Final Report Summary - CCRYSTAL (Innovative equipment and methodologies for APIs continuous crystallization)
The expiry of patents for major blockbuster drugs and the entry of low-priced generic versions are expected to increase the consumption of generic drugs, with the weak pipelines of major pharmaceutical companies amplifying this trend. With many patent expiries occurring during 2010 - 2012, generic consumption is expected to peak, and in turn increase the competition between pharmaceutical producers.
The global pharma players are looking at sourcing cheaper APIs, particularly in the situation of weak purchasing power worldwide. This is because of patent expiry of the blockbuster drugs, cut in medical costs in developed world and rise in R&D spending. Therefore, the opportunity descends upon low cost producers such as Chinese API makers. The project aimed at introducing advanced technologies to complete continuous manufacturing processes of pharmaceutical substances with enhanced, mainly continuous, crystallisation technologies, affording fundamental new approaches to obtain products with high purity levels and constant and reproducible crystalline forms at a competitive price, enabling EU APIs industry to regain his competitiveness against Asian producers. The innovative technologies, jointly with the new processes for four selected APIs developed during the two year of the project lifetime, are based on a new concept crystalliser and on continuous mode operation. From the point of view of drug manufacturing economic analysis, the new technologies are especially important for the market of generic APIs. Generics are changing the companies' strategies. The incidence of drug substance manufacturing cost on the price of the final pharmaceutical specialty was really not relevant for under patent drugs. This generated small or no attention to API manufacturing cost reduction and, as a consequence, process improvements. Now generic companies are experiencing more and more competition, and their business logics resemble those of fine chemical producers. So they have to improve their processes in order to remove the more relevant inefficiencies that are related to organisational aspects as well as to raw material costs reduction. As an example, for what the batch/continuous process comparison is considered, batch processes need to 'stop' because of repeated sampling and isolation at every intermediate step. Such a situation will force companies to innovate and move from 'quality by analysis' to 'quality by design' manufacturing. To reach this goal a complete command of the process has to be reached in order to avoid being victims of 'analysis paralysis'. To get out of this situation; alternate manufacturing scenarios are under consideration. Contract manufacturing organisations (CMO) specialise in certain chemistries, unit processes, and operations and campaign their production. Based on the amount of API needed, some of the products are campaigned in pilot plants to mini-plants. Modular plants are going to be an excellent option for batch or continuous processes. Innovation in manufacturing methods, improved process technologies, along with good manufacturing methods deliver excellent processes and are able to handle low to high production rates. Better manufacturing technologies for API and formulation are expected to improve profits at that level. For API, the improvements come in the way of improved yield, reduced waste, reduced and better solvent use, improved productivity, and improved business processes. For formulations they come with the blending, sise uniformity and uniform admixing, etc. In the supply chain, there are opportunities and one has to pick them. Improved technologies will also significantly reduce their carbon footprint. Companies that are excellent in manufacturing API and formulating drugs are now acting as the innovation driver more than the companies dedicated to invent new active principles.
Project context and objectives:
The CCRYSTAL fits into a context of continuous changes in the market it refers to, namely that of APIs.
The following trends were registered during the last two years:
- evolution of API regulatory requirements;
- expansion of the use of continuous technologies in API production processes;
- increase of technological level of companies producing in emerging countries;
- diffusion of generic drugs, with a push for reduction of manufacturing costs.
For all the previous issues CCRYSTAL studies can give useful outcomes.
Evolution of API regulatory requirements
More and more regulatory agencies (EMEA, FDA and the national agencies) ask for a 'risk management strategy' based on the concept of quality by design (QBD). Risk management means confidence that there is not possibility that the manufacturing complex give off drug substances not complying with authorised standard quality level. The manufacturing complex includes working procedures, manufacturing processes, and production facilities and so on.
It is a recent achievement (see guidelines Q11 of ICH) that a real control of constancy of the quality level of products can be obtained only by a deep knowledge of the chemical and physical principles of the manufacturing process, having a clear vision of relationships between operating variables and process performances. Production techniques able to maintain the value of critical operating variables in a defined range of operability are encouraged.
Continuous operations, which work in stationary conditions of operating variables, assure the simplest and surest method to achieve this fundamental goal.
Expansion of the use of continuous technologies in API production processes
The use of continuous reactions for API synthesis is more and more gaining popularity in the technical literature, in the scientific meetings, in the offer of commercial equipments for synthesis. This impressive increase in the interest of the scientific community is producing the first industrial production based on the proposed technology. Some big companies, like Lonza and DSM, started industrial plants for the manufacturing advanced intermediates for pharma and agrochemicals products: DSM operates an industrial plant for the production on a nitric ester; Lonza presented its proprietary process for Niacin.
The use of continuous technologies for other unit operations, like crystallisation, is the natural evolution expected in the next years. The reason is due to technical, organisational and economic reasons. Technical reasons refer to better performances of continuous in front of batch operations and on the improved control possibilities as previously explained. Organisational reasons derive from the fact that it is very inconvenient joining batch and continuous operations in series, due to the need of inserting storage steps which require complex scheduling of production phases. Economic reasons are due to the increasing plant cost when continuous and batch sections have to be connected together.
CCRYSTAL anticipated an evaluation of the problems connected with continuous crystallisation, highlighting advantages and problems, and indicating what is important to consider for the development of this operation. The obtained knowledge is not exhaustive, but it represents a good starting point for future works.
Increase of technological level of companies producing in emerging countries
Emerging economies were considered a real threat to the European API producers in the past years. Using the low manpower costs and the absence of regulations regarding the environmental impact of productions, Indian and Chinese companies produced a strong impact on the costs of chemicals, including those for APIs, inducing the European companies to exit from several productions and making more difficult the competition, maintaining the high quality standards required by the pharmaceutical industry.
Now the panorama is changing: oriental producers are more conscious of the relevance of the quality requirements (including reliability) and are interested to adopt producing processes with an higher technological level. India and China are now an interesting potential market for the new technologies, acting at the same level of the occidental companies.
Diffusion of generic drugs, with a pull for reduction of manufacturing costs
Reduction of welfare costs, without any reduction of the quality level, is a need for all countries in Europe and over the world. The cost of drugs is a significant addendum to the overall health costs, and its reduction constitutes a goal for all public administrations. Generic drugs comply with this need, and their use if strongly recommended by all governments. Government efforts to promote and reward expanded generic use should continue to pay off, notably in France and Japan. The Swiss market is also performing strongly, although it is limited in size by the country's small population.
Drugs worth USD150 billion will go off-patent between 2010 and 2017, helping to propel the generic-drugs market from revenues of USD123.85 billion in 2010 to an estimated USD 231 billion in 2017. Leading global generic drug markets will grow to USD 221 billion by 2016. US generic growth is forecasted to rise to USD 104.1 billion by 2016, due to expiry of patents on major drugs such as Lipitor and Zyprexa, increased pressure for generic use from Medicare drug plans, and the gradual emergence of the biosimilar market.
On the other side revenues for the pharmaceutical industry have to be quite high, in order to cover the costs of development of new drugs, able to better treat all kind of illness. Reduction of drug prices has to be reached by the reduction of manufacturing costs, using always more efficient production techniques.
The market of generics is based on three fundamental values: quality, reliability and costs. Production efficiency is a necessary requisite in order to compete in that field. Improvement of processes is, may be, the more effective tool in order to gain advantages over competitors in this kind of business. Also from this point of view continuous technologies are a powerful help to reach a high level of efficiency.
In this frame, the CCRYSTAL represent a corner stone by itself, being industrial crystallisation a major problem for the pharmaceutical industry, and its success will represent an unprecedented innovation in the field. The project re-invents the whole value chain of APIs production, providing EU SMEs with competitive solutions against their Asian competitors and guarantying at the same time the high quality and safety usage of medicines for the citizens. At the same time more and more it appears clear that the efficacy of a drug is determined by its bioavailability, depending on its physical form. New polymorphic forms, solvates or cocrystal are considered as suitable tools to improve beneficial effects also of old drugs. As a consequence, new improved crystallisation methods and apparatuses are required to produce the wanted pharmaceutical forms.
In this framework, the objectives of the CCRYSTAL project have been:
- applying improved crystallisation technologies to crystallise pharmaceutical substances, affording fundamental new approaches to obtain products with high purity levels and constant and reproducible crystalline forms at a competitive price, enabling EU APIs industry to regain his competitiveness against Asian producers.
- testing the new technologies on competitive APIs, eventually developing the whole chemical scheme in a continuous manner;
- demonstrating that the use of the improved processes increase the final product reliability.
The workplan has been divided into predominantly experimental / technological (WP1, WP2, WP3 and WP4) and demonstration activities (WP5) that combine and cross-refer synergistically. Other two WPs have been included in the workplan, that regard the dissemination and exploitation WP (WP7), and the management one (WP8). The list of WPs is included hereunder:
WP1: Sizing continuous crystallisation train and operating procedure development (RTD)
WP2: A user-friendly crystallisation process modelling environment for optimal design, operation and control of continuous crystallisation process (RTD)
WP3: Use of high performance crystallisation unit to meet compositional quality requirements of API 1 (RTD)
WP4: Use of high performance crystallisation unit to increase API two recovery yields and to meet physical (RTD)
WP5: Validation Trials (demo)
WP6: Dissemination and exploitation of project results (other)
WP7: Project management (MGT).
Project results:
WP1 have been focused on the design of the repeatable crystallisation unit and on the sizing of the crystallisation train. Then the effort has been put in the development of a methodology to tune actual production. The work performed for this WP has been done mainly by the partners Amtech and Autico, with the collaboration first of all of Serichim and then of the other partners.
Amtech exploitable results are two systems: a lab scale repeatable rig unit and a continuous crystalliser.
Repeatable rig unit: Coflux
The Coflux system allows monitoring online and in a non-invasive way processes such as bacterial growth, reactions and crystallisations. Before the project, large scale system using this technology was available but this project has offered the possibility to apply the Coflux technology to smaller vessels, suitable for lab testing. This both opens a new market but also increases the sales of the larger systems since it will be easier for the users to test the technology prior to committing to the larger scale with a larger investment.
Real time process monitoring aligns with the current FDA directives to run better processes through QbD and process analytical technologies (PAT). Once returned, the lab scale Coflux will be used at a university in the UK to continue testing and generate more case studies in order to market this technology.
