Printer Friendly

Developing cell therapies: enabling cost prediction by value systems modeling to manage developmental risk.

INTRODUCTION

The field of cell therapy is rapidly developing, and many clinical trials have been initiated exploring the use of stem/progenitor cells in the treatment of degenerative diseases and cancer and for the repair of damaged or lost tissues. Cell therapies represent an emerging and rapidly developing industry with a unique opportunity to contribute to both health and economic wealth. As the industry has developed from experimental research to commercial growth over 500 businesses have been established to push cell therapies through to clinical development adoption. (1) In spite of this only a small number of cell therapy products have made it to market with the vast majority of companies still engaged in preclinical and early clinical development. The ability of companies to transition their therapies to market will depend on a successful ability to manage risk and cost while creating value for all the stakeholders involved in the health-care marketplace. A key issue in boosting commercial success rates involves creating the tools to develop an evidence based approach to picking commercial winners. of technologies and processes used in the production of all therapeutic products clinical use. (6) Therapies seeking adoption in the US healthcare market are regulated by the US Food and Drug Administration which dictates when a developer must transition from non-GMP (Good Manufacturing Practise) environments to GMP-validated production during clinical development. Because a phase 1 clinical trial initially introduces an investigational new drug into human subjects, appropriate GMP help ensure subject safety. (7)

This paper is focused on providing analytical and simulation models that assist in the prediction of two of the main drivers of cell therapy product development, cost and risk. We present a value system model that has been assembled to predict the development cost and lead time associated with the clinical and process development activities involved in moving from preclinical testing to completion of Phase III clinical trials.

CELL THERAPIES RISK MANAGEMENT

Risks to the development and market adoption of cell therapies depend on many risk factors. A common approach to risk assessment and management is to understand the probability of the risk occurring and the impact of the harm caused by its manifestation. (2) A risk factor or hazard is defined as a potential source of harm. (3) Developing cell therapies face two types of risk factor;

1. Product risk factors: risks that can harm the patient--including the type of stem cells used, their source, the level of manipulation applied to them and method of use and delivery mechanism. Product risks inform the basis of the regulatory framework that cell therapies must be developed and delivered under4 as has been discussed in the literature. (5)

2. Enterprise risk factors: risks that can harm the introduction of the product and the business that develops it--include financial risks such as cost overruns, market risk, technical risks associated with developing a manufacturing platform and temporal risks associated with completing product development. Products must also be brought to market in a timescale that investors and developers can tolerate.

A thorough evaluation of enterprise and product risk factors, along with their consequences, at the start and during the development of a stem cell based therapy may help to determine the extent and focus of the product development and safety evaluation plans. The differing nature of the two classes of risk defines the amount of effort developers with limited resource will allocate to their management or reduction. The regulatory sys-tem cell therapies are developed within rightly dictates that clinical evidence surrounding safety and efficacy is collected by pre-clinical and clinical trials. In addition regulatory authorities demand stringent validation

How developers choose to conduct these product risk management processes will ultimately influence the enterprise risk however the regulatory system does not account of effects on resource caused by the system. Regulators cannot take account of the difficulty in developing a product over the risks that such a product may pose to a patient population or the cost of developing medicinal products.

THE CELL THERAPY VALUE SYSTEM

It is necessary to include the requirements of developers, investors, healthcare providers and patients along with regulators to gain a true understanding of the enterprise risk associated with cell therapies. These groups represent the actors within the cell therapy value system. We define a value system as the representation of the various activities, actors and resources that are involved in delivering goods (and services) to a market. (8) Resources employed include time, capital, infrastructure and personnel. Actors include, but are not limited to, developers, regulatory authorities, investors, healthcare providers and patients. An overview of the whole value system can be treated as a level of analysis below innovation systems, which often view innovation systems, which often view innovation through the lens of a national, regional or industrial level innovation system (9), (10) as it is centred on individual product markets. How a developing therapy navigates this value system influences when costs are committed into a product (for example when a manufacturing facility is built) and relates cost to business development and value creation. As therapies progress through the value system they will ideally increase in value to all stakeholders, including patients, investors and healthcare systems while having a decreasing level of enterprise and product risk.

