While serialisation schemes around the globe already have much in common, one feature in particular has come to the fore in recent months, and it is not entirely welcome. Following last year’s delayed publication of the Delegated Acts associated with the EU’s Falsified Medicines Directive (FMD), the de facto compliance deadline has effectively moved to January 2019. This news was followed by the announcement in October 2015 that an indefinite halt had been called on Brazil’s serialisation programme, while at the end of the same month, the US Food and Drug Administration (FDA) announced a further four-month delay in enforcement of the track and trace requirements of the Drug Quality and Safety Act (DQSA), which were due to come into force on 1 November 2015.
All this points to the fact that serialisation is proving to be far more complex in practice than had been anticipated – and that the difficulties are being encountered in disciplines hitherto unfamiliar to the pharmaceutical industry. Back in 2011 when the FMD was first published, there was widespread recognition that – insofar as the coding requirements of the Directive were concerned – a large proportion of the installed base was not serialisation-capable. At the time, this was perceived as one of the biggest challenges but taking heart from the successful implementation of France’s CIP 13 programme (which, although not a serialisation scheme in itself, did necessitate updating the coding and vision capability of production lines) many in the industry hoped that investment, however large, in new, serialisation-capable technology would be the biggest hurdle they would encounter.
Serialisation is proving to be far more complex in practice than had been anticipated and the difficulties are being encountered in disciplines hitherto unfamiliar to the pharmaceutical industry
Looking at the reasons for the various hold-ups, however, this is clearly not the case. What is evident is that the pharmaceutical packaging industry continues to excel in the disciplines it knows best – namely, those covered by Good Manufacturing Practice. What is proving more problematic is what happens beyond the factory gate: how to maintain a robust chain of custody in a sector with complex and often global distribution models.
In common with the US, Argentina, South Korea, China and the EU, with legislation no. 11.903/2009 Brazil is aiming to stem the flow of falsified medicines within the supply chain to protect patients from potentially life-threatening criminal activity.
Packaging – and most particularly features such as serialised codes – is the first line of defence in this campaign against the counterfeiters, and until last summer, Brazil’s pharma regulator Agência Nacional de Vigilância Sanitária (ANVISA) was assumed to be making progress in driving the implementation of full serialised track and trace solutions throughout the supply chain.
Since the announcement on 1 October, however, the progress of Brazil’s programme has been halted, with ANVISA’s board of directors agreeing on ‘the need for some adjustments’ to resolution RDC 54/2013. This is being interpreted by some as a potential opportunity to overhaul the entire initiative.
Similar to the EU FMD, the RDC 54/2013 states that the data matrix bar code symbol on each pack should encode the GTIN, ANVISA registry number, serial number, expiration date and batch number. One aspect of this regulation however – the requirement for pharma manufacturers to have serialised and tracked three batches of product through the supply chain by some companies by 10 December – has been suspended. A primary point of contention is the requirement for a medicine’s licence holder to receive and collate traceability data from its trading partners, and then to be wholly responsible for reporting this data to ANVISA. It has now been acknowledged that this approach could give way to ‘predatory behaviour between companies’, in turn leading to the decision that a more centralised reporting model should be adopted.
This is a fundamental shift and one that commentators believe will delay the entire programme by two to three years. So why has Brazil encountered so much difficulty and how does this bode for Europe’s own legislation?
The big data hold up
The Brazilian scheme has been characterised as one of the most complex track-and-trace regulations in the world. Some independent specialists, including US healthcare supply chain consultant Dirk Rodgers, have called Brazil’s data exchange requirements ‘unnecessarily complex’, referring to two major concerns: the amount of data Brazilian manufacturers must provide, and the data protection risk this creates.
If this transactional data is visible to pharmaceutical companies, it could give manufacturers an opportunity to use the data for commercial purposes
The stipulation in Brazil that product manufacturers must capture an extensive set of product movements from across the entire supply chain demands a control of data outside direct business relationships that has room for error at best, and anticompetitive practices and even corruption at worst. As widely discussed throughout the industry, if this transactional data is visible to pharmaceutical companies, it could give manufacturers an opportunity to use the data for commercial purposes, giving way to competition and potential price manipulation.
By contrast, other schemes call for data to be held by a central repository, and this takes away the complex element of data handling through all areas of the supply chain. The European Medicines Verification System (EMVS), an interoperable, pan-European database that enables medicines to be verified instantaneously at the point of dispensing, reduces complexity by creating a single interface as a central repository for manufacturers across all Member States. This European Hub distributes data from the entire supply chain to the relevant National Medicine Verification Organisation (NMVO), such that a unique record for each item entering the supply chain is created.
That record remains available for interrogation through to the point at which the product is dispensed, when the pharmacist or other dispenser scans the product. This action updates and effectively closes the record, such that the product is no longer live and available within the supply chain. The Cloud-based EMVS is widely thought to be a cost-effective and secure solution to data storage – a model that ANVISA now seems keen to adopt. This was indicated through ANVISA’s October announcement, which stipulated that the Brazilian system should instead adopt centralised model tracking, through which production and distribution data will be handled by ANVISA itself.