Continuous crystalliser: Coflore ATR
The Coflore ATR has been mainly applied to synthesis but its ability to handle solids and the possibility to obtain gradient cooling makes it suitable to be used for crystallisation processes too. However, in order to market the ATR as crystalliser, it is necessary to investigate further the solid handling capabilities (in terms of sizes, concentration and composition), the tendency to foul (especially on the colder surfaces) and the ability to influence the crystal size, size distribution and shape. These characteristics will more likely vary on a case by case basis depending on the system of interest but the know how generated by running a series of different processes will provide a list of possible actions and running strategies to cope with different demands.
Continuous crystallisation is a very new application and many end users are showing high interest in this approach.
The ATR produce during CCRYSTAL and the know-how generated will be used to run feasibility studies in order to facilitate user uptake and increase crystalliser sales. These activities will also affect the sales of the same system as a reactor. One of the key requirements of the industry these days is flexibility and multi-purpose equipment. So if it can be proved that the same equipment can be used for different process steps, this will increase chances of sales.
The know-how generated by this project on both systems will be used to make design modifications to improve performances and user experience too.
Concerning Wp2, the main goal is the development of a user-friendly crystallisation process modelling environment for use by the SME's employee in order to streamline their processes and operation of crystallisers.
The RTD PSE released a model-based engineering crystallisation tool, the gCRYSTAL, configured for use in the CCRYSTAL. Within the project this tool have been used to create three case specific model-based engineering tools, one for three of the following SMEs in CCRYSTAL (Flamma, Galentis, Merchav).
Before configuring the gCRYSTAL environment for use in the CCRYSTAL project, user requirements were gathered by interviewing various stakeholders within the SMEs. Relevant stakeholders were not only the envisaged users of a user-friendly tool, but also the people who will use the information provided by this tool for decision support purposes, such as managers, operators, etc.
The foreground resulting from WP2 is:
- A process model and control protocol for the cooling crystallisation of paracetamol (Merchav). Further results (foreground) generated in the process of developing the process model and control protocol are:
i. configured PSE's gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. estimated kinetic parameters using data collected from batch experiments;
iii. reconfigured gCRYSTAL for the same chemical system but continuous processing: describing the equipment and process configurations;
iv. performed predictive simulations of the performance of several process configurations, which differ with respect to:
- number of crystallisation stages;
- temperature in each stage;
- residence time in each stage.
- Process model and control protocol for the anti-solvent crystallisation of paracetamol (Galentis). Further results (foreground) generated in the process of developing the process model and control protocol are:
i. configured PSE's gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. estimated kinetic parameters using data collected from batch experiments;
iii. reconfigured gCRYSTAL for the same chemical system but continuous processing: describing the equipment and process configurations;
iv. performed predictive simulations of the performance of several process configurations, which differ with respect to:
- number of crystallisation stages;
- solvent composition in each stage;
- residence time in each stage.
- Process model for the crystallisation of Ornithine alpha-ketoglutarate (Flamma). Further results (foreground) generated in the process of developing the process model are:
i. configured gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. defined next steps required for:
- estimating kinetic parameters using data collected from manufacturing runs;
- optimising the crystallisation process with respect to yield and purity.
The SMEs will be able to use gCRYSTAL as configured for their case of interest to facilitate a wide range of steps in the workflow for design and optimisation of crystallisation processes: identifying dominant mechanisms, maximising the formation of the required polymorphs, estimating model parameters that allow extrapolation of knowledge, investigation of batch vs continuous operation, process scale-up and optimisation of process operation at the manufacturing scale.
WP3
Use of high performance crystallisation unit to meet compositional quality requirements of API 1 main objective was the development of the purification procedures for Flumethasone and Flumethasone intermediates, capable of separating chlorinated by products, in order to meet the very strict specifications prescribed by the pharmacopoeia. During the first reporting period, the revision of the existing chemical know-how has been performed in order to optimise chemical yields and crudes compositions along the synthetic path, while during the second reporting period the main objective was the development of separation procedures of chlorinated by-products from the main product acting on the solid intermediates occurring in the manufacturing process.
A completely new flumethasone purification process was developed, based on the separation of unwanted impurities at the stage of orthoacetate derivatives of flumethasone and related by-products. It makes use of a milling solid-liquid equilibrium operation. The new purification procedure allows the use of perchlorylfluoride fluorination reagent, which is a very cheap fluorinating agent in front of more expensive chemicals. Perchlorylfluoride is a chemical produced by Galentis and by only a second company all over the world, so that the new process is a valuable asset for Galentis or as a producer of Flumethasone, or as a supplier of perchlorylfluoride to flumethasone producers.
A process book was produced, describing all precess steps, based on Galentis experience and Serichim's new studies.
Purification steps were verified at the lab scale.
Beneficiary of the developed know how is Galentis.
Three different APIs were object of study in WP4, namely Gabapentin, Sulindac and Ornithine ketoglutarate. During the course of the project, for all of them, the research objectives were better specified and new opportunities were identified, to make more competitive the production processes. For all of them, it was recognised that the more important goal is to have a robust process which meets the quality requirements in terms of organic impurity contents.
As far as Flamma Spa, crystallisation process has been developed:
- to have an all continuous process, synthesis and recovery sections, obtaining a high quality filterable crystal suitable for continuous filtration, for Sulindac;
- to have a low cost process based on selective crystallisation steps for di-ornithine ketoglutarate.
Concerning the Sulindac, the previously stated goal aim of producing Sulindac Form II without traces of Form I was reached out of the project, working on drying procedure, while continuous crystallisation aims were focused on improvement of filterability of crude product and to the interconnection between continuous reaction and continuous crystallisation steps, without intermediate storages.
A continuous crystallisation operation was developed, able to obtain a well filterable product. A few percent of the unwanted polymorph was always present, but an improved drying procedure was able to convert it to the right form. Crystallisation was directly joined to the previous synthetic step, so that a complete continuous process was developed, starting from the parent thioether and ending with the filtered solid.
The entire process was documented in the Sulindac process book. Preparation was stepwise tested at the pilot scale, using ATR for the reaction step and a continuous two-stages crystalliser for the crystallisation step.
Foreground beneficiaries are Flamma for the Sulindac production process and AMT for the ATR qualification.
For Di-ornithine ketoglutarate, the new process overcame the typical problems of the the precipitation processes, namely processes where the insoluble product is formed by salt formation from an acid and a base, where crystal size is very fine, and the filtration and drying operation on the formed solid are quite troublesome. In this case crystallisation was carried out so that the crystallisation rate was quite slow, and the crystal size quite high.
While previous goals resulted not so attractive, the possibility of a completely new crystallisation process, significantly more convenient of the existing one, was identified and the process developed, as previously explained.
The new process uses Ornithine Hydrochloride as raw material. The process is based on two crystallisation steps, that can be operated both batchwise and continuously. Only the batch crystallisations were tested at the bench scale.
The new process allows a reduction of raw material cost and avoids a complex electrochemical step in the overall synthetic path. All solvents can be recovered and recycled, so that the process can be classified as environmentally friend.
Beneficiary of the process is Flamma.
For Gabapentin, the main result is the achievement of the right quality profile by a crystallisation step of the raw material and the tuning of the Hofmann reaction operating conditions. Having reached this result, it was possible to confirm that the recovery and purification process, based on continuous l-l extraction and crystallisation operations, can give a significant containment of production costs. Gabapentin is a high volume-low cost API, for which manufacturing cost has to be minimised in order to compete on a very challenging market.
The improved operating conditions were fully examined, and a synthetic process defined and documented.
Samples were produced at the bench scale and qualified by one of the major producer of the API. However it was not possible to complete the process validation, including recycles.
Beneficiary of the developed know how is Merchav ltd.
WP5
Validation trials, was dedicated to validate project results ensuring successive effective exploitation by SMEs. The activities of WP5 have been performed mainly during the second reporting period, with the following objectives:
- finalisation of the repeatable unit and of the train testing and validation;
- process technical validations;
- the scale up feasibility study for the integration of the crystallisation unit into existing plants;
- drug master file guidelines for continuous crystallisation.
Process technical validations carried out in the frame of this WP have to be considered as a part of know how developed for Flamma, Galentis and Merchav.
Operational studies of crystallisation trains, experience of implementation of equipments in a pilot or production environment, as well as methodologies for interfacing batch-to-continuous equipments have to be considered as a know how available for all CCRYSTAL participants and, in general for all companies facing issues derived from continuous crystallisation use in their processes. This is the same for the guidelines worked out for the compilation of drug master files sections dedicated to crystallisation steps.
Operating and installation instructions of ATR and Coflore equipments are now available to help in their use.
Schemes for their connection to the ancillary equipments, feeding systems, thermostatting units and so on were developed, checked and documented, DMF writing criteria elucidated. All participants can have access to the worked documents.
Potential impact:
APIs: a short market presentation
APIs may be defined as substances or mixtures of substances intended to be used in the manufacture of a drug or any medicinal product that, when used in the production of a drug, becomes an active ingredient of that drug or medicinal product. APIs can either be classified as synthetic or biotech, based on the raw materials used in their production, and as innovator or generic, depending upon the customer requirement.
The API forms the most vital part of every formulated end product, and is an important part of the whole pharmaceutical industry.
The overall API market was valued at USD 101.08 billion in 2010, and is expected to grow at a CAGR of 7.9 % from 2011 to 2016. More than half of the global APIs market is dominated by North America and Europe, with Asia-Pacific forming the third largest but fastest growing API market in the world. The Asia-Pacific API market is highly fragmented, with a very large number of players competing for market shares. This market is characterised by an eclectic mix of companies, which range from large, vertically integrated manufacturers to distributors and importers.
Europe is the third largest regional market for APIs by revenue in the world. European API market revenue held a share of around 24.2 % of global API market revenue of USD 108.6 billion in 2011. Total revenue generated by the European API market was USD 26 288.96 million in 2011 and is expected to grow at a CAGR of 6.5 % to reach USD 38 255.67 million by 2017.This will be supported by healthy demand from generic and biotech API sectors.
There has been an increase in influence of API players from emerging economies such as India and China after the economic recession. The recession restricted the growth of Innovative sector in developed economies such as the US and Europe, as the Innovative sector requires huge investments. This has helped fuel the growth of generics market in Asian countries such as India and China.
The API market is facing a period of unprecedented growth as market dynamics have undergone a major change with the expiration of patents pertaining to global blockbuster drugs in the US. The consequences of the economic crisis has hit the Innovative drugs market hard, with less budgets allocated by the major players for the R&D of Innovative drugs. This has led to drying up of pipelines for new drugs, and therefore the market for generic drugs is quickly growing. Thus, the patent expiry factor is slated to drive the API market for the coming years.
European countries such as the UK, France, and Germany are promoting the use of Generics by providing incentives to the doctors for writing prescriptions relating to generic drugs and also to the pharmacists if they offer the generic equivalent of prescribed drugs. Thus, generic API market is expected to grow till the year 2016 as compared to the Innovative API market.
