Introduction
To support supply chain continuity, we recently doubled our liquid media production capacity through an expansion of an existing site in Logan, UT. Both the existing facility and the new liquid media facility, designated “A1X,” are ISO 13485:2016 and 9001:2015 certified and governed by Cytiva’s Quality Management Systems. The A1X expansion effectively doubles this site’s capacity.
The added liquid capacity comes from the addition of three manifold fill lines, three filling manifolds, six mixing tanks, six formulation booths, and a utility building to support large volume liquid media production. Housed in the utility building are a 45,000-liter tank and process water system, a 55,000-liter tank and water for injection system, a clean steam generator, and additional supporting utilities. In addition to the added capacity, we’ve updated several aspects of the manufacturing floor layout and equipment, which will shorten production cycle time, improve safety, and minimize product risk. These updates also establish closed systems for cleaning and a controlled environment for the transport and handling of raw materials and finished goods (Table 1).
Scope of comparison study
Cell culture media variations can have a host of unintended and unwanted effects on processes. To understand and control for such variations in media produced in the two facilities, we analyzed the equipment used in each and compared two types of representative liquid media from each. Note that we used this study to demonstrate comparability between the two facilities; it was not used for facility qualification or validation.
We used a risk-based approach to identify the following primary risks associated with the commissioning of the A1X expansion.
- New product contact material, which may impact product performance
- New equipment for addition of components to the mixing tank, which may impact shear, aeration, and foaming-sensitive components.
Table 1. Comparison of existing and A1X facility features
| Feature | Existing | A1X |
| Cleanroom classification (CNC) | CNC, ISO-5,6,7 | ISO-5,6,7 |
| WFI temperature control | 80ºC, with point of use chilling system | 80ºC, with point of use chilling system |
| Manufactured product classification | Animal derived (AD), animal-derived component free (ADCF) | ADCF only |
| Mix tank composition | SS316L | SS316L and AL6XN SS alloy |
| Powder mixing equipment | Tri-blenders and jet injector | Jet injectors |
| Powder addition equipment | Calibrated weigh-down system | Calibrated weigh-down system |
| Clean out of place (COP) solution preparation | COP sink with manual NaOH cleaning | Dosed, automated NaOH cleaning system/cycles |
| Clean in place (CIP) | Manual formulation in CIP transfer tank | Citric acid/NaOH dosed, direct injection |
| Steaming or sanitization in place (SIP) | Steam at 124ºC for >45 min | Steam at 124ºC for >45 min |
| Batch sizes | 100–10,000 L | 700–13,000 L |
| Packaging capabilities | Bottles, bags, ALLpaQ Genesis bins | Bottles, bags, ALLpaQ Genesis bins |
| Finished goods handling | Dolly and barrel lifter | Gravity-fed roller conveyor system |
| Bag sealing | Radio frequency (RF) sealed bioprocess containers | RF sealed bioprocess containers |
| Filling process | Fill by weight | Fill by weight |
| RM handling/transport | Indoor/outdoor material transfer | Indoor controlled environment material transfer |
New product contact
The A1X facility uses mixing tanks and liquid media transfer lines comprised of AL6XN and 316L stainless steel. This differs from the existing facility equipment, which is comprised solely of 316L stainless steel. AL6XN is a low-carbon, high-purity stainless-steel alloy that is more resistant to wear and corrosion than 316L, representing an upgrade to the product contact layer versus the existing facility equipment. Table 2 highlights the composition differences between the two materials.
Table 2. Differences in composition of 316L vs AL6XN stainless steel
| Element | Amount in 316L (max % or range) (1) | Amount in AL6XN (max % range) (2) |
| Manganese | 0.0–2.00 | 2.00 |
| Chromium | 16.0–18.0 | 20.00–22.00 |
| Nickel | 10.00–14.00 | 23.50–25.50 |
| Molybdenum | 2.00–3.00 | 6.00–7.00 |
| Copper | NA | 0.75 |
| Iron | Balance (62.8–71.5) | Balance (41.4–46.5) |
An investigation was conducted to identify changes in trace metal profile of water for injection (WFI) due to exposure to AL6XN vs 316L. We mixed WFI from production in 4000 L tanks and piping comprised of AL6XN or 316L for 20 h, then analyzed samples for trace metals via inductively coupled plasma-mass spectrometry (ICP-MS) (Figure 1).
Fig 1. Trace metal concentration measured in water exposed to product contact after 20 h.
The trace metal levels detected in each WFI sample reflected the metal composition of each tank. However, when compared to the trace metal concentrations in finished goods, these differences were negligible—measuring approximately two orders of magnitude lower for the same analytes. This result suggests that trace metals from AL6XN pose little risk to the final product.
Manufactured media comparison
New equipment for component addition
We also implemented new equipment for component addition in A1X. Liquid media is typically formulated using dry powdered media and individual wet and dry components. The A1X facility utilizes a jet injector system to add components to the mixing tanks. This system uses the Venturi effect to produce a homogeneous powder-liquid mixture for liquid media manufacture. The existing facility uses tri-blender technology, where an impeller creates a centrifugal force to dissolve powder. We analyzed the impact on physical, chemical, and biological performance of media produced in A1X compared to the existing facility.
