Christopher Cole

In the first half of a two-part deep dive on major pharma-AI partnerships, Richard Jaenisch, senior director of education, outreach, and digital experience at Open Biopharma, breaks down why Roche's raw GPU infrastructure buildout and Sanofi's regional talent hub represent fundamentally different bets on where AI adoption actually stalls in pharmaceutical manufacturing. (

Richard Jaenisch surveys pharma's biggest recent AI deals with Chris Cole in part one of a two-part breakdown of the major moves.

Manufacturing Intelligence: Sizing Up Pharma's Biggest AI Deals, Part One

1:18 - The biggest differences between the six-plus major AI pharma moves in the last month

12:30 - Which of these six is most likely to be quietly walked back or scaled down in 18 months

17:50 - Whether we’re looking at in-house ML training and dedicated datacenter buildout or closed-system subscription expansion with a workforce layer bolted on, and if it matters

24:14 - If the latter, what a genuine in-house build would look like in contrast

30:22 - As far as the end result, be that better drugs or faster timelines, whether it matters if it’s in-house infrastructure versus a closed-system subscription

36:51 - Whether Roche going big on raw GPU infrastructure and Sanofi on a regional talent hub are different best on the same bottleneck in pharma AI

42:47 - If a pharma CEO on a frontier AI lab’s board is governance, oversight, or pharma keeping an eye on what’s coming out of these labs

46:48 - If the embedding of pharma execs into AI labs is something to expect more of

Speaking with Chris Cole, associate editorial director at 

, Jaenisch argues that compute investment and workforce investment solve different bottlenecks entirely. Roche's approach, he explains, functions as a hedge: excess GPU capacity gets monetized as a service to offset cost, and if external demand falls short, the infrastructure simply gets absorbed into internal operations. Sanofi's talent-first strategy, by contrast, targets what Jaenisch sees as the deeper obstacle to AI value capture: not hardware, but people who neither trust nor know how to use the tools they've been handed. Jaenisch notes that when it comes to AI training, "it is rare where I have talked with someone and they've said, 'Yeah, I was completely satisfied with the training that I got from my work on AI.'"

Jaenisch is skeptical of enterprise-wide rollout claims, describing most current deployments as top-down implementations that generate low-quality use cases because employees are forced into tools that don't address their actual pain points. Bottom-up adoption, he argues, produces far better outcomes because it starts from real friction points rather than mandates. He also flags a distinction pharma leadership often glosses over: whether these deals represent genuine in-house ML infrastructure or subscription-based tools with a workforce layer bolted on — a difference he says matters more to medium-sized companies, who lack the scale to benefit from proprietary systems the way large enterprises do.

On the trend of pharma executives joining frontier AI lab boards, Jaenisch is measured. He says the practice is "more than optics" but adds that "I don't think there's going to be a huge influence on how the model gets built." He predicts the real value will come not from executive-level relationships but from embedded, cross-functional project teams working directly with AI labs on specific manufacturing use cases.

Videos

PharmTech recently caught up with Remco Munnik, founder and owner of Arcana Life Sciences, to follow up on predictions he made 

on the 2026 landscape for pharmaceutical development and manufacturing

When PharmTech last spoke with Munnik about 2026, he emphasized the need for investment in data-driven infrastructure to ensure traceability from the point of data creation to final registration. He also highlighted promising innovations, such as the work of the 

 (ROG), which seeks to replace resource-intensive document submissions for lifecycle changes with direct database updates. These advancements, combined with the 

upcoming European pharmaceutical legislation

 at the time, he said, were set to redefine how the industry maintains compliance and achieves speed to market.

Now that 2026 is half over, Munnik offers his thoughts on what predictions came to fruition, which didn't, and why, as well as what's in store for the second half of the year and into 2027.

PharmTech: In December, you noted that advancements from the Regulatory Optimisation Group, combined with the European New Pharmaceutical Legislation, were set to redefine how the industry maintains compliance and achieves speed to market. Has this redefinition occurred?

Munnik: The redefinition is ongoing and a work in progress. Especially with the ROG, we have made significant progress, including a face-to-face (F2F) workshop in June to outline a future state process. We are now in the process of finalizing detailed discussions on specific technology and process requirements. After the summer break, I expect to plan another F2F workshop to finalize these activities, which should set the ROG to achieve tangible success in 2027.