The expiry of patents for major blockbuster drugs and the entry of low-priced generic versions are expected to increase the consumption of generic drugs, with the weak pipelines of major pharmaceutical companies amplifying this trend. With many patent expiries occurring during 2010 - 2012, generic consumption is expected to peak, and in turn increase the competition between pharmaceutical producers.
The global pharma players are looking at sourcing cheaper APIs, particularly in the situation of weak purchasing power worldwide. This is because of patent expiry of the blockbuster drugs, cut in medical costs in developed world and rise in R&D spending. Therefore, the opportunity descends upon low cost producers such as Chinese API makers. The API players in China also see the growth in local market driven by couple of factors such as growth in the generics sector, government's plan to develop multi-level insurance system and the development in biotech drugs.
As far as WP1, crystallisation of high value chemicals has traditionally been performed in batch reactors. By virtue of improved surface to volume ratio, flow reactors offer the potential unprecedented improvements in product quality and yield. The challenges of building flow reactors for this purpose however are formidable. There are two essential elements to developing flow crystallisation processes. Firstly, equipment is required to study crystallisation rate at comparable surface to volume ratios to a flow reactor. In this project, a heat balance reaction calorimeter in glass was developed in glass, Coflux. The second requirement is to have a flow system which can carry slurries in an orderly fashion. A flow reactor was designed for this purpose during the project, Coflore ATR. The technical challenges in terms of equipment development were formidable. In the case of the calorimeter, the challenge was to develop a multi element non-metal limpet jacket with variable area control and heat measuring capabilities. For the flow crystalliser, this required a dynamically mixed tubular system with multi zone temperature control. Testing of these systems was carried out in two phases. The first phase was water testing undertaken by Autico. Both systems performed well in these tests. The second phase involved product testing and this work was carried out by Serichim. A number of problems were identified during this phase which is normal for new technology. As result of the testing work carried out by Serichim, a number of further modifications and improvements have been made to both machines. Further tests carried out after the end of the project suggest that these problems are being addressed. For AMTECH, this project has been extremely valuable in identifying specific design requirements for developing and operating flow crystallisation processes. We are now investing more money in further development and testing.
Concerning WP2, the SMEs will be able to use gCRYSTAL as configured for their case of interest to facilitate a wide range of steps in the workflow for design and optimisation of crystallisation processes: identifying dominant mechanisms, maximising the formation of the required polymorphs, estimating model parameters that allow extrapolation of knowledge, investigation of batch versus continuous operation, process scale-up and optimisation of process operation at the manufacturing scale.
More specific benefits of the use of mechanistic models are:
- increased R&D efficiency:
i. use fewer experiments to adequately characterise a process / processing step (typically the number of experiments can be reduced by 50 - 70 %);
ii. design additional experiments with a higher information content (as compared to statistical techniques for the design of experiments);
- increased efficiency and reduced risk for tech transfer activities by turning data into knowledge that can be applied for extrapolation to other scenarios with known confidence and reduced experimental requirements;
- efficient investigation of the feasibility (and desirability) of batch-to-continuous: design of a continuous process does not have to be supported by continuous experiments, i.e. no need to significantly invest in new lab equipment.
WP3 had the goal of developing crystallisation procedures for flumethasone and flumethasone intermediates capable of separating chlorinated by products, in order to meet the very strict specifications prescribed by the pharmacopeia.
Flumethasone is the basic ingredient of a number of drug substances, like flumethasone acetate and flumethasone pivalate, as well as an intermediate for other steroideal active substances. Galentis, one of the few producers of steroids in Italy and in Europe, will be in the position to start with a new production in its facilities in Marcon. The previous purification method, part of a pre-existing know how, was too much complex, not reliable and costly, so that it was impossible until now to start a regular production.
Opportunity to start with this production has to be verified through an updated market analysis. It has also to be evaluated in the frame of Galentis new strategies. In this perspective a second chance is the valorisation of the knowhow jointly with other companies involved in steroid manufacture in Europe. Sicor (a TEVA subsidiary), Hovione, Aurisco and Farmabios are presently producers of Flumethasone derivatives; other companies, producers of other steroideal drugs, are potentially interested to a technology for flumethasone.
The qualifying element of the developed process for flumethasone is the quality of the final product. It is really better than that required by the European Pharmacopoeia, sixth edition, for flumethasone pivalate, which sets a limit of 1.5 % for chloromethasone pivalate (in front of 0.1 % given by our process). No limits are reported by the current Pharmacopoeia about 6-beta isomers, deriving from the first fluorination step, but also for this by-product the achieved concentration level is less than 0.1 %.
The second point of advantage of the developed process over the other is related to the cost of chemicals. Perchlory fluoride is a cheap reagent in front of other fluorinating agents, like Selectofluor, and it gives less co-products. Galentis is the only producer of perchloryl fluoride, so that it could have a revenue from the cells of this chemical also if the flumethasone should be produced by a different company according to Galentis proprietary technology.
WP4 was dedicated to the development of crystallisation steps for products in the pipeline (already in or under development) of two SME participants, Merchav and Flamma S.p.a. Specifically, the work for Merchav has been done on the Gabapentin, while in the case of Flamma, the Ornithine and the Sulindac have been investigated.
Gabapentin
Gabapentin continues to be one of the biggest drugs for quantities yearly produced. The production volume went over 2000 tons in 2010 and only slightly decreased in the last two years. Reasons of this decrement in the drug consumption were some restrictions on its use for treatment of secondary diseases, like headache, that can be treated with products with less side effects. For the future years a slow increase of volumes is expected, and only for the end of 2020s new molecules, like Pregabalin, might substitute Gabapentin in some uses.
Jointly with the volume increase prices were go down below USD 100/kg. Production costs have to be lowered in order to compete on the drug substance market. A number of producers are now present on the market: Teva, Medichem, ZachSystem, Erregierre were the first producers of the generic drug, while recent new players are mainly Indian or Chinese, like Jiangxi Synergy, Jiangsu Hengrui or Anjan Drug. Recently Teva and Pfizer, the API originator, reached a settlement about their patent litigation on the generic versions of Pfizer's Neurontin capsules and tablets sold by Teva and its subsidiary Ivax. Now Teva sells the generic Gabapentin under license by TEVA.
All the major producers use a process including the Hofmann rearrangement as the last synthetic step. Teva and ZachSystem, which are the biggest generic producers, use this reaction, but different purification strategies.
The shift from batch to continuous approach based on the process developed by CCRYSTAL project is under discussion with both those two major players and could be proposed to the new producers. As the change of technology for a so high capacity product is a very complex matter, it is forecasted that discussions and technical analysis will continue in the next year, but there is chance that a cooperation will start with almost one of the big producers, interested in cost reduction and reliability improvements.
Ornithine
The new process for Ornithine Ketoglutarate (OKG), the direct process, gives the advantage of the direct use of ornithine hydrochloride as raw material instead of that of ornithine free base. This last chemical has two disadvantages in front of the hydrochloride:
- its cost is about the double of that of the hydrochloride, because it is obtained from this last chemical by an electrochemical process;
- it is a very instable molecule, easily forming a lactam by standing.
An inconvenient point of the new process is that it makes use of one chemical more, the trimethylamine, with the related cost. In principle, trimethylamine can be recovered and recycled, so that its cost can be minimised.
It was calculated that the new process affords a raw material cost several Euros lower than that of the process based on the free base, also without the trimethylamine recycle: Flamma will be in the position of adopting the new process, if re-registration operations will be accepted by the regulatory agencies.
Sulindac
The market situation of Sulindac changed dramatically in the last two years. It is a non-steroidal anti-inflammatory drug, which reached a volume of 150 tons/year in the first 2000s. Recently some new APIs with a COX 2 inhibition activity substituted Sulindac in several uses, so that the annual production went down to 50 - 100 tons/year. It is not clear if this trend will continue in the next years, or if it stops and the market tends to grow.
Flamma invested a big effort in the development of a proprietary process, including preparation of the intermediate, SUL4, and, in the frame of CCRYSTAL, its transformation in Sulindac through the continuous steps of oxidation and crystallisation. The overall process is cheaper than traditional processes, but this is not the only condition to go on the market, because of costs of registration of the product joint to the new technology. The industrial use of the Sulindac continuous process depends on the future market trend. In the next few years it will be clear if the new APIs completely substitute Sulindac, or if a category of diseases continues to be treated with the old drug. It will depend on the results of the post-marketing surveillance outcomes carried on by regulatory agencies.
In any case, the development of the new process demonstrated the advantages of a chain of continuous operations, without intermediate storages. It allowed the use of a solvent in the crystallisation step that dissolves very well the product, but that is reactive with it, forming on the long-time an ester. Control of contact time between this solvent and Sulindac is the key to take advantage from the good solvent properties.
WP5 was dedicated to validate project results ensuring successive effective exploitation by SMEs. In order to reach these goals the validation was done taking into account the provisions and international protocols required for APIs production.
The objectives of this WP are mainly three:
- to construct the repeatable unit and the crystallisation train and to test and validate them;
- to scale up the crystallisation unit;
- to perform validation trials for the successive scaling up to commercial amounts production of the APIs;
- to analyse how to write the drug master file.
The outcomes of this WP are both technical and methodological.
The CCRYSTAL project has enabled these new batch and continuous reactor technologies to be developed for full industrial use.
The Coflux repeatable rig unit performed in line with the requirements of the Technical Specification. As expected, heat up / cool down rates for the glass version of the reactors were slower when compared with similarly sized metal jacketed vessels due to the poorer heat transfer coefficients of the reactor material. However, the heating / cooling response rates were much faster than for jacketed glass vessels due the smaller volume of heat transfer fluid in the reactor tubes and high flow rates. The enthalpy, process power and other control functions performed as designed, with the process power function reading values accurately down to 10 Watts. Further development could reduce this figure further to enable use with less energetic reactions. The enthalpy value was useful for tracking crystal growth. From the construction techniques developed during the CCRYSTAL project, it would be possible to make both smaller and larger versions of the Coflux reactor. Developing a scientific glass version has widened the potential market for laboratory and pilot plant use. Larger versions can be constructed from industry standard metals such as stainless steel and Hastelloy or vitreous enamelled (glass-lined) steel reactors. Potential industries for this technology, other than pharmaceutical, includes fine chemicals, food, petrochemical, waste treatment and environmental.