Media selection for facility comparison
We chose one complex basal medium and one concentrated feed medium to assess functional equivalence: ActiPro™ medium and Cell Boost™ feed supplement 7A. ActiPro medium is chemically defined, chosen because its complex formulation lends itself to analytical characterization via quantitation of amino acids, vitamins, and trace metals. We used Cell Boost supplement 7A because its highly concentrated formulation is conducive to comparing the two facilities’ component mixing and addition methods and equipment. We selected these samples as they’re often used in challenging manufacturing conditions incorporating highly complex operations, thus simulating a “worst case” scenario in terms of manufacturing variations. We evaluated both the basal and feed media in cell culture and assessed for cell growth, productivity, and product quality.
Chemical evaluation
Basal medium and feed supplement samples (taken at the beginning, middle, and end of the fill process) manufactured in both facilities met all QC criteria for the standard products, including appearance, pH, osmolality, and endotoxin (Tables 3 and 4). Glucose in the feed supplement was within ±20% across all conditions tested (Table 4).
Table 3. Appearance, pH, osmolality, and endotoxin results for basal medium samples from the existing facility and A1X.
| Facility | Sample | Appearance | pH | Osmolality (mOsm/kg) | Endotoxin (EU/mL) |
| Existing | Beginning | Clear | 7.28 | 310 | ≤1.00 |
| Existing | Middle | Clear | 7.29 | 310 | ≤1.00 |
| Existing | End | Clear | 7.29 | 310 | ≤1.00 |
| A1X | Beginning | Clear | 7.38 | 299 | ≤1.00 |
| A1X | Middle | Clear | 7.36 | 304 | ≤1.00 |
| A1X | End | Clear | 7.40 | 306 | ≤1.00 |
| Specification limits: | Clear | 6.75-7.55 | 270-330 | ≤1.00 |
Table 4. pH, osmolality, and glucose results for Cell Boost 7A samples: existing facility and A1X batches
| Facility | Sample | pH | Osmolality (mOsm/kg) | Glucose (g/L) |
| Existing | Beginning | 6.73 | 281 | 68.2 |
| Existing | Middle | 6.73 | 289 | 77.3 |
| Existing | End | 6.73 | 287 | 73.7 |
| A1X | Beginning | 6.69 | 279 | 67.4 |
| A1X | Middle | 6.67 | 284 | 73.0 |
| A1X | End | 6.66 | 277 | 69.0 |
| Specification limits: | 6.45-6.80 | 247-303 | 61.6-92.4 (expected concentration) |
We measured amino acid, vitamin, and trace metal concentrations in basal medium and feed supplement samples (beginning, middle, end) manufactured in each facility to demonstrate process and product comparability. Amino acid concentrations across all samples tested were within ±20% of expected recovery, with no substantial differences between products manufactured in A1X vs the existing facility (Figures 2, 3). Vitamin concentrations for both basal medium and feed supplement produced in A1X were within ±20% of vitamin concentrations of batches produced in the existing facility, or within the normal method tolerances (Figures 4, 5). Finally, no substantial differences were found in the trace metal profiles of basal medium and feed supplement produced between the two manufacturing facilities (Figures 6, 7).
Fig 2. Amino acid concentrations for basal medium samples produced in the existing facility and A1X. QC specification is 80–120% recovery.
Fig 3. Amino acid concentration for feed supplement samples produced in the existing facility and A1X. QC specification is 80–120% recovery.
Fig 4. Vitamin concentration for basal medium samples: existing facility and A1X. Data are shown relative to the averages for samples manufactured in the existing facility.
Fig 5. Vitamin concentration for feed supplement samples: existing facility and A1X. Data are shown relative to the averages for samples manufactured in the existing facility.
Fig 6. Trace metal yield for basal medium samples: existing facility and A1X. Samples were tested by ICP-MS from the beginning, middle, and end of product fill.
Fig 7. Trace metal yield for feed supplement samples: existing facilities and A1X. Metals 5–8 were not detectable by ICP-MS. Samples were tested from the beginning, middle, and end of product fill.
Biological evaluation
We compared the cell culture performance of basal medium and feed supplement samples (beginning, middle, and end) manufactured in A1X to analogous samples manufactured in the existing facility. The growth and viability profiles were comparable, following similar trends during log, stationary, and decline phases of growth (Figures 8–11). The integral of viable cell density (IVCD), productivity, and protein product quality profiles (charge variant, aggregation, and glycosylation) were consistent between the facilities; among beginning, middle, and end samples; and with a finished goods control lot and an equivalent dry powdered media lot hydrated for comparison (Figures 12–21). (The finished goods control lot was previously produced in the existing facility. The equivalent dry powdered media lot was a lot prepared at 1 L scale with the same dry powdered media lot used to manufacture the media in the A1X and existing facilities.)