One method of adding value to any early stage technology based enterprise is risk reduction by either reducing product or enterprise risk by providing more information relating to risk factors to the value system actors. (11) As outlined above product risk may be reduced by accomplishing a significant process development step (12) or moving through preclinical and clinical trials to demonstrate product safety, utility and efficacy. Enterprise risk may be reduced by the developer proving more evidence surrounding return on investment (ROI) to an investor or shareholder. The extent of the increase in value is sensitive to the amount of information that will accrue (or uncertainty that will be reduced) during development. While the regulatory and scientific communities have provided extensive research and requirements surrounding product risk reduction strategies there is a limited amount of research concentrated on reduction of cell therapy enterprise risk.

This work focuses on the reduction of enterprise risk by prediction of the value cost and price associated with developing cell therapies. This is driven by the need to understand the economics of a product early in the development process. Several recent studies have drawn attention to the increasing need for of early-stage economic modelling for medical products while acknowledging the uncertainties and difficulties intrinsic in such a enterprise. (13), (14)

The timely application of economic evaluation in the product development process can provide the manufacturer with a significant amount of useful information, not just on the future economic viability of their new product. (15), (16) Traditionally, new technologies have been evaluated at market launch, as a one off exercise by decision makers to decide whether to purchase or invest in a new technology. Developers and investors need to be able to identify candidate therapies with the best clinical and commercial potential and communicate their value to potential investors and the healthcare system ideally before significant investment decisions. As the health services continue to develop robust health economic appraisal methods, developers have started to look at their technologies in the same critical way as healthcare decision makers in order to make better investment decisions. (14) Some proposals envisage on-going health economic assessment as an integral part of the development cycle. (13)

As the final commercial success of a proposed product will be largely determined by its rate of adoption which is informed by its cost-effectiveness, it is sensible to conduct such an analysis at the outset. While an early assessment may be limited in the accuracy of information it can provide regarding exact cost or price the analysis will help guide developmental targets in terms of product development timeframes, cost and clinical effectiveness goals. The predicament when it comes to the assessment of any innovative medical technology in early stage development is that the available evidence of clinical effectiveness is still lacking or only available to a very limited extent.

By conducting predictive modelling of price and cost at early stage development, when final effectiveness is unknown, and at key stages throughout product development, predictions about the probability of the product being sufficiently affordable can be established and could prove significant in persuading healthcare systems, patients and investors of its value. (17)

To calculate the potential value of a therapy to investors or healthcare systems, three key numbers must be considered; Value, Cost and Price. The difference between cost and price will dictate the potential return on investment by a therapy, i.e. the value to an investor who must judge this against the risk in developing a new cell therapy.

A method has already been presented for scoping the gross commercial opportunity (or "headroom") by establishing a simple price ceiling available to a developer based on an estimate of clinical effectiveness within a cost-utility model. (18) The aim of this work was to provide a quick method for rapid decision-making that would support, for instance, the selection of promising concepts from a larger pool of options. The drawbacks to the "headroom" method are that it is only applicable to healthcare systems where cost effectiveness is measured using the QUALY (Quality adjusted life years) model and does not provide a method to estimate the potential cost of a cell therapy or medical device. The headroom method can be viewed as price appraisal method. What is needed is a range of companion models for the supply side issues surrounding cost and risk.

LINKING CELL THERAPY DEVELOPMENT AND COST

The total cost of developing, marketing, manufacturing and delivering a cell therapy to a patient will dictate the final Cost of Goods Supplied (COGS). At the early stages of technology development--when sometimes even the nature of the product is unknown--realistic estimates of cost are difficult. Significant technical and financial uncertainty surrounding the product, its manufacturing system and its supply and business model exists. Product and manufacturing system based cost drivers can be identified as likely to be lowered either through technology improvements such as automation or through economies of scale. Understanding all cost drivers allows developers to identify areas for savings. However unanticipated costs of developing cell therapies have the potential to drive the development cost substantially higher than forecast.

VALUE SYSTEMS MODELING FOR CELL THERAPY

For cell therapies, like all medicinal products, the path to a marketed product involves a long and exhaustive journey through basic research, discovery of a therapeutically effective cell type, preclinical development tests, process development, increasingly complicated clinical trials and regulatory approval. Several years and significant financial investment is needed to undertake this process.

