “greener” alternatives for solvents with high hazard potentials which is illustrated in the following.

Green / sustainable solvents A solvent is ‘green’ when it is sourced from a renewable feedstock, has a low carbon footprint, is biodegradable, not soluble in water, is not a VOC (boiling point not too low), and is easy to recycle (boiling point not too high). However, the greenness of a solvent also depends on the application. A safe solvent is typically characterized by its stability, low flammability and moderate toxicity.

The topic of “green solvent” has significantly gained importance in the last decade as illustrated in Figure 1

Figure 1: Results from the SciFinder search for the keyword “green solvents” in Journals from dated 08/01/2013. Novartis, 2013.

Novartis actively participates in researching and publishing on green solvents. There is a list of solvents which are commonly accepted as green and approved as such by authorities. The “International Conference on Harmonisation of Technical Requirements for Registration of



10

Pharmaceuticals for Human Use” (ICH) guidelines differentiate solvents into the categories of preferred, usable, and undesirable. XXXXXXxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

Novartis is an early adopter of the solvent selection guide initiative. Major internal efforts were made, in particular since 2013.

Common Solvent Replacement Table The Common Solvent Replacement Table proposes alternatives for some undesirable solvents. These common replacements are well known and accepted, but there are still many challenges – particularly related to solvents which are difficult to replace due to their unique properties.

Pfizer Guide Green Chemistry Tools is a medicinal chemistry and research chemistry based organization. They assessed, in a thorough and systematic way, solvents in three general areas: worker safety, process safety, environmental and regulatory aspects. The resulting Pfizer Guide is a simple and handy guidance list differentiating between preferred, usable and undesirable solvents.

The American Chemical Society Green Chemistry Institute Pharmaceutical Roundtable (ACS GCIPR) - Solvent Selection Guide Overview The ACS GCIPR developed a Solvent Selection Guide in 2010 for use by member companies. The guide considers safety, health, and environment aspects related to air, water, and waste of solvent selection. Each of these aspects are rated for more than 60 solvents in a scoring system from 1 to 10 (1 is the best) which is also applied in other guides. Furthermore, the aspects are colour coded: green (rating 1-3), yellow (4-7) and red (8-10).

Novartis is a member of ACS GCIPR with a leading and contributing role in all working groups since this initiative started. Novartis is actively engaged in promoting the development of ACS GCIPR, for example by regularly suggesting initiative studies and projects or by contributing to the up-to-date status of the organisation.

Several long-term projects have been started by Novartis in global initiatives such as the ACS Green Chemistry Roundtable, which aims to identify suitable alternatives to polar aprotic solvents.

AstraZeneca (AZ) Guide The AZ Guide adds a life cycle analysis to the safety, health and environment categories. The environmental part is further differentiated into 6 subcategories which are impact on air, VOC potential, impact on water, potential biotreatment plant load, recycling, and incineration. The scoring and colour code is identical to the ACS GCIPR.

GSK Guide The GSK Guide has an extended list of parameters including flammability and explosion related to handling and storage, reactivity and stability factors affecting the solvent, as well as a life cycle score assessing the environmental impacts of producing the solvent. These parameters are also classified in a scoring system and using a colour code, however in contrary order, green is best with (10-8), yellow



11

(7-4), and red (3-1). Furthermore, the melting and boiling points are provided, as well as a potential legislation flag column alerting in case of relevant regulatory restrictions, for example if solvents are classified as carcinogenic, toxic for reproduction or as an ozone depleting substance.

Sanofi Guide The Sanofi guide takes the ICH limits, occupational health, safety, environment and other concerns into account to evaluate the hazard potential. Results are categorized in four classes: recommended, substitution advisable, substitution requested, and banned.

Novartis efforts Novartis is committed to improving the “greenness” of its processes. The Twelve Principles of Green Chemistry are used to guide on designing a chemical process to reduce or eliminate the use and generation of hazardous substances (Warner, J. & P. Anastas, 1998):

1. Prevention: it is better to prevent waste than to treat or clean up waste afterwards.

2. Atom Economy: maximise the incorporation of all raw materials into the final product.

3. Less Hazardous Chemical Syntheses: use and generate substances that minimise toxicity to human health and the environment.

4. Designing Safer Chemicals: design chemical products that minimize their toxicity.

5. Safer solvents and Auxiliaries: use of solvents or separation agents should be avoided wherever possible.

6. Design for Energy Efficiency: minimize the energy requirements of chemical processes and conduct synthetic methods at ambient temperature and pressure if possible.

7. Use of Renewable Feedstock: use renewable raw material or feedstock whenever practicable.

8. Reduce Derivatives: minimise derivatisation (protection/deprotection), which generates waste.

9. Catalysis: catalytic reagents are superior to stoichiometric reagents.

10. Design for Degradation: design chemical products so they break down into innocuous products that do not persist in the environment.

11. Real-time Analysis for Pollution Prevention: develop analytical methodologies avoiding formation of hazardous substances.

12 Inherently Safer Chemistry for Accident Prevention: choose reagents to minimise accidents (releases, explosions, and fires).

In addition, the Twelve Principles of Green Engineering are considered for the development and commercialisation of industrial processes that are economically feasible and reduce the risk to human health and the environment (Anastas, P. & J. Zimmerman, 2003):

1. Inherent Rather Than Circumstantial: designers need to strive to ensure that all materials and energy inputs and outputs are as inherently nonhazardous as possible.

2. Prevention Instead of Treatment: it is better to prevent waste than to treat or clean up waste after it is formed.

3. Design for Separation: separation/purification operations should be designed to minimise energy consumption and materials use.



12

4. Maximize Efficiency: products, processes, and systems should be designed to maximise mass, energy, space, and time efficiency.

5. Output-Pulled Versus Input-Pushed: products, processes, and systems should be “output pulled” rather than “input pushed” through the use of energy and materials.

6. Conserve Complexity: embedded entropy and complexity must be viewed as an investment when making design choices on recycle, reuse, or beneficial disposition.

7. Durability Rather Than Immortality: targeted durability, not immortality, should be a design goal.

8. Meet Need, Minimize Excess: design for unnecessary capacity or capability (e.g., “one size fits all”) solutions should be considered a design flaw.

9. Minimize Material Diversity: material diversity in multicomponent products should be minimised to promote disassembly and value retention.

10. Integrate Material and Energy Flows: design of products, processes, and systems must include integration and interconnectivity with available energy and materials flows.

11. Design for Commercial “Afterlife”: products, processes, and systems should be designed for performance in a commercial “afterlife”.

12. Renewable Rather Than Depleting: material and energy inputs should be renewable rather than depleting.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

Novartis intensified additional research projects to identify SVHC solvent alternatives in 2013 with significant organisational and financial means. SVHC are categorised in appropriate sub-classes, e.g. diglyme is part of the glymes sub-class. These classes are prioritized before initiating long-term replacement efforts and education by the internal Chemistry Network. More than 10 long-term research projects focusing on SVHC solvent alternatives have been started by Novartis so far. Two of these projects specifically aim to identify alternatives to polar aprotic solvents. As a result, the number of synthetic steps using SVHC solvents in the pilot plant was successfully reduced by 80% in 2014. In addition, SVHC are banned from all new Novartis projects when an alternative has been identified.

Furthermore, Novartis continuously conducts case studies to further improve greenness of the production processes in various fields, including but not limited to: alternative reagents, solvent recovery, purification, alternative routes, convergent synthesis, and telescoping. Improvements and best practices identified by Novartis’ expert chemists are shared with the chemical community in order to develop a sustainable chemical synthesis.



13

Novartis - diglyme specific research and development This section describes Novartis’ activities to find an alternative for diglyme, and other glymes. They are used due to their superior properties such as good solvating capabilities, high stability under basic conditions and moderate stability under acidic conditions; making them good ligands, leading to improved selectivity and reactivity.

The alternative for diglyme must be able to solubilize poorly soluble substrates, as it is often the case for advanced complex intermediates and APIs. Further requirements are:

- Avoidance / minimisation of genotoxic impurities to prevent adverse quality impact of the final product;

- Sufficient yield; - Reduced hazard and risk of alternative compared to diglyme; and - Economic feasibility.

From a technical point of view, there are currently limited alternatives to polar aprotic solvents, undesirable solvents being often substituted with other undesirable solvents which tend to display the same hazard factors. It is all the more difficult for pharmaceutical targets, as a large proportion of Novartis’ portfolio of intermediates and APIs display poor or limited solubility in classical and desirable organic solvents. Recourse then has to be made to the polar aprotic solvents, with very few of them being effective.

Novartis is developing an aprotic solvent industry position paper and provides therein a variety of specific information on diglyme including:

- Background information on the use of diglyme; - Information on uses, including the use as a solvent for the production of API of medicinal

products subject to this application; - Exposure control; - Status quo on technical alternatives; and - Limit values.

However, it will take many years to demonstrate feasibility and to implement them into projects.

Novartis contributes to and partially leads this development as a member of ACS GCIPR since the start of the initiative.

4.1.2. Data searches

As illustrated in section 4.1.1, Novartis has substantial expertise related to the substitution of SVHC solvents in general and in particular to the substitution of diglyme in-house.

