There is new evidence that cold storage systems used for clinical products and samples is at risk from poorly designed, installed or maintained refrigeration systems. Here's what to do about it.
Traxx sensors on a laboratory freezer. Credit: KLATU
Over the past 5—10 years, the life sciences industry has significantly upgraded its standards for transporting research and commercial materials between labs, investigator sites and commercial depots or warehouses. This activity culminated, for the moment, in the Good Distribution Practices (GDP) documents originating in the European Union; these are now becoming a global best practice. 
Now it is time to say, “We know what is happening to products in transit from Point A to Point B. But what is happening at Points A and B?” In other words, are the life sciences industries able to monitor conditions in their laboratory and scale-up facilities adequately to maintain the quality and integrity of the highly valuable materials they are working with?
An infamous incident occurred two years ago at laboratories associated with Harvard University’s research facilities. Human brains that constituted a decades-long repository of research materials for Autism lost their refrigerated storage over a weekend, resulting in damaged or destroyed tissue samples.  The dollar value of this loss is incalculable; it might have set Autism research back by years.
Research and site surveying by our firm indicates that this loss, while unusually prominent in terms of value and consequences, is not a one-of-its-kind occurrence. Refrigerated storage at both industrial and academic laboratory settings are frequently running close to failure. The management response, generally, is to have multiple backups in case a storage unit fails—and this precaution has rescued upsets in many instances. But there is an unnecessary cost, both in capital equipment and in the intense power demands of laboratory refrigeration systems that the industry is paying. This should not be the standard practice going forward.
The situation is further complicated by the inability of clinical operations managers to isolate the cause of spoiled trial materials. In a typical scenario, a sample arrives from the field but is found to be adulterated, necessitating its discard. Trial managers build in a safety factor of 5—10% in the operation of a trial, in part to account for such accidents. Perhaps the clinical trial manager looks to the specialty logistics firm that handled the shipment for an explanation; perhaps not. But rarely is the refrigeration equipment at the field site suspected. This, too, is an unacceptable standard of practice, and when it happens rarely will the CRO know about it.
While new GxP best-practices have arrived in the management suite, little change has taken place on the shop floor in terms of the way we store and ship products and specimens in clinical trials. We accept, as if no problem exists, repeated occurrences of spoiled or damaged products upon arrival at their destination.
There is strong evidence that many product losses that occur during clinical trials are attributable to cold-chain storage equipment, such as ULTs (ultra-low temperature freezers), refrigerators and walk-in cold-rooms. Our firm’s research indicates that 40—50% of even well-maintained ULTs in the US are operating out-of-spec, wasting an average 22% of all energy consumed. An average of 12% of all ULTs exhibit signs of imminent mechanical failure even while maintaining temperature.
ULTs are an overlooked source of product integrity problems, particularly when these assets are located in foreign markets with limited visibility by US-based CRO managers. Problems thought to be due to shipping operations may instead be due to cold-storage problems at the beginning or end-points of the shipment. The power of the Internet has collapsed previous barriers of time and distance. Yet, few solutions exist that can monitor a ULT in India as easily and as cost effectively as one down the hall.
A perfect storm is brewing as clinical trials move to foreign markets that are more difficult to manage and as new product development shifts from synthetics to vaccines and cell therapies. A recent study from UPS reported that 67% of surveyed cold-chain executives cited concerns for product damage and spoilage connected with Asian operations—twice the rate of domestic and western European operations. 
The majority of GxP and research facilities monitor their cold-storage equipment using an existing building automation system. These systems are great for controlling lighting and HVAC, but they lack the database analytics and diagnostic power to detect problems in refrigeration systems before they result in equipment breakdown. Nor are most systems deployable as a cloud application—enabling monitoring of a ULT in China for example, by a quality control or logistics manager in Cambridge, MA.
It does not help that current FDA regulations on facilities equipment in laboratories is unaware of new game-changing monitoring technologies. Title 21 CFR 58.61, Subchapter A, “Equipment Design,” states:
Equipment used in the generation, measurement, or assessment of data and equipment used for facility environmental control shall be of appropriate design and adequate capacity to function according to the protocol and shall be suitably located for operation, inspection, cleaning, and maintenance. 
This is a dated standard largely unaware of new computing technologies which have already transformed many other industries from a fail-and-fix to a predict-and-prevent management philosophy.
Standard, laboratory-scale freezers and refrigerators are generally well-made, reliable products. But the conventional approach of facilities managers and R&D directors is to run the equipment until it fails, or to schedule routine maintenance which rarely can identify underperforming assets. In reality, refrigeration systems should get constant surveillance, and managers should employ predictive failure analytics. Preventive maintenance, and its corresponding technical requirement of continuous condition-based monitoring are fairly standard practices in commercial manufacturing; the value of in-process materials is too high to allow them to be damaged by equipment that suddenly loses performance.
Employing new tools that can also manage the effectiveness of the repair process makes it possible to hold equipment service and repair companies accountable for results. For lack of appropriate tools, most companies do not know that one-third of all ULT repairs are ineffective after six months. 
Condition-based monitoring of mission critical equipment in laboratories is now possible on a routine basis through use of web-connected sensors and alarms. The KLATU system, TRAXX-EKG, is one example. The system incorporates battery-powered, Wi-Fi-enabled sensors that are easy to deploy by leveraging the customers’ existing Wi-Fi network. More than six sensor inputs are monitored on each freezer to determine its state-of-health relative to its peers. The TRAXX-EKG system also has the ability to monitor the integrity of its own Wi-Fi connections; in effect, monitoring itself and thereby avoiding one problem that caused the specimen losses at Harvard; a complete communications failure.
When ULTs fail at overseas locations, the products they contain are moved, exposing them to handling risk and product losses, and it is unlikely that regional or US-based supply chain managers with responsibility for clinical trial operations will ever know or hear about it for lack of visibility and analytical software. In the end, problems will likely be seen in the scientific data and in most cases never attributed to the source event that caused the discrepancy.
To address industry’s needs, we can begin by first acknowledging that ULT-related cold-storage issues represent a serious problem in cold-chain integrity, and that it is not just a facilities department issue. We must go beyond today’s simple definition of cold-chain management that is limited in focus to shipping operations, to a broader definition that includes continuous monitoring inclusive of the cold-storage units at either end of the shipment.
It is an inconvenient truth that the vast majority of ULT failures are both predictable and avoidable. The avoided operating costs in terms of energy savings, reduced maintenance and repair costs will, in most instances, pay for a state-of-the-art monitoring system in three years or less. The avoidance of intangible costs associated with loss of products or specimens is priceless—as Harvard will no doubt acknowledge.
 Directive 2001/83/EC; accessed at http://ec.europa.eu/health/human-use/good_distribution_practice/index_en.htm
 “Freezer failure at brain bank hampers autism research,” Boston Globe, June 11, 2012.
 UPS Pain in the (Supply) Chain survey.
 KLATU research.
ABOUT THE AUTHOR
Rick Kriss is founder and president of KLATU Networks, Inc. (Poulsbo, WA; www.klatunetworks.com). Previously, he was CEO of Xsilogy, a wireless networking company that was recognized as a Frost & Sullivan winner of Most Innovative Product of the Year.