With regard to the full impact of the European Union's New Pharmaceutical Legislation (NPL), we need to be patient. Working in a highly regulated environment requires thoroughness. I expect that during the second half of 2026 we will get more information with regard to the impact of the NPL and the actual implementation acts.

You also noted a widening skills gap caused by the digital evolution, advocating for on-the-job rotations and a leadership mindset to bridge this gap. Have you seen such rotations and mindsets within pharma so far this year?

Many organizations still treat digital evolution as a technical upgrade rather than a leadership challenge. The companies moving fastest are those that frame transformation as capability building, not compliance, investing in cross-functional rotations, data literacy, and new operating models.

But there's another factor slowing progress: the economic reality. Global political uncertainty, pricing pressures, and constrained margins are forcing teams to prioritize short-term compliance deliverables over long-term capability development. When budgets tighten, rotations and skill-building are often the first to be postponed, even though they are exactly what is needed to close the gap.

In December, you identified fragmented legacy systems and limited data standardization as barriers to AI adoption, noting that successful AI adoption requires technical tools plus robust data governance and a significant culture shift within organizations. Have any of these occurred so far this year?

I do see progress, but it remains slow. Several large pharma organizations have now put data governance frameworks in place, especially those linked to identification of medicinal products (IDMP). There's a strong appetite from management to “solve everything with AI,” but the reality is unchanged: AI only works when the underlying data governance is solid. Without consistent standards, ownership, and quality controls, even the best tools can't deliver meaningful outcomes. Many companies are still working on the basics: establishing trusted data, using it consistently, and making decisions based on it.

The second half of the year should also be a real opportunity for the EU network (EMA, National Congress on Autonomous Systems, and industry) to make tangible progress with regard to post-market surveillance (PMS) qualification.

What's the most unexpected development in pharma manufacturing and development so far in 2026?

I'm genuinely excited by the level of support and momentum behind PMS and IDMP this year. Key stakeholders across the network have been far more aligned and vocal than in the past. That alignment is finally creating the conditions for real progress: clearer priorities, faster decision-making, and a shared understanding that structured data is no longer optional but foundational.

What do you see as the biggest challenge facing pharma manufacturing and development for the remainder of 2026?

The biggest challenge for the remainder of 2026 is digitalization with consistency. As more processes across manufacturing and development become data-driven, the pressure to maintain clean, aligned, and interoperable data increases dramatically. Digitalization only works when the underlying data is stable, governed, and consistent across regulatory, quality, and manufacturing systems. Today, many organizations still operate with fragmented and legacy systems, mismatched data definitions, and stand-alone processes. This creates friction, slows automation, and undermines trust.

The companies that will succeed are those that treat digitalization as a data consistency challenge, as opposed to a technology project, ensuring that what is submitted, manufactured, released, and reported is all driven from a reliable source.

Articles

Mid-2026 signals incremental, not transformational, progress in regulatory modernization and data-centric operating models. Regulatory Optimisation Group work has advanced through in-person workshops and requirements definition, aiming for measurable outcomes in 2027. The EU New Pharmaceutical Legislation impact remains pending, with implementation acts expected to clarify second-half priorities. Digital transformation continues to be treated as a technical project in many companies, while leaders invest in cross-functional rotations and data literacy. Data governance progress, often IDMP-linked, remains the prerequisite for credible AI adoption.

Arcana Life Sciences' Remco Munnik revisits his 2026 predictions on EU regulatory reform, AI adoption, and data governance, and previews pharma's biggest challenges for H2 2026.

Munnik Revisits His 2026 Predictions: A Midyear Check-In on Regulatory Reform, Data, and AI in Pharma

High-speed autoinjector assembly integrates automated component feeding, orientation, spring loading, energy module integration, and housing assembly, followed by functional testing and quality inspection

Peptide or protein drug substances are produced under cGMP in sterile areas meeting 21 CFR Parts 210

 Class 5 standards. Active ingredients and excipients are mixed using validated procedures for uniformity and stability, and then sterile filtered through 0.2 μm filter per USP <797>.

Sterile solutions are filled into prefilled glass syringes in ISO Class 5 environments, often within isolators or RABS, following EU GMP Annex 1 and FDA guidance. Syringes are prepared per ISO 11040 with validated washing, siliconization, and depyrogenation, meeting USP <85>

 Accurate doses are dispensed with validated systems (ICH Q2,

), then stoppered with closures compliant with ISO 8362

 and container closure integrity verified per USP <1207>.