As a result critical decisions are often made with imperfect information. This can result in the need to redevelop or "rework" parts of the cell therapy development processes causing an iteration of enterprise activity. An example of this would be the need to redo some non-clinical and clinical testing following a change in manufacturing system, process or input. Iteration is a fundamental characteristic of complex and highly regulated product development projects. (19) However, cost predication techniques that rely on past experience or heuristics have very limited capabilities in coping with iterations. The majority of process modelling literature and software is oriented toward production or business processes, where the goal is to repeat a chain-like process without interwoven iterative loops. Shortcomings of standard flowchart presentations of development processes in clearly representing many feedbacks are seldom exposed. However, much of the waste and inefficiency in iterative development processes stems from these interactions and feedbacks--i.e., having to repeat activities because of changes in the information and/or assumptions upon which they were initially executed, or an increase or change in the regulatory environment.

A fruitful way to increase understanding of a process is to look at its parts and their relationships. Decomposition is a possible approach to addressing sys-tem complexity--it is generally possible to make more accurate estimates about simpler elements within the system. However, it is generally more difficult to make accurate estimates of the effects on the overall system of relationships between simpler elements. Similarly "bottom up" production orientated cost models often rely on activity based costing that requires a large amount of historic or product specific information to be availably a characteristic that limits its usefulness in maturing industries such as cell therapy. The relationships among elements are an important characteristic that differentiates a system from a mere grouping of elements. As a value system, the development of cell therapies products are defined not only by their decomposition into activities but also by how they interact together. (20), (21)

In practice, most product development definitions and models account for a minimal amount of information about the element relationships or interfaces. A single input and output for each activity is often considered sufficient. However, especially in the early stages of product development, people and the activities they execute tend to provide and require a great deal of information to and from each other. (22) A large number of activity interfaces are necessary to document the full range of information flow and dependency. Most process models do not attempt to elicit and represent the actual information flow, even though it is a major driver of product development competence and predictability. (23)

Steward (24) developed the design structure matrix (DSM) method for such purposes. The DSM provides a compact representation of a complex system by showing information dependencies in a square matrix. The DSM method is based on the earlier work in large-scale system decomposition. Eppinger et al. (19) extended Steward's work by explicitly modelling information coupling among tasks and investigating different strategies for managing task procedures. Some researchers have used the DSM framework to design iteration modelling to extend its information-based structuring analysis to schedule analysis. (20) Work by Browning (11) shows its increasing use in various application areas including product development, project planning, project management, systems engineering, and organization design in other highly regulated industries such as aerospace.

A design structure matrix (DSM) can be used to represent a process such as product development. The DSM shows activities and interfaces in a concise for-mat. A DSM is a square matrix in which a cell on the diagonal represents each activity. Activity names are usually given to the left of the matrix. A mark in an off-diagonal cell indicates an activity interface. For each activity, its row shows its inputs and its column shows its outputs. When activities are listed in temporal order, super-diagonal marks denote a feeding of deliverables forward in the process, from upstream activities to downstream activities, while sub-diagonal marks indicate feedback. The DSM provides a simple way to visualize the structure of an activity network and to compare alternative process architectures.

We present in this paper the application of a DSM-based simulation model, building on work by Cho et al. (25) that illustrates how model-based design process analysis may be used as an early stage assessment tool applicable to development cost prediction for a cell therapy product.

MODEL CONSTRUCTS

We follow the information-based view (26) of design projects in which a task is the information-processing unit that receives information from other tasks and trans-forms it into new information to be passed on to subsequent tasks. The information exchanged between tasks includes both tangible and intangible types such as materials, documentation, learning, etc. Model inputs characterize behaviours of individual tasks and interactions among the tasks from a schedule perspective. The duration of a task is used to model uncertainty and complexity within the domain of the task. Precedence and resource constraints determine the start times of tasks. Iterations are modelled to depict the patterns of workflows caused by dynamic information exchanges among the tasks.

