Sustainable medicines development and use: Challenges and opportunities in the sustainable production of active pharmaceutical ingredients

The development of medicines has had a positive impact on the quality and longevity of our lives, and it is of central importance to society that we maintain and enhance our ability to treat disease. However, to continue to deliver life-changing medicines for patients, we must minimize the impact of these activities on the planet. A shared blueprint for peace and prosperity for people and the planet, now and into the future, has been outlined by the UN in a collection of 17 sustainable development goals (SDGs).¹ It is self-evident that the production and development of medicines, like every other commercial enterprise, will have to play its part in ensuring that these goals are met.

The sustainable production of medicines is a very broad topic, encompassing equitable access to healthcare, ethical research and production practices and product environmental stewardship, to name just a few. This commentary centres on the key environmental impacts of active pharmaceutical ingredient (API) production—which can account for a significant percentage of the overall environmental footprint of a medicine (vide infra)—and highlights activities and practices that help to minimize this footprint.

In the production of medicines, synthetic chemists and process engineers have always been conscious of developing the most efficient processes, in terms of use of resources and cost. However, since the advent of ‘Green Chemistry’, these considerations have become ever more central.² To this end, we employ multiple tools and metrics to help us understand and quantify our environmental footprint.^{3, 4} However, arguably, the most important of these is lifecycle assessment (LCA). We employ this standardized, science-based methodology to assess the magnitude and significance of key environmental impacts, for example, global warming potential, ozone depletion, resource intensity and freshwater ecotoxicity. This approach can be applied right across the pharmaceutical value chain, from raw material extraction to use by the patient. Employing a holistic tool such as LCA ensures that we take a ‘systems level’ viewpoint, thus avoiding burden-shifting, where one impact is reduced but another impact is increased or created in its stead. The LCA approach also allows us to identify ‘hotspots’ of environmental impact, empowering us to make targeted improvements, allocating the greatest resource to the areas of most concern.

According to a WHO report in 2019, global health spent as a percentage of GDP is increasing year-on-year worldwide, while the number of new drugs gaining approval each year is also following an upward trajectory.⁵ Due to an increasing number of regulatory requirements around sustainable production, it is ever more important that environmental impacts are decoupled from the manufacturing processes of medicines. Climate change, impacted biodiversity and pollution threaten the long-term survival of many species on this planet, including our own. In absolute terms, the pharmaceutical industry is considered a ‘medium-impact’ sector.⁶ However, in order to address climate change, pollution and loss of biodiversity, every industry needs to take responsibility for the environmental impacts of their business activities.

There is increased momentum from governments to implement policy proposals under the banner of a ‘Green New Deal’ to facilitate the transition to low carbon, low waste, zero-pollution societies. For example, the EU has included the European Green Deal as one of six priorities⁷ and has recently recognized the challenges of the Environment Action Programme and the ambition for zero pollution.⁸ The United States has also introduced legislation known as the Inflation Reduction Act, pledging over $350 billion dollars of investment over the next 10 years to reduce emissions.⁹ The recent publication of the pharmaceutical strategy for Europe makes clear reference to the ambitions articulated in the EU Green Deal and outlines specific initiatives to embed environmental sustainability in legislative instruments relevant to the Pharma Industry.¹⁰

But where are the challenges and opportunities in small molecule APIs? APIs are manufactured through multi-stage, often complex synthetic processes. These processes involve a wide variety of chemical reactions that are necessary to ‘build’ the API. Chemical reactions are energetic processes bound by the physical laws of thermodynamics. To put it plainly, the reactants must be reactive, otherwise they may not react in a beneficial manner. Substances that are highly reactive or have the potential to be highly reactive under specific conditions will invariably be classed as highly hazardous due to their inherent physical and chemical properties. Hazardous chemicals may be toxic, bioaccumulative, persistent and mobile in the environment and are controlled under legislation to protect people and the environment. However, synthetic organic chemistry, the scientific discipline employed to build ‘small molecule drugs’, often requires the use of hazardous chemicals and solvents. The distinction here between Hazard and Risk is particularly relevant, because when appropriate controls, safety measures and fail-safes are ‘designed-in’ to a process, even the most Hazardous materials can pose minimal Risk to the environment or to the people operating the process.

In process development, we follow what has become known as the ‘waste hierarchy’ approach: prevention, reduction, recycling, recovery and as a last resort, disposal. The preference is for prevention, or avoidance of the use of a particular material. But sometimes, this is not possible. Currently, we depend on oil-derived compounds to produce many of our solvents and starting materials. This is likely to be the case for the foreseeable future, so we must endeavour to reduce and recycle these precious materials where possible, while we invest in research to develop alternatives.