In-process controls include fill volume, stopper position, and particulate inspection per USP <788>

 Environmental monitoring is conducted per EU GMP Annex 1,

 with media fills for sterility assurance. Filled syringes undergo QC for sterility, endotoxins (USP <71>,

), dose accuracy, and visual quality before integration into autoinjectors built under ISO 13485

 Class 7 environments using validated equipment, and each unit is tested for injection force, delivery time, and safety per ISO 11608

 and FDA guidance. Stability studies follow ICH Q1A.

 Release occurs only after all sterility, functional, and quality checks pass.

Formulations maintain stability, dose accuracy, and device compatibility under ICH Q6B,

 Buffers, pH, stabilizers, and surfactants control degradation and adsorption. High peptide concentration increases viscosity, affecting filling, syringe performance, and needle injection, evaluated per ISO 11608.

 Particulates are controlled per USP<788>,

 and container compatibility verified via adsorption, extractables/leachables (USP<1663>,

), ensuring stability and safety throughout shelf life.

Having outlined formulation requirements, regulatory standards, and quality controls, the following sections discuss how these factors influence autoinjector performance, high-speed assembly, and factory acceptance testing. Table 1 summarizes relevant standards and guidance across the lifecycle of a peptide formulation in a prefilled syringe (PFS) and its integration into the autoinjector system.

Table 1-1 Lifecycle Stage Guidance, Regulations, and Standards Mapping

Table 1-2 Lifecycle Stage Guidance, Regulations, and Standards Mapping

Table 1-3 Lifecycle Stage Guidance, Regulations, and Standards Mapping

Factors Influencing Autoinjector Performance

Peptide formulations in PFS directly impact autoinjector performance. For PFS-based autoinjectors (Figure 1), viscosity, driven by peptide concentration, affects injection force, filling accuracy, and dose delivery. Buffers and pH must be compatible with syringe materials to prevent degradation, while surfactants like polysorbates minimize adsorption and particle formation (USP<788>14, USP<790>15, USP<787>28). Plunger lubrication ensures smooth motion, and formulations must withstand shipping and handling stresses.

Injection efficiency depends on the interaction of formulation, syringe mechanics, and device design. Fine needles improve comfort but increase resistance (Hagen–Poiseuille law; Shire, et al. 2004, 1390–1402).

 Autoinjectors use springs to push the plunger, overcoming viscosity and plunger friction (ISO 11040

). Shear stress, air bubbles, and temperature also influence performance. Design verification follows ISO 13485,

Engineers balance spring force, needle size, plunger friction, and viscosity to ensure reliable full-dose delivery. Typical volumes range from 0.3–1.0 mL, with springs generating 20–80 N and injection times of 5–10 seconds. Early modeling integrates these factors to ensure safe, consistent, and comfortable administration. Understanding how formulation and device design affect injection performance informs the assembly process, where precise alignment and integration of components are critical to ensuring consistent function and patient safety.

1. Outer Cap 2. Needle Shield 3. Needle 4. PFS System 5. Plunger Stopper 6. Flange 7. Plunger Rod 8. Spring 9. Main Body 10. Activator / Trigger

Autoinjector assembly begins with automated machines feeding and aligning components such as housings, springs, plungers, and caps. Sensors and vision systems inspect parts for correct orientation and defects, ensuring only compliant components proceed, in line with ISO 13485,

 Prefilled syringes are loaded using robotic or vacuum systems designed to prevent damage and maintain sterility. Vision inspection confirms syringe quality and alignment per ISO 11040,

 while environmental controls follow 21 CFR Part 211.

Springs are inserted and compressed using automated stations, with sensors verifying proper force and placement according to safety and GAMP 5

 validation requirements. Internal components, including the syringe holder, plunger, and trigger, are assembled into the housing using precision fixtures and secured via snap-fit or welding. Alignment and design specifications are confirmed with sensors and measurements.

Completed autoinjectors are labeled and packaged, with vision systems checking accuracy and appearance. Serialization tracks each unit for full traceability, ensuring compliance with 21 CFR Part 820,

This process supports consistent quality, reliable device function, and regulatory compliance, forming the basis for factory acceptance testing of high-speed autoinjector equipment.