The ATR crystallisation train unit was challenging to build, especially with respect to the heat transfer system developed for the reactor tubes. The heat transfer capabilities exceeded the requirements of the technical specification and the techniques used may be applied to other new technology systems. The addition by Serichim of temperature probes on the inlet and outlet heat transfer lines is a function that will be considered for future reactors. Further development of a multi-temperature zone version will widen the number of potential applications. Following the construction of the CCRYSTAL 1 litre ATR, further units have been built and successful process trials run. Scaling up of the ATR for larger applications is made easier by the development of the novel drive mechanism for the CCRYSTAL reactor which could be adapted for larger mass reactors. The Hastelloy reactor tubes can accommodate a wide range of chemicals and compounds but more exotic metals, such as monel and tantalum, could be used for reactor tubes for more demanding processes.
The use of an industry standard control system, with good analogue signal handling and control functions, is a key to developing prototype machines for industrial use. This approach saves time and reduces costs. Often, a specially designed or non-industrial control system is used for development which means that the whole system needs to be transposed on to another platform before it can be marketed. The development of the Honeywell HC900 control system, using the new 900 series operator control station, has enabled the reactor technologies to move forward from a control perspective due to the increased range of available functions and ease of programming, thus saving development time and costs. Data logging and reporting functions were developed for both reactor systems and included a useful screenshot download option. The operator was able to download data files in CSV format for reports.
Continuous crystalliser: Coflore ATR
The Coflore ATR has been mainly applied to synthesis but its ability to handle solids and the possibility to obtain gradient cooling makes it suitable to be used for crystallisation processes too. However, in order to market the ATR as crystalliser, it is necessary to investigate further the solid handling capabilities (in terms of sizes, concentration and composition), the tendency to foul (especially on the colder surfaces) and the ability to influence the crystal size, size distribution and shape. These characteristics will more likely vary on a case by case basis depending on the system of interest but the know how generated by running a series of different processes will provide a list of possible actions and running strategies to cope with different demands.
Continuous crystallisation is a very new application and many end users are showing high interest in this approach.
The ATR produce during CCRYSTAL and the know-how generated will be used to run feasibility studies in order to facilitate user uptake and increase crystalliser sales. These activities will also affect the sales of the same system as a reactor. One of the key requirements of the industry these days is flexibility and multi-purpose equipment. So if it can be proved that the same equipment can be used for different process steps, this will increase chances of sales.
The know-how generated by this project on both systems will be used to make design modifications to improve performances and user experience too.
Validation trials
Four APIs were object of study during the course of the project:
1. Ornithine ketoglutarate;
2. Sulindac;
3. Gabapentin;
4. Flumethasone.
For each one of them a suitable production process was developed, according to the SME's requirements, characterised by technical and economic improvements in respect to the state of the art. All developed processes, but one, contain one crystallisation step which determines the qualifying performance of the entire process. Exceptions are:
- Flumathasone, for which the qualifying step is a digestion step, a unit operation based on solid-liquid equilibrium, as crystallisation, but in a reverse sense, namely partial dissolution of impurities;
- Gabapentin, for which two critical crystallisation steps are present, required to reach the wanted quality and cost goals.
Batch and continuous operations were used, according to the physical characteristics of systems and to the process needs. For continuous operations performances of the new continuous crystallisers, designed and built up as a part of this project, were checked.
Some general teachings were derived from the performed studies. They can be categorised as physical, mechanical and control issues.
Physical issue regards the thermodynamic behaviour of solid-liquid systems involving APIs and related impurities. We have to start from the consideration that quality profile is the fundamental property of every API. No other achievements deserve value if the impurity contents of the final product do not comply with the pharmacopoeia standards or with the ICH rules. As drug substances are usually solid, crystallisation is the pivotal operation which generates the product quality. Other properties of the solid state, like the crystal habit or the crystal size distribution are less important.
Impurities often have molecular structures similar to that of the drug substances, so that they can arrange in the same crystalline reticulum, forming solid solutions. Separations in this case are not so sharp as in the case of eutettoid systems, and crystallisation is effective only if the reaction crudes have limited concentrations of impurities: reaction and crystallisation optimisations have to be carried out as a unique activity to obtain acceptable products. This was the reason why validation tests generally included both reaction and crystallisation steps.
Mechanical issue regards the rheological characteristics of the obtained solid. Continuous crystallisers have to assure the transfer of solids along the equipment, but it is heavily conditioned by the morphology of crystals. It is not matter of quality of the product; it is just a problem of operability of equipment. Particularly robust equipment have to be used when the solid is sticky: in some cases continuous crystallisation is not convenient, and only batch units can be used. For this reason, operability of ATR was checked with paracetamol, which is a 'good' crystalline product, not included in the 4 APIs for which the process was studied, while a more robust crystalliser was used to validate the continuous crystallisation of Sulindac, a very 'bad' solid.
Validation of ornithine ketoglutarate (OKG) processes was carried out batch wise. It was focused on the check of recycles, required in order to reach an economical process yield. A sequence of four consecutive trials was used in order to test the influence of recycles. Their effect was assessed by batch operations. Trials demonstrated that the developed procedure is able to consistently give a product with the wanted quality, with a process yield (including recycles) of about 80 %. Experiments showed that by a fine tuning of experimental conditions this figure should be also improved. Crystallisation operations gave well filterable products. All data needed for a cost analysis were collected, and a first evaluation showed that the new process saves a 20 % of the raw material cost in front of the current process.
The flumethasone critical operation, namely digestion, is inherently batch operation, which does not make sense to transform in continuous. In this case, the main problem resides in an accurate control of conditions and on the quality of the obtained solid. Standard analytical procedures were used to validate the critical operations.
Gabapentin process has two critical crystallisation steps. The first one is the purification of raw material by an extraction-crystallisation operation: it is a part of preparation of feeding material and has to be performed batch wise. The second one is the continuous crystallisation of the final product, with recycle of mother liquor: it should be performed in continuous mode; however it was validated batch wise because of unavailability of enough raw materials. Several preliminary tests were carried out, in order to tune the operative conditions, mainly for the Hofmann reaction step, and a final run was carried out, producing a 10 g sample of product with the following purity profile:
It was impossible to repeat the validation trials because of lack of raw material, and also recycles were not tested. However the main information on the process, on its needs in terms of operating conditions, control methods and mass balances were collected. Experimental data were congruent with the expected values on the basis of the design data.
Sulindac continuous process was tested at the pilot scale using ATR for the reaction step and a two stages continuous crystalliser for the crystallisation step. First and second stages operated at different temperatures, affording a final crystal of reasonable filterability characteristics. The validation trials indicated that a full continuous approach, with reduced residence times, is required in order to avoid the formation of a byproduct due to the reaction between Sulindac and the crystallisation solvent.
Flumethasone validation trials were limited to the check of the milling step, starting from a flumethasone ortoacetate crude produced in Galentis. Tests were carried out at the lab scale, obtaining samples of Flumethasone with chloromethasone content less than 0.2 %. It was confirmed that the new process can reach the desired quality, simply increasing the number of wet milling steps.
DMF guidelines
A full discussion of criteria to be adopted in writing the drug master file sections treating crystallisation steps was elaborated, both for batch and continuous operations. It was done taking into account the recent perspectives regarding the risk management presented in the recent Q11 ICH guidelines for the QbD process control. The document discusses both procedural and equipment related issues, so that it can be taken as an aid when a DMF in the CTD format has to be written down.
The CCRYSTAL project is present on the web thanks to its official website: http://www.ccrystal.eu The website contains the full description of the project, of its consortium and described the results achieved. Moreover, an intranet has been developed only for the members of the consortium, where they can find the documentation produced during the project lifetime.
The website will be kept alive on the web for other five years after the end of the project.
The coordiantor contact details are:
Gilda Gasparini
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: gilda.gasparini@amtechuk.com
Robert Ashe
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: robert.ashe@amtechuk.com
Whittington Jane
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: jane.whittington@amtechuk.com
The other beneficiaries contact details are the following:
Merchav Engineering Ltd
33 Frieschman St
Kiryat Ata 28110 Israel Zvi Merchav
Tel: +972(0) 4 8454428
Fax: +972(0) 4 8454429
Mobile: +972(0)54 4454428
Email: zvi@merchav.com
Flamma Fabbrica Lombarda Ammino Acidi
via Bedeschi, 22
24040 Chignolo d'isola (BG) Italy Negrisoli Gianpaolo
Tel: +39 035 4991811
Fax: +39 035 4991812
Email: mngdir@flamma.it
Canevotti Renato
Tel: +39 035 4991846
Fax: +39 035 4991812
Email: resdevt@flamma.it
Maurizio Goffredo
Tel: +39 035 4991846
Fax: +39 035 4991812
Rossella Pedroncelli
Tel: +39 035-4991811
Fax: +39 035 4991812
Email: accountdept@flamma.it
Galentis
Via delle industrie11
30020 Marcon (Venice) Italy
Zambelli Luca
Tel: +39 041 4567349
Fax: +39 041 5958987
Email: lzambelli@galentis.it
De Pieri Donatella
Tel: +39 041 4567349
Fax: +39 0414567349
Email: ddepieri@galentis.it
Antonio Paulon
Tel: +39 041 4567349
Fax: +39 0414567349
Serichim
Piazzale Marinotti, 1
33050 Torviscosa (UD) Italy
Delogu Pietro
Tel: +39 0431-381308
Fax: +39 0431-381400
Email: pietro.delogu@serichim.it
Ferrazzi Faustino
Tel: +39 0431 381415
Fax: +39 0431-381400
Email: fausto.ferrazzi@serichim.it
Giuseppina Vassallo
Tel: +39 0431 381420
Fax: +39 0431 381400
Email: giuseppina.vassallo@serichim.it
Sabrina De Rosa
Tel: +39 0431 381413
Fax: +39 0431 381400
E-Mail: sabrina.derosa@serichim.it
Paolo Ferrario
Tel: +39 0431 381410
Fax: +39 0431 381400
E-Mail: paolo.ferrario@serichim.it
Process Systems Enterprise
6th Floor East, Hammersmith Grove, 26-28
W6 7HA, London
United Kindom
Bermingham Sean
Tel: +44 20 8563 0888
Fax: +44 20 8563 0999
Email: s.bermingham@psenterprise.com
Brian Mckenzie
Tel: +44 20 8563 6245
Fax: +44 20 8563 0999
Email: b.mckenzie@psenterprise.com
Autico
Unit 8, Monument View
Chelston Business Park
Wellington,Somerset, TA21 9ND,
United Kingdom
Poole Kevin
Tel: +44 1823 618086
Mobile: +44 7960 918080
Email: kevin.poole@autico.co.uk
LTD International
Via Camperio, 9
20123 Milano, Italy
Garavaglia Antonio
Tel: +39 02 89409392
Fax: +39 02 89421104
Mobile: +39 338 6006038
Email: agaravaglia@intltd.it
The global pharma players are looking at sourcing cheaper APIs, particularly in the situation of weak purchasing power worldwide. This is because of patent expiry of the blockbuster drugs, cut in medical costs in developed world and rise in R&D spending. Therefore, the opportunity descends upon low cost producers such as Chinese API makers. The project aimed at introducing advanced technologies to complete continuous manufacturing processes of pharmaceutical substances with enhanced, mainly continuous, crystallisation technologies, affording fundamental new approaches to obtain products with high purity levels and constant and reproducible crystalline forms at a competitive price, enabling EU APIs industry to regain his competitiveness against Asian producers. The innovative technologies, jointly with the new processes for four selected APIs developed during the two year of the project lifetime, are based on a new concept crystalliser and on continuous mode operation. From the point of view of drug manufacturing economic analysis, the new technologies are especially important for the market of generic APIs. Generics are changing the companies' strategies. The incidence of drug substance manufacturing cost on the price of the final pharmaceutical specialty was really not relevant for under patent drugs. This generated small or no attention to API manufacturing cost reduction and, as a consequence, process improvements. Now generic companies are experiencing more and more competition, and their business logics resemble those of fine chemical producers. So they have to improve their processes in order to remove the more relevant inefficiencies that are related to organisational aspects as well as to raw material costs reduction. As an example, for what the batch/continuous process comparison is considered, batch processes need to 'stop' because of repeated sampling and isolation at every intermediate step. Such a situation will force companies to innovate and move from 'quality by analysis' to 'quality by design' manufacturing. To reach this goal a complete command of the process has to be reached in order to avoid being victims of 'analysis paralysis'. To get out of this situation; alternate manufacturing scenarios are under consideration. Contract manufacturing organisations (CMO) specialise in certain chemistries, unit processes, and operations and campaign their production. Based on the amount of API needed, some of the products are campaigned in pilot plants to mini-plants. Modular plants are going to be an excellent option for batch or continuous processes. Innovation in manufacturing methods, improved process technologies, along with good manufacturing methods deliver excellent processes and are able to handle low to high production rates. Better manufacturing technologies for API and formulation are expected to improve profits at that level. For API, the improvements come in the way of improved yield, reduced waste, reduced and better solvent use, improved productivity, and improved business processes. For formulations they come with the blending, sise uniformity and uniform admixing, etc. In the supply chain, there are opportunities and one has to pick them. Improved technologies will also significantly reduce their carbon footprint. Companies that are excellent in manufacturing API and formulating drugs are now acting as the innovation driver more than the companies dedicated to invent new active principles.
