Complex targeted therapies are dominating drug development pipelines and represent a considerable challenge for manufacturers of active pharmaceutical ingredients (APIs)
The increasing complexity of new chemical entities (NCEs) is leading to a rise in hazardous chemistries that form key steps in the synthetic manufacturing processes. As a result, API manufacturers are faced with an increased manufacturing risk relating to product quality, worker safety and environmental impact. In many cases, it is proving beneficial for pharmaceutical companies to reduce their risk by employing contract manufacturing organisations (CMOs) to perform the hazardous steps. As such CMOs are increasingly faced with the need to offer a wide variety of chemistries and manufacturing capabilities at varying scales.
In this article, Paul Moscrop, Process Engineering Manager at Sterling Pharma Solutions, outlines the current trend for hazardous chemistry in the API manufacturing space and discusses the necessity of process hazard evaluation. He also explores the importance of integrating process engineering and chemistry expertise within an organisation to ensure project success, whilst still maintaining a safe plant, ensuring workforce safety and protecting customer confidence.
Hazardous chemistry is often essential to access a particular functionality within a specific molecule. Although not always necessary, the technique can also result in additional benefits such as cleaner chemistry with fewer or no side reactions, or provide a more direct route to a molecule with fewer processing stages. These advantages reduce cost by consuming less material, providing easier purification, producing less waste and reducing capital equipment and processing time.
The drive for more efficient manufacturing processes and to enable the complex routes demanded by NCEs has pushed development chemists to look at chemical routes involving simple but energetic molecules such as ethylene and propylene oxide, epichlorohydrin, hydrazine, hydroxylamine, as well as at chemistry involving nitrated species such as nitroethane and aromatic nitrates, and metal catalysed reactions (palladium, lithium and zinc).
Around 15 years ago, API outsourcing activities were finding a home in emerging markets in the East because of perceived cost efficiencies. However, as molecules become more complex and the potential risks associated with hazardous chemistry become more prevalent, the current trend is for these activities to move back to the West where higher standards are enforced through more stringent regulations, protecting individual workers and the reputation of pharmaceutical manufacturers.
In the UK, the Health and Safety Executive (HSE) is responsible for enforcing safety within the chemical manufacturing sector whilst the UK Environment Agency (EA) is responsible for enforcing environmental law. Like many other similar agencies around the world, its focus is on reducing or minimising the risk of injury to personnel or the impact on the environment during product manufacture.
Key to understanding the level of risk a process may impose during manufacture is identifying the reaction hazards it possesses. It is recommended that every process manufactured, be it at 20 litre scale or 10000 litre, undergoes rigorous hazard evaluation. The process should be subjected to a full run through in a reaction calorimeter (RC1) to identify heat outputs from reactions and quantify gas generation. Samples of reaction mixtures at key parts of the process must be taken (including raw materials when necessary) and subjected to differential scanning calorimetry (DSC) to screen for thermal activity and decomposition temperatures. When deemed necessary, accelerated rate calorimetry (ARC) can be performed, which will identify thermal onset temperatures for runaway reactions and any pressure rises associated with these. A vent sizing package (VSP) can then be used to assess gas-generating reactions and emergency venting requirements during scale-up.
Many of the compounds used in API production are powders. Powder handing is still a significant cause of incidents within the process industries despite recent widespread publicity. Tests available to identify hazards associated with powders include minimum ignition energy (MIE) identification, 20 litre sphere tests for dust explosion classification and electrostatic charge relaxation identification.
The information produced from the hazard evaluation is assessed by a process engineer and used to design the plant to manufacture the product. Reaction heat outputs define the control strategy for material additions, reactor cooling systems and gas evolution quantities, and rates define the size of vents and exhaust gas abatement systems. An understanding of equipment and instrument failure modes should be used to define the emergency cooling and venting requirements to prevent a loss of containment, as well as to confirm the emergency shutdown systems and response procedures.
The process engineer does not do this in isolation. Regular discussions with a development chemist and multidisciplinary process review meetings can ensure details are communicated and incorporated into the design. Does the reaction mixture thicken? Do solids suspend easily? If agitation is stopped, does the mixture layer? Where significant hazards are identified, the engineer may propose changes that aim to reduce the consequences of the hazard; however, this may require further investigation and approval by the chemist.
Every process must be subject to a rigorous hazard identification and risk assessment. This can be conducted on the final plant design using the fixed process and piping and instrumentation diagrams (P&IDs). The hazard evaluation work is again relied upon to determine the potential consequences of identified process upsets and hazards. Where Major Accident Hazards (MAHs) are identified, with the potential to injure people on or off site, or the ability to result in a significant environmental release, these are subject to more detailed risk assessments that are proportional to the severity of the hazard and may include a layer of protection analysis (LOPA) or a quantified risk assessment (QRA). When instruments are required as protection, their safety integrity level (SIL) is assessed and defined.
