Production Cost Analysis: Economic Analysis as a tool for Process Development: Harvest of a High Cell-Density Fermentation

The number of biotechnology-based human therapeutic products in the late-stage pipeline, and the average cost to commercialize a biotech product, have steadily increased.^1,2 This has required biotech companies to use economic analysis as a tool during process development and for making decisions about process design. Process development efforts now aim to create processes that are economical, as well as optimal and robust.^3-6

Novais et al. recently performed an economic comparison of conventional versus disposables-based technology for the production of an antibody fragment from an E. coli fermentation.⁷ The authors concluded that the capital investment required for a disposables-based option is substantially reduced—less than 60% of that for a conventional option. The disposables-based running costs were 70% higher than those of the conventional equivalent. However, the net present value of the disposables-based plant was found to be positive and within 25% of that for the conventional plant. More recently, the economic feasibility of using disposables has been examined for facility design, highlighting the need to perform a thorough analysis for the application at hand.^8,9

Harvesting biotechnology products from cell culture or fermentation process streams is often performed by a combination of several-unit operations. Centrifugation, depth filtration, and microfiltration are commonly used. In a recent publication, different harvest approaches were investigated for a case study involving recovery of a therapeutic protein from Pichia pastoris fermentation broth.¹⁰

Quick Recap

This article, the seventh in the "Elements of Biopharmaceutical Production" series, describes how economic analysis can be used to compare different processes and assist in designing an "economical" option.

Figure 1. Schematics for options 1 and 2 that are examined in this economic analysis

BACKGROUND

Figure 1 illustrates the two options that will be examined in this economic analysis. Option 1 involves a three-unit operation harvest process: centrifugation, followed by depth filtration, and completed with a concentration and buffer exchange via tangential flow ultrafiltration–diafiltration (UF–DF). Option 2 involves a two-unit operation process: microfiltration followed by a concentration and buffer exchange via tangential flow filtration (UF–DF). Table 1 presents a comparison of process performance under the two options. Under optimal conditions, both options can deliver the desired product recovery (> 80%), harvest time (<15 hours including sequential UF–DF), and clarification (< 6 NTU). In this study, we perform a cost comparison of these options under different scenarios, presented in Table 2. Scenarios 1 and 2 involve cost estimation with new equipment but without discounting for depreciation. These scenarios describe what happens when a company needs to buy all of the processing equipment and is not likely to be able to use it for other products. An example would be a small company with a single product that is planning to make a few lots of a product for clinical trials. The use of disposables is examined in scenario 2. Scenarios 3 and 4 involve cost estimation with new equipment and with discounting for depreciation. These are more typical scenarios, where a company has several products in its pipeline, and will utilize the equipment, with possible minor modifications, for other product(s). Examples of these would be most medium and large biotech companies and contract manufacturing organizations.

Table 1. Comparison of process performance for option 1 and option 2. Adapted from reference 10.

The following assumptions were undertaken for the economic analysis:

1. Since both of the options described above incorporate a UF–DF unit operation using the same type and number of membranes, this step is not included in the cost comparison.

Table 2. Different scenarios analyzed in the economic comparison

2. Raw material costs include consumables and are obtained from vendors. It is assumed that the depth filter is single use and that the microfiltration media can be recycled for 10 cycles. The reuse number is assumed to be low as the feed material for the microfiltration step is very crude and for our application resulted in significant membrane fouling.

3. Purified water (PW), clean steam, and process air are considered utilities, and are included in the cost of operating the facility. The facility cost also includes the cost of labor and other costs that are incurred to run the facility.

4. For estimating capital cost, an annual discount rate of 10% is assumed for all equipment (assumes a 10-year equipment life).¹¹ Capital cost also includes the cost for commissioning new equipment, and validation.

RESULTS AND DISCUSSIONS

Scenario 1: As mentioned earlier, this scenario represents a situation in which a small biopharmaceutical company plans to conduct a manufacturing campaign for a single product. The equipment cost is not discounted for the depreciation. Figure 2 shows a comparison of the different cost components for the two options. The total cost is similar for the two options. The capital cost is higher for option 1, since it requires a centrifuge. However, the facility cost is higher for option 2, due to higher usage of purified water during microfiltration. Raw material costs for both options are relatively minimal.

Figure 2. Cost comparison of options 1 and 2 under scenario 1

Scenario 2: Scenario 2 is similar to scenario 1 except that it explores the use of disposable technology. The use of disposable diafiltration pods, instead of depth filters and housings, and buffer bags, instead of tanks, resulted in lower costs for equipment cleaning, commissioning, and validation. Figure 3 shows a comparison of the different cost components for the two options. The total cost for both options is approximately 10% lower than those in scenario 1. The raw material costs for option 2 are ~70% higher because of the use of disposable consumables, which is similar to what has been reported in the literature.⁷ However, the reduction in capital cost (~25%) more than makes up for this increase for this application. Disposables are an economically attractive alternative, particularly when a company wants to limit its capital investment.

