Managing Product Supply Risks

Published on: 
Pharmaceutical Technology, Pharmaceutical Technology-04-02-2012, Volume 36, Issue 4

How to use geographic diversification and legacy technology transfers to avoid product shortages.

In 2011, there were 267 drug shortages in the US alone, up from 178 in 2010; if unchecked, even more shortages could occur in the coming years (1, 2). In response to this situation, President Obama issued an Executive Order in October 2011 that directs FDA and the pharmaceutical industry to take appropriate steps to prevent and reduce drug shortages (3). Patients who rely on these products, especially drugs that prevent life-threatening conditions, are seriously affected by the shortages. In addition, disruption in product supply can negatively impact a company's revenue stream, share price, and reputation. Bio/pharmaceutical companies must seriously consider how they would react to unforeseen problems that could disrupt product supply. The ultimate goal is to have a robust product supply chain and to ensure that product reaches the patients safely every time.

Factors that cause product shortages

Several factors can cause product shortages, including:

  • regulatory nonapprovals

  • contamination (e.g., viral, mycoplasma)

  • natural disasters

  • raw material issues.

Regulatory nonapprovals due to major failings in quality systems (e.g., consent decree) can result in a manufacturer being unable to meet product demand, as occurred in 2010 and 2011 with Johnson & Johnson McNeil Tylenol (4). More recently, shortages occurred with two critical cancer therapies, methotrexate and Doxil, due to substantial product quality and safety issues at Ben Venue Laboratories (2).

Multiple instances of contamination have been reported in recent years, such as the vesivirus contamination at Genzyme's Allston facility in 2009 and Merck's porcine circovirus (PCV1) contamination in 2010 (5, 6). The former resulted in shortage of supply of Fabrazyme and Cerezyme to patients, which undoubtedly impacted Genzyme's reputation, and also resulted in a loss in market share to a competitor's product (Shire's Replagal), (7).

Natural disasters such as tsunamis, earthquakes, and fires can have catastrophic effects as well. For example, the earthquake and tsunami that hit Japan in March 2011 and the subsequent crisis at the Fukushima nuclear power station resulted in concerns about irradiation of critical raw materials that are used in drug substance manufactruing.

There have been increasing reports of serious issues with raw materials, including cases of product adulteration, counterfeiting, and a general lack of control in the raw material supply chain (e.g., heparin contaminated with an oversulfated form of chondroitin sulfate), (8, 9).

Dealing with product supply risks

With limited resources of people, money, time, and facilities, it is not possible to mitigate all risks. A key first step to addressing risks is to comprehensively identify and document risks, and rank them based on the severity of the disruption and its probability of occurrence. How and when a company addresses these risks depends on the company's tolerance for risk. Critical or high risks should be addressed, while the low or medium risks could be accepted and dealt with in a reactionary mode if and when they occur. Although many of these mitigation projects may have a negative return on investment, they should be considered as the insurance premium that a company is willing to pay to mitigate the risk.


Once the decision is made to deal with a risk, the team can consider several mitigation strategies, as shown in Figure 1. These are:

  • infrastructure

  • preventive and analytical technology

  • inventory control

  • diversification.

Figure 1: Four strategies can be used to mitigate product supply risk. (ALL FIGURES ARE COURTESY OF THE AUTHOR)

Infrastructure can be designed to mitigate the effects of natural disaster. For example, buildings can be constructed with seismic controls or fire-proofing.

Technology can be used to help prevent the risk of adventitious agents (e.g., viruses, mycoplasma, or prions). For example, media can be treated with high-temperature, short-time (HTST) or ultra-violet C (UVC) radiation or viral filtration, or a manufacturer can use raw materials that are animal-free, recombinant, or gamma-irradiated. In addition, analytical methods for early and rapid detection should be considered. Analytical tools that enable early detection (e.g., quantitative polymerase chain reaction assays for mycoplasma/viral contamination) might allow a manufacturer to isolate the problem and take corrective action before the contamination spreads through a facility.

Using inventory control, the combined shelf-life of drug substance intermediate (DSI), drug substance (DS) and drug product (DP) can be leveraged to minimize product shortfalls. By maintaining sufficient levels of DSI and DS, the impact of a sudden DP shortage could be mitigated by converting the DSI into DS, and then forward processing to DP. In addition, multisite storage of the DSI, DS and DP should be considered to avoid a stock-out situation in case of a disruption at the storage facility or with the transportation lanes. However, this needs to be balanced with the cost of inventory.

Finally, diversification by implementing backup manufacturing is an effective mitigation strategy. Diversification can be achieved using an in-house manufacturing network, partnership, or contract manufacturing organization (CMO) manufacturing capability. Recent industry surveys suggest that microbial and mammalian manufacturing networks are not fully utilized, which shows that diversification remains a viable option for companies to pursue to ensure robust product supply. Manufacturers should consider which products should be diversified and when this should be implemented, whether at late-stage clinical development or postcommercial approval of the product. Before employing diversification, a company should first consider whether the impact of disruption can be mitigated effectively using other options and if the company has the resources required for the desired technology transfers.