For the assessment of alternatives, relevant literature and in-house test reports have been assessed.



14

Identification of possible alternatives

Based on the existing Novartis’ expertise, 16 standard solvents were identified as potential alternatives to diglyme. These 16 solvents listed in Table 4 were investigated with basic tests at laboratory scale.

Table 4: List of tested diglyme alternatives

Category No. Alternative

Category 1 alternatives 1 n-butyl acetate

2 n-butanol

Category 2 alternatives

3 Xylene

4 Sulfolane

5 1,2,3,4-tetrahydronaphtalene

Category 3 alternatives

6 Diethyl carbonate

7 Ethylene carbonate

8 Butane-1,3-diol

9 2-ethylhexyl acetate

10 Propane-1,2-diol

11 2-butoxyethyl acetate

12 Pentane-1,5-diol

13 Ethylene glycol

14 Propane-1,3-diol

15 Cyclohexanol

16 4-methyl-1,3-dioxolan-2-one

In the first step all alternatives were tested for yield and the ones that revealed (significantly) reduced yield or failed other key requirements such as high melting points were therefore not investigated further (category 3).

Based on the test results for yield, promising alternatives were further investigated. Isolated yields were thus determined for alternatives that gave good selectivity in the reaction mixture (category 1 and 2).

Category 2 alternatives are potential candidates related to impurity profile and yield but have constraints regarding their ecological innocuousness.

Category 1 alternatives are suited to substitute diglyme and were tested for the parameters impurity profile, yield and ecological innocuousness. Category 1 alternatives are evaluated in detail in section 4.2.3.



15

4.2. Assessment of shortlisted alternatives 4.2.1. Category 3 alternatives

All possible alternatives have been investigated regarding their calculated yield of the interim product in the reaction, and the area of integrated chromatographic peak of substance in the HPLC response-time-diagram as a percentage of total area of all peaks. The minimum yield of 65% (benchmark is diglyme) was not achieved by 11 out of 16 alternatives. Furthermore, alternatives were screened out due to a high melting point. These were declared as category 3 alternatives. The results for category 3 alternatives are presented in Table 5.

Table 5: Overview on yield and area results for category 3 alternatives.

Category 3 alternative Yield, calculated* threshold = 65%

Diethylcarbonate significantly lower than threshold

Ethylene carbonate significantly lower than threshold

Butane-1,3-diol lower than threshold

2-ethylhexylacetate lower than threshold

Propane-1,2-diol significantly lower than threshold

2-butoxyethyl acetate significantly lower than threshold

Pentane-1,5-diol lower than threshold

Ethylene glycol significantly lower than threshold

Propane-1,3-diol significantly lower than threshold

Cyclohexanol1) meets threshold

4-methyl-1,3-dioxolan-2-one lower than threshold * Calculated yield based on a HPLC w/w-analysis of the reaction mixture. 1) No further optimization due to high melting point (processing issues foreseen).

Substances with a yield of a minimum of 65% or more have been further investigated as illustrated in the following.

4.2.2. Category 2 alternatives: xylene, sulfolane, and 1,2,3,4-tetrahydronaphtalene

The remaining possible alternative solvents had the most promising results in the first basic tests for yield. These were further investigated to determine their suitability as alternative for diglyme. The remaining 5 candidates meet the requirements related to impurity profile and yield but 3 of them, xylene, sulfolane and 1,2,3,4-tetrahydronaphtalene, have constraints regarding their ecological innocuousness.

Xylene is an aromatic solvent, which can contain benzene and therefore constitutes no significant shift towards a less hazardous substance.

For sulfolane, the isolated yield was significantly lower. Moreover, sulfolane has an unfavorable impurity profile and can be problematic in waste streams as the substance has a low melting point



16

ranging between 20-26°C. Consequently, the risk of solidification is increased in e.g. drums, containers or pipes at ambient temperature.

1,2,3,4-tetrahydronaphtalene is suspected to be carcinogenic and would therefore not constitute a significant shift towards a less hazardous substance.

Therefore these solvents have been declared as category 2 alternatives. Their test results are shown in Table 6. In general, they provided good selectivity in the reaction mixture.

Table 6: Overview on feasibility of category 2 alternatives compared to the benchmark of diglyme. Parameter Parameter description Xylene Sulfolane 1,2,3,4-tetrahydro-

naphtalene

Yield, calculated, threshold >= 65%?*

Amount of FIP synthesized in the reaction

process. Yield is calculated based on a HPLC w/w-analysis of the reaction mixture.

OK OK OK

Area

Area of integrated chromatographic peak of substance in the HPLC

response-time-diagram as a percentage of total area

of all peaks

significantly lower lower lower

Yield, isolated

Amount of FIP which can be isolated from the

reaction chamber lower significantly

lower significantly lower

Purity at 230nm

Peak purity at 230nm wavelength similar similar similar

Bis-Adduct

Percentage of Bis-impurity formed during the

synthesis significantly higher similar similar

Regio-isomer

Percentage of regioisomer impurities formed during

the synthesis similar significantly

higher similar

*The minimum threshold for calculated yield is 65%.

The remaining two possible alternatives n-butyl acetate and n-butanol were further investigated in more detail (category 1 alternatives). A comprehensive assessment of both category 1 alternative solvents is provided in the following chapters. Category 1 Alternative: n-butyl acetate

4.2.2.1. Substance ID, properties, and availability

N-butyl acetate is an available standard solvent in the pharmaceutical industry, which can be obtained from several providers. General information and physicochemical properties of n-butyl acetate are provided in Table 7:



17

Table 7: Substance IDs and properties of n-butyl acetate (European Chemicals Agency: www.echa.europa.eu)

Parameter Value Physicochemical properties Value

Chemical name and composition

n-butyl acetate Mono constituent substance

Melting point -78 °C

EC number 204-658-1 Boiling point (at 1013 hPa)

126 °C

CAS number 123-86-4 Relative density 0.8812 g/cm³ (at 20 °C)

IUPAC name n-butyl acetate Vapour pressure 10.7 hPa at 20 °C

Molecular formula C6H12O2 Water solubility Soluble (1,000–10,000 mg/L) 5.3 g/L (at 20 °C)

Molecular weight 116.16 g mol-1 Surface tension 61.3 mN/m (at 20 °C) Concentration: 1 g/L

Molecular structure

Flash point 27 °C (at 1013 hPa)

Physical state (at 20°C and 1013 hPa)

Clear and colourless organic liquid, fruity odour

Auto-ignition temperature 415 °C (at 1010 hPa)

4.2.2.2. Technical feasibility of n-butyl acetate

The assessment criteria as well as the detailed scoring results of n-butyl acetate in comparison with diglyme according to the ACS GCIPR system are shown in the following Table 8:



18

Table 8: Scoring system evaluation of diglyme and n-butyl acetate in comparison.

The comparison illustrates that n-butyl acetate is an improvement for the categories safety, health, environment (water) and environment (waste). It is rated only slightly worse for environment (air). Significant improvements are evident for the parameters peroxide former (1 to 0), health hazards (reduction from 5 to 2), biodegradation (3 to 1), water solubility (4 to 1) and the boiling point (4 to 3).

Novartis investigated n-butyl acetate at lab scale. N-butyl acetate gave good selectivity in the reaction mixture. Therefore, isolated yields were determined and compared to the benchmark of diglyme as well as other relevant parameters (Table 9).



19

Table 9: Isolated yields and impurities for n-butyl acetate compared to the benchmark of diglyme.

Parameter Parameter description n-butyl acetate

Yield, calculated, threshold >= 65%?*

Amount of FIP synthesized in the reaction process.

Yield is calculated based on a HPLC w/w-analysis of the reaction mixture.

OK

Area Area of integrated chromatographic peak of

substance in the HPLC response-time-diagram as a percentage of total area of all peaks

lower

Yield, isolated Amount of FIP which can be isolated from the reaction chamber similar

Purity at 230nm Peak purity at 230nm wavelength similar

Bis-Adduct Percentage of Bis-impurity formed during the synthesis significantly higher

Regioisomer Percentage of regioisomer impurities formed during the synthesis similar

Genotoxic Impurities Genotoxic impurities (GTI);

shall be minimized; target threshold value is <100ppm

OK


Out of the 5 alternatives from category 1 and 2, n-butyl acetate proved to have the best results related to impurity profile, yield and ecological innocuousness. N-butyl acetate was used for further process optimization.

The substitution of diglyme with n-butyl acetate is a like-for-like replacement and does not change the overall process described in section 3.1.3. The same equipment train can be used after process optimization. However, some changes will need to be implemented.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxXXxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.



20

In summary, n-butyl acetate is a technically feasible alternative to diglyme at laboratory scale. The next step will be the scale-up to verify technical feasibility at production scale. It is expected that the process step, where diglyme is currently used at Novartis, can technically be performed with n-butyl acetate.

4.2.2.3. Economic feasibility and economic impacts of n-butyl acetate

The same equipment train can be used after process optimization and successful implementation. Therefore, the solvent replacement process will not require a capital investment. Accordingly, no further significant costs related to equipment, maintenance, training etc. are expected. There will also be no recurring costs.