In order to build such a rich process model, we employ numerical simulation methods. Simulation techniques are effective for the two analytical purposes: sampling of task duration from the known distribution function and modeming of the dynamic progress of a project. We employ the parallel discrete-event simulation method for modelling the progress of a project as a dynamic system, where system variables evolve over time. Note that modelling non-Markovian transitions is impossible to represent as a Markov chain.

1. Task durations

A variety of distributions have been used to represent stochasticity of task duration. This model chooses the triangular probability distribution to represent task durations since this distribution is simple and familiar to many project managers. (27) For each task, the model receives three estimated values for the expected duration of one-time execution--optimistic, most likely and pessimistic. These values represent the duration of a task from the start to the end of its continuous work, even though the task may later be repeated after its initial completion. Remaining duration decreases over time as the model simulates the project's progress.

The model uses the Latin Hypercube Sampling (LHS) method28 to incorporate the uncertainty of the expected duration of each task based on the three estimated durations. The LHS method divides the range between them into n strata of equal marginal probability, where n is the number of random values for the expected duration rep-resenting the triangular probability distribution function. Then, it randomly samples once from each stratum and sequences the sampled values randomly.

2. Precedence constraints

From a schedule perspective, we consider two types of information flow in a task: 1) information flow at the beginning or at the end of the task and 2) information flow in the middle of the task. Accordingly, we define two types of information flow between two tasks. The first type represents the case that the task requires final output information from the upstream task to begin its work. The second type represents the case that the task uses final output information from the upstream task in the middle of its process or begins with preliminary information but also receives a final update from the upstream task.

The first type of information flow is translated to a "finish-to-start" precedence constraint between two tasks, while the second type is translated to a "finish-to-start-plus-lead" constraint. With lead time, two tasks are overlapped so that a successor task starts before a predecessor task is finished.

3. Resource constraints

The model assumes that there exists a fixed, renewable resource pool throughout the entire project duration. It consists of specialized resources and/or resource groups of which constituents exhibit the same functional performance. Each task has its own resource requirement which is assumed to be constant over the entire period the task is processed. The resource requirement for the costing model is represented as a "cash-burn" associated with each specific activity.

4. Iteration

Iteration is defined as the repetition of tasks to improve an evolving development process. It is generally accepted that iteration improves the quality of a product in a design project while increasing development time. Managers must control the project to address this time-quality tradeoff. (15) In this paper, iteration is the rework of a task caused by the execution of other tasks. This definition excludes any repetitive work within a single task's execution (that being considered within the variance in the task's duration contained within the task distribution function). This includes all planned and unplanned iterations that can be modelled probabilistically. Some unplanned iterations cannot be considered because they result in structural changes to the project. For example, a major project failure or addition of different activities imposed by the regulator would involve re-structuring the entire process, not simply reworking the established tasks.

An event is defined as the completion of an active task instead of any information transfer. Thus, when any active task in the current state is completed, the model makes a transition to the next state. The duration of state is defined as the minimum remaining duration of active tasks in the state. Before making a transition to the next state, the model subtracts the duration of the current state from the remaining durations of all active tasks. If all the remaining durations of tasks are zero (the termination condition), one simulation run is complete and the lead time is calculated as the sum of all the state durations. The cumulative cost of the completion of all tasks at the end of the simulation run is calculated by the sum of all the products of individual task duration and cash burn level. After all simulation runs are complete, the probability distribution of lead time and cost can be constructed.

CASE STUDIES OF CELL THERAPY COMPANY DEVELOPME0NT

Creation of the value systems model required additional information surrounding development costs and time-frames that could not be extracted from the literature. These were needed to provide the initial triangular probability function outlined above and define a cell therapy new product development process to model. Case studies of four cell therapy companies were compiled by recording their historic stock values and outstanding share levels. Company newsflow in the form of press releases, annual reports and analyst coverage were examined to determine key points in the product development process and company development. Instances of financing by licensing agreements, stock offerings and private investment were recorded and examined to determine the strategies adopted by cell therapy companies in financing development and value creation activities. In order to assess the commercial valuation and financial records of these organisations it was necessary to con-fine the companies studied to those listed on a US stock exchange. This allowed for access to publically avail-able financial information filled with the Unites States Securities and Exchange Commission (SEC).

