Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Masters of multi-tasking

Share this on social media:

Contract research organisations are using HPC to get the best out of pharmaceutical data. Sophia Ktori investigates

The pharmaceutical industry’s drive to tighten its R&D belt and outsource aspects of its drug discovery, preclinical, and clinical development has spawned a multibillion dollar, global contract research market. It’s a highly competitive environment where pharma is looking for maximum value from dwindling budgets. Contract research organisations (CROs) are vying for the business, and are heavily reliant on cutting-edge informatics to manage the inevitable multi-tasking, and ensure that clients get maximum value from the resulting data – fast, securely, and according to the contract specifications.  
While pharma has traditionally outsourced preclinical development and clinical trials management to CROs, the industry is now increasingly partnering upstream research and drug discovery, including high-throughput screening (HTS), suggests Andy Vines, product manager at IDBS: ‘Rather than just providing off-the-shelf ADME, DMPK or profiling services, for example, some CROs, particularly those in the BRIC regions, are offering early discovery screening services underpinned by their own domain expertise and specialist knowledge. These organisations are establishing long-term partnerships with pharma.’


The three ‘ilities’

Such partnerships require a robust, flexible mechanism to deliver three ‘ilities’ – portability, incorporability, and reproducibility, Vines says. ‘The issue of portability is centred on ensuring that the data packages generated by the CRO can be transferred seamlessly to the pharma client, in a form that ensures the depth and quality of information is the same as if it had been generated by pharma in house,’ he explains. ‘Portability then has to be coupled with incorporability, to ensure that the pharma client can utilise the data within their own informatics systems. The CRO must also ensure that the entire workflow is carried out to exact specifications, and that it is reproducible. The client may only be outsourcing a portion of its screening workflow, so the procedures carried out by its contract research partner must exactly mirror those carried out in house.’
IDBS has developed its ActivityBase screening data management system as a highly adaptable and flexible solution that pharma and CROs can use to optimise and facilitate these three ‘ilities’, whilst maintaining data richness and security of plate-based screening assays. ActivityBase allows every aspect of workflow, assay, data management and transfer, quality control and custody chain to be customised to requirements, down to the definition of any container-type dimensions.
‘What we provide is a solution that offers the ability to create a data analysis template that encodes the entire data analysis process, such that putting raw data into the system will generate results according to the precise data analysis methods specified,’ Vines says. The pharma company or CRO can develop workflows for the screening assay, results presentation and data analysis, along with supporting methodology and reagents, to exact specifications, and then transfer that whole package of information between partners, where it can easily be imported into an in-house system. ‘In an ideal world, both the CRO and pharma client would have ActivityBase installed in house, to make the transportation of data totally straightforward. However, the flexibility of the system means that information can be transferred in the correct format to other informatics platforms, without loss of relevance.’

A strategic imperative
CROs need to address a number of emerging issues in today’s dynamic environment: increasing emphasis on precision medicine, rapid deployment, flexibility, interoperability and scalability, says Rick Mandahl, VP for business development at UNIConnect. The firm’s software platform, UNIFlow, is a web-based solution that is designed to support the depth and complexity of genomic science, which Mandahl describes as ‘a strategic imperative as blockbuster approaches yield to more targeted, nimble, fast moving frameworks, supporting adaptive clinical trials.’
UNIConnect’s UNIFlow system is also ideally suited to the high throughput screening (HTS environment), and particularly for CROs that may have to implement new assays rapidly, from multiple clients with differing requirements, he adds. ‘However, conventional off-the-shelf laboratory information management systems (LIMS) aren’t necessarily customisable, whereas custom systems are time-consuming to build and reconfigure when requirements change. In contrast, UNIFlow has been designed to be ultimately configurable, based on the changing requirements of laboratories. The platform puts the control firmly back in the hands of the lab, rather than the lab being at the mercy of the configurability and complexity of the LIMS.’
UNIFlow is based around the concept of content flow, explains Bill Harten, UNIConnect’s founder and CEO, who specialises in the application of genealogical principles to laboratory problems. ‘At its most basic level we model, in the database, the contents of what goes into, say, a tube, and then when we aliquot something from one tube into another, we copy the record representing those contents from the source container into the destination container.’