The current market costs based on publically available information are more than 4 times lower with 36€/L for n-butyl acetate compared to 156 €/L for diglyme. The productivity and yield is slightly lower for n-butyl acetate and requires a higher quantity of process related substances. The latter minimises the economic benefit based on chemical costs. In general, chemical costs are minor compared to other process related costs. Thus, minimal cost impact is expected on the final API.

Further costs will be associated with the Novartis Commercial production division technical transfer, and registration.

In summary, the use of n-butyl acetate instead of diglyme is economically feasible.

4.2.2.4. Availability of n-butyl acetate

The alternative solvent n-butyl acetate is a standard solvent in the pharmaceutical industry. It is commercially available in sufficient quantities and affordable. Novartis has experience in the utilization of n-butyl acetate.

However, the suitability of n-butyl acetate for the use as solvent for the synthesis of indacaterol needs to be evaluated and documented carefully. A timeframe of 7 years after the sunset date of diglyme (22 August 2017) is needed to finalise the solvent substitution process and to obtain regulatory approval globally. For more detailed information the reader is referred to the SP in chapter 7.

4.2.2.5. Hazard and risk of n-butyl acetate

Information on classification and labelling of n-butyl acetate is summarized in Table 10:

Table 10: Classification and labelling of n-butyl acetate (European Chemicals Agency: www.echa.europa.eu).

Substance Name

Hazard Class and Category Code(s)

Hazard Statement Code(s) (labelling)

Number of Notifiers

Additional classification and labelling comments

Regulatory and CLP status

n-butyl acetate

EC: 204-658-1

CAS: 123-86-4

Flam. Liq. 3

H226 (Flammable liquid and vapour.)

65 -

REACH registered.

Included in CLP Regulation, Annex VI – harmonized classification (index number 607-025-00-1)

STOT SE 3

H336 (May cause drowsiness or dizziness.)



21

Based on the classification shown in Table 10, the alternative n-butyl acetate shows the same level of flammability (Flam. Liq. 3) as observed for diglyme but has neither reprotoxic nor teratogenetic effects. The alternative n-butyl acetate is less toxic to human health than diglyme.

With regard to environmental concerns, neither diglyme nor n-butyl acetate are classified according to Harmonized classification - Annex VI of Regulation (EC) No 1272/2008 (CLP regulation).

Although n-butyl acetate is not listed in the “preferred” or “usable” category of the DD solvent selection guide, it compares very well with ethyl acetate, designated as a preferred solvent:

Ethyl acetate and n-butyl acetate can be considered to be structurally similar: Both substances are esters of acetic acid with a short-chain linear aliphatic alcohol. The ethyl and the n-butyl fraction differ by two C-atoms. These substances share a common functional group (ester bond) and differ only by the length of the esterified alcohol moiety. They can therefore be considered to show similar toxicological and environmental properties. Indeed, neither ethyl acetate nor n-butyl acetate are classified for any environmental or health hazards. n-butyl acetate has a low partition coefficient (log Kow = 1.82; Table 8), is readily biodegradable, and is of low toxicity to aquatic organisms. With respect to human health, n-butyl acetate can likewise be considered as a substance with low hazard, as reflected by its non-classification: It is not acutely toxic, not irritating skin and eyes, not sensitizing, not genotoxic and carcinogenic, and not toxic to reproduction. Therefore, it can be concluded that n-butyl acetate and ethyl acetate show a very similar hazard profile.

In summary, the use of n-butyl acetate instead of diglyme constitutes a substitution with a less hazardous substance.

4.2.2.6. Conclusions on n-butyl acetate

The alternative solvent n-butyl acetate is a standard solvent in the pharmaceutical industry. It is commercially available in sufficient quantities and affordable. Novartis has experience in the utilization of n-butyl acetate.

N-butyl acetate gave good selectivity in the reaction mixture and offers the best results related to impurity profile, yield and ecological innocuousness from all investigated alternatives at laboratory scale. Technical feasibility needs to be verified after scale-up which has not yet been performed.

The substitution of diglyme with n-butyl acetate is a like-for-like replacement and does not change the overall process significantly.

The same equipment train can be used after successful implementation. Therefore, the solvent replacement process will not require a capital investment. The cost impact will be driven by cost difference and quantity of n-butyl acetate versus diglyme and is expected to be minimal on the final API.

n-butyl acetate is less toxic to human health than diglyme and its use constitutes a significant shift to a less hazardous substance. It compares very well with ethyl acetate which is designated as a preferred solvent. N-butyl acetate is also an improvement compared to diglyme for the categories safety, health, environment (water) and environment (waste) according to the ACS GCIPR solvent selection guide.



22

n-butyl acetate cannot yet be used as solvent in the synthesis for indacaterol as upscaling activities (including the respective tests and documentation) and approval by the health authorities are still required. These efforts are expected to take a total of 7 years after the sunset date in 2017 and are described in detail in the SP (chapter 7).

4.2.3. Category 1 Alternative: n-butanol

4.2.3.1. Substance ID, properties, and availability

N-butanol is an available standard solvent in the pharmaceutical industry which can be obtained from several providers. n-butanol is listed as a “preferred” solvent in the solvent selection guide from Green Chemistry (refer to section 4.1.1). General information and physicochemical properties of n-butanol are provided in Table 11:

Table 11: Substance IDs and properties of n-butanol (European Chemicals Agency: www.echa.europa.eu)

Parameter Value Physicochemical properties Value

Chemical name and composition

n-butanol Mono constituent substance

Melting point -89.8 °C

EC number 200-751-6 Boiling point (at 1013 hPa)

119 °C

CAS number 71-36-3 Relative density 0.8095 g/cm³ (at 20 °C)

IUPAC name butan-1-ol Vapour pressure 5.6 hPa at 20 °C

Molecular formula C4H10O Water solubility Miscible, 66 g/L (at 20 °C)

Molecular weight 74.12 g mol-1 Surface tension 69.9 mN/m (at 20 °C) Concentration: 1 g/L

Molecular structure

Flash point 35 °C (at 1013 hPa)

Physical state (at 20°C and 1013 hPa)

Clear and colourless organic liquid, alcoholic odour

Autoignition temperature 355 °C (at 1019 hPa)



23

4.2.3.2. Technical feasibility of n-butanol

The assessment criteria as well as the detailed scoring results of n-butyl acetate in comparison with diglyme according to the ACS GCIPR system are shown in the following Table 12:

Table 12: Scoring system evaluation of diglyme and n-butanol in comparison.

The comparison illustrates that n-butanol is an improvement for the categories safety, health, and environment (waste). It is rated slightly worse for environment (air) and worse for environment (water). Significant improvements are evident for the parameters peroxide former (reduction from 1 to 0), health hazard (5 to 3), biodegradation (3 to 2), water solubility (4 to 2 thus less soluble in water) and the boiling point (4 to 1).



24

Novartis investigated n-butanol at lab scale. N-butanol showed good selectivity in the reaction mixture. Therefore, isolated yields were determined and compared to the benchmark of diglyme (Table 13).

Table 13: Isolated yields and impurities for n-butanol compared to diglyme.

Parameter Parameter description n-butanol

Yield, calculated,

threshold >= 65%?*

Amount of FIP synthesized in the reaction process. Yield is calculated based on a HPLC w/w-analysis of the

reaction mixture. lower**

Area Area of integrated chromatographic peak of substance in the HPLC response-time-diagram as a percentage of total

area of all peaks lower

Yield, isolated Amount of FIP which can be isolated from the reaction chamber lower

Purity at 230nm Peak purity at 230nm wavelength significantly lower

Bis-Adduct Percentage of Bis-impurity formed during the synthesis significantly higher

Regioisomer Percentage of regioisomer impurities formed during the synthesis significantly higher


**n-butanol is listed as a “preferred” solvent in the solvent selection guide from Green Chemistry and was therefore further investigated as potential alternative although the minimum threshold of 65% is not met.

Out of the 5 alternatives from category 1 and 2, n-butanol proved to have the second best results related to impurity profile, yield and ecological innocuousness. The second rating is due to the higher portion of regioisomers which is significantly higher compared to the best alternative n-butyl acetate.

Genotoxic impurities have not been tested for n-butanol and no further assessment on required changes to the process was conducted as n-butyl acetate was ranked as best alternative. However, it would be also a like-for-like replacement. .

4.2.3.3. Economic feasibility and economic impacts of n-butanol

The same equipment train can be used after process optimization and successful implementation. Therefore, the solvent replacement process will not require a capital investment. Accordingly, no further significant costs related to equipment, maintenance, training etc. are expected. There will also be no recurring costs.

The current market costs, based on publicly available information are 3 times lower with 50€/L for n-butanol compared to 156€/L for diglyme. The productivity and yield is slightly lower for n-butanol requiring a higher quantity of this substance and thus reducing effectiveness of the production process. Latter minimises the economic benefit based on chemical costs. In general, chemical costs are minor compared to other process related costs. Thus, minimal cost impact is expected on the final API manufacturing process.

Further costs will be associated with the Novartis Commercial production division technical transfer, and registration.



25

In summary, the use of n-butanol instead of diglyme is economically feasible.