A 100,000-unit FAT may be conducted to verify autoinjector assembly, especially for high-value, complex lines. Large FAT runs are often required when the purchase agreement or User Required Specification (URS) mandates a throughput demonstration, when the system is first-of-a-kind or high-risk, or when post-installation troubleshooting would be costly or disruptive, such as installing directly in a cGMP environment or overseas. Very expensive equipment (often $5M+) also justifies extended FAT to confirm reliability, throughput, and reject rates. Standard projects typically use smaller FAT batches of 5,000–20,000 units, with a 100,000 run reserved for commissioning or performance qualification at the manufacturing site. High-speed systems record force curves, cycle data, and alarms in the HMI, enabling engineers to detect assembly or transport deviations.

 The following sections describe two critical autoinjector assembly stations and their sampling requirements.

This station is responsible for precisely inserting the PFS into the auto-injector housing.

 The syringe typically contains the drug product and includes key components such as the glass syringe barrel, plunger stopper, needle, and needle shield. A servo-controlled robot or linear axis performs the pick-and-place operation to transfer the syringe from a tray or feeder. The system usually uses a vacuum gripper or custom end-effector designed to handle the fragile glass syringe safely. The servomotor ensures accurate positioning, smooth motion, and correct alignment during insertion into the injector body or cartridge holder. Sensors verify syringe presence, orientation, and proper seating inside the housing before the assembly proceeds to the next station. The servo operates primarily in position control mode, emphasizing precision and repeatability. If the syringe is misaligned, not fully seated, or cracked, the injector may not function correctly. Problems at this stage can lead to needle deployment issues, incomplete dose delivery, or device activation failures during later functional or destructive testing (DT). Therefore, this station ensures that the drug container (syringe assembly) is correctly positioned within the auto-injector mechanism for reliable operation.

The press-fit station assembles critical autoinjector components, such as the outer housing, spring module, actuator parts, trigger mechanism, and syringe carrier, using a servo press system.

 The servo press applies precise force over a controlled distance to join each component correctly. For every part pressed, the machine records a force–displacement curve showing how the applied force changes as the component move into position. Each component has its own expected profile due to differing mechanical properties.

If the press force is too low, components like the spring module or trigger may not lock properly. If too high, parts such as the housing, syringe holder, or internal mechanisms can deform or be damaged. The system compares each curve against validated limits, flagging or rejecting devices that fall outside tolerance. Curves are stored in the HMI or quality database for traceability and process validation. Monitoring these curves helps engineers detect misaligned parts, incorrect tolerances, or damaged components, which could later cause trigger failure, incomplete needle extension, or improper drug delivery during destructive testing. By ensuring correct assembly and secure integration of mechanical and activation components, this station supports reliable autoinjector function and consistent drug delivery. To ensure the assembly processes at critical stations perform consistently across the production run, representative sampling strategies are applied.

Statistical Sampling for Critical Stations

For a FAT of 100,000 autoinjector units, stratified sampling ensures representative testing across the production run.

 The run is typically divided into five strata of ~20,000 units each, capturing variations at the start, middle, and end. Non-destructive testing (NDT) is performed on 1–2% of units per stratum, verifying assembly integrity, syringe placement, press-fit quality, and vision inspection without destroying the product. Force-displacement curves for pressed components are included to monitor mechanical consistency, and NDT samples are evenly distributed to detect drift or anomalies. DT is performed on 0.1–0.5% of units per stratum, focusing on critical functions such as trigger activation, spring release, needle deployment, and dose accuracy. DT samples are evenly distributed across all strata to ensure consistent functional performance. Together, NDT and DT provide statistically significant evidence that the line produces conforming devices.

 guidelines, balancing regulatory confidence with efficiency while avoiding full-batch destructive testing. Each stratum acts as a control segment to detect process drift or setup issues, and sample sizes align with ANSI/ASQ Z1.4

 functional testing requirements. For autoinjector assembly, five well-distributed strata are sufficient because the process is highly controlled, critical quality attributes are limited, and continuous monitoring reduces the risk of undetected variation. Adding more strata increases inspection effort with minimal additional benefit.

) analysis measures how well a process produces parts within specification by combining variability and centering of key parameters.

 In a 100,000-unit autoinjector run, NDT samples 1–2% of units per stratum and DT samples 0.1–0.5%, across five 20,000-unit strata are typically sufficient. Continuous monitoring via force-displacement curves, sensors, and vision systems supports both C

 calculations and functional assurance. Each stratum captures temporal process variation.

Autoinjector molds may have up to 16 cavities, so sampling represents all cavities to detect cavity-specific variation. NDT samples are distributed evenly across cavities, and DT includes units from each cavity to identify functional failures.