Resource efficiency is a cornerstone of chemical process development, and we make sure that our processes are as ‘mass efficient’ as possible. To measure this, we use a metric known as process mass intensity (PMI). This number is the sum of the masses (kilogram) of all the materials required in a chemical process to produce 1 kg of desired API; by applying this metric, we strive to minimize the amount of waste that we generate per kilogram of API produced. Despite its simplicity, there is a welcome correlation between reduction in PMI and carbon footprint reduction. As process chemists, we can realize reductions in waste in several ways. By simplifying the synthesis of the API through reducing the number of chemical steps/transformations, we can minimize the amount of solvent and reagents (and therefore raw materials) we use to run our chemical reactions. This can increase the chemical yields and throughput of our processes to develop more efficient work-up and isolation, reducing the number of unit operations required as well as the volumes of solvent used. LCA data for the production of oral solid dose drugs have shown us many times that API production is the largest contributor to the carbon footprint. The large footprint for API production is largely down to the ‘single use’ and incineration of solvents. Therefore, one way to reduce waste and environmental impact significantly is to use the same solvent multiple times, that is, solvent recycling.

Solvent recycling and reuse is an area of much untapped potential in API production. It has been an industry-standard approach in pharmaceuticals, to incinerate the bulk of the waste produced in chemical processes, although there are many exceptions where solvent recycling is carried out for certain stages of API processes. Accepting concerns for the potential toxicity of the waste streams, part of the reason for this practice was that, historically, as well as there being much less focus on the circularity of chemical processes, the technology required to separate complex waste streams was also underdeveloped. However, with more sophisticated separation science, together with the application of computational modelling, the isolation of high purity solvent suitable for reuse is now a possibility for relatively complex mixtures, providing safe, economic and environmentally viable access to more circular chemical processes. To this end, it is enabling us to ‘bake-in’ simplified recovery of materials by ‘beginning with the waste in mind’. For example, where possible, reactions should use a single organic solvent, with post-reaction operations (e.g., work-up and isolation) also using only this solvent plus another immiscible solvent—water being the most common—thus enabling more straightforward recycling and reuse. With appropriate testing and controls in place, widespread solvent recycling in API production processes has the potential to dramatically reduce the carbon footprint of API production. As an industry, we need a collective change of mindset to a point where we no longer refer to waste solvent but speak instead of used solvent, highlighting the inherent societal value of the material and its potential for reuse. Even in cases where the quality of solvent recovered is not appropriate for the production of medicines, the solvent could find use beyond the pharmaceutical industry. The use of certain bio-renewable solvents, generated from agricultural waste products, could also lead to low carbon, or even carbon-negative API production processes.

Apart from solvent recycling, what else could be recycled? It may not be apparent to the reader, but the pharmaceutical industry often employs significant quantities of precious metals, for example, palladium, rhodium and ruthenium as catalysts to carry out our chemical reactions. One way to consider these catalysts is that they facilitate reactions; they are ‘breakers’ of weaker chemical bonds in the starting materials and ‘formers’ of materials with stronger bonds in the products. These metals allow us to carry out chemical transformations that would have seemed fanciful in the first half of the twentieth century, but they do this at a cost. The mining of these metals is associated with a large carbon footprint and extremely high water use per kilogram of pure isolated metal. Currently, these metals are central to the production of many of the most important drugs available to humankind and when we do employ these remarkable ‘match-makers’, we must ensure that we do our research to minimize the quantities used and ensure that our recycling or recovery technologies allow for highly efficient use and reuse of these most precious natural resources.

It is also crucial that we minimize any losses during final purification of API, formulation and packaging, due to the high embodied environmental impact of this material, and this is critical for waste stream management.

If we can reduce the impact and material costs of small molecule API production, what about biological APIs, are there sustainability opportunities in that field? While there are multiple types of biologic APIs, the most important and numerous are the monoclonal antibodies (mAbs). Some mAbs have been approved by the FDA, and they are among the top-selling drugs worldwide. Production processes for these molecules are very different to those employed in small molecule API synthesis, and thus, the production of monoclonal antibodies comes with its own particular sustainability challenges. A typical production process involves the fermentation of host cells to produce mAbs, which are then extracted, purified and formulated into the final API or drug substance. These processes require a controlled environment to avoid microbial contamination. According to a recent study, the energy required to maintain this controlled environment or ‘clean room’ represents a large fraction of the overall carbon footprint of these processes.¹¹ Efforts to improve energy efficiency, plant time efficiency and the procurement of renewable energy can help in reducing the impact of this aspect of production.