The case for aggregation
Another key concern in Brazil lies within the requirement not only to serialise products at item level and case level, but also to aggregate relationships between case, unit and globally unique serial numbers. As manufacturers know, moving unique products around the world presents its own problems. Items don’t move as single units, but in batches on pallets or in cases, and keeping track of what is included inside the secondary and tertiary level of packaging is the next major challenge for the industry.
Aggregation offers a relatively straightforward means for customs officials and importers to interrogate the contents of each shipment
Aggregation, or accurately recording serialised pack data and building that into a hierarchical structure, records detailed information about each pack’s progress through the supply chain. While not mandatory according to the current FMD framework, there are significant benefits to aggregation where goods are to be shipped around the world, as it offers a relatively straightforward means for customs officials and importers to interrogate the contents of each shipment.
With three years before the EU FMD comes into force, there is enough time for manufacturers to consider aggregation within their solutions, without necessarily including it in the pilot programmes.
For Brazil, however, aggregation is mandated, making serialisation and data handling more complex and time-consuming from the outset. But with much of Brazil’s supply chain stretching globally, once aggregation is implemented it will potentially save the South American industry time and money with a more efficient and accurate packaging line. In fact, properly implemented aggregation can effectively eliminate the requirement for a big data storage database – but getting it right is complex.
The nature of the market
In a statement released by ANVISA following the programme’s suspension, it was suggested that the size of Brazil’s market makes serialisation a much bigger challenge than for smaller countries that have successfully implemented traceability. Within the EU, we have the security of very similar programme elements within the Union, which bodes well for smooth supply chain management and data handling. In fact, the EMVS hub went live last year, when it was also announced that Germany’s SecurPharm database would join forces with the European Stakeholder Model, enabling proof-of-principle of the interoperability between national and European repositories to be demonstrated.
Until 1 October, there was still the expectation that to achieve compliance, Brazilian manufacturers would need to have serialised and tracked three batches of product through the supply chain by December. This appears to have been the sticking point, with many claiming that the scheme makes it impossible for manufacturers to re-package or add serialised codes within Brazil – the codes must be traced right from the point of manufacture, even if outside the country. The integration of the European market means that we mostly avoid this issue, although it is important to ensure that serialisation solutions are suitable for integration with global markets.
Tight timelines
The timeline ANVISA imposed on the Brazilian industry was eye-wateringly tight, the three-batch requirement putting huge pressure on packagers – particularly those with smaller budgets and fewer resources.
Serialisation programmes are notoriously complex, and ample time is critical to ensuring pilots and that programmes run smoothly. If implementation is too rushed in Brazil, it will raise the risk of medicinal shortages if the entire supply chain is not ready to comply by the time the legislation is implemented. While the EU’s legislation now has a predicted implementation date of 2019 following a delay to the announcement of the safety features Delegated Act, Brazil is now looking at a huge delay. A bill tabled in the Brazilian Senate at the end of June is seeking an extension of the three-year deadline for full implementation to 10 years, a clear sign that the complexities of serialisation are being acknowledged.
An update on progress in the EU
Notwithstanding the recent delay, there is much to applaud in the EU’s approach. Clarity around the requirements, combined with a realistic timeline for their deployment, has created a window of opportunity for manufacturers to properly plan and implement serialisation infrastructures that are robust and fit for purpose. The task is still daunting, but now the industry has confirmation of what is expected of it, serialisation roll-outs should be well underway.
The EU FMD at a glance:
- The security feature must include a GTIN, batch number, expiry date and serialised number
- Only one machine readable code per pack will be allowed
- Medicines released for sale or distribution before the 2019 deadline without the safety features can still be dispensed until their expiry date
- All prescription medicines and products are on the blacklist – all other medicines are not allowed to bear the safety features
- Tamper-evident features are required but not yet specified
- A minimum of C Grade is required at dispensation, against ISO 15415:2011
- The character sequence has to stay unique and remain readable until one year after the expiry date or five years after the pack has been released for sale (whichever is the longer period)
- The structure of the unique identifier needs to have a coding scheme with application identifiers, and should not be longer than 50 characters
- Probability to guess the unique number has to be less than 1/10000
- Only the GTIN and the serial number (and for relevant countries the National Reimbursement Number) needs to be printed next to the 2D code. The expiry date and batch code may be printed on the other side of the pack
- No human readable text is required when the sum of the two longest dimensions of the pack is equal or less than 10cm
- Response time in pharmacy less than 300ms in 95% of cases
- Data to be kept during the shelf life of the product plus one year, or five years after the release for sale (whichever is the longer period)
- Data to be uploaded before the product is released for sale or distribution
- The upload must contain:
- All data elements from the 2D code
- The coding scheme used
- The name of the medicine, the common name, strength, package size and type
- The member state where the product is intended to be placed on the market
- Name and address of manufacturer
- Name and address of MAH
- List of wholesalers that will store and distribute the product