 Process stability is checked using control charts, ensuring no special causes of variation. C

 is calculated from the mean and standard deviation of mechanical parameters, either per stratum or pooled across strata and cavities. Comparing C

 across strata and cavities detects drift, outliers, or localized issues. This stratified, cavity-aware approach ensures C

 reflects both temporal and mold-related variation. Combined with DT results, C

 provides robust evidence that the autoinjector production run consistently meets quality and functional specifications.

A coordinated approach integrating peptide formulation, PFS design, high-speed autoinjector assembly, and lifecycle-aligned regulatory compliance ensure consistent product quality, dosing accuracy, and patient safety. Applying FAT, C

 analyses, and adherence to industry standards provides a robust framework for reliable commercial manufacturing and effective delivery of peptide therapies via autoinjectors.

1. 21 CFR Part 210 - Current Good Manufacturing Practice in Manufacturing, Processing, Packing, or Holding of Drugs; General; 2026, 

https://www.ecfr.gov/current/title-21/chapter-I/subchapter-C/part-210

2. 21 CFR Part 211 - Current Good Manufacturing Practice for Finished Pharmaceuticals, 2026, 

https://www.ecfr.gov/current/title-21/chapter-I/subchapter-C/part-211

3. ISO 14644-1, Cleanrooms and associated controlled environments - Part 1: Classification of air cleanliness by particle concentration, 2015, 

4. USP General Chapter <797> Pharmaceutical Compounding – Sterile Preparations, 2023, 

https://www.usp.org/compounding/general-chapter-797

5. USP General Chapter<85> Bacterial Endotoxins Test, 2024, 

https://www.usp.org/harmonization-standards/pdg/general-methods/bacterial-endotoxins

6. USP General Chapter<1228> Depyrogenation, 2017, 

https://doi.usp.org/USPNF/USPNF_M8097_03_01.html

7. USP General Chapter<1229> Sterilization of Compendial Articles, 2022, 

https://doi.usp.org/USPNF/USPNF_M7408_01_01.html

8.ICH Q2 Validation of Analytical Procedures, 2024, 

https://www.ema.europa.eu/en/ich-q2r2-validation-analytical-procedures-scientific-guideline

9. Process Validation: General Principles and Practices, 2011,

https://www.fda.gov/files/drugs/published/Process-Validation--General-Principles-and-Practices.pdf

10. ISO 8362-2, Injection containers and accessories - Part 2: Closures for injection vials, 2024, 

https://www.iso.org/obp/ui/es/#iso:std:iso:8362:-2:ed-4:v1:en

11. USP General Chapter<381> Elastomeric Components in Injectable Pharmaceutical Product Packaging/Delivery Systems, 2024, 

https://doi.usp.org/USPNF/USPNF_M99140_60201_01.html

12. USP General Chapter<382> Elastomeric Component Functional Suitability in Parenteral Product Packaging/Delivery Systems, 2025, 

https://doi.usp.org/USPNF/USPNF_M11355_04_01.html

13. USP General Chapter<1207> Package Integrity Evaluation—Sterile Products, 2017, 

https://doi.usp.org/USPNF/USPNF_M99926_01_01.html

14. USP General Chapter<788> Subvisible Particulate Matter in Injections, 2026,

https://www.usp.org/harmonization-standards/pdg/general-methods/particulate-contamination

15. USP General Chapter<790> Visible Particulates in Injections, 2017, 

https://doi.usp.org/USPNF/USPNF_M7197_01_01.html

16. EU GMP Annex 1, Manufacture of Sterile Medicinal Products, 2022, 

https://health.ec.europa.eu/system/files/2022-08/20220825_gmp-an1_en_0.pdf

17. ISO 14698-1, Cleanrooms and associated controlled environments, 2003, 

https://www.iso.org/obp/ui/#iso:std:iso:14698:-1:ed-1:v1:en

18. USP General Chapter<71> Sterility Tests, 2016,

https://www.usp.org/harmonization-standards/pdg/general-methods/sterility-test

19. ISO 13485, Medical Devices - Quality Management Systems, 2016, 

20. ISO 11608, Needle-based Injection Systems for Medical Use - Requirements and Test Methods 2022, 

21. ISO 10993, Biological evaluation of medical devices, 2025, 

22. ICH Q1A Stability Testing of New Drug Substances and Drug Products, 2003,

https://www.ema.europa.eu/en/ich-q1a-r2-stability-testing-new-drug-substances-drug-products-scientific-guideline

