How automation can drive a competitive advantage for CROs & CDMOs
The heart and key to success of contract research organisations (CROs) and contract development and manufacturing organisations (CDMOs) is to ensure every process and procedure is as meticulous as possible.
The problem: the processes and procedures of complicated experiments and the samples, equipment, and data management necessary to support them require step-by-step documentation, often done manually via hundreds of paper pages of standard operating procedures or on the “primitive” electronic side with PDFs and spreadsheets. In addition, manual entry may have created errors, and results are incomplete due to having disparate data across silos that must be found or completely reformatted before it can be analysed.
All these steps still require strict documentation to ensure regulatory compliance with GxP (GCP, GLP, and GMP). Each manually generated audit trail needs to include the who, what, when, where, why, and how to ensure reproducibility and ensure that the CRO/CDMO has fulfilled all obligations to its clients.
When the CRO/CDMO grows, the complexity and the paperwork grow concurrently. The CROs/CDMOs who don’t start to turn a critical eye toward how things are traditionally done are going to fall behind. It isn’t possible to be quick and efficient, deliver high-quality results for their clients, and or meet regulatory requirements without automation. Automation is key to ensuring long-term success.
Taking the right approach to automation
Embracing automation is a critical business decision; it will streamline operations, increase efficiency, and improve client service. The key to automation is connectivity: inventory, samples, equipment, experiments, and regulatory activities need to be connected and orchestrated, with all data combined into a single source of truth.
At many CROs/CDMOs, protocols equal pages and pages (and pages) of paper; either real paper or PDFs and spreadsheets that go on for megabytes. While following protocols to the letter mean the difference between success and failure and keeping the regulators and clients happy, the protocols and all related activities can be easily automated using existing technologies, including sample and inventory management platforms, electronic lab notebooks, and laboratory information systems.
These cloud-based technologies accelerate the pace of automation. It’s simple to create forms that include every protocol step. The automated forms automatically improve processes, procedures, and results as their simple set up minimises manual entry errors. For example, an experiment generates results ranging 5-10o Celsius. If the scientist inputs anything above or below that number, a message can be automatically generated, asking the researcher to check the data entry, or even the results themselves.
Automating protocols streamlines regulatory compliance and the client’s ability to track activities. Electronic lab management platforms have access controls and logging built in, so the who, what, when, where, and how of every action is tracked. Every piece of equipment used, the assay templates, samples, time between steps, and who performed the activities can easily be seen. The platforms usually come with a client portal, as well, ensuring the CRO/CDMO can communicate directly, receiving and sharing information with the client without having to spend hours processing requests and creating reports manually.
One aspect of automation that needs to be emphasised is having a single source of truth for all data, both structured and unstructured. Either a single database can be used to capture and collate all information for later analysis, or a data lake can be set up, ensuring everything is stored for immediate and future use.
Automation supports corporate growth, as well. New equipment can be easily added, as most solutions integrate with common equipment. Scaling the workload is simplified, also. It’s easy to modify existing electronic protocol templates to support new experiments and activities.
Stronger security is a byproduct of electronic automation, as only authorised personnel can access designated parts of the platforms, and each client’s project is co-located, but not accessible to other clients.
Using automation insights to strengthen the business
Not only can reports be run for the regulators and the clients, but automation makes internal reporting and thus business decision making easier as well. Reports can be run across a variety of parameters to deliver previously inaccessible insights within the manual environment. For example:
- Streamline internal decision making by building activity dashboards as they relate to:
- Client
- Equipment
- Personnel
- Inventory
- Protocols
- Speed ordering by tracking reagents and supplies. Many lab management platforms now include automated ordering, so when specific supplies run low or near expiration, the vendor is automatically contacted with the next order.
- Implement preventative maintenance by tracking equipment usage
- Improve scheduling by automating the equipment usage request calendar
- Simplify protocol version control
- Maximise the bottom line by tracking out-of-scope client requests and correlating them with additional service fees
- Better service clients with complete transparency via an automated client portal
- Make communications more efficient by having a single place to share and store notes
- Fulfill regulatory compliance with automated, real-time reporting
By eliminating manual processes and smoothing operations, more time is available to focus on profitable and creative activities.
Embracing automation with manual processes
In an ideal world, automation itself is automated. As we don’t live there, the first step to implementing automation is actually quite manual. A champion project manager, with senior management 100% on board, needs to run the hands-on process. The champion needs to recruit a small team across the entire organisation, including representatives from researchers and purchasing to regulatory compliance and IT and anyone else who is going to be affected by the process.
The team needs to work closely within each department to map existing manual processes to determine what can and should be automated, the priorities, and the expected timeline for automation. Without mapping and prioritisation, automation is set to fail.
The key is to go for quick wins versus tackling major challenges. For example, while translating a 100+ page protocol into an electronic version may take a few days of data entry, once it’s done, it can already be shared and used.
Eventually, though, everything will be automated.
Why automation needs to come before AI
Recently, Cenevo surveyed more than 150+ scientists across industries and saw that 45% plan to implement next-generation lab technologies such as AI and advanced robotics within two years; 25% stated that they either have no current plans or expect it to be longer than five years to implement next-generation lab technologies.
AI has become ubiquitous quickly. Software A, B, C, D, and E come with AI analysis; equipment F, G, H, and I use AI for predictive maintenance; liquid handler J and freezer K use AI to analyse inventory usage patterns. While this is all well and good, if none of this data is available within a single address, the time it takes to find the information and integrate it within a single platform outweighs any benefits AI may have delivered.
Automation is the future
Without automation, multiple employees will be spending their weeks only generating reports. Human errors will continue to hinder the ability to deliver real results, and trying to find data across silos and doing paperwork will continue to waste more and more time.
Automation is a must have, not a nice to have. Without a clear path to automation, CROs and CDMOs will find themselves left behind to competitors who have made automation standard operating procedure for their standard operating procedures. Clients are going to demand access.
The digital lab is already here. Automation is the first step.
About the author
Ben Harpazi holds an MSc in Plant Sciences and Genetics and is Labguru ELN product manager at Cenevo. He works closely with contract research organisations and contract development and manufacturing organisations to ensure platform customisations meet their individual requirements.
