Advanced technologies can help personalized medicine firms minimize their environmental impact, but only if used as part of a strategic plan that takes all aspects of production, storage, and distribution into account. This advice comes from Gabrielle Samuel, PhD, senior research fellow at the Wellcome Centre for Human Genetics at Oxford University, who says the environmental impact of making patient-specific medicines should not be underestimated.

“While personalized medicine might sound more environmentally friendly as they are more targeted, it will be important to mitigate the environmental impacts of personalized medicine manufacturing as much as possible,” she tells GEN. “This mitigation must not only be carbon-centric, but must also consider, for example, the use of water, where the manufacturing equipment comes from, who produces it, are they remunerated appropriately, and do they work in adequately regulated conditions, as well as disposal?”

Samuel and co-authors set out recommendations in a new paper advising that, in addition to using renewable energy, personalized medicines firms should minimize relying on temperature-controlled facilities and develop repurposing or recycling plans for old equipment.

Key aims

Technologies that make production more efficient should also be considered, Samuel says, citing areas like reduced energy and water consumption as key aims.

“I think automation, AI, etc., can and will be invaluable in helping to ensure we minimize the environmental impacts of personalized medicine production,” she continues. “Given the high proportion of impacts that come from running facilities, we can look to other fields to see how the use of AI is helping. For example, Google has used AI to reduce data center cooling by 40%, and AI is being used for precision farming to help reduce water use.”

However, Samuel says, industry’s applying advanced technologies to reduce environmental impact needs to be carefully considered as, left unchecked, efficiency gains can promote process intensification.

“It is important that we do not view AI as a technological solution. Technology alone cannot solve the problems; it can only help as a tool,” she points out. “For example, AI can make processes more efficient, but without constraints, as systems become more efficient, the ‘space’ saved through efficiency gains is likely to be filled quickly, running more power-hungry algorithms as machines become more efficient and so consumption will increase rather than decrease.”

In addition, making digital technologies impacts the environment. According to Samuel, the digital sector accounts for 2.1–3.9% of global emissions—roughly the same as the aviation industry—and generates significant amounts of technology waste each year.

To minimize environmental impact, personalized medicine makers need to consider all the technologies, facilities, and distribution networks involved in their products, including storing information about each element in a database.

“These databases need to include life cycle assessments for as many technologies as possible,” she says. “The process of manufacturing a personalized medicine is complex and has a range of steps and variables, for example single vs. multi use.

“When a group wants to calculate the environmental impacts of this process, they need to account for each of these steps and their associated variables. Inventories hold data to help them do this, but at present there are many data gaps for personalized medicine manufacturing, making it harder for researchers to calculate the overall impact of the process. The more information we can compile the better.”