23. ICH Q6B Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products, 1999, 

https://www.ema.europa.eu/en/ich-q6b-specifications-test-procedures-acceptance-criteria-biotechnological-biological-products-scientific-guideline

24. ICH Q8 Pharmaceutical Development, 2006, 

https://www.ema.europa.eu/en/ich-q8-r2-pharmaceutical-development-scientific-guideline

25. ICH Q9 Quality Risk Management, 2023, 

https://www.ema.europa.eu/en/ich-q9-quality-risk-management-scientific-guideline

26. ICH Q10 Pharmaceutical Quality System, 2008, 

https://www.ema.europa.eu/en/ich-q10-pharmaceutical-quality-system-scientific-guideline

27. ICH Q5C Stability Testing of Biotechnological/Biological Products, 1996, 

https://www.ema.europa.eu/en/ich-q5c-stability-testing-biotechnological-biological-products-scientific-guideline

28. USP General Chapter <787> Subvisible Particulate Matter in Therapeutic Protein Injections, 2020, 

https://doi.usp.org/USPNF/USPNF_M6497_02_01.html

29. USP General Chapter<1663> Assessment of Extractables Associated with Pharmaceutical Packaging/Delivery Systems, 2020, 

https://doi.usp.org/USPNF/USPNF_M7126_03_01.html

30. USP General Chapter<1664> Assessment of Drug Product Leachables Associated with Pharmaceutical Packaging/Delivery Systems, 2024, 

https://doi.usp.org/USPNF/USPNF_M7127_03_01.html

31. Shire, S.J; Shahrokh, Z.; and Liu, J. Challenges in the development of high protein concentration formulations,

https://pubmed.ncbi.nlm.nih.gov/15124199/

32. ISO 11040-8, Prefilled Syringes: Requirements and Test Methods for Finished Prefilled Syringes, 2026, 

33. ISO 14971 Medical devices - Application of Risk Management to Medical Devices, 2019, 

34. 21 CFR Part 820, Quality Management System Regulation, 2026, 

https://www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-820

35. ISPE GAMP 5: A Risk-Based Approach to Compliant GxP Computerized Systems, 2

https://guidance-docs.ispe.org/doi/book/10.1002/9781946964571

36. 21 CFR Part 11, Electronic Records; Electronic Signatures, 2026, 

https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-11

37. J. Sprovieri, Servo Presses Aid Assembly of Autoinjectors, 2022, Assembly Magazine, 

https://www.assemblymag.com/articles/97398-servo-presses-aid-assembly-of-autoinjectors

38. Gerresheimer’s, Adapter Aids Positioning of Syringes in Autoinjectors, Pharmaceutical Technology, 2014, 

https://www.pharmtech.com/view/adapter-aids-positioning-syringes-autoinjectors

39. ASTM E2910-12, Standard Guide for Preferred Methods for Acceptance of Product, 2018, ASTM International, 

https://www.ema.europa.eu/en/human-regulatory-overview/marketing-authorisation/compliance-marketing-authorisation/sampling-testing

41. ANSI/ASQ Z1.4, Quality Standards, 2003, 

https://asq.org/quality-resources/z14-z19?srsltid=AfmBOookYlqMIDtOTau_oZbzycHPOl7a1VyxcIUfGLZ8qw01ryll5iw1

42. ISO 2859-1, Sampling Procedures for Inspection by Attributes, 2026, 

43. Engineering Statistics Handbook: Process Capability Indices, Chapter 6, NIST, 2012, 

https://www.itl.nist.gov/div898/handbook/

44. ISO 20457 Plastics Moulded Parts - Tolerances and Acceptance Conditions, 2018, 

Peptide/protein drug products in prefilled syringes require tightly controlled aseptic manufacture (ISO 5), validated fill–finish, and comprehensive QC for sterility, endotoxin, particulates, dose accuracy, and container-closure integrity. Formulation attributes—especially viscosity, pH/buffer selection, surfactants, and lubrication—directly determine autoinjector injection force, delivery time, and robustness to handling. High-speed assembly relies on automated alignment, sensor/vision inspection, servo-press force–displacement profiling, and end-to-end traceability. Large-scale factory acceptance testing, stratified sampling, and CpK analyses provide evidence of sustained throughput and functional reliability.