Information flow
With each portion of the overall workflow being thought of as a container through which information flows, this concept can be used to model and track all kinds of associations, and any object, process, activity or relationship from the assay steps and the ordering, acquisition and usage of individual reagents, instrumentation and samples, to results input, analysis, reporting and document management, and the whole relationship between CRO and pharma.
Implementing an informatics platform based on this concept enables the notion of flow to be expanded to include authority flow, accountability flow, and even monetary flow, such that the UNIFlow platform can act as a complete process management system, Harten continues. ‘What’s key here is that it becomes very easy and fast to implement and modify specifications. Qualified individuals simply need to define the process at the level of the new information that is being input or changed, and the system captures that data in terms of its flow through whichever containers it’s relevant to. It’s like building with bricks, rather than with grains of sand.’ For the scientists at the front end of the CRO operation, new assays requested by clients can be added into the UNIFlow system by staff in hours or a few days rather than months, the firm claims.

A hundred assays
UNIConnect’s director of implementation, Clinton Maddox, was previously a biochemist at Southern Research Institute, and says the latter’s HTS centre has added over 100 assays to its UNIFlow LIMS in just two years. ‘Largely contracted by the US government, the HTS centre would get new assays assigned every month or so, and we had to have a system that was flexible enough to enable the scientists to easily create the specifications for every step and reagent of the assay, run the experiment, and then collate, analyse, and report the data. Prior to installing UNIFlow that was about a three-month process. Using UNIFlow, this task becomes much simpler.
‘Doing it this way we could input tens of assays within a few months, and by the end of our second year we had done over a hundred assays. In fact, by six months we had undertaken the new assay procedure enough times that we had enough steps to match just about any assay that might come in to the HTS unit.’
UNIFlow negates the requirement for multiple supporting systems that all have to be customised to interface, and upgraded when situations change, Mandahl stresses: ‘When you factor everything into one system there are not only tremendous cost- and time-savings, but auditing becomes simpler, your processes are more transparent, compliance can be achieved and maintained more easily, and every other logistic and data transfer aspect of the CRO-client relationship can be managed seamlessly.’

Data format and downstream use
The delivery of data from CRO to client, in the format requested, and without loss of content or its usability downstream, can be a major headache for CROs. Germany-based BSSN Software has developed its Contract Research Manager (CRM) process management system specifically with the CRO in mind, to manage workflows, capture all workflow, process and analytical data using the AnIML data standard, and automatically generate the requested deliverable for each client.
BSSN has pioneered the establishment of AnIML, explains Burkhard Schaefer, BSSN’s president and the lead architect of the data standard, which is based on an open XML format that is human-friendly. ‘CRM and AnIML in effect allow the CRO and client to maximise the full richness of their data. It enables the storage and sharing, in a single space, of all experimental data – such as screening, spectroscopy, chromatography, biological and medical imaging – including sample and method information, along with analytical results, instrumentation details/logs and software, and the workflow itself.’

A switchboard between CRO and client
Featuring user-configurable dashboards, CRM sits between the CRO and the pharma client, acting as a kind of switchboard that patches disparate data through to each client as part of the workflow, Schaefer claims. ‘CRM effectively captures all the workflow data, and following review and approval of the data by the CRO, automatically generates the required deliverable according to the terms of the contract, providing the client with the same quality of data that would be generated by them in house.’ The platform also manages work orders, requests and supplementary information sent to the CRO by the customer, allows clients to keep track of costs and monitor the progress of their projects, and provides access to the raw data as well as the specified electronic data deliverables (EDDs). Importantly, the CRM platform strictly controls the visibility of data, such that native data systems on either side are never exposed to the outside.
‘CRM doesn’t supplant any LIMS, ELN or instrument management system that the CRO already has installed,’ he adds. ‘Instead, the platform establishes a bridge between the CRO’s and client’s systems, integrating with them and enforcing a well-defined, secure data flow pattern and maintaining all data originating from these different sources in a standard format and with no loss of content. What’s particularly exciting for us at BSSN is that at Pittcon later this year we’ll be launching the capability for clients to access and take off data wherever they are, on a MAC, PC, web browser, tablet or smartphone.’
For clients orchestrating a network of preferred partners for a range of outsourced functions, this flexibility to access data at any time, and any place, will give the CRO a competitive advantage, Schaefer believes. ‘We want to position ourselves in the middle of these networks, and provide the tools that will allow pharma and its CROs to be confident in the data that they transfer, and to manage and oversee their relationships.’