4.2.3.4. Availability of n-butanol

The alternative solvent n-butanol is a standard solvent in the pharmaceutical industry. It is commercially available in sufficient quantities and affordable. Novartis has experience in the utilization of n-butanol. No detailed availability assessment has been conducted as n-butyl acetate was identified as best alternative.

4.2.3.5. Hazard and risk of n-butanol

Information on classification and labelling of n-butyl acetate is summarized in Table 14:

Table 14: Classification and labelling of n-butanol (European Chemicals Agency: www.echa.europa.eu).

Substance Name

Hazard Class and Category Code(s)

Hazard Statement Code(s) (labelling)

Number of Notifiers

Additional classification and labelling comments

Regulatory and CLP status

n-butanol EC: 200-751-6 CAS: 71-36-3

Flam. Liq. 3 H226 (Flammable liquid and vapour.)

63 -

REACH registered. Included in CLP Regulation, Annex VI -harmonized classification (index number 603-004-00-6)

Acute Tox. 4 H302 (Harmful if swallowed.)

Skin Irrit. 2 H315 (Causes skin irritation.)

Eye Dam. 1 H318 (Causes serious eye damage.)

STOT SE 3 H335 (May cause respiratory irritation.)

STOT SE 3 H336 (May cause drowsiness or dizziness.)

Based on the classification shown in Table 14, the alternative n-butanol shows the same level of flammability (Flam. Liq. 3) as observed for diglyme, but shows acute toxic effects without chronic effects. Therefore, the alternative n-butanol is less toxic to human health than diglyme.

With regard to environmental concerns, neither diglyme nor n-butanol is classified according to Harmonized classification - Annex VI of Regulation (EC) No 1272/2008 (CLP regulation).

Furthermore, n-butanol is listed as a “preferred” solvent in the solvent selection guide from Green Chemistry.

In summary, the use of n-butanol instead of diglyme constitutes a substitution with a less hazardous substance.



26

4.2.3.6. Conclusions on n-butanol

The alternative solvent n-butanol is a standard solvent in the pharmaceutical industry. It is commercially available in sufficient quantities and affordable. Novartis has experience in the utilization of n-butanol.

N-butanol showed good selectivity in the reaction mixture but offers only the second best results as an alternative to diglyme due to the slightly worse impurity profile (higher portion of regioisomers) compared to n-butyl acetate. Therefore, tests regarding genotoxic impurities have not been performed and no further assessment on required changes to the process was conducted.

Cost impact driven by difference in price and quantity of n-butanol versus diglyme is expected to be minimal on the final API.

N-butanol is a less hazardous substance than diglyme. It is also an improvement compared to diglyme for the categories safety, health, and environment (waste) according to the ACS GCIPR solvent selection guide.

No detailed substitution plan has been developed to assess the time needed for industrial implementation as n-butyl acetate has been identified as the best alternative due to its better impurity profile.

4.3. The most likely non-use scenario To replace diglyme, development and validation is ongoing, which finally requires health authority regulatory approvals for the respective countries in which the products are sold. In the following, the non-use scenario considers the most likely actions to be taken by Novartis, if authorisation would not be granted for the bridging period.

The production of indacaterol using diglyme would have to be stopped at the sunset date as long as the health authorities’ approvals for the use of an alternative are pending (please consider the SP in chapter 7 for more details).

A switch of the production to a non-EEA site (EEA: European Economic Area) is not possible. Novartis itself has two chemical production facilities in non-EEA countries. As the production facility in Asia is a large volume facility and the here mentioned indacaterol production is a low volume one, a transfer to China would require a new plant to be built. Prerequisite health authority approvals would be required before using the site. Therefore, relocation of the production is technically not possible and building a new plant not an economically viable option. In the past it was expected that the Swiss ordinance would adopt EU authorisation measures, therefore the Basel site was not considered as an option. However, the switch of the production to the Basel site would also first require a health authority review and approval and cannot be considered as a short term option for a bridging period.

For the same regulatory approval timeline reasons as stated above, subcontracting the production to a non-EEA third party is not an option, regardless of whether it would be possible to identify in a short period of time a manufacturer with the capacity, technical capability and the quality standards



27

required to produce the intermediate. In addition, Novartis would have concerns about performing a technical transfer and disclosing proprietary process and pipeline information to a third party.

In case the bridging period would not be granted, stock building of the interim product for which diglyme is used could be a feasible non-use scenario, but very limited due to the rather short shelf-life for the drug product (currently 18 months for climate zone III/IV countries). Some countries – for example Brazil -–do not even accept products with less than 12 month. In the case of stock building Novartis would try to forecast the demand until regulatory approval of the new reaction synthesis containing the replacement solvent is in place.

In any case, building bridging stocks would result only in postponing supply shortages, because it will not be possible to create bridging stocks for the entire 7 year period. The impact therefore will be worldwide stock-out of Novartis medicinal products containing indacaterol.



28

5. IMPACTS OF GRANTING AUTHORISATION

5.1. Economic impacts

Novartis prides itself as a reliable supplier of innovative medication to meet patient need. Interruption to supply can result in patients not having access to their medication and a consequent loss in confidence by patient and physicians in Novartis as a reliable supplier. This cannot be quantified but is expected to have far-reaching consequences for all Novartis products.

Continuing production and supply without interruption allows Novartis to keep and expand their availability of our medicines in the COPD disease area. Otherwise the medicine market presence could be at risk and patients would need to switch to other therapeutic options (see also 3.2).

Value Added

As explained before, the most realistic non-use scenario is that the EU facility would have to stop the production related to indacaterol, if an authorisation is not granted. For this reason, as a measure of this impact, the value added foregone is calculated for the 7 years review period assessed in this document. In order to calculate the value added foregone, the costs of inputs (except capital and labour) are subtracted from the turnover (net sales).

𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑣𝑣𝑎𝑎𝑎𝑎𝑣𝑣𝑎𝑎 𝑓𝑓𝑓𝑓𝑓𝑓𝑣𝑣𝑓𝑓𝑓𝑓𝑓𝑓𝑣𝑣 = 𝑡𝑡𝑣𝑣𝑓𝑓𝑓𝑓𝑓𝑓𝑣𝑣𝑣𝑣𝑓𝑓 (𝑓𝑓𝑣𝑣𝑡𝑡 𝑠𝑠𝑣𝑣𝑣𝑣𝑣𝑣𝑠𝑠)− 𝑐𝑐𝑓𝑓𝑠𝑠𝑡𝑡𝑠𝑠 𝑓𝑓𝑓𝑓 𝑖𝑖𝑓𝑓𝑖𝑖𝑣𝑣𝑡𝑡𝑠𝑠 (𝑣𝑣𝑒𝑒𝑐𝑐𝑣𝑣𝑖𝑖𝑡𝑡 𝑐𝑐𝑣𝑣𝑖𝑖𝑖𝑖𝑡𝑡𝑣𝑣𝑣𝑣 & 𝑣𝑣𝑣𝑣𝑙𝑙𝑓𝑓𝑣𝑣𝑓𝑓)

In the period from 2012 to 2014, the sales revenue of the Novartis facility in Ringaskiddy for sales of the API necessary for the production of COPD drugs, varied, in Swiss Franc, from CHF 20 million to CHF 65 million annually. Due to confidentiality reasons, the exact amount of sales revenue in 2014 will not be disclosed and therefore the minimum amount in the range of CHF 20 million will be considered for the calculation of the value added foregone. The amount of CHF 20 million when converted to Euro1 is equivalent to approximately EUR 16.4 million, see Table 15.

Table 15. Minimum sales revenue in Euro.

Minimum sales revenues (in the period between 2012 and 2014) - CHF

Average exchange rate (EUR/CHF) 2012-2014

Equivalent amount in EUR of the minimum sales revenue

20,000,000 1.217 16,433,854

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx2xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.

1 EUR/CHF=1.217. This is the average of the daily exchange rates between January 1st 2012 and December 31st 2014 according to data from the European Central Bank (https://www.ecb.europa.eu/stats/exchange/eurofxref/html/eurofxref-graph-chf.en.html).

2 xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.



29

Considering that the sales revenues are expected to be kept at least constant (as described, more probably they are expected to increase given by higher market demand) during the whole assessment period, the net present value (NPV) of the 7 years of value added foregone or the value added given by a granted authorization is calculated as follows in Figure 2 below.

Figure 2: NPV of 7 years of value added foregone due to production shutdown.

5.2. Human Health or Environmental Impact

As discussed in section 3.4 the human health and environmental impacts are considered to be zero. This is given by the fact that all exposure values are by far below the DNEL, for further details please consider the corresponding CSR. The production takes place in a contained system and no open handling of diglyme takes place, even not for charging. Therefore, none of the workers is exposed directly. All waste streams are incinerated.

5.3. Social impacts

The social benefit for Novartis Ringaskiddy Limited to further use diglyme for the production of the medicinal products before the substitution plan is completed will be that the patients can continually have the access to the medication without interruption. Furthermore, the substantial labour force employed for indacterol-containing medicinal product production (e.g. drug and drug containing inhalation devices) in various countries e.g. Spain, Switzerland and Ireland could be impacted.