Company value was measured using the market capitalization of each organisation. Market capitalization (market cap) is a measurement of size of a business enterprise and is equal to the share price times the number of shares outstanding of a publicly traded company. As owning stock represents ownership of the company, including all its equity, market capitalization represents company's net worth.

This value was plotted alongside historic market capitalisation to determine if they had influence on the publically perceived value of each company. This study focused on four companies: Two developing allogeneic therapies and two developing autologous treatments. All are using cell types or products that can be targeted against multiple indications. All companies selected where using adult derived stem cells. This remove any influence US public policy on embryonic stem cell research has on the study.

A cross-case analysis was performed to search for patterns and themes that cut across the individual cases. Results revealed large amounts of NPD rework or iterative development undertaken within the companies studied. A distinctive feature of the cell therapy NPD process is the importance of adherence to regulatory frameworks that dictate the order of clinical and process development milestones. As a result any rework or iterations of tasks that place within tasks during NPD potentially required the rework of tasks both preceding and subsequent to the task that causes the iteration.

Results from the case studies allowed collection of data for development programs surrounding both "Orphan" and "Non-orphan" cell therapies. Orphan therapies refer to therapies with a much narrower market segment resulting in lower numbers of patients recruited to clinical trial activities and possibly higher market prices if the target indication has significant unmet clinical need.

APPLICATION OF VALUE SYSTEMS MODEL TO CELL THERAPY CASE STUDY

The results of the case studies allowed construction of a candidate new product development process for cell therapies (Figure 1). The process has eight tasks, seven feed forward dependencies and thirteen feedback dependencies. This process has been illustrated using input data from both Orphan and Non-Orphan cell therapy development case studies. The structuring of the tasks was directed by rework loops and iteration observed in the companies studied. The case studies highlighted the feed-forward and feedback dependencies and iteration loops experienced by cell therapy companies.

The case studies also provide triangular probability distributions of the duration of the NPD tasks and monthly "cash burn" levels associated with each development task (See Figure 3), allowing for estimation of development cost within the model. The triangular distributions of duration and cash burn levels were developed from financial reports of the four companies and normalised for company headcount and patient recruitment levels in clinical trials. The rework probabilities and impact factors are shown in Figure 2. The inputted task durations and cash burn levels differed for the Orphan and Non-orphan development pathways. The number of simulation runs was kept high due to the large probability distributions for time and cost--to ensure that the sampled task durations closely follow the inputted triangular distributions.


Dependancies

Name                    1  2  3  4  5  6  7  8

Pre-clinical '       1     X

Process Development  2  X     X

Investment           3     X     X

Phase 1              4  X  X  X     X

Scale-up (1-2)       5     X     X     X

Phase 2              6           X  X     X

Scale Up (2-3)       7     X           X     X

Phase 3              8                 X  X

Rework Proabilities

Name                     1    2    3    4    5    6    7   8

Pre-clinical '       1

Process Development  2            0.9

Investment           3       0.1         1

Phase 1              4  0.1  0.1  0.1         1

Scale-up (1-2)       5       0.1       0.1         1

Phase 2              6                 0.1  0.1         1

Scale Up (2-3)       7       0.1                 0.1       1

Phase 3              8                           0.1  0.1

Rework Impacts

Name                     1    2    3    4    5    6    7    8

Pre-clinical '       1       0.5

Process Development  2  0.5       0.9

Investment           3       0.5       0.9

Phase 1              4  0.5  0.5  0.5       0.9

Scale-up (1-2)       5       0.5       0.5       0.9

Phase 2              6                 0.5  0.5       0.9

Scale Up (2-3)       7       0.5                 0.5       0.9

Phase 3              8                           0.5  0.5

Figure 2: Design structure matrix, rework probability matrix
and rework impact matrix for cell therapy new product development
process

Input Data - Non-Orphan

                        Durati ons

   Name                     Min     Likely  Max  Learn  $k/Month

1  Pre-clinical '               12      16   24    0.3     428.2
2  Process Development          10      16   20    0.5     440.7
3  Investment                    1       3    6    0.9     333.2
4  Phase 1                       8      10   12    0.9     578.7
5  Scale-up (1-2)                2       3    6    0.5     618.7
6  Phase 2                       9      10   12    0.5     784.3
7  Scale Up (2-3)                1       5    9    0.5     708.3
8  Phase 3                      10      24   38      1    1520.7