Planning a legacy molecule transfer

Once a company makes the decision to transfer a legacy or mature process to a backup manufacturing site, it should consider the key aspects of the process illustrated in Figure 2. A successful technology transfer must have a successful conformance campaign (also referred to as process performance qualification or process validation lots), in which the company demonstrates process consistency and product comparability and gains regulatory licensure. The first and foundational strategy is thus to carefully scope process and method changes and improvements that might affect the success of the conformance campaign. It is generally accepted that the primary and backup manufacturing sites should have the same process chemistry, and that changes should be minimized to avoid process drift. Any process improvements should be deferred to after implementation of the site transfer, and even then should be done in concert between the primary and backup sites to ensure process alignment. Ideally, the process scale should also be kept the same, but in some instance this may be challenging based on facility and equipment constraints and a desire to minimize the capital outlay for such a site-transfer project. Method changes are another concern. If a company decides to use this as an opportunity to remediate and modernize methods, it should allocate sufficient resources and time for these activities. The newer methods may be more sensitive and may reveal the presence of impurities, which, once identified, should be characterized. Care should be taken in changing specifications or process controls, as this could make the site transfer more challenging.

Figure 2: Several aspects must be considered in the legacy molecule transfer process.

The company should consider several key questions that may impact the scope, cost and timeline for implementing the project and establishing engineering design objectives:

  • Will the secondary facility be used for regular commercial production or purely as a backup?

  • How many lots (i.e., engineering runs) will be required to shake-down the process?

  • How will the process be validated?

  • What is needed to maintain commercial licensure status?

Finally, the company should consider the regulatory landscape. Typically, site transfers are submitted using a prior approval supplement (PAS). Alternate regulatory strategies, such as a comparability protocol or Changes-Being-Effected-30, may be considered, depending on the complexity of the transfer and scope of potential changes. Bundling changes, such as method remediation or specification changes, with the site transfer could increase the complexity and make first-time approval more difficult to obtain. Legacy molecule transfers must deal with the challenge of staying current with evolving regulatory, quality, and operating standards.

The following section describes in more detail a few of the major challenges of legacy molecule transfers and strategies to deal with them.

Challenges for mature molecules transfers

Transferring legacy molecules and establishing backup manufacturing capability as part of a geographic diversification strategy to avoid product disruption raises several challenges.

Design objectives. The strategy for how the backup site will be used impacts the engineering design objectives. A company may decide to use both the main and backup site for production, or use the backup site only in an emergency. For example, tax incentives or other financial considerations may favor the use of the primary site. In this instance, the backup site would need to run a limited number of runs as part of the technology transfer to confirm that the product and process operates in a comparable manner to the primary site. The backup site would also need to perform a few runs every 3 to 4 years, or as needed, to maintain the regulatory license. However, in the event of a disruption at the primary manufacturing site, the backup site would need to ramp up very quickly and be able to maintain the product supply to meet product demand. This emergency scenario requires fast run-rates and ramp-up times. In addition, while it may be most efficient to run longer campaigns, the need to meet product demand may necessitate shorter campaigns with more product or host changeovers. Manufacturers should consider these factors when making design decisions, such as equipment sharing or dedication, facility layout, air handling, personnel and material flows, use of clean-in-place equipment, and level of automation. A clear guideline is critical for implementing a design that balances the emergency and regulatory readiness scenarios.

Process characterization. Legacy molecules may have less extensive characterization data than do current molecules that have adopted the design space concept in the QbD approach. The manufacturer can use historical data, however, from both past characterization studies and historical manufacturing experience. Good knowledge management systems, by which the techology transfer teams have easy access to protocols, reports, laboratory notebooks, process monitoring summary reports, campaign summary reports, and annual product reviews are important. The team should learn from deviations and failed lots at the primary site to understand process sensitivities and identify where improved controls are required for robustness. The team may need to make minor changes due to facility constraints. In addition, supplemental characterization studies may be needed to understand the impact of the change on process performance and product quality. These studies should be carefully designed because they could potentially implicate the existing process design at the primary manufacturing site.