30

5.4. Wider economic impacts

A granted authorisation will avoid that Novartis would be unable to fulfil multiple supply, co-marketing and partnering contracts for the COPD portfolio, which require continued supply.

5.5. Distributional impacts

Indacaterol-containing medicinal products can be continually supplied provided diglyme can be used on an ongoing basis until a substitution plan is completed. Thus the existing demand for COPD medicinal products can be continually covered without shortages. In addition, patients’ trust in Novartis as an innovative medicines supplier will be maintained.

Further, Novartis can guarantee the continued supply of medicinal products for COPD treatment and thus avoid the possible negative consequences for COPD patients worldwide. Otherwise physicians would have to switch medication of their patients which might impact the therapeutic impact of treatment. Furthermore, this medicinal product is widely reimbursed in the European markets, whereas alternative treatments may not be.

5.6. Uncertainty analysis

The ECHA Guidance on SEA (1) proposes an approach for conducting the uncertainty analysis. This approach provides three levels of assessment that should be applied if it corresponds.

- Qualitative assessment of uncertainties - Deterministic assessment of uncertainties - Probabilistic assessment of uncertainties

The ECHA guidance further states: level of detail and dedicated resources to the assessment of uncertainties should be in fair proportion to the scope of the SEA. Further assessment of uncertainties is only needed, if assessment of uncertainties are of crucial importance for the overall outcome of the SEA.

Hence, a qualitative assessment of uncertainties has been conducted. Table 16 illustrates the systematic identification of uncertainties related to economic impacts.

Table 16: Uncertainties on economic impacts.

Identification of uncertainty (assumption) Classification Evaluation

Criteria and scaling (contribution to total uncertainty)

Turnover remains constant over the review period

Parameter uncertainty

Due to growing market volume indeed an increase over time is expected: underestimation

Low

Cost of inputs Range The exact value cannot be assessed therefore a range is provided

Low

Uncertainties related to human health and environmental impacts can be excluded as exposure measurements and protection measures clearly proof an exposure value below the DNEL. These impacts are therefore assessed to be zero.



31

6. CONCLUSIONS

6.1. Comparison of the benefits and risk

To summarise the previous assessment and to estimate the overall costs and benefits of a decision to grant or deny this Application for Authorisation (AfA), a combined assessment of impacts is set out here. The uncertainty analysis (see section 5.6) demonstrates that the effects of uncertainties on the overall result of the SEA are low.

Table 17 summarises the effects described in section 5.

Table 17: Comparison of impacts for the applied for use and the non-use scenario.

Type of impact Applied for use scenario Non-use scenario

Human health - All exposure values are below DNEL - Not applicable

Environmental impacts - Negligible environmental impacts - No environmental impacts related to diglyme

use in the EEA

Economic impacts

- Uninterrupted drug supply to COPD patients

- Value added given by an uninterrupted use of diglyme

- Supply stop of indacaterol including medicinal products and related value added foregone

- Loss of existing availability - Hampering further development in the COPD

therapeutic field - Loss of trust

Social impacts - Maintenance of jobs and payments of salaries - Impact on labour force

Wider economic impacts

- Continuous fulfilment of multiple co-marketing and partnering contracts - Novartis cannot fulfil contracts

Distributional impacts

- No supply shortages worldwide - Growing demand can be covered - Further expansion in the COPD

market keeping innovative market leader position

- Patients need to switch medication - Reduced treatment options for COPD patients

worldwide, difficulties to switch some patients

Table 18: summarises the quantitative impacts for the review period of 7 years. The value added has to be considered as a minimum impact based on conservative assumptions. It should be mentioned that impact on trust in Novartis’ ability to deliver innovative medicines are not monetised and consequently the economic impacts for a non-use scenario are expected to be much higher. Health and environmental impacts have to be considered as zero over the whole review period, therefore even low socio-economic impacts clearly outweigh the risks.



32

Table 18: Impacts for the applied for use and the non-use scenario for a review period of 7 years.

Type of impact EUR million

Value added XX(45-75)

Health and environmental impacts 0

Net benefits of a granted authorisation XX(45-75)

6.2. Information for the length of the review period

Considering all the impacts described in section 5 and the robust outcome presented in Table 18: as well as the explained efforts made to find an alternative (see section 4.1), including the substitution efforts explained in detail in the following chapter, a review period of 7 years is clearly justified.



33

7. SUBSTITUTION EFFORT TAKEN BY THE APPLICANT

7.1. Summary Novartis uses diglyme as a solvent in the manufacturing process of the Active Pharmaceutical Ingredient (API), indacaterol. Indacaterol is a long-acting β2-agonist (LABA). Indacaterol is an active ingredient in a number of Novartis medicinal products. All products are approved for the treatment of chronic obstructive pulmonary disease (COPD).

Novartis products containing only indacaterol as active ingredient and are approved in more than 110 countries under different trade names. A Novartis fixed-dose combination with indacaterol and another active ingredient is currently approved in more than 70 countries under different trade names. Additionally, Novartis currently has two development products for which Indacaterol is a component and thus impacted by the authorisation application.

The responsible use of chemicals is key for Novartis and extensive efforts are taken to continuously improve all steps of pharmaceutical manufacturing. In the course of these activities, Novartis searched for replacements of diglyme in the production process of indacaterol. As outcome of these efforts, n-butyl acetate has been identified as the most promising alternative based on a number of criteria, including technical, quality, toxicological and economic suitability. The detailed documentation of the Analysis of Alternatives (AoA) is described in the respective document.

The implementation of changes within a pharmaceutical manufacturing process, such as the replacement of a solvent in the manufacturing process of APIs is a complex procedure with significant lead-time required. It requires a complete revalidation cycle to ensure quality, safety and efficacy of the changed process and of the final product. This change is classified as regulatory relevant by the vast majority of national health authorities., This means that before product produced according to the new process can be supplied to the market in a particular country, it is a prerequisite for most national health authorities where the product is approved to first evaluate then approve this change. To do so, a separate variation procedure/submission must be filed in each applicable country, each procedure following local regulatory requirements. As the change impacts the above-mentioned medicinal products which are registered and marketed in more than 110 countries around the world, the complexity is particularly high for a change to the manufacturing process of an API like indacaterol.

Novartis will be technically ready to replace diglyme reaction synthesis scheme for indaceterol by the sunset date of August 2017 but there is a risk of technical issues arising during the process validation which could delay implementation of the alternative solvent beyond the sunset date. While all efforts are being made by the project team to assure a successful introduction of a new process, a degree of uncertainty exists regarding the technical complexities that may be encountered when production using the replacement solvent is performed at commercial scale, an example of which may include, problems removing side products and/or impurity generation.

In addition due to the lengthy and complex regulatory process, the required regulatory approval to support marketing of the impacted medicinal products will not be in effect before the sunset date for diglyme in August 2017. Hence, a review period of 7 years following sunset date of 22 August 2017 is deemed necessary to provide sufficient time to address potential drawbacks and safety margins in



34

the replacement process of diglyme. Five years is estimated as the timeline required for regulatory approval at a global level in an optimal case, an additional two year contingency has been included to support:

(1) delays in responses from health authorities/request for additional information;

(2) technical uncertainties from the validation campaign; and

(3) the need to provide a review report to ECHA 18 months before the end of the review period: the review period for diglyme substitution is mainly based on the timeframes for global regulatory approval. As such, a meaningful update on the regulatory status can only be provided after 5 years earliest. Consequently, a review period of 7 years is needed to cope with these circumstances.

In the following chapters, the process and the expected timeframe for diglyme substitution, including milestones and important tasks and time-line with respect to the technical and regulatory requirements is presented.



35

7.2. Factors affecting substitution

Important aspects affecting the suitability of possible substitutes can be found in the AoA in previous chapters. The substitution plan is based on the considerations of the identified milestones for the implementation of the alternative and how these factors may influence the actions and the timing needed for the transfer to the substitute. Whereas the suitability of a substance as synthesis solvent depends on various physical and chemical quality parameters, toxicological and ecotoxicological properties also have to be addressed prior to choosing appropriate solvents in industrial synthesis processes.

7.2.1. International supply chain of indacaterol containing products

The technical implementation, regulatory requirements and the final approval by the different national health authorities define the main activities for the replacement of diglyme and dictate the time needed to successfully implement the substitution plan. These activities are performed by different operational units in different countries within the Novartis group, which then have to be approved by national health authorities worldwide with varying data requirements and evaluation timelines.

The supply chain for the pharmaceutical products associated with indacaterol is split among three business groups within Novartis: Commercial production division (CPD), Pharmaceutical Operations (PharmOps) and Distribution Operations. CPD are responsible for the manufacture, testing and release of the API. One step of the reaction synthesis of the API requires the use of diglyme as a solvent. The production pipeline of this multi-step synthesis is split across three countries (two EU countries and one non EU country). The supply chain of indacaterol is illustrated in Figure 3.