Input Data - Orphan
                        Durati ons

                            Min     Likely  Max  Learn  $k/Month

l  Pre-clinical '               12      16   24    0.3     435.7
2  Process Development          12      16   24    0.5     398.7
3  Investment                    1       3    6    0.9     295.6
4  Phase 1                       8      10   12    0.9     458.6
5  Scale-up (1-2)                1       3    6    0.5     618.7
6  Phase 2                      18      20   22    0.5     641.3
7  Scale Up (2-3)                1       5    9    0.5     708.3
8  Phase 3                      12      24   36      1    1208.3

Figure 3: Triangle probability function and cash burn rates for
cell therapy new product development model


As with Soo-Haeng Cho, 2005 (25) the computer pro-gram was written in Visual Basic and subsequently added into a Microsoft Excel 2011 spreadsheet which simplifies model input and control and is used to display analysis results. Extensive numerical experimentation was undertaken to test the simulation program and validate the initial results. Small scale test scenarios were run on individual simulation runs to validate the model code. The input data collected from the case study work outlines above was inputted and ran over 10,000 simulation runs.

RESULTS--APPLICATION OF VALUE SYSTEMS MODELLING TO CELL THERAPY CASE STUDY

The 10,000 model runs for each scenario, orphan and non-orphan produced a frequency distribution of both cost and time required to complete the NPD process. This allow a cumulate probability curve to be drawn that marks the probability of the process completing within a given duration or cost. For a desired probability of completing the NPD process this allows a cost or duration to be generated as seen in Figure 4.


            Probability       20%         50%        80%
            of success

Non-Orphan  Duration      122 Months      155  204 Months
                                       Months

            Cost             $146.4M  $176.6M     $227.5M

Orphan      Duration       114Months      143  191 Months
                                       Months

            Cost             $128.0M  $157.6M   $203.8.5M

            [DELTA]         8 Months       12   13 Months
            Duration                   Months

            [DELTA] Cost      $18.4M     $19M      $27.3M

            Probability     99%
            of success

Non-Orphan  Duration           351
                            Months

            Cost             $365M

Orphan      Duration           338
                            Months

            Cost             $319M

            [DELTA]       13Months
            Duration

            [DELTA] Cost      $46M

Figure 4: The probability of completing the NPD process
"success" is expressed against cumulative cost and duration
for Acute Myocardial Infarction when developed under orphan
and non-orphan processes


The frequency distributions in Figures 5-8 illustrate the frequency distribution of completed simulation runs and the results and duration and costs for each process. Figure 6 summarises the expected costs and durations from the accompanying cumulative probability curves. These results illustrate the leas time (duration) and cost incurred in taking a product from start of pre-clinical research to completion of Phase III clinical trials for a given probability.

This level of investment and duration--while significant--aligns with the current timescales and investment levels seen in the cell therapy community and current expenditure recorded in the input case studies. The probability distribution of the lead time and cost shown in Figures 4 and Figure 5 is skewed to the right because the lead time and cash burn becomes larger as more iteration loops occur and probabilistic sampling will lead to a small number of scenarios with multiple cases of large iteration loops.

Due to the subjective nature of interpreting the rework and impact probabilities associated with the cell therapy case studies and transferring these into the model framework additional work was undertaken to assess the impact of changing the rework probability on overall duration. Rework probability was varied for each of the thirteen feedback loops from 10% to 70%.

CONCLUSION

There are two key conclusions of this paper.

1. The model presented here should be developed to form part of a larger structured framework that aids in the segregation and estimation of COGS and price for cell therapies early in the development cycle. To develop a comprehensive understanding of the factors that impact cost of goods supplied (COGS) for cell therapies a developer must understand how cost is influences by the entire value system surrounding a cell therapy. Use of the developed framework simulation model can guide this process. Overall, the model provides a framework in which to examine the impacts of a variety of effects on process cost, duration, and risk--yielding several important decision making capabilities. Plus, the basic model is extensible toward providing additional realism, analyses, and insights. Organizations developing new products will benefit especially from being able to illustrate to investors that their cell therapy product development process has an acceptable or at least quantified level of risk.