Project context and objectives:
The CCRYSTAL fits into a context of continuous changes in the market it refers to, namely that of APIs.
The following trends were registered during the last two years:
- evolution of API regulatory requirements;
- expansion of the use of continuous technologies in API production processes;
- increase of technological level of companies producing in emerging countries;
- diffusion of generic drugs, with a push for reduction of manufacturing costs.
For all the previous issues CCRYSTAL studies can give useful outcomes.
Evolution of API regulatory requirements
More and more regulatory agencies (EMEA, FDA and the national agencies) ask for a 'risk management strategy' based on the concept of quality by design (QBD). Risk management means confidence that there is not possibility that the manufacturing complex give off drug substances not complying with authorised standard quality level. The manufacturing complex includes working procedures, manufacturing processes, and production facilities and so on.
It is a recent achievement (see guidelines Q11 of ICH) that a real control of constancy of the quality level of products can be obtained only by a deep knowledge of the chemical and physical principles of the manufacturing process, having a clear vision of relationships between operating variables and process performances. Production techniques able to maintain the value of critical operating variables in a defined range of operability are encouraged.
Continuous operations, which work in stationary conditions of operating variables, assure the simplest and surest method to achieve this fundamental goal.
Expansion of the use of continuous technologies in API production processes
The use of continuous reactions for API synthesis is more and more gaining popularity in the technical literature, in the scientific meetings, in the offer of commercial equipments for synthesis. This impressive increase in the interest of the scientific community is producing the first industrial production based on the proposed technology. Some big companies, like Lonza and DSM, started industrial plants for the manufacturing advanced intermediates for pharma and agrochemicals products: DSM operates an industrial plant for the production on a nitric ester; Lonza presented its proprietary process for Niacin.
The use of continuous technologies for other unit operations, like crystallisation, is the natural evolution expected in the next years. The reason is due to technical, organisational and economic reasons. Technical reasons refer to better performances of continuous in front of batch operations and on the improved control possibilities as previously explained. Organisational reasons derive from the fact that it is very inconvenient joining batch and continuous operations in series, due to the need of inserting storage steps which require complex scheduling of production phases. Economic reasons are due to the increasing plant cost when continuous and batch sections have to be connected together.
CCRYSTAL anticipated an evaluation of the problems connected with continuous crystallisation, highlighting advantages and problems, and indicating what is important to consider for the development of this operation. The obtained knowledge is not exhaustive, but it represents a good starting point for future works.
Increase of technological level of companies producing in emerging countries
Emerging economies were considered a real threat to the European API producers in the past years. Using the low manpower costs and the absence of regulations regarding the environmental impact of productions, Indian and Chinese companies produced a strong impact on the costs of chemicals, including those for APIs, inducing the European companies to exit from several productions and making more difficult the competition, maintaining the high quality standards required by the pharmaceutical industry.
Now the panorama is changing: oriental producers are more conscious of the relevance of the quality requirements (including reliability) and are interested to adopt producing processes with an higher technological level. India and China are now an interesting potential market for the new technologies, acting at the same level of the occidental companies.
Diffusion of generic drugs, with a pull for reduction of manufacturing costs
Reduction of welfare costs, without any reduction of the quality level, is a need for all countries in Europe and over the world. The cost of drugs is a significant addendum to the overall health costs, and its reduction constitutes a goal for all public administrations. Generic drugs comply with this need, and their use if strongly recommended by all governments. Government efforts to promote and reward expanded generic use should continue to pay off, notably in France and Japan. The Swiss market is also performing strongly, although it is limited in size by the country's small population.
Drugs worth USD150 billion will go off-patent between 2010 and 2017, helping to propel the generic-drugs market from revenues of USD123.85 billion in 2010 to an estimated USD 231 billion in 2017. Leading global generic drug markets will grow to USD 221 billion by 2016. US generic growth is forecasted to rise to USD 104.1 billion by 2016, due to expiry of patents on major drugs such as Lipitor and Zyprexa, increased pressure for generic use from Medicare drug plans, and the gradual emergence of the biosimilar market.
On the other side revenues for the pharmaceutical industry have to be quite high, in order to cover the costs of development of new drugs, able to better treat all kind of illness. Reduction of drug prices has to be reached by the reduction of manufacturing costs, using always more efficient production techniques.
The market of generics is based on three fundamental values: quality, reliability and costs. Production efficiency is a necessary requisite in order to compete in that field. Improvement of processes is, may be, the more effective tool in order to gain advantages over competitors in this kind of business. Also from this point of view continuous technologies are a powerful help to reach a high level of efficiency.
In this frame, the CCRYSTAL represent a corner stone by itself, being industrial crystallisation a major problem for the pharmaceutical industry, and its success will represent an unprecedented innovation in the field. The project re-invents the whole value chain of APIs production, providing EU SMEs with competitive solutions against their Asian competitors and guarantying at the same time the high quality and safety usage of medicines for the citizens. At the same time more and more it appears clear that the efficacy of a drug is determined by its bioavailability, depending on its physical form. New polymorphic forms, solvates or cocrystal are considered as suitable tools to improve beneficial effects also of old drugs. As a consequence, new improved crystallisation methods and apparatuses are required to produce the wanted pharmaceutical forms.
In this framework, the objectives of the CCRYSTAL project have been:
- applying improved crystallisation technologies to crystallise pharmaceutical substances, affording fundamental new approaches to obtain products with high purity levels and constant and reproducible crystalline forms at a competitive price, enabling EU APIs industry to regain his competitiveness against Asian producers.
- testing the new technologies on competitive APIs, eventually developing the whole chemical scheme in a continuous manner;
- demonstrating that the use of the improved processes increase the final product reliability.
The workplan has been divided into predominantly experimental / technological (WP1, WP2, WP3 and WP4) and demonstration activities (WP5) that combine and cross-refer synergistically. Other two WPs have been included in the workplan, that regard the dissemination and exploitation WP (WP7), and the management one (WP8). The list of WPs is included hereunder:
WP1: Sizing continuous crystallisation train and operating procedure development (RTD)
WP2: A user-friendly crystallisation process modelling environment for optimal design, operation and control of continuous crystallisation process (RTD)
WP3: Use of high performance crystallisation unit to meet compositional quality requirements of API 1 (RTD)
WP4: Use of high performance crystallisation unit to increase API two recovery yields and to meet physical (RTD)
WP5: Validation Trials (demo)
WP6: Dissemination and exploitation of project results (other)
WP7: Project management (MGT).
Project results:
WP1 have been focused on the design of the repeatable crystallisation unit and on the sizing of the crystallisation train. Then the effort has been put in the development of a methodology to tune actual production. The work performed for this WP has been done mainly by the partners Amtech and Autico, with the collaboration first of all of Serichim and then of the other partners.
Amtech exploitable results are two systems: a lab scale repeatable rig unit and a continuous crystalliser.
Repeatable rig unit: Coflux
The Coflux system allows monitoring online and in a non-invasive way processes such as bacterial growth, reactions and crystallisations. Before the project, large scale system using this technology was available but this project has offered the possibility to apply the Coflux technology to smaller vessels, suitable for lab testing. This both opens a new market but also increases the sales of the larger systems since it will be easier for the users to test the technology prior to committing to the larger scale with a larger investment.
Real time process monitoring aligns with the current FDA directives to run better processes through QbD and process analytical technologies (PAT). Once returned, the lab scale Coflux will be used at a university in the UK to continue testing and generate more case studies in order to market this technology.
Continuous crystalliser: Coflore ATR
The Coflore ATR has been mainly applied to synthesis but its ability to handle solids and the possibility to obtain gradient cooling makes it suitable to be used for crystallisation processes too. However, in order to market the ATR as crystalliser, it is necessary to investigate further the solid handling capabilities (in terms of sizes, concentration and composition), the tendency to foul (especially on the colder surfaces) and the ability to influence the crystal size, size distribution and shape. These characteristics will more likely vary on a case by case basis depending on the system of interest but the know how generated by running a series of different processes will provide a list of possible actions and running strategies to cope with different demands.
