The internet: moving in on drug research

The pharmaceutical industry has been one of the slower industries to take advantage of electronic technologies like the internet. But, as Sarah Houlton reports, its benefits in supply chain management and r&d are now being realised

The internet and related technology are having a dramatic effect on most, if not all, areas of business. Yet, according to a report from KPMG, despite being one of the most high-tech industrial sectors, take-up of e-business in the pharmaceutical industry has been slow. The report attributed this to the lack of direct retail customers, allied to the procurement of supplies representing a relatively minor expense.

However, in the past few months, pharmaceutical executives have begun to realise the potential benefits of using electronic processes. Time and money can both be saved by moving much of the clinical trials process on line. Selling to distributors and procuring supplies can both be done on line. And promotional activities such as setting up disease-management websites or even sponsoring major healthcare sites can give access to important information about the consumers of their drugs.

The KPMG report highlights the future expectations of pharmaceutical companies with regards to doing business on line. In around a year's time, 82% expect to offer at least limited transactions through their websites, up from the 21% who were offering such a service when the report was prepared last autumn. In the same time-frame, on-line sales are projected to rise from 3% to 14%.

Large increases in the use of industry-specific on-line exchanges for supply chain management are also foreseen, with a rise from 6% to 49%. Multi-industry exchange usage is expected to grow from 12% to 36%.

Small biotech companies start out largely as virtual companies, with a strong research foundation. However, it is not just costs that drive choices of suppliers and partner companies, but confidentiality issues. Service providers will manage the supply chain through better information exchange or sharing. So future outsourcing decisions will not necessarily be made purely on a financial basis, but with the company they feel most happy sharing confidential information with.

Perhaps the most dramatic changes will be seen in the way drugs are discovered and developed. There is the potential for great time savings, leading to financial gains because drugs will be on the market with their patent protection intact for longer. The pharmaceutical companies have set themselves huge growth targets, and for a top-ten company to keep up with an annual industry growth rate of 10% it will have to bring at least five important NCEs a year. A mid-sized organisation will have to deliver three. Yet figures from Accenture (the former Andersen Consulting) say that the top companies have, on average, launched only 0.45 significant NCEs a year from 1990;94, and only 8% of launched products reached the target of sales of US$350m. Moreover, claims Accenture, the current r&d operating models will be unable to deliver the number of excellent NCEs required to meet the targets.

Improving the success rate is not merely going to happen by throwing more money and more people at research; strategies will have to be fundamentally rethought as well. The increase in high-tech techniques such as high-throughput screening and bioinformatics, which are ideally suited to the implementation of internet strategies to speed up the process, will also make research more effective.

Pharmaceutical companies are hoping to use these modern drug discovery techniques to save 2;4 years on the drug discovery process, cutting it back to around 8;10 years. The use of electronic methods is more widespread earlier in the drug discovery timeline, where new techniques are being deployed increasingly in preclinical studies, for example toxicology screens.

As Shankar Hammady, ceo of PharmQuest, explained at the recent AAPS congress in Indianapolis, using e-r&d techniques can cut the potential for problems with both GLP and GMP, and he suggests that process mapping and workflow automation could well prove to be the best solution. Improving communication, documentation and tracking will necessarily speed up the process. As long as data security is ensured, using a computer documentation system should aid regulatory compliance. And being able to detect failure earlier will lead to a higher success rate.

According to the KPMG report, the pharmaceutical industry currently spends US$30bn a year on clinical trials, with much of this being in administrative areas such as paper-based collection of data and coordinating the whole process.

Using the internet within the clinical trials process will make keeping track of the trials much easier. As there are only a finite number of patients with a particular condition or disease available for trials, companies working in the same therapeutic area are in competition for them. Making data collection from a more geographically wide spread of centres instantly widens the pool of available patients for trials, making patient recruitment less of a problem.