Peptide autoinjector manufacturing demands precise formulation-device coordination, rigorous Cpk-driven quality control, and FAT validation for reliable, compliant dosing.

A Lifecycle Approach to Peptide Formulation in Prefilled Syringes for High-Speed Autoinjector Assembly

Screenshot 2026-06-12 at 1.20.08 PM.png

CAI's Jon Thompson explains how digital twins cut cleanroom rework by up to 50% and pressure excursion investigation time by up to 70%.

Manufacturing Intelligence Episode 4: Cleanroom Digital Twins

, co-hosts Chris Cole and Richard Jaenisch explore the emerging role of digital twins in cleanroom design and qualification in pharmaceutical manufacturing with guest Jonathon Thompson, global director of the digital center of excellence at CAI.

To view the full fascinating, practical, engaging conversation, click here!

Digital twins of cleanrooms are offering measurable reductions in commissioning time, rework, and pressure excursion investigation, but only when paired with rigorous real-time data verification, explains Thompson. He adds that the technology's core value lies not in replacing IQ/OQ/PQ, but in catching design flaws before construction locks them in.

The traditional cleanroom qualification workflow exposes problems late, Thompson notes, often after equipment is already installed and early-phase processes are running. Digital twins, fed with 3D building information modeling (BIM) models, HVAC specifications, personnel flow data, and physics-based fluid dynamics simulations, allow manufacturers to stress-test facility designs upstream. Thompson points to tangible returns: 10–30% reductions in commissioning time and resource hours, 20–50% reductions in rework, and 30–70% faster root cause identification during pressure excursions.

Regulatory acceptance, Thompson argues, depends on how the twin is used. "If you're using this digital twin as your single source of truth without verifying that data in real time with what's actually happening out in the field," he warns, "they're going to start to question it." Continuous process verification—cross-referencing the model against live environmental monitoring data, particle counts, and pressure differentials—is what earns FDA and EMA confidence in the approach.

For CDMO and modular facilities, where equipment configurations and product mixes change frequently, the stakes around model maintenance are especially high. Thompson observes that "we've got to change the mindset" to treat the BIM model as a living asset subject to change management, not a static deliverable from the capital project phase. As generative AI and light detection and ranging-based reality capture mature, automated model updating may eventually reduce that burden, making the digital twin a true operational constant rather than a commissioning tool.

CAI's Jon Thompson details how digital twins cut cleanroom commissioning time by up to 30% and accelerate GMP deviation investigations.

Manufacturing Intelligence: How Digital Twins Are Changing Cleanroom Validation

In a rapidly shifting global marketplace characterized by post-pandemic onshoring and an increasingly complex tariff environment, the pharmaceutical manufacturing industry is searching for stability and speed, according to Delaware Governor Matt Meyer. At the 

 session on June 2, 2026, Meyer detailed his state’s aggressive strategy to become a premier hub for biopharmaceutical innovation. While his remarks focused on Delaware, the principles of collaboration, workforce pipeline development, and regulatory speed provide a vital roadmap for pharma manufacturing across the United States and the globe.

For many in pharmaceutical manufacturing, the relationship with government is often defined by regulatory friction. However, the Delaware model suggests that local governments must evolve into active partners to move technology forward. Governor Meyer emphasized that a state’s role should be to collaborate on research, help secure funding, and reduce the friction between development and market entry.

For professionals globally, this highlights the growing importance of public-private ecosystems. In Delaware, this is anchored by institutions like the 

National Institute for Innovation in Manufacturing Biopharmaceuticals

 (NIIMBL), where global partners collaborate to improve how medicines are made, tested, and delivered.

Scaling Innovation Through Specialized Infrastructure

Delaware is positioning itself as a biopharmaceutical manufacturing hub amid onshoring pressures and tariff uncertainty by reframing government as an enabling partner rather than a source of regulatory drag. The strategy emphasizes public–private ecosystems anchored by NIIMBL, specialized infrastructure to bridge molecule-to-scale challenges, and integrated capabilities spanning precision materials science, analytics, and next-generation modalities. A long-horizon workforce plan starts in high school with synthetic biology training and extends through a new in-state medical school. Competitive advantage is framed around permitting speed, refundable R&D credits, and cost-effective proximity to major markets.

At CPHI Americas, Delaware governor Matt Meyer outlined the state's biopharma vision: fostering workforce talent, industry collaboration, and regulatory speed to market.