Continuous crystallisation is a very new application and many end users are showing high interest in this approach.
The ATR produce during CCRYSTAL and the know-how generated will be used to run feasibility studies in order to facilitate user uptake and increase crystalliser sales. These activities will also affect the sales of the same system as a reactor. One of the key requirements of the industry these days is flexibility and multi-purpose equipment. So if it can be proved that the same equipment can be used for different process steps, this will increase chances of sales.
The know-how generated by this project on both systems will be used to make design modifications to improve performances and user experience too.
Concerning Wp2, the main goal is the development of a user-friendly crystallisation process modelling environment for use by the SME's employee in order to streamline their processes and operation of crystallisers.
The RTD PSE released a model-based engineering crystallisation tool, the gCRYSTAL, configured for use in the CCRYSTAL. Within the project this tool have been used to create three case specific model-based engineering tools, one for three of the following SMEs in CCRYSTAL (Flamma, Galentis, Merchav).
Before configuring the gCRYSTAL environment for use in the CCRYSTAL project, user requirements were gathered by interviewing various stakeholders within the SMEs. Relevant stakeholders were not only the envisaged users of a user-friendly tool, but also the people who will use the information provided by this tool for decision support purposes, such as managers, operators, etc.
The foreground resulting from WP2 is:
- A process model and control protocol for the cooling crystallisation of paracetamol (Merchav). Further results (foreground) generated in the process of developing the process model and control protocol are:
i. configured PSE's gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. estimated kinetic parameters using data collected from batch experiments;
iii. reconfigured gCRYSTAL for the same chemical system but continuous processing: describing the equipment and process configurations;
iv. performed predictive simulations of the performance of several process configurations, which differ with respect to:
- number of crystallisation stages;
- temperature in each stage;
- residence time in each stage.
- Process model and control protocol for the anti-solvent crystallisation of paracetamol (Galentis). Further results (foreground) generated in the process of developing the process model and control protocol are:
i. configured PSE's gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. estimated kinetic parameters using data collected from batch experiments;
iii. reconfigured gCRYSTAL for the same chemical system but continuous processing: describing the equipment and process configurations;
iv. performed predictive simulations of the performance of several process configurations, which differ with respect to:
- number of crystallisation stages;
- solvent composition in each stage;
- residence time in each stage.
- Process model for the crystallisation of Ornithine alpha-ketoglutarate (Flamma). Further results (foreground) generated in the process of developing the process model are:
i. configured gCRYSTAL for this system: defining components, specifying physical properties and describing the equipment used for batch experiments;
ii. defined next steps required for:
- estimating kinetic parameters using data collected from manufacturing runs;
- optimising the crystallisation process with respect to yield and purity.
The SMEs will be able to use gCRYSTAL as configured for their case of interest to facilitate a wide range of steps in the workflow for design and optimisation of crystallisation processes: identifying dominant mechanisms, maximising the formation of the required polymorphs, estimating model parameters that allow extrapolation of knowledge, investigation of batch vs continuous operation, process scale-up and optimisation of process operation at the manufacturing scale.
WP3
Use of high performance crystallisation unit to meet compositional quality requirements of API 1 main objective was the development of the purification procedures for Flumethasone and Flumethasone intermediates, capable of separating chlorinated by products, in order to meet the very strict specifications prescribed by the pharmacopoeia. During the first reporting period, the revision of the existing chemical know-how has been performed in order to optimise chemical yields and crudes compositions along the synthetic path, while during the second reporting period the main objective was the development of separation procedures of chlorinated by-products from the main product acting on the solid intermediates occurring in the manufacturing process.
A completely new flumethasone purification process was developed, based on the separation of unwanted impurities at the stage of orthoacetate derivatives of flumethasone and related by-products. It makes use of a milling solid-liquid equilibrium operation. The new purification procedure allows the use of perchlorylfluoride fluorination reagent, which is a very cheap fluorinating agent in front of more expensive chemicals. Perchlorylfluoride is a chemical produced by Galentis and by only a second company all over the world, so that the new process is a valuable asset for Galentis or as a producer of Flumethasone, or as a supplier of perchlorylfluoride to flumethasone producers.
A process book was produced, describing all precess steps, based on Galentis experience and Serichim's new studies.
Purification steps were verified at the lab scale.
Beneficiary of the developed know how is Galentis.
Three different APIs were object of study in WP4, namely Gabapentin, Sulindac and Ornithine ketoglutarate. During the course of the project, for all of them, the research objectives were better specified and new opportunities were identified, to make more competitive the production processes. For all of them, it was recognised that the more important goal is to have a robust process which meets the quality requirements in terms of organic impurity contents.
As far as Flamma Spa, crystallisation process has been developed:
- to have an all continuous process, synthesis and recovery sections, obtaining a high quality filterable crystal suitable for continuous filtration, for Sulindac;
- to have a low cost process based on selective crystallisation steps for di-ornithine ketoglutarate.
Concerning the Sulindac, the previously stated goal aim of producing Sulindac Form II without traces of Form I was reached out of the project, working on drying procedure, while continuous crystallisation aims were focused on improvement of filterability of crude product and to the interconnection between continuous reaction and continuous crystallisation steps, without intermediate storages.
A continuous crystallisation operation was developed, able to obtain a well filterable product. A few percent of the unwanted polymorph was always present, but an improved drying procedure was able to convert it to the right form. Crystallisation was directly joined to the previous synthetic step, so that a complete continuous process was developed, starting from the parent thioether and ending with the filtered solid.
The entire process was documented in the Sulindac process book. Preparation was stepwise tested at the pilot scale, using ATR for the reaction step and a continuous two-stages crystalliser for the crystallisation step.
Foreground beneficiaries are Flamma for the Sulindac production process and AMT for the ATR qualification.
For Di-ornithine ketoglutarate, the new process overcame the typical problems of the the precipitation processes, namely processes where the insoluble product is formed by salt formation from an acid and a base, where crystal size is very fine, and the filtration and drying operation on the formed solid are quite troublesome. In this case crystallisation was carried out so that the crystallisation rate was quite slow, and the crystal size quite high.
While previous goals resulted not so attractive, the possibility of a completely new crystallisation process, significantly more convenient of the existing one, was identified and the process developed, as previously explained.
The new process uses Ornithine Hydrochloride as raw material. The process is based on two crystallisation steps, that can be operated both batchwise and continuously. Only the batch crystallisations were tested at the bench scale.
The new process allows a reduction of raw material cost and avoids a complex electrochemical step in the overall synthetic path. All solvents can be recovered and recycled, so that the process can be classified as environmentally friend.
Beneficiary of the process is Flamma.
For Gabapentin, the main result is the achievement of the right quality profile by a crystallisation step of the raw material and the tuning of the Hofmann reaction operating conditions. Having reached this result, it was possible to confirm that the recovery and purification process, based on continuous l-l extraction and crystallisation operations, can give a significant containment of production costs. Gabapentin is a high volume-low cost API, for which manufacturing cost has to be minimised in order to compete on a very challenging market.
The improved operating conditions were fully examined, and a synthetic process defined and documented.
Samples were produced at the bench scale and qualified by one of the major producer of the API. However it was not possible to complete the process validation, including recycles.
Beneficiary of the developed know how is Merchav ltd.
WP5
Validation trials, was dedicated to validate project results ensuring successive effective exploitation by SMEs. The activities of WP5 have been performed mainly during the second reporting period, with the following objectives:
- finalisation of the repeatable unit and of the train testing and validation;
- process technical validations;
- the scale up feasibility study for the integration of the crystallisation unit into existing plants;
- drug master file guidelines for continuous crystallisation.
Process technical validations carried out in the frame of this WP have to be considered as a part of know how developed for Flamma, Galentis and Merchav.
Operational studies of crystallisation trains, experience of implementation of equipments in a pilot or production environment, as well as methodologies for interfacing batch-to-continuous equipments have to be considered as a know how available for all CCRYSTAL participants and, in general for all companies facing issues derived from continuous crystallisation use in their processes. This is the same for the guidelines worked out for the compilation of drug master files sections dedicated to crystallisation steps.
Operating and installation instructions of ATR and Coflore equipments are now available to help in their use.
Schemes for their connection to the ancillary equipments, feeding systems, thermostatting units and so on were developed, checked and documented, DMF writing criteria elucidated. All participants can have access to the worked documents.
Potential impact:
APIs: a short market presentation
APIs may be defined as substances or mixtures of substances intended to be used in the manufacture of a drug or any medicinal product that, when used in the production of a drug, becomes an active ingredient of that drug or medicinal product. APIs can either be classified as synthetic or biotech, based on the raw materials used in their production, and as innovator or generic, depending upon the customer requirement.
The API forms the most vital part of every formulated end product, and is an important part of the whole pharmaceutical industry.
The overall API market was valued at USD 101.08 billion in 2010, and is expected to grow at a CAGR of 7.9 % from 2011 to 2016. More than half of the global APIs market is dominated by North America and Europe, with Asia-Pacific forming the third largest but fastest growing API market in the world. The Asia-Pacific API market is highly fragmented, with a very large number of players competing for market shares. This market is characterised by an eclectic mix of companies, which range from large, vertically integrated manufacturers to distributors and importers.
Europe is the third largest regional market for APIs by revenue in the world. European API market revenue held a share of around 24.2 % of global API market revenue of USD 108.6 billion in 2011. Total revenue generated by the European API market was USD 26 288.96 million in 2011 and is expected to grow at a CAGR of 6.5 % to reach USD 38 255.67 million by 2017.This will be supported by healthy demand from generic and biotech API sectors.
There has been an increase in influence of API players from emerging economies such as India and China after the economic recession. The recession restricted the growth of Innovative sector in developed economies such as the US and Europe, as the Innovative sector requires huge investments. This has helped fuel the growth of generics market in Asian countries such as India and China.
The API market is facing a period of unprecedented growth as market dynamics have undergone a major change with the expiration of patents pertaining to global blockbuster drugs in the US. The consequences of the economic crisis has hit the Innovative drugs market hard, with less budgets allocated by the major players for the R&D of Innovative drugs. This has led to drying up of pipelines for new drugs, and therefore the market for generic drugs is quickly growing. Thus, the patent expiry factor is slated to drive the API market for the coming years.
European countries such as the UK, France, and Germany are promoting the use of Generics by providing incentives to the doctors for writing prescriptions relating to generic drugs and also to the pharmacists if they offer the generic equivalent of prescribed drugs. Thus, generic API market is expected to grow till the year 2016 as compared to the Innovative API market.
