samedan logo
home > ebr > spring 2012 > storage and processing technology
European Biopharmaceutical Review

Storage and Processing Technology

A little over a decade ago the drug development activities of most major pharmaceutical companies were focused around small molecule drug discovery (SMDD); today, however, the situation is very different. Drug discovery has become much more ‘industrialised’, with the widespread use of automation and robotics. At the same time, drug discovery is bringing different technologies to the fore, with the storage and processing of biological materials becoming increasingly important when attempting to understand disease causes and treatment.

Many of the techniques and tools developed for SMDD are now finding applications in these new areas. In many cases, there are new challenges cropping up owing to the nature of the samples being handled. Three key challenges are sample storage at ultra low temperatures; the processing of inherently variable biological samples; and quality control. This article will uncover ways these challenges are being met, encompassing -20oC and -80°C storage, liquid handling, blood processing and fractionation, as well as emerging quality control technologies.

Sample Storage

In the early to mid 1990s, compound libraries in pharmaceutical companies expanded rapidly, partly as a result of the application of combinatorial chemistry techniques. This expansion was accompanied by an increase in the application of automation techniques, both for sample storage and preparation (sample management) and high throughput screening.

For the most part, these compounds were stored: neat as dry powder or gels, gums or films in glass vials under dry, ambient conditions; and solubilised in dimethylsuplfoxide (DMSO) in microtubes or plates.

Throughout the late 1990s and the early part of the new millennium, there was a significant debate at pharma conferences regarding the best conditions for storage of these solubilised samples. Storage temperature, sample concentration, storage format (consumables), water content/water uptake and freeze/thaw cycles were the subject of many studies and presentations.

The automation suppliers responded to the concerns of the industry by developing solutions which allowed storage:

  • Under a range of temperatures (generally between room temperature and -20°C)
  • Under a range of atmospheric conditions (for example dry air or inert atmospheres)
  • In different labware formats (for example, single shot storage versus multisip microtubes)
  • With integrated sample processing of varying degrees of complexity designed to minimise freeze/thaw cycles or exposure of samples to undesirable environmental conditions

Automated storage and cherry picking of various consumables at temperatures down to -20°C is now commonplace.

While these storage technologies were generally developed for SMDD operations, where -20°C is considered sufficiently low to effectively eliminate sample degradation, millions of biological samples are collected and stored every year for diagnostic and research purposes. In the case of these samples, while there is some overlap of storage temperatures, and therefore with existing technology (many organisations, for example, currently store DNA at -20°C), a significant proportion of the biosamples collected – which include not only DNA but also RNA, serum, plasma, buffy coat, cells, urine, saliva, hair, nails, tumour samples, stools, and so on – require storage under controlled conditions (typically low, ultra-low or cryogenic temperatures) to minimise degradation and maximise sample integrity.

The majority of these samples are currently stored in manual freezers at -80°C/-150°C, or under liquid nitrogen (vapour phase/liquid phase) storage conditions. Many are manually labelled, leading to concerns over sample degradation from temperature fluctuations as freezers are opened and closed to allow samples to be removed or replaced, often for extended periods as the operator searches for the required sample(s), or as non-required samples are removed to allow the requested samples to be accessed/ picked. Sample degradation from nonideal storage conditions, and sample tracking and chain of custody also pose problems, including when samples are lost or unidentifiable if labels become detached. Locating samples can also be difficult in large collections if they are misplaced within the freezer bank; simple spreadsheet solutions for sample tracking do not guarantee a robust chain of custody.

There are claims that existing practices render a significant proportion of stored biological samples unusable or unsuitable for specific research tasks, and there is a growing concern that erroneous or poor quality results are contaminating the scientific literature.

The same debate around storage conditions is now heard in the context of biobanks and biorepositories. Many of the detailed topics are familiar – storage temperature, frosting, integrity and stability – but new topics have also been added – controlled rate freezing and the glass transition temperature of water. Added to the discussions around capital expense associated with purchase of automated storage is the issue of running costs. This can partly be attributed to the awareness of ‘green issues’, but it is also partly a reflection of rising running costs with increasing ΔT (the difference between store temperature and the ambient environment). Generally in the biobanking world, the lower the storage temperature the better. But this has to be balanced against the capital expense and ongoing operating costs of the storage solution.

In principle, the techniques and solutions developed for SMDD can be applied in these applications. However, the challenge of providing robust, reliable automation in a costeffective manner has led most suppliers to offer hybrid solutions, with the samples stored at -80°C, but with the automation running at -20°C and with only passive components entering the -80°C environment to retrieve blocks of samples for output or cherry picking. Early solutions were based around one of two approaches:

  • Chest freezer accessed from above by robots
  • Vertical freezer banks either side of a central aisle robot

These technologies continue to develop, with both costs and efficiencies having improved significantly in recent years. Innovations such as the use of a moveable ‘tile wall’ to separate the -20°C and -80°C environments, while allowing the -80°C evaporator to be optimally located directly above the stored samples, combine the benefits of each of the solutions.