Figure 3: XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



36

7.2.2. Identification of alternatives and process change

As part of the activities of the Development division (DD) function, Novartis experts organise and streamline efforts with the objective to phase-out all Substances of Very High Concern (SVHCs) in CPD. This included the proactive revision of SVHC related risks and the implementation of new business processes. Novartis intensified additional research projects to identify SVHC solvent alternatives in 2013. SVHC are categorised in appropriate sub-classes, e.g. diglyme is part of the glymes sub-class prior to trigger long-term replacement efforts including education by the internal Novartis Green Chemistry Network. More than 10 long-term research projects have been started in this context by Novartis so far. Two of these projects specifically aim to identify alternatives to polar aprotic solvents.

Generally a substitute solvent should comprise the following properties:

(i) comes from a renewable feedstock; (ii) has a low carbon footprint; (iii) is biodegradable; (iv) is not soluble in water; (v) is not a volatile organic compound (VOC) (boiling point not too low); and (vi) is easy to recycle (boiling point not too high).

Novartis initiated the diglyme substitution process in 2013 in the context of its ongoing ‘Green Solvent’ approach of continuously improving manufacturing processes. Since 2013, Novartis has tested 16 different solvents as possible alternatives for diglyme for the manufacture of indacaterol. In 2014 n-butyl acetate was identified as the most promising potential substitute.

Solvents in the synthesis process of indacaterol have to fulfil minimum criteria regarding yield, amount of by-products, regio-control, purity and the occurrence of genotoxic impurities. The use of possible replacement solvents must lead to comparable quality results with regard to safety, stability and efficacy to obtain approval for the changed manufacture procedures of the API by the different health authorities. The new process and the resulting pharmaceutical products have to be validated as being compliant with legislation in force in each country where medicinal products containing indacaterol are registered. To fulfil these requirements an integrated team approach is necessary for the revalidation of the synthesis. This includes expertise from a variety of disciplines (e.g., process engineering, industrial chemistry, analytical chemistry, microbiology, statistics, manufacturing, and quality assurance). The availability of a substitute solvent is of high importance to ensure an uninterrupted production of the API and to fulfil the duty to supply. For the substitution of diglyme with n-butyl acetate, no concerns are expected regarding the availability of this substitute solvent. n-butyl acetate is registered by 17 companies in accordance with REACH regulations, with an EU-wide annual tonnage of 100,000 to 1,000,000 tonnes. The current market costs for the substitute substance based on publicly available information are more than 4 times lower with 36 €/L for n-butyl acetate compared to 156 €/L for diglyme.



37

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

7.2.3. Technical and organisational

The modification of the manufacturing process of indacaterol is accompanied by manifold organisational changes, starting with revising the internal information systems, including work instructions, quality information documents, inspection plans, costing estimation, shipment preparation and various other tasks which were to be implemented in the databases. The project team for the substitution project consists of a multi-disciplinary team across different countries and areas of specialisation including Research and Development, Supply Chain, Purchasing, IT: SAP master data, Production, Quality Assessment (QA), Health, Safety and Environment (HSE) and others.

Further organisational activities include the purchasing of raw material, dispensing and vendor approval activities. This includes the identification of specific material dispensing activities that may be important for production and formulation processes. Health, Safety and Environmental Risk Assessment activities have to be performed.

Results from technical development performed in the PD Lab have to be verified for their industrial scalability and therefore performed prior to a technical scale-up campaign. This quasi-prototype process will ensure that the diglyme substituting solvent n-butyl acetate does not cause technical problems in larger scaled production facilities. However unforeseen technical difficulties can be encountered when the validation campaign is manufactured at commercial scale.

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX With a Quality Risk Assessment (QRA) process, the risk accompanied by the change will be assessed and documented. This involves a thorough safety assessment of the modified process and defines engineering controls and containment - process measures needed to implement the modified process.

Change control activities are a well-known concept of cGMP and focus on managing changes to prevent unintended consequences. Certain major manufacturing changes require regulatory filings



38

and prior regulatory approval. Effective change control activities (e.g., quality planning and control of revisions to specifications, process parameters, procedures) are key components of quality systems.

Production campaigns for all relevant intermediates of indacaterol synthesis have to be performed. Here, the results from PD Lab and scale-up campaigns are implemented as realistic production of the API. This information is an essential part of the core regulatory documentation that has to be submitted to the health authorities. The production campaigns include a complete blank batch, the production of the API itself and cleaning steps. Altogether the information from this large-scale manufacturing process builds the basis for the change application.

Execution of all steps of the production campaign including process validation will take at least one year. This timeframe is required due to the large number of production steps in the synthesis. Following production of the validation batches of the API and Drug product, stability data is required to be generated and a significant amount of registration work. In summary, for successfully completing the process change and the technical activities, a duration of seven years is anticipated.

7.2.4. Regulatory Challenges

A medicinal product may only be placed on a market when a marketing authorisation has been issued by the local health authority. To obtain one, the applicant must first submit an application for marketing authorisation to each health authority. The requirements of format and content of an application dossier are different in each country. The CTD (Common Technical Document) is the most commonly used format, it is an internationally agreed format for the preparation of applications to be submitted to health authorities in the three ICH (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use) regions of EU, USA and Japan, regrouping all the non-clinical, clinical and quality documentation pertaining to a product. In the CTD, the quality documentation including the manufacturing process of the drug is contained in Module 3.

Throughout the life of a medicinal product, the marketing authorisation holder is responsible for the product which is placed on the market and is also required to make any amendments to the approved dossier, including to its quality part, for instance any change to the manufacturing process. Such a procedure is a variation to the terms of the marketing authorisation which is strictly regulated. Variations to medicinal products can be classified in different categories, depending on the level of risk to public health and the impact on the quality, safety and efficacy of the medicinal product concerned. Changes which have the highest potential impact require a complete scientific assessment before being implemented, in the same way as for the evaluation of new marketing authorisation applications. This system of marketing authorisations ensures that the medicinal product is assessed by each health authority to ensure compliance with local requirements of safety, quality and efficacy.

In Europe, the centralised procedure is a harmonised process to register a product in all EU/EEA countries via a single assessment. However, there is no global harmonised approach. Each region has its own set of specific regulatory requirements that must be fulfilled by the marketing authorisation holder. For instance, the appropriate level of detail on manufacturing process and controls to be included in each local core regulatory dossier can vary between countries; and supplementary documentation can be requested in specific ones.



39

Also, when comparing the local legislations for post-approval changes, the same change may not be classified in a consistent manner, resulting in the necessity to manage multiple dossiers for the same variation. If the local legislation classifies a change as “minor”, no prior examination by the health authority is required before it can be implemented. However, the marketing authorisation holder must afterwards still submit to the health authority the updated core regulatory documentation corresponding to the relevant change. For a “major” change, the documentation must be provided to the health authority in a variation package which must be approved before implementation can occur. The health authority reviews the dossier and may request additional information to the marketing authorisation holder before approval can be granted. Provision of this information may require additional experimental work or issuance of commitments to the health authority to provide supportive data at a later stage. In some countries, the procedure is suspended until supplementary information is provided which can even further delay the approval timeframe, delaying therefore the implementation step that much longer.

This lack of harmonization in change classifications, requirements for the core submission documents and huge variance in health authorities’ review timelines are the main factors d the planned diglyme substitution implementation timelines.

Additionally, the manufacturing process has to follow GMP practices, which means that the marketing authorisation holder must comply with numerous requirements, including documentation and quality control procedures. This includes maintenance of suitable rooms, certified facilities and trained personnel. These requirements are not only valid for the first time a drug is synthesised but for the whole life-cycle of a drug or pharmaceutical procedure. Any changes on the originally approved drug manufacturing process requires therefore the execution of Change Control programs which are essential elements of any pharmaceutical quality assurance system. Once the technical feasibility of a change can be demonstrated on a laboratory scale the proposed production modifications have to be performed on a larger scale (production campaigns) with various standardised test procedures being performed to provide proof that the change has no adverse quality impact or influence on the quality of the final product. All of these tasks are managed via a Change Request (CR). Details on this legislation in Europe can be found e.g. in Annex 15 of the EU GMP Guideline and in Chapter 5.23 of the EU GMP Guideline. The structure of Change Control procedures are described by the FDA as a process which includes QM statements, operating instructions for Change Control and Change Request documents (FDA News, 2015). In line with applicable law, the type of change, the evaluation of the change, time-frames and measures are presented to the health authorities. This documentation must prove that the change was assessed appropriately, e.g. by risk analysis. For diglyme substitution, a CR has to be developed for all synthesis steps of indacaterol following the particular step where diglyme is being replaced.

Furthermore, stability testing of the API and DP is an essential part of the documentation to be generated. Various guidelines including ICH Topic Q1 A (R2) and the WHO Technical Report Series, No. 953, 2009 Annex 2 describe in detail the testing procedures for stability tests on APIs. The purpose of these testing procedures is to provide information on how the quality of DP and API varies over time under the influence of specific storage conditions with varying environmental factors. As



40

important quality criteria, degradation of API and other ingredients, appearance, physical attributes and functional tests (e.g. hardness, re-suspendibility) are evaluated.