Process changes. As mentioned, some process changes may be required due to equipment or facility constraints. The "copy and paste" model (i.e., the identical process scale, process chemistry, and procedures are implemented at the receiving site) can be used to avoid changes that significantly impact the process or product. A rigorous and systematic process for making changes will minimize the risk of failure. First, teams at both the sending and receiving site should identify potential differences or gaps through a "facility fit" or "virtual batch" exercise, in which they review in detail how the process would operate at the backup facility as compared to the primary site. The teams rate the differences based on potential impact to product quality and process (i.e., severity) and the likelihood of occurrence. This risk assessment can take the form of a preliminary hazard analysis (PHA) or failure-mode-and-effects-analysis (FMEA). Once the risks are identified, the team needs to actively manage and mitigate these risks. The team should document any actions taken to reduce the risk level, and review any new risks identified during the technology transfer. Any changes made should have clear reasons and preferential changes should be avoided. Changes based on facility/equipment constraints should be considered against options for additional investment or characterization studies. A key question is when to introduce process improvements. We suggest that the first order of business is to run the secondary site's process as closely as possible to that of the primary manufacturing site. Once the secondary site has been successfully established, process improvements could be made at both sites in tandem, thus avoiding potential process drift problems. When changes are required, teams can leverage generic data, and also generate supplemental characterization data at the bench, including wet testing at scale prior to the cGMP runs.

Meeting current regulatory/validation standards. Another potential challenge in the transfer of legacy molecules is meeting current quality and operating standards at the secondary site, such as expectations for viral clearance and segregation. While the primary facility may have been "grandfathered-in" at the time of the original licensure, today's standards require separation between pre- and post-viral areas and at least one step that provides major viral reduction. Newly licensed sites must also adopt current validation requirements, as published in the FDA's 2011 updated guidance on the lifecycle approach to process validation. While the original validation package may not have included statistical approaches to justify the level of sampling or the number of validation lots, for example, the new validation process must comply.

Establishing statistical process controls. Continual process verification (i.e., process monitoring) and use of statistical process controls (SPC) provide an early indication of process trends and ensure continued control of a robust commercial process. Traditional methods for establishing meaningful statistical control limits, however, typically require greater than 20 runs. If only a few runs are performed every 3 to 4 years at a secondary site, it could take 10 years to establish control limits. Since this is obviously unacceptable to regulatory agencies, an alternative approach is to consider establishing control limits using the combined data set from the primary (main) and secondary (backup) manufacturing sites.

Meeting inspection requirements. At a secondary site, there may be only be a limited window for regulatory inspectors to see the process in operation. The Executive Order issued by the President, however, does require FDA to work with companies to expedite approvals and avoid product shortfalls. It is hoped that some relief may be granted with respect to pre-approval inspections. FDA could adopt, for example, an approach similar to that used by Australia's Therapeutic Goods Adminstration with GMP certification of a facility. This approach would allow the licensure of the facility for different products to be focused on the supplemental product filings related to the site transfer. Regulatory inspection requirements can potentially result in an increase in resources required and hence on the cost and timeline for the successful implementation of the backup facility. Additionally, the team should carefully think through the impact of inspection requirements on the already licensed primary facility.


Systematically identifying risks to product supply and taking the necessary steps to mitigate these risks allows a company to proactively avoid product shortages. Multiple options to mitigate risk are available. Diversification by establishing backup manufacturing facilities in geographically distinct areas is one strategic approach to avoid product shortages in the event of an emergency situation at the primary manufacturing site. By carefully laying out the strategy for and potential challenges of conducting legacy technology transfers, companies can successfully implement diversification. It remains to be seen how companies and regulatory agencies such as FDA can work together to move forward in preventing and avoiding such product shortages and thus ensuring safe treatments for patients.


The author wishes to express sincere thanks to Arleen Paulino, Bill Keefe, Dan Rathman (Amgen), Ali Siahpush (Dendreon), and David Clark (MedImmune) for their active support and insightful comments in preparation of this paper.

Sushil Abraham is director of technology transfer at MedImmune, 636 Research Drive, Frederick, MD 21703, tel. 301.228.5183, At time of acceptance of this paper, the author was at Amgen, 4000 Nelson Road, Longmont, CO 80503.


1. J. Wechsler, BioPharm Intl., 24 (11), 12–14 (2011).

2. A. Federgruen, The Wall Street Jrnl., Mar. 2, 2011.

3. The White House, "Executive Order 13588–Reducing Prescription Drug Shortages," Press Release (Washington, DC, Oct. 31, 2011).

4. Johnson and Johnson, "Johnson & Johnson Provides Update on McNeil Consumer Healthcare Remediation; Announces Completion of Internal Assessment Phase of Comprehensive Action Plan," Press Release (New Brunswick, NJ, Jan. 14, 2011).

5. L. Bush, BioPharm Intl Bulletin, June 17, 2009.

6. M. Wiebe, presentation at West Coast Chapter of the PDA meeting (June 17, 2010).

7. B. Obsoj-Cardwell, The Pharma Letter online,, July 7, 2010.

8. J. Woodcock, MD, Statement, House Committee on Energy and Commerce Hearing, Apr. 29, 2008.

9. FDA, Information on Heparin,, accessed March 5, 2012.