Sample Processing

In contrast to the high throughput sample preparation and processing demands of the SMDD environment, biorepositories and biobanks typically handle fewer samples, but of an inherently more variable nature. This presents a number of challenges for automation: robotic systems are good at doing the same thing repeatedly, usually outperforming humans in terms of throughput and accuracy, but are generally less suited to tasks where the input is unpredictable or variable. This is one reason why many sample processing activities in biobanks have remained largely manual, or at best semi-automated.

There has, however, been progress in automating a number of common processes, including DNA extraction, blood fractionation, prep for sequencing and PCR, and sample sub-aliquoting. Fractionation of blood samples is a particularly challenging example and worthy of further description. Many different types of organisation store, process and analyse blood or blood products. Fractioning blood is not required in all cases. However, for certain types of biobanks, particularly prospective biobanks, or those conducting longitudinal or genome wide association studies, fractionation can offer certain benefits.

For many biobanks, the traditional process of storing and receiving biosamples leads to:

  • Heavy ‘up front’ investment in DNA extraction, since this process has to be carried out on all samples at the time of collection
  • High DNA extraction costs owing to expensive ‘large volume’ DNA extraction kits
  • A requirement for a large number of samples per patient (if additional analysis is required) because the whole sample is used for the extraction of the DNA

In contrast, a fractionation approach (see Figure 1) is intrinsically cheaper:

  • DNA extraction from lower volume samples (buffy coat only) is less costly
  • DNA is extracted only from the samples required at the time of sample request
  • Plasma and red blood cell fractions are available for further/ subsequent analysis

Blood fractionation is a means to make the most effective use of precious blood samples. Plasma can be harvested and stored for analysis of analytes, biomarkers, and so on. Buffy coat can be recovered for immediate or future extraction of DNA, and red blood cells can also be stored, if required.

Storage of buffy coat for future DNA extraction offers distinct advantages. DNA extraction is only carried out on samples requested by researchers (some estimates claim that maybe 95 per cent of samples in a longitudional study will never be analysed). The cost of DNA extraction is deferred until samples are used; depending on throughput, this may equate to several thousand dollars of deferred expenditure per day. The lower volume required for DNA extraction also means that the cost of extraction is reduced. Of course, these benefits can only be realised if the blood samples can be reliably, cost effectively and safely fractioned.

The nature of the samples handled in biobanks and biorepositories therefore provides a different set of drivers for process automation when compared with the throughput drivers for HTS:

  • Human safety/wellbeing: handling unscreened blood samples exposes operators to health risks
  • Sample integrity: biological samples degrade at room temperature. Requesting operators to perform complex or skilled tasks at reduced temperature over extended periods is not practical
  • Reproducibility and chain of custody: the samples processed are often genuinely unique and irreplaceable (samples donated by volunteers, samples taken in hospitals or GP surgeries, and so on) and a demonstrable chain of custody and confidence that the sample has been skilfully processed are important

So, while the drivers for automation are clear, the barriers are significant. Even ‘standard’ samples (for example blood collection into a 10ml evacuated blood tube) will vary in volume (not all donors will provide the full 10ml sample), colour and viscosity (depending on the age of the sample, the health of the patient and even what they ate for breakfast). In addition to these bulk sample properties, the relative proportions and ‘locations’ of the sample components (such as plasma/serum, buffy coat and red blood cells) will also vary from tube to tube, depending on the state of health of the patient and other factors.

These variations in sample properties present a number of challenges to automation which have to be addressed by adding some ‘intelligence’ to the automation so that it may react to changing input conditions. Machine vision techniques developed for vision guided robotic packaging or end of line quality assurance (QA) applications have been harnessed to address these issues. The automated blood fractionation system illustrated is an example of such a liquid handling solution. Designed to process biological samples, the device uses vision technology to identify layers within a centrifuged blood sample. The system then uses this information to calculate the position of the various layers (and hence the volumes of the different components), then drives the liquid handling device to separate the contents of the input tube into various output cryovials or plates for downstream analysis and/or storage. Decisions regarding the number of aliquots can also be made based on the vision data, with a full chain of custody guaranteed via 1D/2D barcode read and deck audits combined with an interlocked enclosure which prevents access to the samples during processing.

Sample quality is of utmost importance. Blood samples degrade over time and, to slow this process, blood samples are often chilled immediately after collection and are then transported and stored at low temperatures to minimise sample degradation. Fractioning the samples manually generally requires processing at higher temperatures (typically room temperature, for the sake of the technicians), but some commercially available automated fractionation systems allow the whole fractionation process to be carried out in a temperature controlled +4°C enclosure, thereby minimising exposure of the sample to elevated temperatures and maximising sample quality.