For the substitution of diglyme, technical line functions within CPD develop the technical source documentation which contains reports comprising key aspects: the substitution of the solvent is reported and justified. Possible toxic effects of n-butyl acetate as the substitute have to be investigated and compared with diglyme using appropriate studies. The reports must clearly and evidentially demonstrate the absence of critical effects on product quality of the API. Major quality aspects in this context are for example the residual solvent profile and the impurity profile which has to be comparable for both solvents. This needs a detailed investigation and documentation of the properties of both solvents. The change in the manufacturing process of indacaterol will entail several technical adaptations and the impact of these changes is assessed by means of process validation, with the preparation of applicable validation reports. This technical source documentation can only be drafted once the API, from the new synthesis has been industrially produced (manufactured), validated, tested and released by QA. The lead time for the production of the API is approximately two years. The API testing monograph and release specification are required to be amended to reflect the solvent change. The current testing monograph for indacaterol requires diglyme analysis. Following the production of the API, stability data is required to be generated. This data generation requires 12 months. The technical source documentation is used to write the core regulatory documentation which will be used for submission by Novartis to each local competent health authority together with justification that the changes have not adversely impacted the quality, efficacy or safety of the product.

For most regions, approval within two years of submission is expected, however some countries could take longer to approve a change due to additional requirements and unpredictable timelines for their internal review process. Furthermore, there is no guarantee that health authorities will approve the proposed changes and additional data could be requested to support the approval. The lead time for provision of such data could be lengthy (range of 6 to 18 months).



41

7.3. List of actions and timetable with milestones

The challenges of experimental and regulatory tasks associated with the substitution of diglyme in the manufacturing process of indacaterol are presented in Chapter 7.2. In this chapter corresponding tasks, milestones and the expected timeline of the substitution are presented in more detail. At Novartis, most steps of the change process are separated between the two different business units, the development division and Commercial production division. These tasks include the most critical steps before the core regulatory documentation can be prepared and submitted to the different national health authorities for evaluation. XXXXXXXXXXXXXXXXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

7.3.1. Initiation and implementation of substitution process

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Table 19: XXXXXXXXXXXXXXXXXXXXXX

Task Working days Start Date End Date

XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX

XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX

XXXXXXXX XXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



42

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX


Task Working Days Start Date End Date

XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX XXXXXXXXXXXX





XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX


Task Working days Start Date End Date




XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

7.3.1.1. DD activities

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



43

XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX


Task Start Date End Date









7.3.1.2. CPD activities

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



44








XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX







XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



45







7.3.2. Regulatory tasks

Manufacturing processes of API are highly controlled and regulated. The same applies for changes to these manufacturing procedures, which have to be reviewed and approved by each local health authority ensuring that the process still meets quality, efficacy and safety requirements (post-approval change).

A variety of different legislation has to be considered resulting in an extensive set of experimental and documentation efforts. Section 1.4 gives a brief overview of the steps and regulatory requirements that have to be followed after or in parallel of a process change and prior to product market launch (Validation Campaigns, Analytical Method Transfer, Change Control and CR activities). Stability tests are to be performed before the submission in some countries. After successful submission of the core regulatory documentation (documentation summarising all gathered information and being presented for application at the national authority level), the applicant has to wait for the approval of each submitted procedure before implementation.

XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Overall, the substitution of diglyme has to be performed under conditions that fulfil the numerous regulatory requirements from all countries where medicinal products containing indacaterol are approved, regarding reproducibility, product quality and documentation.

7.3.2.1. Analytical method transfer

Scale-up and technology transfer processes are standard activities in the pharmaceutical industry which must be performed in accordance with pharmaceutical quality systems and standard procedures which ensure a high robustness of the manufacturing process. This includes critical-to-quality product



46

attributes and process parameters, process capability, manufacturing and process control technologies. Robustness can be seen as the ability of a process to demonstrate acceptable quality and performance, while tolerating variability in inputs. Robustness depends on formulation and process design. Technology transfer is e.g. described in the ICH guideline Q10 on pharmaceutical quality systems. Table 26 illustrates the time frame of the analytical method transfer process.




7.3.2.2. Change Control and Change Request activities

In order to follow GMP, manufacturers must comply with various requirements. Change control is a core concept of cGMP that focuses on managing changes to prevent unintended consequences and ensures that changes are implemented in a controlled and coordinated manner. XXXXXXXX XXXXXXXX XXXXXX XXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX




7.3.2.3. Other tasks

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.



47


Task Working Days Start Date End Date

XXXXXXXXXXXX XXXXXXXXX XXXXXXXXXX XXXXXXXXXXXX






7.3.2.4. Stability testing

Some health authorities request long-term real time stability data (6-12 months) on API and/or DP as part of the submission of a change which significantly increases the time until approval can be granted. Different regulations including ICH Topic Q1 A (R2) or the WHO Technical Report Series, No. 953, 2009 Annex 2 describe in detail the testing procedures for stability tests on API. Various analyses have been done to identify suitable testing conditions for WHO Member States based on climatic data. Those Member States that have notified WHO of the long-term stability testing conditions they require. The purpose of these testing procedures is to provide information on how the quality of DP and DS varies over time under the influence of specific storage conditions with varying environmental factors. In order to be able to reduce the amount of stability testing required, the number of different long-term testing conditions was reduced to four, which match with the climatic conditions of the target markets categorised in four different climatic zones. This concept is described in regulatory guidelines and pharmacopoeias and has become an established standard in developing finished pharmaceutical products (FPPs). As important quality criteria, degradation of API and other ingredients, appearance, physical attributes and functional tests are evaluated.

7.3.2.5. Core Documentation preparation and submission

• XXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX



48


Task Working Days Start Date End Date XXXXXXXXXXXX XXXXXXXXX XXXXXXXXXX XXXXXXXXXXXX











• XXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX


Task Working Days Start Date End Date XXXXXXXXXXXX XXXXXXXXX XXXXXXXXXX XXXXXXXXXXXX







49

7.3.2.6. Core Regulatory Documentation review period

As a general remark and described in chapter 7.2.4., the worldwide pharmaceutical regulatory landscape varies significantly in terms of classification of changes, data requirements and evaluation timelines from one health authority to the other. In globally acting companies as Novartis, an extensive staggered approach must be carried out by regulatory departments to meet all country specific requirements, meaning that before being submitted, the global documents are first being locally adapted to each health authority-specific need.

The goal is to ensure a compliant drug supply for patients all over the world. Final implementation of post-approval changes, like the substitution of a solvent in a manufacturing process, is limited in a global environment until the last global approval is obtained.

In the following paragraphs, some major regional differences in terms of data requirements and evaluation timelines are discussed.

As a general rule for EMA, for major variations of Type II, it can take up to 2 weeks to perform the validation of the dossier; then a 60-day evaluation period will apply, during which the CHMP, the committee assessing the regulatory application for EU, will review the dossier. This procedure can be delayed, if the CHMP request further clarification during the review of the documents, meaning that the evaluation period will not be continued until the additional information is provided by the company (clock stop). As a general rule, a clock-stop of up to 1 month applies. Then assessment of responses will take up to 30 or 60 days depending on the complexity and amount of data provided by the MAH. Upon receipt of the responses from the MAH, the procedure will be re-started according to the same principles as the ones applied at the initial start of procedure. This means that EMA/CHMP can request several rounds of clarifications as needed. According to industry and associations surveys (European Federation of Pharmaceutical Industries and Associations) survey, the time until decision of the health authority for a Type II variation can take more 6-8 months (with clock stop). (Escher report 2014, Frey-Stanislawski, 2007)

Importantly, EMA/CHMP only requires limited data on stability testing as part of the documents, and would be one of the first countries where the submission would take place. However other health authorities request long-term real time stability data (6-12 months) on API and/or DP as part of the submission, which significantly increases the time until submission can occur and approval can be granted. Stability testing requirements differ significantly depending on the region and climate conditions of the relevant country. For example, the Association of South East Asian Nations (ASEAN, comprising different Asian countries, including Indonesia, Thailand, Singapore) adopted the ICH guidelines on stability testing to the climate conditions in their countries (ASEAN Guideline on Stability Study of Drug Product) and request in their ASEAN variation guideline for pharmaceutical products that long-term and accelerated stability studies covering a minimum time period of 6 months are provided for the submission documents of a major variation. Several other countries request data on DP stability covering a time period of 12 months for the submission documents. As the production campaign for the interim product with the replacement solvent is foreseen to be carried out in Jan 2016, synthesis of the API will be finished by Nov/Dec 2016 Availability of the DP can be estimated for June 2017 at the earliest. Consequently, stability testing of API will not start before this time point, while testing of the DP will start even later, presumably



50

not before July 2017. As a consequence, results for long-term stability of API will not be available before November 2017, while for DP stability data will be obtained not before July 2018. Submission of the relevant documents in these countries and evaluation by the health authorities can only start after these dates.

The changes to the core regulatory documentation will have to be submitted for each registered medicinal product containing indacaterol (e.g. some countries where both medicinal products are registered will have to file distinct procedures). Prior to submission to local health authorities, often further customisation/ translations of the core regulatory documentation is required, which can take from 3 to 6 months in some countries.

Also, even though the regulatory package would be ready for local submission to the health authority, in some countries Novartis must wait for finalization and approval of another procedure still ongoing locally for the concerned product (e.g. other variation ongoing, renewal) before they can proceed to file this new CR.