Coordinating the process on the internet could cut the cost of the whole clinical trials process by at least 30%, and more so in the US where the FDA is accepting electronic submissions. It will inevitably make communication between different centres in multiple countries more straightforward.

faster trial results

An additional time-saving benefit is that entering data by patient as it is generated, rather than awaiting data from all patients in a trial before submitting them to the drug company, means they can be constantly monitored on the internet, and there is a good likelihood that a statistically-significant result will be seen sooner than might otherwise have been the case. Equally, any problems with side-effects can be picked up on much more quickly and the trial stopped sooner if necessary, again saving time and money in the long run.

Using e-r&d techniques could prove to be the means to unclogging the development bottleneck, which has increased from the order of 10,000 leads to one product to more like 1,000,000 leads to one product. This is being seen as a result of the implementation of modern drug discovery techniques like pharmacogenomics, high throughput screening, combinatorial chemistry, bioinformatics and their ilk.

A number of companies are working on software solutions to facilitate the move to e-r&d for pharmaceutical companies. For example, management consultancy Pricewaterhouse Coopers and the software provider Oracle have joined forces to create a system to manage the clinical trials information that is spread across a number of locations, both internally and externally. Oracle's Juan Rada claims the key to the system is a central data repository. 'This accommodates and integrates all the relevant data, irrespective of source and format. Significant benefits accrue as this critical resource is exploited and interrogated to improve decision-making throughout the drug development lifecycle and beyond.'

Software models have also been evolving. The traditional model sees a large up-front licensing fee allied to implementation costs, ongoing maintenance and upgrade fees. In future, it is likely that 'applications for rent' will be available, which will be licensed on a per month or a per user basis. This will lead to more predictable costs, with no upgrade or maintenance costs being due. Ultimately, it could even go as far down the line as a usage-based fee per molecule. The advantage of this for the pharmaceutical company is that it shifts the financial risks from them to the supplier.

Connection speed is another issue that must be addressed. 'Universal broadband access will dramatically affect the internet,' says Pradip Banerjee of Accenture. He explained at the AAPS meeting that netsourcing means computer software can be hosted at a remote location, and can then be accessed through secure lines, providing a more cost-effective method for keeping information up-to-date. 'Using tracking systems, comparable to those used by companies like Fedex for tracking deliveries, will make it possible to tell precisely where a drug is in the trials process.'

regulatory compliance

Another reason why the take-up of e-r&d has been sluggish is the need to comply with the requirements of the regulators. A traditional drug filing will involve a mountain of paperwork, but regulatory bodies like the FDA are now beginning to be modified in order to accommodate electronic developments. As Randy Levin, associate director for electronic review at the FDA, explained at the AAPS, his organisation is undertaking electronic initiatives for submission, review, inspection, information dissemination and leveraging. 'Congress has mandated that by 2002 all submissions must be accepted in electronic format instead of paper.'

It has been possible to submit an e-NDA since February 1999. Levin said that the majority of submissions now have an electronic component, and 15% of original applications are now completely electronic. 'This has led to a 50% reduction in paper,' he said. In addition to NDAs, in the near future, the FDA will begin to accept DMF, IND and ANDA submissions electronically. In an IND submission, for example, Levin said that being able to create an XML-based cumulative table of contents will make it much easier to keep track of all information regarding the submission.

Drug regulation and listing information are also being considered, which would lead to an up-to-date and complete drug listing being able to interface with field inspection information and import information, so it will contain such details as where it was packed.

But why is e-r&d a practical proposition? PharmQuest's Reed Benet explained at the AAPS that, as r&d is a process, there is much scope for electronic improvement. 'Processes can be mapped, automated, rationalised and integrated, and all this can be done on line,' he said. And the internet can be a powerful tool in achieving this. 'Web browsers can overlay legacy systems,' said Benet, 'and provide hooks for new software systems integration. The internet connects 100,000 decision makers in a US$300bn industry. In the end, it is just another tool.'