Figure 3. Cost comparison of options 1 and 2 under scenario 2

Scenario 3: This scenario depicts a medium or large biopharmaceutical company that has several products in the pipeline and thus, the equipment cost in this scenario can be discounted for depreciation. Figure 4 shows a comparison of the different cost components for the two options. The capital cost for scenario 3 is drastically lower versus scenarios 1 and 2, because of the equipment costs discounted for depreciation. Further, option 1 (centrifuge + depth filtration) appears to be a more desirable option, with a 20% lower cost in comparison to option 2 (microfiltration), primarily because of the larger amount of purified water required for the microfiltration step.

Figure 4. Cost comparison of options 1 and 2 under scenario 3

Scenario 4: This scenario also assumes discounting of equipment costs for depreciation and models the cost for a high volume product (50 lots per year). Figure 5 shows that, like scenario 3, the total cost for option 1 is about 25% lower than that for option 2. The cost of raw materials and the capital cost are an insignificant portion of the total cost.

Figure 5. Cost comparison of options 1 and 2 under scenario 4

OVERALL COMPARISON

Figure 6 presents a comparison of the total cost for the four different scenarios under consideration. Several key observations can be made upon reviewing the data presented. First, option 1 (centrifuge + depth filtration) and option 2 (microfiltration) are comparable under scenarios 1 and 2. For a small company that wants to minimize the capital cost, pursuing the microfiltration route might be more attractive. The situation changes for scenarios 3 and 4, primarily because of the higher utility (purified water) costs for microfiltration; option 1 becomes significantly more attractive. Second, the use of disposables is certainly an economically viable option, especially for a smaller company that wants to minimize capital investment. Third, the total cost per lot is significantly reduced once we start discounting the equipment cost for depreciation. This is the correct way to perform an economic analysis for the cases where the equipment will be likely used by other products in the pipeline. Fourth, in our case we did not see a significant cost reduction for a high-volume product because we limited the use of microfiltration membranes to 10 cycles. However, for other applications (such as chromatography), where media can be reused for a much higher number of cycles, significant cost reduction would be expected for a high volume product.

Figure 6. Comparison of total cost per lot under the four scenarios

CONCLUSION

For the biotechnology industry to continue being profitable, it is necessary to consider economics, along with the traditional targets of process development—recovery, consistency, and product quality. Conducting a cost analysis of various process options while the process is being developed can help ensure creation of economic processes. If process performance of the different options under consideration is comparable, as in the case presented here, the optimal choice would depend on other "non-process" factors. These include the size of the company and its product pipeline, the volume of its product, and the clinical advancement of the program (clinical supplies versus commercial supplies).

While the results of the economic analysis are expected to vary with the application, the approach presented here can be useful for biochemists and engineers performing cost-analyses in the biopharmaceutical industry.

ACKNOWLEDGMENTS

The authors would like to acknowledge Oliver Kaltenbrunner and Darrell Lewis-Sandy, both from Amgen Inc., for useful discussions.

Anurag S. Rathore , PhD, is the director of process development at Amgen, One Amgen Center Dr., 30W-2-A, Thousand Oaks, CA 91320, 805.447.4491, arathore@amgen.comMatthew Karpen is a senior engineer, process development, at Amgen.

REFERENCES:

1. Agres T. Better Future for Pharma/Biotech? Drug Disc Dev. 2002 Dec 15.

2. Curling J, Baines D. The cost of chromatography. IBC conference on Production and Economics of Biopharmaceuticals. 2000 Nov 13–15; La Jolla, CA

3. Rathore AS, Levine H, Latham P, Curling J, Kaltenbrunner, O. Costing Issues in Production of Biopharmaceuticals, BioPharm Int. 2004 Feb; 17(2):46–55.

4. Ernst S, Garro OA, Winkler S, Venkataraman G, Langer R, Cooney CL, et al. Process Simulation for Recombinant Protein Production: Cost Estimation and Sensitivity Analysis for Heparinase I Expressed in Escherichia coli, Biotechnol Bioeng. 1997; 53: 575–582.

5. Sadana A, Beelaram AM. Efficiency and Economics of Bioseparation: Some Case Studies. Bioseparation. 1994; 4:221–235.

6. Warner TN, Nochumson S. Rethinking the Economics of Chromatography: New Technologies and Hidden Costs. BioPharm Int. 2003 Jan; 18(1): 58–60.

7. Novais JL, Titchener-Hookner NJ, Hoare M. Economic comparison between conventional and disposables-based technology for the production of biopharmaceuticals, Biotechnol Bioeng. 2001; 75:143–153.

8. Sinclair A, Monge M. Biomanufacturing for the 21st Century: Designing a Concept Facility Based on Single-Use Systems, BioProcess International. Part 1: 2004 Oct supp; 2–12; Part 2: 2005 Oct supp; 51–55.

9. Langer E, Ranck J. The ROI Case: Economic Justification for Disposables in Biopharmaceutical Manufacturing. BioProcess Int. 2005 Oct supp; 46–50.

10. Wang A, Lewus R, Rathore AS. Comparison of Different Options for Harvest of a Therapeutic Protein Product for High Cell Density Fermentation Broth. Biotechnol Bioeng. 2005;94:91–104.

11. Collentro A. Current Design Economics for USP Purified Water Generation Systems. Pharm Processing. 2004 Dec.