2. The value systems model accounts for a number of PD process characteristics, including interdependency, iteration, uncertain activity cost and duration, rework probability and impact. The model is used to explore the effects of varying the process risk distribution. This highlights that securing early stage investment is crucial for developing cell therapy companies. It also highlights how critical process development (for the product) is as rework of process development requires rework of clinical trials--with the associated duration and cost penalty. These critical risk points are unlikely to change due to the structure of the cell therapy NPD being dictated by regulatory requirements. The level of potential cost gains is also highlighted in the analytical model presented at the end of the paper and highlights how early decision support tools can highlight areas for high cost saving.

The simulation model provides a tool to assist informed discussion and projection of development task cost and duration including concurrency, iteration and rework, and can take account of learning. Results of the use of the simulation program can be used to compare the relative merits of alternative development and manufacturing strategies and the associated impacts on time to market, cash burn and return on investment. Current limitations of the value system model include reliance on case study input data and a limited resolution view of the development process which limits the information of specific risks that can be highlighted.

The DSM approach discussed in this paper represents an activity based view of the development process. The activities relate to each other as shown in Figure 1. This architecture has a large influence on the appropriate structure of the product development organization as each activity will require different types and levels of organisational resource since organizational elements are typically assigned to develop various product components. This established development architecture can constrain the consideration of alternative product development strategies. The development architecture and product development strategy relationship can affect an enterprise in several dimensions. Better understanding the relationship between product architectures and organization structures is a promising area for further research which may highlight more effect methods of brining cell therapies to market as the industry develops. DSMs will prove helpful in comparing and contrasting development architecture and product development strategy configurations.

The structure of a cell therapy product offering--including manufacturing considerations, supply chain constraints, regulatory approval route--affects how a development process can and should be configured. That is, the product offering structure determines the process (activity) structure. If separate design activities develop separate but coupled aspects of this offering, as in cell therapy, then the need for these activities to exchange information should be noted when designing the design process. It would be interesting to contrast how established NPD processes deal with novel product development when contrasted with new development processes that may take a change in regulatory environment to approve. Again, the DSM can be a useful tool in such research provided adequate input information is available.

Future work will move to collect a higher resolution view of the activities within each development step and will use accepted costs and timescales where possible--for instance regulatory authorities now specific the time that certain regulatory approval steps take.

Increased understanding of the underlying development processes and their interaction with enterprise risk will help develop more efficient development processes for cellular therapies.

REFERENCES

(1.) Parson AB. Stem Cell Biotech: Seeking a Piece of the Action. Cell 2008 02/22;132(4):511-513.

(2.) ISO/IEC Guide 51:1999.

(3.) ICH Q9: Quality Risk Management.

(4.) Redmond K. The US and European regulatory systems: a comparison. J Ambul Care Manage 2004;27(2):105-114.

(5.) Herberts C, Kwa M, Hermsen H. Risk factors in the development of stem cell therapy. Journal of Translational Medicine 2011;9(1):29.

(6.) Burger S. Current regulatory issues in cell and tissue therapy. Cytotherapy 2003;5(4):289-298.

(7.) Department of Health and Human Services, U.S. Guidance for Industry CGMP for Phase 1 Investigational Drugs. 2008.

(8.) Hergert M, Morris D. Accounting data for value chain analysis. Strategic Manage J 1989;10(2):175-188.

(9.) Rampersad G, Quester P, Troshani I. Managing innovation networks: Exploratory evidence from ICT, biotechnology and nanotechnology networks. Industrial Marketing Management 2010 7;39(5):793-805.

(10.) Senker J. National systems of innovation, organizational learning and industrial biotechnology. Technovation 1996 5;16(5):219-229.

(11.) Browning TR, Deyst JJ, Eppinger SD, Whitney DE. Adding value in product development by creating information and reducing risk. Engineering Management, IEEE Transactions on 2002;49(4):443-458.