Fig 8. CHO adaptation passage viable cell density (VCD) and terminal growth profile for basal medium samples: existing facility and A1X.
Fig 9. CHO growth profile in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture.
Fig 10. CHO viability profile in basal medium across three adaptation passages and a terminal growth curve A: existing facility and A1X.
Fig 11. CHO viability profile in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture.
Fig 12. Final IVCD of CHO cells grown in basal medium samples: existing facility and A1X. Cells were fully adapted into their respective basal medium condition over three passages prior to the final terminal growth curve.
Fig 13. Final IVCD of CHO cells growth in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture.
Fig 14. Day 8 batch IgG titer for CHO cells in basal medium samples: existing facility and A1X. Cells were fully adapted into their respective basal medium condition over three passages prior to the final terminal growth curve and titer collection.
Fig 15. Day 14 fed-batch IgG titer for CHO cells in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture and titer collection.
Fig 16. Protein charge variant profile across basal medium samples: existing facility and A1X. Cells were fully adapted into their respective basal medium condition over three passages prior to the final terminal growth curve and titer collection.
Fig 17. Protein quality aggregation profile across basal medium samples: existing facility and A1X. Cells were fully adapted into their respective basal medium condition over three passages prior to the final terminal growth curve and titer collection.
Fig 18. Protein quality glycosylation profile across basal medium samples: existing facility and A1X. Cells were fully adapted into their respective basal medium condition over three passages prior to the final terminal growth curve and titer collection.
Fig 19. Protein quality charge variant profile of CHO cells in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture and titer collection.
Fig 20. Protein quality aggregation profile of CHO cells in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture and titer collection.
Fig 21. Protein quality glycosylation profile of CHO cells in a single basal medium lot supplemented with feed supplement samples produced in existing and A1X facilities. Cells were fully adapted into basal medium over three passages prior to the fed-batch culture and titer collection.
Summary
This study demonstrates physical, chemical, and functional parity of liquid media manufactured in two facilities in Logan, UT. The ADCF liquid media expansion facility (A1X) has larger mixing tanks than the existing facility, supporting batch sizes up to 13 000 L. Several tanks in the new expansion facility are made of a different stainless-steel alloy, and our evaluation indicates that this new alloy does not adversely impact product quality or integrity during contact. We also compared samples of complex, chemically defined media and feed formulations manufactured in each facility to assess any differences in performance or composition. For this, we used ActiPro™ basal medium and Cell Boost™ feed supplement 7A, and found that they were comparable across physical, chemical, and biological tests. All media produced in A1X were within the expected variations for all testing performed. The demonstrated comparability of these media formulations produced at both the A1X and existing facilities provides the critical assurance needed to fully leverage the expanded batch size and plant capacity for liquid media manufacturing – reinforcing Cytiva’s unwavering commitment to reliability, instilling confidence in our customers, and ensuring consistent, high-quality supply for every step of their cell culture journey.
References
Summary report: Appendix A
Regarding: Logan, UT liquid media expansion facility high-concentrate comparability study
Introduction
To further demonstrate the full capabilities of the Cytiva A1X liquid media (LM) expansion facility, Cytiva conducted an additional comparability study using a high‑concentration glucose product. The objective was to expand upon the previous evaluation by assessing the performance and consistency of high‑concentrate solutions manufactured at the Logan LM expansion site. High‑concentration products are a key component of the Cytiva manufacturing portfolio, and 60% glucose presents a particularly rigorous test case due to its substantially higher solids content and dissolution demands compared to typical high‑concentrate feeds. This study provides additional confirmation of the facility’s ability to reliably produce complex, high‑density solutions.
Scope of glucose comparison study
This study demonstrates comparability between high concentration glucose liquid media products produced in the existing manufacturing facility vs the A1X expansion facility. The study described herein was designed only to evaluate the mixing capabilities of high concentrate products between the two facilities and was not used for facility qualification or validation.
Expansion facility manufacturing comparison evaluation
The 60% glucose samples (beginning, middle, and end) manufactured in both the existing manufacturing facility and in the A1X expansion facility were within ± 2% of the expected concentration (Fig A1).
Summary
This study confirms the physical equivalence of high‑density liquid solutions produced in the newly expanded A1X facility compared with those manufactured at the established Logan, UT site. A 60% glucose solution was selected as a representative high‑density formulation commonly used in biologics manufacturing to demonstrate functional comparability. Concentration measurements showed no meaningful differences between the two facilities. The demonstrated consistency of these high‑concentration solutions across both A1X and the existing manufacturing facility provides the assurance needed to fully leverage increased batch sizes and expanded plant capacity for liquid media production. This result reinforces the commitment of Cytiva to reliability, strengthens customer confidence, and supports a dependable, high‑quality supply throughout the cell-culture workflow.
Fig A1. Measured glucose concentrations of 60% glucose solutions manufactured in the A1X and existing manufacturing facilities (n = 3 for each condition)