Finally, some health authorities require variations to be approved in another country first (named the reference country) before the submission to the local health authority can actually take place.

A staggered submission would consist of sequentially submitting to reference countries first (like Europe, USA, Japan), then only to other ROW countries from climatic zone I to climatic zone IV. (Climatic zones IV would submit the variation package last, due to the necessity to wait for the availability of lengthy real-time stability data locally required).

See in Table 31 the current projections for such changes to the manufacturing process based on local intelligence (local regulatory requirements and experience).



51

Table 31: Tasks and time-frames from regulatory submission to the competent health authority until approval

Changes

Regions (overall 110 countries)

Classification

Eventual delay in submission (following first submission in any region)

Estimated approval timelines (following submission to health authority)

Necessity to wait for dispatch until finalization of a local ongoing procedure

Need for a reference country approval prior to submission

Best case Delayed case

New solvent, to be bundled with change of API and starting materials supplier, as well as a change in specifications of a API intermediate

EU: Grouped Type IB + Type II changes

No No 60 days

Up to 150 days in case of 1 round of request for supplementary information, several rounds possible

US: CBE-30 No No

If no response from FDA after 30 days, the change can then be implemented at risk.

Involves a review time of 6 months from date of receipt by the FDA.

ROW countries:

Regulatory relevant and approval needed in a majority of countries before implementation

Yes, expected around 30% of overall ROW countries

Yes, expected around 25% of overall ROW countries

Less than 12 months after submission in 50% of countries.

After submission, can take up to 24 months in some countries.

Each country approves the regulatory changes independently of other countries, so the timelines can vary significantly and are often hard to predict. As described above, evaluation by the EMA may take up to 8 months, while in other countries, the evaluation period predicted by the health authorities can take at least one year (e.g. Japan) or might take even longer. All factors considered, experience from previous submissions showed that the regulatory process from the time point when the core regulatory documentation will be available in January 2017, it can take up to seven years until the approval is granted in the last country where the medicinal product is marketed due to the aforementioned country specific data requirements, different health authority work flows and unpredictable timelines. Generally, these time-frames can only be achieved if the approval is granted without changes or additional information requests, which could prolong the duration of the total procedure by another 18 months, especially if additional tests have to be performed. The overall time-frame for the regulatory tasks is illustrated in Figure 4.



52

Figure 4: Summary and timeframe until global approval process is accomplished



53

7.3.3. Overall time-line of diglyme substitution

The most important milestones of the diglyme substitution process are presented in Figure 5. The production of the validation batches in CPD will not be complete prior to the latest application date, as a result a degree of uncertainty will exist regarding the technical success of the solvent replacement. As the global regulatory framework for post-approval changes to medicinal products are of high complexity and differ drastically, extensive efforts have to be carried out by Novartis to maintain a compliant supply of medicinal products to patients worldwide. First submissions of the core regulatory documentation to national health authorities are planned for the beginning of 2017. Initiation of the stability testing will take place as soon as the API and DP are available. When the country specific information are obtained and included into the respective submission documents, Novartis shall continue with its staggered submission approach. According to previous experience, it is foreseen that it will take seven years (until 2024) after the core regulatory documentation is submitted to the first health authorities until approval is granted in the last country where indacaterol is marketed, if no major obstacles occur. There is no guarantee that the approval of a process change will be granted and additional information or testing might be required by the health authorities. As implementation is limited until the last of the global approvals is obtained, and to in order maintain drug supply worldwide, the review period should not be shorter than 7 years.

Novartis is requesting authorisation to use diglyme until August 2024. Five years is required for the regulatory filing at a global level in a best case scenario. The requested review period of 7 years refers to the above described worst case scenario which includes additional stability data requests of individual health authorities, support for unforeseen variables relating to technical issues and delays form health authorities. Furthermore, Novartis needs to provide a review report to ECHA 18 months before the end of the review period in case a prolongation of the authorisation is required. The review period for diglyme substitution is mainly based on the timeframes for global regulatory approval. As a meaningful update on the regulatory status can only be provided after 5 years earliest, a review period of 7 years is needed. Figure 5 illustrates the overall timeline of diglyme substitution. Novartis will manufacture the n-butyl acetate variant of Indacaterol in line with regulatory approval.

As regulatory approval for the n-buyl acetate variant of indacaterol occurs, Novartis will increase production of this variant, whilst subsequently decreasing production of the diglyme variant. Novartis plans to stop the use of diglyme earlier in case the approval by the respective health authorities will be granted earlier than currently anticipated.



54

Figure 5: Overall timeline of diglyme substitution



55

7.4. Monitoring of the implementation of the substitution process

The supply chain for medicinal products containing indacaterol is split among different business groups and countries. For an efficient and regulatory conforming diglyme substitution, an integrated team approach is necessary for the organisational tasks, the revalidation of the synthesis and the fulfilment of all regulatory requirements. This includes expertise and close collaboration from a variety of disciplines and business units within Novartis.

The progress has to be monitored very closely to ensure the success of the substitution within the expected time-frame. Process monitoring and documentation are also essential for fulfilment of regulatory requirements, as already described in detail. Therefore, Novartis developed an exhaustive project charter to

(a) Identify all tasks required to support the solvent replacement; and

(b) Track the status of the activities.

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.

Figure 6: XXXXXXXXXXXXXXXXXXXXXXXXX



56

7.5. Conclusions

The EU REACH Committee voted for inclusion of diglyme in REACH Annex XIV in February 2014 and the EU Commission added diglyme to REACH Annex XIV in August 2014. At Novartis, diglyme is used as a solvent in one particular synthesis step during the manufacturing of the API indacaterol. The active substance is used in medicinal products which provide effective maintenance bronchodilator treatment for adults with COPD. Novartis is currently the MAH of a number of products containing indacaterol; the first is which are currently marketed in over 110 countries; the second in the US and the third in over 70 countries.

Significant effort has been undertaken by Novartis since 2013 according to their ‘Green Solvent Policy’ with the goal to identify and replace solvents with a high hazard potential, including diglyme. n-butyl acetate was identified as the most promising substitute for diglyme and a substitution campaign was initiated to implement the alternative in the manufacturing process. The substitution process includes manifold organisational, technical and regulatory tasks over a long time-period of several years, in accordance with the requirements of extensive and diverse regulatory requirements worldwide associated with a change in a pharmaceutical manufacturing process. Numerous technical and regulatory tasks need to be performed in order to verify and document, under GMP conditions, that the change has no influence on the product in terms of quality, efficacy and safety.

Taking the technical readiness level and the extensive regulatory evaluation process into account, Novartis’ conclusion is that the substitution cannot be fully implemented before the sunset date, especially because of the timeframes for the approval of the health authorities where Novartis has no influence on. Since the sunset date for diglyme is in August 2017, a review period of 7 years until 2024 is necessary to validate the adapted manufacturing process and to get the necessary approval from health authorities worldwide. This review period takes into account unforeseen drawbacks such as additional stability data requests of individual health authorities, support for unforeseen variables relating to technical issues, delays from health authorities. Final implementation of this post-approval changes in a global environment is pending until the last of the global approvals to ensure compliant drug supply for patients.

The requested review period of 7 years refers to the above described realistic scenario which included safety margins in the replacement process of diglyme. Furthermore, Novartis needs to provide a review report to ECHA 18 months before the end of the review period in case a prolongation of the authorisation is required. The review period for diglyme substitution is mainly based on the timeframes for global regulatory approval. As a meaningful update on the regulatory status can only be provided after 5 years earliest, a review period of 7 years is needed.



57

8. REFERENCES

1. Anonymous, 2015. Application for authorisation: DNEL setting for reprotoxic properties of digylme. RAC/33/2015/08 rev 1, final, European Chemicals Agency, Helsinki, Finland, 05 June 2015.

2. ASEAN VARIATION GUIDELINE FOR PHARMACEUTICAL PRODUCTS, FINAL DRAFT 7.2 2013.

3. ASEAN Guideline on Stability Study of Drug Product, Updated 15 May 2013

4. CHAD Chemistry Network, 2011: CHAD green solvent guide.

5. Escher report (2014) Improving the EU system for the marketing authorisation of medicines - Learning from regulatory practice

6. www.fda.gov

7. FDA News: Pharma Change Controls, [accessed: 22.01.2015]; http:--www.fdanews.com-ext- resources-files-The_Food_And_Drug_Letter-2013-Pharma-Change-Control-Peither-ExecSeries.pdf

8. Frey-Stanislawski E. (2007) Revision of the Variation Regulations Commission Regulation (EC) No. 1084/2003 and No. 1085/2003 Industry-Proposals and Consultation Paper from the European Commission, Impact on Industry and Health Authorities. Master-Thesis.

9. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

10. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

11. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

12. Warner, J. & P. Anastas, 1998: Theory and Practice, Green Chemistry.

13. Anastas, P. & J. Zimmernman, 2003: Design through the Twelve Principles of Green Engineering.

14. Global Initiative for Chronic Obstructive Lung Disease, 2011. Pocket Guide to COPD Diagnosis, Management, And Prevention, 2011

analysis of alternatives - europa. analytical method transfer ... declaration. analysis of...

Documents