Managing samples and supply
Sample ordering, management, allocation and client confidentiality are key issues for CROs that work on multiple projects for a range of clients. Titian Software recently announced the installation of its Mosaic sample management software at four major CRO sites in China, and the appointment of a distributor for the product in the country. UK-based Titian’s modular and scalable Mosaic sample management platform supports sample supply workflows, and orchestrates the tracking, ordering and preparation of small molecule compounds, biological samples and reagents, explains David Booth, regional sales manager for Europe. The platform also integrates with a range of automated stores and liquid handling systems to further streamline and automate workflow, and reduce human error.
Sample management is obviously key for compound providers, but it’s also important for CROs that offer preclinical assay services or carry out clinical trials for the pharma industry, Booth says. ‘CROs will be handling samples from or for multiple customers, and they need to make sure that the samples, and the associated data, are strictly segregated. The same safeguards need to be in place whether the material is synthetic or biological in nature, and whether for preclinical or CFR21 part 11-compliant studies.’

Safeguarding exclusivity
Compound providers in particular are restricted contractually with respect to whom they can issue a compound out to, and for what target. Mosaic includes safeguards to ensure that exclusivity isn’t breached. ‘And unlike other software that may offer inventory management but not necessarily ordering mechanisms, shipping, or integration with the liquid handling instrumentation, Mosaic provides the complete package that will handle every step, and can be interfaced with a range of informatics and instrumentation providers,’ Booth states.
The platform allows customers to place a request for samples in the required form, even from a global inventory that could span multiple sites.
Booth adds: ‘Mosaic then determines whether the samples are available for immediate dispatch and, if not, initiates workflows to create them. Inventory tracking information is maintained in a database that can be accessed at any time. Staff are guided through the individual operations, and data is automatically passed to robotic workstations. This ensures that all the correct methods and procedures are executed on the workstations, through every step of the sample preparation process.’

Keeping data live
One major hurdle that any CRO must overcome is the inability to collate compounds and project work in a consistently ‘live’ and readable electronic package, says Ryan Sasaki, director of global strategy at ACD/Labs. ‘This challenge is due to the vast array of different analytical instrumentation and informatics options and combinations employed by different laboratories around the world. Without consistent, analytically enabled environments, data is often only made available in a ‘dead’ format like a PDF, spreadsheet, or Word document. As a result, contextual information can be lost, as is the ability to search the data in the future.’
Toronto-based ACD/Labs’ software platform for analytical and chemical knowledge management has been developed to address this issue. ‘To provide maximum value to their customers the CRO must be able to do more than just distribute reports in dead data formats,’ Sasaki continues. ‘Of course that is easier said than done when the data is acquired on different types of instrumentation and processed in a variety of different software packages that may not be employed at the client organisation. What we offer is a vendor-neutral platform for handling analytical data.’

Keeping data contextual
ACD/Labs’ software can sit between a CRO and its clients and existing informatics platforms, such as ELN (electronic laboratory notebook) and LIMS, keeping the data in a live environment rather than archiving it into silos. ‘Archiving data is necessary, and exists primarily for auditing and regulatory purposes, but the problem is that archived data then essentially becomes locked down in a silo and can only be viewed in whatever format it was stored,’ Sasaki notes. ‘Again, it’s effectively dead data; the ability to search and mine what could be valuable information resources is lost. Even if the data is accessible it is usually in its raw, instrument data form which lacks the context around which conclusions were made.’
This is a real drawback for CRO and client alike, particularly if the client needs to revisit and reanalyse the analytical data. It also significantly reduces the return on investment for clients, Sasaki adds. ‘The client wants to ensure that the money they are saving through outsourcing won’t adversely impact on the delivery of a product of equivalent value. Our platform essentially gets around this problem by keeping the data live, and allowing the CRO or client to look at all that data with its chemical context, and keep the information on any compound searchable by assignments or annotations, structure, substructure or spectral parameters. This capability transforms data into knowledge.’