The expiry of patents for major blockbuster drugs and the entry of low-priced generic versions are expected to increase the consumption of generic drugs, with the weak pipelines of major pharmaceutical companies amplifying this trend. With many patent expiries occurring during 2010 - 2012, generic consumption is expected to peak, and in turn increase the competition between pharmaceutical producers.
The global pharma players are looking at sourcing cheaper APIs, particularly in the situation of weak purchasing power worldwide. This is because of patent expiry of the blockbuster drugs, cut in medical costs in developed world and rise in R&D spending. Therefore, the opportunity descends upon low cost producers such as Chinese API makers. The API players in China also see the growth in local market driven by couple of factors such as growth in the generics sector, government's plan to develop multi-level insurance system and the development in biotech drugs.
As far as WP1, crystallisation of high value chemicals has traditionally been performed in batch reactors. By virtue of improved surface to volume ratio, flow reactors offer the potential unprecedented improvements in product quality and yield. The challenges of building flow reactors for this purpose however are formidable. There are two essential elements to developing flow crystallisation processes. Firstly, equipment is required to study crystallisation rate at comparable surface to volume ratios to a flow reactor. In this project, a heat balance reaction calorimeter in glass was developed in glass, Coflux. The second requirement is to have a flow system which can carry slurries in an orderly fashion. A flow reactor was designed for this purpose during the project, Coflore ATR. The technical challenges in terms of equipment development were formidable. In the case of the calorimeter, the challenge was to develop a multi element non-metal limpet jacket with variable area control and heat measuring capabilities. For the flow crystalliser, this required a dynamically mixed tubular system with multi zone temperature control. Testing of these systems was carried out in two phases. The first phase was water testing undertaken by Autico. Both systems performed well in these tests. The second phase involved product testing and this work was carried out by Serichim. A number of problems were identified during this phase which is normal for new technology. As result of the testing work carried out by Serichim, a number of further modifications and improvements have been made to both machines. Further tests carried out after the end of the project suggest that these problems are being addressed. For AMTECH, this project has been extremely valuable in identifying specific design requirements for developing and operating flow crystallisation processes. We are now investing more money in further development and testing.
Concerning WP2, the SMEs will be able to use gCRYSTAL as configured for their case of interest to facilitate a wide range of steps in the workflow for design and optimisation of crystallisation processes: identifying dominant mechanisms, maximising the formation of the required polymorphs, estimating model parameters that allow extrapolation of knowledge, investigation of batch versus continuous operation, process scale-up and optimisation of process operation at the manufacturing scale.
More specific benefits of the use of mechanistic models are:
- increased R&D efficiency:
i. use fewer experiments to adequately characterise a process / processing step (typically the number of experiments can be reduced by 50 - 70 %);
ii. design additional experiments with a higher information content (as compared to statistical techniques for the design of experiments);
- increased efficiency and reduced risk for tech transfer activities by turning data into knowledge that can be applied for extrapolation to other scenarios with known confidence and reduced experimental requirements;
- efficient investigation of the feasibility (and desirability) of batch-to-continuous: design of a continuous process does not have to be supported by continuous experiments, i.e. no need to significantly invest in new lab equipment.
WP3 had the goal of developing crystallisation procedures for flumethasone and flumethasone intermediates capable of separating chlorinated by products, in order to meet the very strict specifications prescribed by the pharmacopeia.
Flumethasone is the basic ingredient of a number of drug substances, like flumethasone acetate and flumethasone pivalate, as well as an intermediate for other steroideal active substances. Galentis, one of the few producers of steroids in Italy and in Europe, will be in the position to start with a new production in its facilities in Marcon. The previous purification method, part of a pre-existing know how, was too much complex, not reliable and costly, so that it was impossible until now to start a regular production.
Opportunity to start with this production has to be verified through an updated market analysis. It has also to be evaluated in the frame of Galentis new strategies. In this perspective a second chance is the valorisation of the knowhow jointly with other companies involved in steroid manufacture in Europe. Sicor (a TEVA subsidiary), Hovione, Aurisco and Farmabios are presently producers of Flumethasone derivatives; other companies, producers of other steroideal drugs, are potentially interested to a technology for flumethasone.
The qualifying element of the developed process for flumethasone is the quality of the final product. It is really better than that required by the European Pharmacopoeia, sixth edition, for flumethasone pivalate, which sets a limit of 1.5 % for chloromethasone pivalate (in front of 0.1 % given by our process). No limits are reported by the current Pharmacopoeia about 6-beta isomers, deriving from the first fluorination step, but also for this by-product the achieved concentration level is less than 0.1 %.
The second point of advantage of the developed process over the other is related to the cost of chemicals. Perchlory fluoride is a cheap reagent in front of other fluorinating agents, like Selectofluor, and it gives less co-products. Galentis is the only producer of perchloryl fluoride, so that it could have a revenue from the cells of this chemical also if the flumethasone should be produced by a different company according to Galentis proprietary technology.
WP4 was dedicated to the development of crystallisation steps for products in the pipeline (already in or under development) of two SME participants, Merchav and Flamma S.p.a. Specifically, the work for Merchav has been done on the Gabapentin, while in the case of Flamma, the Ornithine and the Sulindac have been investigated.
Gabapentin
Gabapentin continues to be one of the biggest drugs for quantities yearly produced. The production volume went over 2000 tons in 2010 and only slightly decreased in the last two years. Reasons of this decrement in the drug consumption were some restrictions on its use for treatment of secondary diseases, like headache, that can be treated with products with less side effects. For the future years a slow increase of volumes is expected, and only for the end of 2020s new molecules, like Pregabalin, might substitute Gabapentin in some uses.
Jointly with the volume increase prices were go down below USD 100/kg. Production costs have to be lowered in order to compete on the drug substance market. A number of producers are now present on the market: Teva, Medichem, ZachSystem, Erregierre were the first producers of the generic drug, while recent new players are mainly Indian or Chinese, like Jiangxi Synergy, Jiangsu Hengrui or Anjan Drug. Recently Teva and Pfizer, the API originator, reached a settlement about their patent litigation on the generic versions of Pfizer's Neurontin capsules and tablets sold by Teva and its subsidiary Ivax. Now Teva sells the generic Gabapentin under license by TEVA.
All the major producers use a process including the Hofmann rearrangement as the last synthetic step. Teva and ZachSystem, which are the biggest generic producers, use this reaction, but different purification strategies.
The shift from batch to continuous approach based on the process developed by CCRYSTAL project is under discussion with both those two major players and could be proposed to the new producers. As the change of technology for a so high capacity product is a very complex matter, it is forecasted that discussions and technical analysis will continue in the next year, but there is chance that a cooperation will start with almost one of the big producers, interested in cost reduction and reliability improvements.
Ornithine
The new process for Ornithine Ketoglutarate (OKG), the direct process, gives the advantage of the direct use of ornithine hydrochloride as raw material instead of that of ornithine free base. This last chemical has two disadvantages in front of the hydrochloride:
- its cost is about the double of that of the hydrochloride, because it is obtained from this last chemical by an electrochemical process;
- it is a very instable molecule, easily forming a lactam by standing.
An inconvenient point of the new process is that it makes use of one chemical more, the trimethylamine, with the related cost. In principle, trimethylamine can be recovered and recycled, so that its cost can be minimised.
It was calculated that the new process affords a raw material cost several Euros lower than that of the process based on the free base, also without the trimethylamine recycle: Flamma will be in the position of adopting the new process, if re-registration operations will be accepted by the regulatory agencies.
Sulindac
The market situation of Sulindac changed dramatically in the last two years. It is a non-steroidal anti-inflammatory drug, which reached a volume of 150 tons/year in the first 2000s. Recently some new APIs with a COX 2 inhibition activity substituted Sulindac in several uses, so that the annual production went down to 50 - 100 tons/year. It is not clear if this trend will continue in the next years, or if it stops and the market tends to grow.
Flamma invested a big effort in the development of a proprietary process, including preparation of the intermediate, SUL4, and, in the frame of CCRYSTAL, its transformation in Sulindac through the continuous steps of oxidation and crystallisation. The overall process is cheaper than traditional processes, but this is not the only condition to go on the market, because of costs of registration of the product joint to the new technology. The industrial use of the Sulindac continuous process depends on the future market trend. In the next few years it will be clear if the new APIs completely substitute Sulindac, or if a category of diseases continues to be treated with the old drug. It will depend on the results of the post-marketing surveillance outcomes carried on by regulatory agencies.
In any case, the development of the new process demonstrated the advantages of a chain of continuous operations, without intermediate storages. It allowed the use of a solvent in the crystallisation step that dissolves very well the product, but that is reactive with it, forming on the long-time an ester. Control of contact time between this solvent and Sulindac is the key to take advantage from the good solvent properties.
WP5 was dedicated to validate project results ensuring successive effective exploitation by SMEs. In order to reach these goals the validation was done taking into account the provisions and international protocols required for APIs production.
The objectives of this WP are mainly three:
- to construct the repeatable unit and the crystallisation train and to test and validate them;
- to scale up the crystallisation unit;
- to perform validation trials for the successive scaling up to commercial amounts production of the APIs;
- to analyse how to write the drug master file.
The outcomes of this WP are both technical and methodological.
The CCRYSTAL project has enabled these new batch and continuous reactor technologies to be developed for full industrial use.
The Coflux repeatable rig unit performed in line with the requirements of the Technical Specification. As expected, heat up / cool down rates for the glass version of the reactors were slower when compared with similarly sized metal jacketed vessels due to the poorer heat transfer coefficients of the reactor material. However, the heating / cooling response rates were much faster than for jacketed glass vessels due the smaller volume of heat transfer fluid in the reactor tubes and high flow rates. The enthalpy, process power and other control functions performed as designed, with the process power function reading values accurately down to 10 Watts. Further development could reduce this figure further to enable use with less energetic reactions. The enthalpy value was useful for tracking crystal growth. From the construction techniques developed during the CCRYSTAL project, it would be possible to make both smaller and larger versions of the Coflux reactor. Developing a scientific glass version has widened the potential market for laboratory and pilot plant use. Larger versions can be constructed from industry standard metals such as stainless steel and Hastelloy or vitreous enamelled (glass-lined) steel reactors. Potential industries for this technology, other than pharmaceutical, includes fine chemicals, food, petrochemical, waste treatment and environmental.