Quality Control

In SMDD, complex IT systems are used to keep track of the locations and volumes of compounds in storage. These systems are, however, less than perfect, as they cannot account for the effects of evaporation, hydration and liquid handling errors on the sample volumes actually in the tube. Nor can they track sample degradation such as precipitation. Such errors have a real and measurable effect on the screening process, with a recent study indicating that on average, across both pharma compound management departments and biobanks:

  • Seventy-five per cent of samples had a recorded volume discrepancy of more than 25μl
  • Nearly 13 per cent of newly registered samples exhibited some degree of precipitation
  • Nearly five per cent of microtubes in storage were actually empty

In SMDD, the absence of a sample in a plate, either through incorrect sample volumes, or because the compound has precipitated out of solution, has a clear and direct impact on the screening process. One pharma company identified that nearly one in eight boughtin compounds had in fact precipitated out of solution on receipt, resulting in an estimated annual cost wastage in excess of $1 million.

In biobanking, the financial calculation is less clear. However, the need for QA in biobanking is certainly at least as great. In the pharma world, if a compound is missing, it is usually relatively straightforward to replace or regenerate it; in the bio world it may not be possible to go back to the original donor.

Biobanking studies can be viewed as consisting of a number of phases: recruitment, storage, supply and analysis. Depending on the type of organisation and, indeed, the type of study, these phases may run sequentially or in parallel. Furthermore, the process may be wholly carried out by the biobank or may involve a number of different departments or even different organisations. For example, biological samples may be delivered to the biobank from an external organisation for storage and analysis, or a sample within the biobank may be retrieved from storage and shipped (in whole or in part) to third parties to satisfy research requests.

Whether samples are processed wholly in-house or received from/ shipped to internal or external clients, exchange of samples without some form of quality check can ultimately lead to problems downstream when the samples are analysed, including (potentially) commercial consequences. In both application areas, there are traditionally several methods available to monitor or measure liquid levels in tubes; see Table 1 for some examples.

More recently, the use of vision and related spectroscopic techniques in this area has grown. Vision-based sample inspection systems, initially developed for small molecule drug discovery compound QA, offer the possibility of reliable, non-contact goods-in and goods-out QA checks for biorepositories. A recent audit (using this technology on one leading biobank’s DNA sample collection of over 80,000 samples) indicated that, on average, its laboratory information management system (LIMS) had overestimated the sample volume by 70μl.

Furthermore, although precipitation is not such an issue in biobanking, the technique used to detect precipitation in compound collections can be used to determine the presence (or absence) of mixing balls in DNA samples. This information can then be used to adjust the volume calculation accordingly.


Automation tools and techniques originally developed to address the throughput and logistics challenges of compound management and high throughput screening are now being adapted to solve sample integrity, chain of custody and human safety challenges in a range of related life science industries. Developments of technology and creative system designs have enabled current product solutions to be adapted to operate in new environments, while machine vision techniques have enabled highly variable processes to be automated and non-contact QA measurements to be introduced.

Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:

There are no comments in regards to this article.

Simon Sheard is Product Manager at Brooks Life Science Systems. Simon graduated with a BSc in Physics with Physical Electronics from Bath University. Remaining at Bath for his doctoral research, he gained his PhD in 1989 for work on Metallic Glasses for Pulse Compression. Since then Simon has worked in a number of industries, joining RTS (now part of Brooks) in 2001, initially as an applications consultant and subsequently in project management, sales and business development roles before taking up his current role. Simon has 25 years of experience of R&D, product development, project management and sales, with over 10 of these in life science automation applications.
Simon Sheard at Brooks
Print this page
Send to a friend
Privacy statement
News and Press Releases

Medovate announces distribution agreement with LTR Medical for sales of SAFIRA® in Australia and New Zealand

08 September 2020: Medovate, the medical device development company, has concluded a distribution agreement with Brisbane-based LTR Medical which will provide exclusivity for SAFIRA® (SAFer Injection for Regional Anaesthesia) in Australia and New Zealand.
More info >>

White Papers

Key to Outsourcing Method Development and Validation: A Pragmatic Approach


In an industry that is seeing an increasing level of work being outsourced, the Contract Research Organisation (CRO) of choice needs to have proven experience in both the pragmatism and flexibility of the method developer’s mind set and a regulatory background in validation. As companies are focussing on achieving ever shorter times of drug to market, it is vital that a tailored, pragmatic approach is adopted when engaging in both method development and validation activities for an Active Pharmaceutical Ingredient (API) or drug product (DP). Although methods still require a high degree of robustness, the overall strategy should encompass a full evaluation of the regulatory requirements applicable to the particular phase of the drug life-cycle; this is pivotal in Key to Outsourcing Method Development and Validation A Pragmatic Approach order to ensure a successful regulatory submission, where the applicant must demonstrate suitable validation of all methods used to support the filing.
More info >>

Industry Events

12th World Meeting on Pharmaceutics, Biopharmaceutics and Pharmaceutical Technology

8-11 February 2021, Vienna, Austria

In continuation of the very successful scientific meetings in Budapest, Paris, Berlin, Florence, Geneva, Barcelona, Malta, Istanbul, Lisbon, Glasgow and Granada, the 12th PBP World will be held in Vienna from 8-11 Feb 2021. This every two year held conference has gained an ever increasing impact among the pharmaceutical scientists: With up to 1000 submitted abstracts and about 1300 participants it has become a well-established major meeting, attracting scientists from all over the world.
More info >>



©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement