The Impact of Disposables on Project Economics in a New Antibody Plant: A Case Study

ABSTRACT

This paper compares two project alternatives for the construction of a new 100 kg antibody plant, based on a real case. The chosen hybrid approach, which integrates disposable and stainless steel technologies, is compared with an equivalent stainless-steel model, in terms of capital investment, operating costs, and net present value.

Disposable technology has gained importance in recent years. One of its significant advantages is that it reduces the initial capital burden and spreads out capital costs more evenly over the life of a manufacturing project.¹ A fully disposable plant is still a technological dream, however. That concept has some potential applications in small-scale production, or in pilot plants for process development, but seems less likely to have a significant impact on the current paradigm for large-scale biotech plants that have fermentation capacity greater than 100,000 L.²

(Tom Grill/Getty Images)

There is less doubt nowadays, however, that a combination of disposable and reusable equipment is beneficial for process economics in both new and existing manufacturing plants. Such hybrid models can be applied with different degrees of disposable integration, and that degree generally depends mainly on process scale and the biotech manufacturers' experience with disposable technology. Mid-size plants will probably benefit the most from the use of this hybrid model, but data from real case studies are not yet available to validate this hypothesis.

The Center of Molecular Immunology (CIM), in Havana, Cuba, has built a strong pipeline of new antibodies and cancer vaccines. Some of these products are in early research, and others are in clinical development or commercial distribution. Several years ago, to meet this growing demand to produce therapeutic grade antibodies, CIM began the design and construction of a new plant with an annual capacity of 100 kg.

CIM has integrated disposable technologies into its biopharmaceutical processes for more than 10 years, and the use of disposable technology has been a key element in CIM's approach to process innovation, cost control, and regulatory robustness. Current applications include plastic culture ware and disposable systems for formulation operations and medium handling up to 2,500 L. Based on this practical experience, a new antibody active pharmaceutical ingredient (API) production plant was designed to integrate disposable technology to the extent that it has been validated in our processes. This approach was chosen to maximize our antibody production capacity in a context of limited capital availability.

In contrast to most large facility investment projects carried out by biotech companies, CIM's manufacturing project is based on continuous perfusion culture rather than the more widespread fed-batch fermentation. Because of the higher productivity of perfusion, production levels similar to those of the fed-batch mode can be achieved with smaller fermentation volumes. However, the high medium turnover required for the continuous operation of perfusion requires the use of several medium and harvest storage vessels for each production fermenter. A large-scale perfusion operation also involves almost daily purification runs (usually hundreds of lots per year) with intensive buffer preparation and storage operations. In our case, all this led to a smaller, but more sophisticated, manufacturing facility compared to a typical biotech plant.

This paper discusses the economic comparison of two project alternatives for the construction of a new 100 kg antibody plant, based on a real case. The antibody plant, with its significant integration of disposable technology, will be referred to as the "hybrid project" (HYB) and will be compared with an equivalent stainless-steel (SS) project based on reusable equipment. The two project alternatives will be compared in terms of capital investment, operating costs, and net present value.

THE NEW ANTIBODY PLANT

The basic production process used in the new plant is very similar to standard reported antibody production schemes. Cell culture and fermentation are carried out in a continuous perfusion mode that lasts for several months, with harvests being collected at regular intervals. The purification scheme starts with a capture step based on Protein A chromatography, followed by several chromatography steps for contaminant removal and product polishing. Adequate viral reduction is achieved by viral (nano) filtration as well as by viral inactivation steps along with the chromatographic operations.

The plant was designed with two segregated production trains, which each have 2,000 L of fermentation capacity. This setup allows two products to be manufactured simultaneously. The layout was organized to facilitate liquid handling in bags at large scale, based on experience with our current process. Cell culture medium and buffers are stored in a central service corridor, from where all the process rooms are fed through a closed liquid distribution system (Figure 1). This open space allows for flexible allocation of movable pallet tanks with bag containers up to 2,500 L. Larger bags are not available, however, so fixed stainless-steel tanks are used for volumes of 3,000 L and greater. A significant reduction of highly controlled cleanroom space was achieved with this approach.

Figure 1

Liquid storage in bags accounts for the most extended use of disposable elements in the hybrid facility. Table 1 describes the level of substitution of stainless steel tanks by disposable containers in the various steps of the manufacturing process in this model. Disposable bags replaced 73% of the total liquid storage capacity of the plant, estimated at 117 m³ and distributed over more than a hundred process vessels. Other disposable elements were also introduced, such as chromatographic membranes and disposable viral filtration, but those had a limited effect on equipment cost savings.

Table 1. Vessel types used in the stainless steel (SS) and hybrid (HYB) models.

ESTIMATE OF CAPITAL INVESTMENT

Figure 2 shows the total capital costs for constructing a 100 kg antibody plant under each project alternative. This calculation started with a detailed estimate of equipment purchase costs (PC) based on recent price quotes. Prices were corrected for volume and currency depreciation where needed. To avoid significant distortions of the results because of land prices and construction costs in Cuba, the total capital costs were calculated using Lang coefficients, as updated by Petrides for the biotech sector.^3,4 This method makes it possible to calculate the various items of the capital cost based on the equipment purchase cost using several multipliers (or coefficients). The use of these coefficients for disposable and stainless-steel technology has been discussed in the past by various authors.^1,4

Figure 2

After making the calculations for our case study using the combination of the selected Lang coefficients, we found that direct fixed capital was seven times greater than the equipment purchase cost for the stainless steel project alternative, but only five times greater in the hybrid alternative, similar to the estimates obtained by Farid.¹ These results correspond to our experience that a significant reduction in facility complexity is achieved when disposable technology is used extensively throughout the manufacturing process.

As can be seen in Figure 2, equipment purchase costs in the hybrid project were 34% lower than in the stainless steel project, which in turn reduced the total capital cost by 54% compared to the stainless steel alternative. This result demonstrates the significant capital savings that can be achieved through a broad integration of disposables in a medium-sized biopharmaceutical plant.

ESTIMATE OF OPERATING COSTS

To calculate operating costs based on raw materials and consumables, a detailed mass and energy balance evaluation was carried out to estimate the consumption of chemicals, culture medium, disposable materials, and pharmaceutical water for both technology alternatives, according to a manufacturing schedule of 75% plant capacity usage involving more than 200 purification cycles. Figure 3 shows the profile of daily pharmaceutical water consumption over 90 days. This analysis was used to estimate utility costs and the capacity of water production equipment. As can be observed, the daily water consumption in the hybrid project was significantly lower than in the stainless steel alternative, as a result of much lower demand for clean-in-place/steam-in-place (CIP/SIP) operations after 90 process vessels were replaced with disposable bags (as shown in Table 1).

Figure 3

Table 2 shows the cost of various elements included in the calculation of the total annual expenses for both projects. Because the breakdown of costs in biopharmaceutical production operations is often different from the breakdown in other industries, the second column in this table shows percentage values of cost items reported elsewhere, for comparison with our calculations.⁴ In our case study, the cost of raw materials and consumables were obtained separately for the two technology alternatives from the mass balances and adjusted to account for the prices of each item. As expected, the hybrid alternative had a 19% higher cost of consumables and raw materials compared to the stainless steel project.

Table 2. Estimated operating costs for both project alternatives.

To simplify the comparative analysis, the operating costs for labor, quality control/quality assurance (QC/QA), and waste treatment were assumed to be the same for both project alternatives because in previous studies we conducted, the use of disposable materials did not have a significant effect on these costs. Utility costs, however, were 50% lower in the hybrid alternative than in the stainless steel model, as a result of a 57% decrease in water and steam demand.

Equipment-dependent costs have been reported to be one of the largest cost elements in biopharmaceutical production because of the highly sophisticated nature of biopharmaceutical manufacturing facilities.⁴ Our estimate showed this cost item to be 30% of annual manufacturing expenses for the stainless steel alternative, which is roughly in the middle of previously reported ranges of 10–70%. In contrast, the integration of disposable technology in the hybrid project reduced equipment-related costs to 18% of annual operating expenses.

Overall, the data in Table 2 show that for this plant, with an annual production volume of 75 kg, the use of the disposable technology lowered total annual operating expenses by 16% compared to the stainless steel alternative. This represents a reduction in the cost of goods from $768/g to $647/g.

ECONOMIC EVALUATION OF PROJECT ALTERNATIVES

To compare technologies, several profitability indicators can be used. In addition to the commonly used net present value (NPV), other indicators that can be used include internal rate of return (IRR), discounted break-even point (DBEP) and discounted interest-recovery period (DIRP).⁵ As pointed out by Sinclair, NPV can be used effectively to estimate the long term economic benefits of introducing disposable technology in the context of a project alternative evaluation.⁶

Figure 4 shows the evolution of NPV of the hybrid and stainless steel investment projects. For this calculation, the start-up time, market penetration, financial terms, and product selling price were all assumed to be the same for both models. The introduction of disposable elements reduced the time required to recover the investment by almost two years. If calculated over a ten year period, the hybrid project alternative showed an NPV almost US $92 million higher than the stainless steel project, because of the lower capital expenses and operating costs.

Figure 4

CONCLUSION

This case study demonstrates that integrating disposable and stainless steel technology in a hybrid model has economic benefits. Using this approach in a 100 kg antibody production plant yielded significant savings in capital expenses, a key factor that may determine the feasibility of an investment project for a small- or medium-sized company. The economic model also showed lower operating costs for the hybrid project, compared to a project based on stainless steel, mainly as a result of lower equipment-related costs and reduced demand for pharmaceutical water. The net present value of the hybrid project, calculated over 10 years, was 68% higher than a similar project based on stainless steel technology.

ACKNOWLEDGEMENTS

We would like to thank Danilo Ponassi of Sartorius Stedim Biotech for his significant contribution to the introduction of disposable technology in CIM processes over more than 10 years.

José M. País Chanfrau is the plant design engineer, Katia Zorrilla is a chemical engineer, and Ernesto Chico is the technical director, all in the Technical Direction department at the Center of Molecular Immunology, Havana, Cuba, +53 7 271 3357, chico@cim.sld.cu

REFERENCES

1. Farid SS, Washbrook J, Titchener-Hooker NJ. Decision-support tool for assessing biomanufacturing strategies under uncertainty: Stainless steel versus disposable equipment for clinical trial material preparation. Biotechnol Prog. 2005;21:486–97.

2. Darby N. Trends in biological manufacturing: How will our industry change in the next 10 years? 2nd Annual Bioprocess Technology Seminar & Exhibition, Europe; 2008 June 23–25; Stockholm, Sweden. Available from http://www.asmeconferences.org/BioprocessEurope08/ASMEDarby.pdf

3. Lang HJ. Simplified approach to preliminary cost estimates. Chem Eng. 1948;55:112–113.

4. Petrides D. Bioprocess design and economics. In Harrison RG, Todd PW, Rudge SR, Petrides D, editors. Bioseparations science and engineering. London: Oxford University Press; 2000.

5. Holland FA, Wilkinson JK. Process Economics. In Perry R, Green DW, Maloney JO, editors. Perry´s chemical engineers' handbook. 7th ed. New York: McGraw-Hill Co., Inc.; 1999.

6. Sinclair A, Monge M. How to evaluate the cost impact of using disposables in biomanufacturing. Biopharm Int. 2008; 1(6):26–30.