(12.) Hourd PC, Williams DJ. Results from an exploratory study to identify the factors that contribute to success for UK medical device small-and medium-sized enterprises. Proceedings of the Institution of Mechanical Engineers Part H-Journal of Engineering in Medicine 2008;222(H5):717-735.

(13.) Vallejo-Torres L, Steuten L, Buxton M, Girling AJ, Lilford RJ, Young T. Integrating health economics modelling in the product development cycle of medical devices: a Bayesian approach, International Journal of Technology Assessment in Health Care 2008; 24 (4):459-64

(14.) Girling A, Young T, Brown C, Lilford R. Early-stage valuation of medical devices: the role of developmental uncertainty. Value Health 2010; 13(5): 585-59

(15.) Pietzscha J, Pate-Cornella M. Early technology assessment of new medical devices. International Journal of Technology Assessment in Health Care 2008; 24(1): 36-44

(16.) Shemer, J., I. Abadi-Korek, and A. Seifan, Medical Technology Management: Bridging the Gap between Theroy and Practice Israeli Medical Association Journal 2005. 7: p. 211-215.

(17.) Archer R, Williams DJ. Why tissue engineering needs process engineering. Nat Biotechnol 2005 NOV;23(11):1353-1355.

(18.) McAteer H, Cosh E, Freeman G, Pandit A, Wood P, Lilford R. Cost-effectiveness analysis at the development phase of a potential health technology: examples based on tissue engineering of bladder and urethra. Journal of Tissue Engineering and Regenerative Medicine 2007;1(5):343-349.

(19.) Eppinger SD. A model-based method for organizing tasks in product development. 1993(3569-93.).

(20.) Browning TR, Eppinger SD. Modeling impacts of process architecture on cost and schedule risk in product development. Engineering Management, IEEE Transactions on 2002;49(4):428-442.

(21.) Burt JM, Garman MB. Conditional Monte Carlo: A Simulation Technique for Stochastic Network Analysis. Management Science 1971 November 01;18(3):207-217.

(22.) Cooke P, Gomez Uranga M, Etxebarria G. Regional innovation systems: Institutional and organisational dimensions. Research Policy 1997 12;26(4-5):475-491.

(23.) Jin Y, Levitt RE, Kunz JC, Christiansen TR. The Virtual Design Team: A Computer Simulation Framework for Studying Organizational Aspects of Concurrent Design. Simulation 1995 March 01;64(3):160-174.

(24.) Steward DV. Partitioning and Tearing Systems of Equations. Journal of the Society for Industrial and Applied Mathematics: Series B, Numerical Analysis 1965;2(2):345-365.

(25.) Soo-Haeng Cho, Eppinger SD. A simulation-based process model for managing complex design projects. Engineering Management, IEEE Transactions on 2005;52(3):316-328.

(26.) Oloufa AA, Hosni YA, Fayez M, Axelsson P. Using DSM for modeling information flow in construction design projects. Civ Eng Environ Syst 2004 06/01; 2012/03;21(2):105-125.

(27.) Williams TM. Practical Use of Distributions in Network Analysis. J Oper Res Soc 1992 Mar.;43(3): 265-270.

(28.) Mckay MD, Beckman RJ, Conover WJ. A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code. Technometrics 2000 Feb.;42(1, Special 40th Anniversary Issue):55-61.

Received: November 11, 2012

Mark Joseph McCall is a PhD Researcher in the Doctoral Training Centre in Regenerative Medicine at Loughborough University

David John Williams is Professor of Healthcare Engineering, Wolfson School of Mechanical and Manufacturing Engineering and Director of the Research School of Health and Life Sciences at Loughborough University

Correspondence: Mark Joseph McCall, Loughborough University. Email: m.mccall@lboro.ac.uk
COPYRIGHT 2013 thinkBiotech LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Original Article
Author:McCall, Mark Joseph; Williams, David John
Publication:Journal of Commercial Biotechnology
Article Type:Report
Geographic Code:4EUUK
Date:Apr 1, 2013
Words:6426
Previous Article:The financial ecosystem available to early-stage biotechnology firms and its misalignment with interests of these firms, of the biotechnology...
Next Article:A patient centric commercial model for cancer care.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters