What's Next In: Information Technology

Published on: 
Pharmaceutical Technology, Pharmaceutical Technology-12-02-2007, Volume 31, Issue 12

I foresee dramatic future changes in five distinct areas of the pharmaceutical industry: Research, Regulatory Approval, Production, Sales & Marketing, and IT Infrastructure.


I foresee dramatic future changes in five distinct areas of the pharmaceutical industry: Research, Regulatory Approval, Production, Sales & Marketing, and IT Infrastructure.


Research (30 years out) Thirty years from now we will not make much of a distinction between biotechnology and synthetically driven drug development. The biologics will be considered just another way to synthesize chemical entities. We will have much more knowledge about the biological effects of specific substructures within molecules. By analyzing a molecular structure, we will be able to accurately predict a chemical entity's bioavailability, stability, toxicity, and efficacy to specific tissue, organ, or nerve targets. Companies will use their IT infrastructure to maintain massive chemical structure databases with proprietary metadata about the interoperability of specific molecular substructures. Most chemical syntheses will be created by that suggests optimized synthetic steps in much the same way a GPS system generates the shortest route between two addresses. Network security will become paramount as corporations embrace the hacker and virus model to steal confidential information.

Research (100 years out) Advancements in our scientific understanding of both chemical and biological processes will give rise to sophisticated modeling and programming in both areas. Models of organic and inorganic chemistry will allow computer systems to suggest new chemical entities that have specific therapeutic properties. The computer systems will also recommend the steps necessary to synthesize those entities. The trick will be engineering the equipment to physically reproduce theoretically possible chemical processes. On the biologics side, advancements in our understanding of DNA, stem cells, and cloning will allow the emergence of the new discipline of Genetic Programming. Genetic Programmers will use their education in programming and biology to structure the DNA of proposed organisms from scratch. These DNA sequences will be created in the laboratory, inserted into stem cells, and ultimately cloned into organisms designed for a specific purpose. These organisms might be designed to create a chemical entity that cannot feasibly be produced by synthetic means. Or the organism may be designed to ingest specific unwanted bacteria, viruses, or cells in the living body. These programmers will, of course, make mistakes. But in that case they really will be bugs.

Regulatory Approval (30 years out) Approval of new medicines will change dramatically. No longer will approval be something that "happens": It will be something that is "managed" on a continual basis. We are all familiar with adverse event management systems. New systems will also track efficacious events. Each use of a medicine will be audited and will add to the background of information. Adverse events will subtract from a medicine's approval for use, and each efficacious event will add to it. The more efficacious events, the broader the approval and more likely the medicine will be prescribed. Each adverse event will have the opposite effect. The profile of efficacious and adverse events will be matched against the profile of a patient to determine whether the patient is a good fit for a particular medicine. Automated prescribing systems will suggest the combination of medicines and procedures with the best likelihood for the optimal patient outcome. These changes will have a profound effect on pharmaceutical marketing practices, perhaps even eliminating it.

Regulatory Approval (100 years out) The only change is to the regulatory bodies themselves. Our notion of the state evolves so that multinational corporations wield as much political influence and have similar infrastructure profiles as current day governments. The profit motive behind regulatory oversight becomes paramount. Medicines and procedures have wide approval profiles if they reduce the overall cost-of-living. People will have to make choices between living longer and living better, and people will have wildly different notions of what "better" means.

Sales and Marketing (30 years out) The current day conception of a sales force with advertising and brand management goes away. Automated prescribing systems diminish the influence of the sales force. The Marketing department becomes a scientific discipline and functions to increase the prescribing profile of a product by performing deep-diving research into every adverse event to make sure that the impact on the prescribing profile is minimized.


Sales and Marketing (100 years out) No need for a prediction. See 30 years out.

Production (30 years out) The catch phrases of the past 20 years will become a reality: Manufacturing Execution (MES), Electronic Batch Records (EBR), and Process Analytical Technology (PAT) become commonplace in the industry. QbD is adequately described by a validated MES where production is continuously improved via PAT, resulting in an EBR so that Quality by Design fades away as it is absorbed by the other three disciplines. Technology workers will be beside the chemical engineers in creating the production process. S88 evolves and becomes a standard language by which production processes are described.

Production (100 years out) Pharmaceutical production will become much more automated, of course, but it will also become much more customized. Think of it like automobiles. Cars on the lot come in different models, colors, and come equipped with different accessories. You can pick one off the lot that best meets your needs. For a little more money, and a little more time, you can choose among available options and have the car built for you. And for a lot more money and time, you can have a car pretty much customized. Pharmaceuticals will be the same way. Emergency room patients will have to pick a medicine off the shelf. Those patients with not so acute conditions may order a medicine with certain common adaptations for their race, family history, or geographic location. Elective procedure patients might order a customized pharmaceutical. The factory will be equally efficient at making a large batch of off-the-shelf product or a single prescription quantity of a special order. This will be possible through the use of automated controls with supply chain optimization databases.

Infrastructure (30 years out) The role of IT will be enhanced to become a bigger part of the "product." Information that is typically simply printed on the product package will be encoded on RFID chips on the packaging, and that information will be available for at-home or in-clinic computing. For example, the home computer will automatically scan the label of a bottle put in the bathroom medicine cabinet. It will know what the medicine is, for whom it has been prescribed, and how often they are to take it. The computer will know the other medicines the patient has in the cabinet and whether there exist contraindications, and if so, it will automatically alert the head of the household. People will value pharmaceutical products that have complete pedigree information on the label. IT departments will be relied upon to make sure the data are accurate. The typical in-house IT department will be quite small. Most IT work will be farmed out to service providers located around the world. Hosted applications will be commonplace.

Infrastructure (100 years out) Information throughout the society will be ubiquitous, like electricity is now. Every person will have a wireless personal digital assistant. When in the office, the PDA will activate the equipment needed to perform useful work. When one is at home, the PDA will activate household appliances. When one is in the car, it will monitor the engine as well as the music player and phone. And when one is at the physician's office or in the pharmacy, the PDA will be able to provide historical data on blood pressure, heart rate, medicine consumption, and will download these data to the physician or pharmacist upon demand. Each person will have their medical history on their PDA, available to the physician at the user's discretion.

Software engineers will be managing this infrastructure by issuing voice commands to software design servers that will configure needed functionality. Development at the pharmaceutical company will involve hardly any traditional programming skill at all-only a deep understanding of data and process models. A combined process and data language will emerge, similar to the SQL language we have for relational databases. Engineers will support the hardware and communications infrastructure with much the same discipline as current-day electrical engineers support the power grid.

Fortunately for us, even a hundred years out, we will still not see the likes of Gort, Robbie, R Daneel Olivaw, HAL, or Skynet. Thank goodness! Not only that, with the advances in medical science, we might just be there not to see them.

Herschel Kenney, senior director, manufacturing and quality systems, Purdue Pharma

IT Data Management

IMS Health projects worldwide pharmaceutical prescription sales to more than double over the next decade and a half, from $600 billion in 2005 to $1.3 trillion in 2020. Generics account for around 53% of prescription drug sales in the US and about 55% in the UK, 41% in Germany, and 16% in Japan. Patents on many drugs are expiring, and there are fewer blockbuster drugs with the potential to replace them. The sale of generics is expected to rise steeply over the next five years (2007-2012) as patents begin to expire on branded drugs with more than $60 billion in combined annual sales. The overall market share of generic products is expected to rise as drugs go off patent and governments across the world try to control their countries' expenditure on medication, favoring the use of generics over branded drugs. The major pharmaceutical companies will need to move away from the "blockbuster drug" model and adopt new strategies that cater to specific diseases and populations-making targeted medicines. For example, the over-65 population in America is growing four times faster than the population as a whole. As the number of Americans older than age 65 increases from 1-in-8 today to 1-in-6 by 2020, the pharmaceutical industry needs to target old-age diseases like Alzheimer's, which is currently the seventh-leading cause of death in the US. Drugs that address rising multifactor disorders such as cancer, as well as lifestyle disorders such as obesity, are also likely to experience strong revenue growth. The industry must shift its investment focus toward more research and away from sales and marketing. It appears unlikely that pharmaceutical companies can sustain a growth model whereby approximately one third of revenues are spent on sales and marketing costs while just one sixth of revenues is invested in R&D. With recent announcements that top pharmaceutical companies such as Pfizer, Boston Scientific, and Novartis are reducing their sales forces as a means to reducing the cost of sales and marketing, the industry is already signaling a change in spending priorities.

From a technology perspective, the pharmaceutical industry is dealing with an explosion of data that must be integrated, processed, analyzed, and shared-particularly with recent developments in genomics, proteomics, and emerging areas such as systems biology. Grid technology helps to deal with the computational and data integration challenges and enables researchers to collaborate more effectively, but the technology is still maturing. Grids enable organizations to run increasingly complex applications that require ever-larger datasets and multi-step analysis. They enable scientists within a single organization to gain access to the company's research data and/or enable researchers from different organizations to collaborate on a single-drug development effort. Grids also create ways for people or companies to contribute computational power to the study of diseases that we know little about (for instance, the Intel collaboration with Cambridge University to examine anthrax).

Specific applications working in collaboration with patient data and therapeutic efficacy of molecules alone will not suffice. Applications will have to have some artificial intelligence to learn from all experiences and automatically build the knowledge set. Physical examination of patients will become less necessary as remote sensing and data acquisition technology make remote diagnosis easier. A larger collaboration will need to exist between government, insurance companies, patient data systems, and diagnostic instrument data streams.

Another key issue for the industry is the counterfeit drugs entering the supply chain. Electronic tracking technologies such as RFID and bar coding are being evaluated. RFID solutions are also being evaluated to increase productivity and accuracy in shipping and receiving, reduce stock-outs, improve expiration management, and simplify the product recall and sample-management processes. Technology improvements will also ultimately benefit the patients. According to research, prescription errors resulted in 72,000 deaths in the UK. The prescription error could be cut by up to 70 percent using the deployment of bar code technology. The use of robotic dispensing units linked to prescriptions through bar codes helps to reduce fatal medical errors, control cost, keep a closer track on stocks, and order new drugs when required. The robot has an error rate of one in 10,000. The robot will save more than 1,200 hours of staff time per year. The newest generation of these devices can securely store hundreds of medications, and maintain full patient profiles. Thus the pharmacy department role in providing product services will be greatly reduced.

Reducing R&D Time, Slashing Costs The industry focus also has been to reduce the cost and time for new drug development. The role of genetically based diagnostics in the development of personalized medicines has already shortened the R&D cycle for some products. These new technologies will improve understanding of diseases and link genomic and clinical data. The convergence of therapeutics and medical devices, which started in earnest with the drug-releasing stents, will continue and will become increasingly sophisticated, improving efficacy and reducing the risk profile of many existing therapeutic agents.

The ultimate beneficiaries of the changes in the industry and the technology that supports them will be the patients. Patients will benefit from improved treatments and will get them more quickly and less expensively than they do now. The reliability, safety, and efficacy of medicines and practices will increase multifold. Researchers can look forward to much higher processing speed, data crunching technologies, real-time data acquisition, and data synchronization.

The current R&D model of four-phase clinical trials that typically end in submission for a drug license and market approval will be replaced by collaborative in-life testing and "live licenses" being issued based on the performance of the drug during the trial. The industry will conduct smaller, more focused clinical trials, continuously sharing results with regulators. If testing confirms that a medicine is safe and effective, a live license will be issued permitting the company to market the drug on a restricted basis. Further in-life testing will extend the license to cover a larger number of patients or a different patient population. The use of biological models, bioinformatics, and biomarkers to identify toxic or ineffective drugs early in the development process will help to drive down development costs, increase revenues, and improve overall industry productivity. It may be possible that the time between target identification and market launch will be reduced from almost 12 years to less than 5. We might also anticipate an increase in the success rates of bringing products to market by a factor of four, while the costs of clinical studies are slashed from an average $800 million to as little as $200 million. Already, several national and regional regulators have begun to collaborate by sharing safety and efficacy data. There may well be one global regulatory system by 2020, administered by national or federal agencies responsible for ensuring that new treatments meet the needs of patients within their respective domains. Such a system would help to reduce the spiraling costs of regulatory compliance and reduce time to market.

Milind Joshi, vice president,

Life Sciences Practice, Patni Computer Systems

Industry experts give their predictions for the next 30 years.