The ATR crystallisation train unit was challenging to build, especially with respect to the heat transfer system developed for the reactor tubes. The heat transfer capabilities exceeded the requirements of the technical specification and the techniques used may be applied to other new technology systems. The addition by Serichim of temperature probes on the inlet and outlet heat transfer lines is a function that will be considered for future reactors. Further development of a multi-temperature zone version will widen the number of potential applications. Following the construction of the CCRYSTAL 1 litre ATR, further units have been built and successful process trials run. Scaling up of the ATR for larger applications is made easier by the development of the novel drive mechanism for the CCRYSTAL reactor which could be adapted for larger mass reactors. The Hastelloy reactor tubes can accommodate a wide range of chemicals and compounds but more exotic metals, such as monel and tantalum, could be used for reactor tubes for more demanding processes.
The use of an industry standard control system, with good analogue signal handling and control functions, is a key to developing prototype machines for industrial use. This approach saves time and reduces costs. Often, a specially designed or non-industrial control system is used for development which means that the whole system needs to be transposed on to another platform before it can be marketed. The development of the Honeywell HC900 control system, using the new 900 series operator control station, has enabled the reactor technologies to move forward from a control perspective due to the increased range of available functions and ease of programming, thus saving development time and costs. Data logging and reporting functions were developed for both reactor systems and included a useful screenshot download option. The operator was able to download data files in CSV format for reports.
Continuous crystalliser: Coflore ATR
The Coflore ATR has been mainly applied to synthesis but its ability to handle solids and the possibility to obtain gradient cooling makes it suitable to be used for crystallisation processes too. However, in order to market the ATR as crystalliser, it is necessary to investigate further the solid handling capabilities (in terms of sizes, concentration and composition), the tendency to foul (especially on the colder surfaces) and the ability to influence the crystal size, size distribution and shape. These characteristics will more likely vary on a case by case basis depending on the system of interest but the know how generated by running a series of different processes will provide a list of possible actions and running strategies to cope with different demands.
Continuous crystallisation is a very new application and many end users are showing high interest in this approach.
The ATR produce during CCRYSTAL and the know-how generated will be used to run feasibility studies in order to facilitate user uptake and increase crystalliser sales. These activities will also affect the sales of the same system as a reactor. One of the key requirements of the industry these days is flexibility and multi-purpose equipment. So if it can be proved that the same equipment can be used for different process steps, this will increase chances of sales.
The know-how generated by this project on both systems will be used to make design modifications to improve performances and user experience too.
Validation trials
Four APIs were object of study during the course of the project:
1. Ornithine ketoglutarate;
2. Sulindac;
3. Gabapentin;
4. Flumethasone.
For each one of them a suitable production process was developed, according to the SME's requirements, characterised by technical and economic improvements in respect to the state of the art. All developed processes, but one, contain one crystallisation step which determines the qualifying performance of the entire process. Exceptions are:
- Flumathasone, for which the qualifying step is a digestion step, a unit operation based on solid-liquid equilibrium, as crystallisation, but in a reverse sense, namely partial dissolution of impurities;
- Gabapentin, for which two critical crystallisation steps are present, required to reach the wanted quality and cost goals.
Batch and continuous operations were used, according to the physical characteristics of systems and to the process needs. For continuous operations performances of the new continuous crystallisers, designed and built up as a part of this project, were checked.
Some general teachings were derived from the performed studies. They can be categorised as physical, mechanical and control issues.
Physical issue regards the thermodynamic behaviour of solid-liquid systems involving APIs and related impurities. We have to start from the consideration that quality profile is the fundamental property of every API. No other achievements deserve value if the impurity contents of the final product do not comply with the pharmacopoeia standards or with the ICH rules. As drug substances are usually solid, crystallisation is the pivotal operation which generates the product quality. Other properties of the solid state, like the crystal habit or the crystal size distribution are less important.
Impurities often have molecular structures similar to that of the drug substances, so that they can arrange in the same crystalline reticulum, forming solid solutions. Separations in this case are not so sharp as in the case of eutettoid systems, and crystallisation is effective only if the reaction crudes have limited concentrations of impurities: reaction and crystallisation optimisations have to be carried out as a unique activity to obtain acceptable products. This was the reason why validation tests generally included both reaction and crystallisation steps.
Mechanical issue regards the rheological characteristics of the obtained solid. Continuous crystallisers have to assure the transfer of solids along the equipment, but it is heavily conditioned by the morphology of crystals. It is not matter of quality of the product; it is just a problem of operability of equipment. Particularly robust equipment have to be used when the solid is sticky: in some cases continuous crystallisation is not convenient, and only batch units can be used. For this reason, operability of ATR was checked with paracetamol, which is a 'good' crystalline product, not included in the 4 APIs for which the process was studied, while a more robust crystalliser was used to validate the continuous crystallisation of Sulindac, a very 'bad' solid.
Validation of ornithine ketoglutarate (OKG) processes was carried out batch wise. It was focused on the check of recycles, required in order to reach an economical process yield. A sequence of four consecutive trials was used in order to test the influence of recycles. Their effect was assessed by batch operations. Trials demonstrated that the developed procedure is able to consistently give a product with the wanted quality, with a process yield (including recycles) of about 80 %. Experiments showed that by a fine tuning of experimental conditions this figure should be also improved. Crystallisation operations gave well filterable products. All data needed for a cost analysis were collected, and a first evaluation showed that the new process saves a 20 % of the raw material cost in front of the current process.
The flumethasone critical operation, namely digestion, is inherently batch operation, which does not make sense to transform in continuous. In this case, the main problem resides in an accurate control of conditions and on the quality of the obtained solid. Standard analytical procedures were used to validate the critical operations.
Gabapentin process has two critical crystallisation steps. The first one is the purification of raw material by an extraction-crystallisation operation: it is a part of preparation of feeding material and has to be performed batch wise. The second one is the continuous crystallisation of the final product, with recycle of mother liquor: it should be performed in continuous mode; however it was validated batch wise because of unavailability of enough raw materials. Several preliminary tests were carried out, in order to tune the operative conditions, mainly for the Hofmann reaction step, and a final run was carried out, producing a 10 g sample of product with the following purity profile:
It was impossible to repeat the validation trials because of lack of raw material, and also recycles were not tested. However the main information on the process, on its needs in terms of operating conditions, control methods and mass balances were collected. Experimental data were congruent with the expected values on the basis of the design data.
Sulindac continuous process was tested at the pilot scale using ATR for the reaction step and a two stages continuous crystalliser for the crystallisation step. First and second stages operated at different temperatures, affording a final crystal of reasonable filterability characteristics. The validation trials indicated that a full continuous approach, with reduced residence times, is required in order to avoid the formation of a byproduct due to the reaction between Sulindac and the crystallisation solvent.
Flumethasone validation trials were limited to the check of the milling step, starting from a flumethasone ortoacetate crude produced in Galentis. Tests were carried out at the lab scale, obtaining samples of Flumethasone with chloromethasone content less than 0.2 %. It was confirmed that the new process can reach the desired quality, simply increasing the number of wet milling steps.
DMF guidelines
A full discussion of criteria to be adopted in writing the drug master file sections treating crystallisation steps was elaborated, both for batch and continuous operations. It was done taking into account the recent perspectives regarding the risk management presented in the recent Q11 ICH guidelines for the QbD process control. The document discusses both procedural and equipment related issues, so that it can be taken as an aid when a DMF in the CTD format has to be written down.
The CCRYSTAL project is present on the web thanks to its official website: http://www.ccrystal.eu The website contains the full description of the project, of its consortium and described the results achieved. Moreover, an intranet has been developed only for the members of the consortium, where they can find the documentation produced during the project lifetime.
The website will be kept alive on the web for other five years after the end of the project.
The coordiantor contact details are:
Gilda Gasparini
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: gilda.gasparini@amtechuk.com
Robert Ashe
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: robert.ashe@amtechuk.com
Whittington Jane
Tel: +44 (0) 1928 515454
Fax: +44 (0) 1928 515453
Email: jane.whittington@amtechuk.com
The other beneficiaries contact details are the following:
Merchav Engineering Ltd
33 Frieschman St
Kiryat Ata 28110 Israel Zvi Merchav
Tel: +972(0) 4 8454428
Fax: +972(0) 4 8454429
Mobile: +972(0)54 4454428
Email: zvi@merchav.com
Flamma Fabbrica Lombarda Ammino Acidi
via Bedeschi, 22
24040 Chignolo d'isola (BG) Italy Negrisoli Gianpaolo
Tel: +39 035 4991811
Fax: +39 035 4991812
Email: mngdir@flamma.it
Canevotti Renato
Tel: +39 035 4991846
Fax: +39 035 4991812
Email: resdevt@flamma.it
Maurizio Goffredo
Tel: +39 035 4991846
Fax: +39 035 4991812
Rossella Pedroncelli
Tel: +39 035-4991811
Fax: +39 035 4991812
Email: accountdept@flamma.it
Galentis
Via delle industrie11
30020 Marcon (Venice) Italy
Zambelli Luca
Tel: +39 041 4567349
Fax: +39 041 5958987
Email: lzambelli@galentis.it
De Pieri Donatella
Tel: +39 041 4567349
Fax: +39 0414567349
Email: ddepieri@galentis.it
Antonio Paulon
Tel: +39 041 4567349
Fax: +39 0414567349
Serichim
Piazzale Marinotti, 1
33050 Torviscosa (UD) Italy
Delogu Pietro
Tel: +39 0431-381308
Fax: +39 0431-381400
Email: pietro.delogu@serichim.it
Ferrazzi Faustino
Tel: +39 0431 381415
Fax: +39 0431-381400
Email: fausto.ferrazzi@serichim.it
Giuseppina Vassallo
Tel: +39 0431 381420
Fax: +39 0431 381400
Email: giuseppina.vassallo@serichim.it
Sabrina De Rosa
Tel: +39 0431 381413
Fax: +39 0431 381400
E-Mail: sabrina.derosa@serichim.it
Paolo Ferrario
Tel: +39 0431 381410
Fax: +39 0431 381400
E-Mail: paolo.ferrario@serichim.it
Process Systems Enterprise
6th Floor East, Hammersmith Grove, 26-28
W6 7HA, London
United Kindom
Bermingham Sean
Tel: +44 20 8563 0888
Fax: +44 20 8563 0999
Email: s.bermingham@psenterprise.com
Brian Mckenzie
Tel: +44 20 8563 6245
Fax: +44 20 8563 0999
Email: b.mckenzie@psenterprise.com
Autico
Unit 8, Monument View
Chelston Business Park
Wellington,Somerset, TA21 9ND,
United Kingdom
Poole Kevin
Tel: +44 1823 618086
Mobile: +44 7960 918080
Email: kevin.poole@autico.co.uk
LTD International
Via Camperio, 9
20123 Milano, Italy
Garavaglia Antonio
Tel: +39 02 89409392
Fax: +39 02 89421104
Mobile: +39 338 6006038
Email: agaravaglia@intltd.it