An output from the safety review may be to consider alternative chemistry, such as a change of solvent system, different order of operation or alternative reagents. Again, the chemist and engineer must work together to further eliminate or reduce the consequences of the identified hazards.
When MAHs are identified, a report is submitted to the UK HSE and EA. The report summarises the process hazards, their consequences and control measures and demonstrates the process is safe to operate within the nationally recognised risk levels. When the materials used in the process are responsible for the MAH, justification must be made that the chemical route is the most appropriate for the desired molecular transformation — taking into account the environmental impact as well as the hazardous nature of the materials. Hazard evaluation of the process may again be used to demonstrate that alternative routes are more hazardous or that the consequences of the proposed route are as low as practicable.
The routine testing of all processes before manufacture has been used to identify significant reaction hazard potential in a number of processes, including those that are already used in the industry where the hazards have not been fully understood. Clearly, identifying the reaction hazards early in the development phase of the product can be used to influence the choice of reagent to eliminate or at least minimise the consequences. It is easier to do this when a product is in the clinical trial phase of development, rather than after validation when the chemistry is generally fixed and changes are more difficult.
A case study example is presented below whereby a significant hazard was recognised and a solution was engineered to make the processing possible. Nitroethane is an explosive precursor and was used in a reaction at 4000 litre scale using a quantity equivalent to 962 kg of TNT. The nitroethane was used in a reaction with toluene as the solvent at 100% excess. The reaction requires temperatures of around 95 °C and for the water generated as a by-product to be removed during the reaction to allow it to go to completion. A higher jacket temperature was therefore required to distil out the water.
Initial hazard evaluation work identified the thermal decomposition of the nitroethane at temperatures above 122 °C and a recommendation that reactor temperatures should not exceed 100 °C was made. Time to catastrophic failure of the reactor was determined at 270 minutes.
MAHs were identified in the safety studies through loss of reaction temperature control, resulting in product decomposition and catastrophic failure of the reactor. Vapour cloud explosion consequences were modelled using trinitrotoluene (TNT) equivalence and The Netherlands Organisation (TNO) for Applied Scientific Research multi-energy methodologies.
QRAs were performed by external consultants and measures were put in place to prevent the initiating events. The basis of safety was determined through safety instrumented systems at safety integrity level (SIL) one, and a double mechanical steam reduction system with pressure relief was introduced to limit steam temperatures and ensure the identified onset temperatures were not achievable. In total, 21 weeks of hazard evaluation were conducted involving three RC1s, 11 DSCs and 15 ARC runs (two arc bombs were destroyed, design pressure 270 barg). The basis of safety and residual risk was demonstrated in a report to the HSE and accepted. Following the hazard evaluation process, 280 mT of active product was successfully manufactured in 347 batches during a 4-year period.
In some cases, hazardous chemistry may be the only economically viable and environmentally responsible route to a particular molecule. However, there are a number of advantages that can be achieved, including improved process efficiencies. Ultimately, manufacturers must aim to achieve complex synthesis in high yields with high selectivity in as few steps as possible.
Traditional batch processes remain the most commonly used in the pharmaceutical and specialty chemical sectors; however, inventories are high and, as demonstrated above, have the potential for catastrophic MAH consequences should things go wrong.
This leads to extra work and longer project implementation times because of the time taken to identify and implement the measures needed to eliminate or mitigate the consequence of the hazards and justifying to the regulators the safety of the plant. In addition, there are ongoing costs associated with maintaining the protective systems to the required reliability, adding to product costs.
The recent advances in the continuous production of pharmaceuticals may be used to eliminate or significantly reduce the risks involved in the processing of hazardous materials. Continuous manufacturing is not new to the process industry. In fact, it has been used extensively in less quality regulated sectors for many years. The continuous flow of materials can significantly reduce inventory during critical steps and so limit the consequences should the control systems allow the materials to reach onset temperatures. Whilst initial investment may still be comparable with a new batch reactor, there may be benefits in risk reduction, such as cost savings linked to the maintenance of safety systems and in supporting regulatory audits.
In short, hazardous chemistry is an extremely specialised area that requires integrated hazard evaluation, engineering, chemistry and analytical expertise. Careful planning will ensure the precise and careful control of process conditions; however, finding a contract manufacturer with the expertise and the capabilities in this area can be challenging. The role of thorough hazard evaluation when defining the process should not be underestimated.