A further issue to consider is that these processes tend to require large volumes of high purity water, known as ‘water for injection’ (WFI). As a result, the impact of this water-intensive process is more significant when manufacturing is performed in water-scarce locations. According to the UN, water use is outstripping population growth rates, and a growing number of regions are reaching their limit in being able to supply water sustainably.¹²

It is not just the quantity, but the quality of water used throughout the mAb production processes, that comes with potential consequences for the environment. Generation of this highly pure WFI and ‘clean steam’ is extremely energy intensive and also generates some ‘waste’ water that does not meet the criteria for use in the process. The industry is acting on this, increasing water efficiency at sites and adopting water recycling more widely, for example, by diverting the condensed water from ‘clean steam’ generation, or the rejected purified WFI, to facility cooling towers.

Traditional biologic API manufacturing facilities were designed to house large, stainless-steel equipment, including bioreactors that hold perhaps 20 000 L of cell culture, spanning several storeys of the buildings in which they are housed. All stainless-steel equipment undergoes water-intensive cleaning steps in between each production campaign, further increasing the water demand. The systems required to maintain temperature and humidity levels not only are power hungry but also produce further water expenditures, which increase with the size of the manufacturing space. In order to address these particular challenges, the industry is shifting from ‘batch processes’ in large vessels to ‘continuous bioprocessing’, which replaces the large stainless-steel equipment, with small single-use systems, which act as flow-through tanks that generate a constant output stream.^{11, 13} While the increased use of mixed, difficult to recycle plastics poses a significant challenge from a waste stream perspective, continuous processing supports a significant increase in API produced per campaign, which increases the annual output of total API produced at a site, ultimately making a site more efficient. Additionally, the decrease in the physical size of the equipment needed allows for much more mobile/modular facility design within a smaller building footprint and reduces the energy and water demand to maintain control of the cleanrooms; these changes in aggregate could lead to a significant reduction in the impact of these processes overall. As these changes take place and become embedded, we will better understand the size of these savings.

Finally, by way of illustration of the requirement for a holistic approach to assessing the environmental impact of the production of a medicine, we have observed recently, from our own unpublished studies, that the carbon footprint generated by the patient travelling to the pharmacy regularly to pick up their prescription can make up a large fraction of the overall carbon footprint for the total healthcare pathway and treatment of a patient. More research is required to understand if and how this impact varies across different medicines, as well as working in partnership with governments and wider stakeholders to develop systems that minimize this impact.

Within the pharma sector, there is growing momentum for sustainability improvements, with many companies having signed up to ‘the science based targets initiative’ and made a commitment to achieve carbon zero emissions within the next 10–20 years. The NHS in the United Kingdom has vowed to ‘no longer purchase from suppliers that have not aligned with our trajectory towards net zero carbon’,¹⁴ thus providing further impetus, if one was needed, to strive for a lower carbon future in our industry. As noted earlier, environmental sustainability is not limited to a single environmental issue. It is critically important for any meaningful sustainability strategy to look beyond carbon and address other important issues such as pharmaceuticals in the environment, animal welfare, water stewardship, waste disposal, hazardous, persistent, bioaccumulative, or toxic chemicals of concern, air pollution and habitat maintenance and creation. There are significant challenges ahead, and recent calls for a systems approach to healthcare will rely on effective LCA.¹⁵ However, there is cause for optimism, as progress in these key areas is currently being accelerated through collaboration across the pharmaceutical industry, sharing best practice on how the environmental credentials of our processes can be improved. Groups like the American Chemical Society Green Chemistry Institute Pharmaceutical Roundtable (ACS GCIPR) bring members of the world's biggest pharmaceutical companies together to share best practice and to generate ‘call to arms publications’, identifying the scientific challenges to be solved and providing some research grants that help researchers to carry out key proof-of-concept experiments toward innovative solutions to current sustainability challenges. The European Federation of Pharmaceutical Industries and Associations (EFPIA) can also serve to educate both the public and legislation developers on the key challenges that come with certain proposed changes in legislation. We believe that regulators and others are also interested in how the development of medicines can become more sustainable. Open discussion and bilateral learning such as this can help to ensure that our industry strides onward sustainably, secure in the knowledge that we are not only making innovative and life-changing medicines but also acting as responsible and mindful custodians of our fragile planet.

The authors are employees of and/or shareholders in AstraZeneca.