Much has been written about a perceived lack of innovation in the pharmaceutical industry, as it relates to productivity. While some of the criticism may be justified, much is not. Clearly, you can question the investments that companies made in the 1990s to develop de novo high-throughput technologies. More recently, you can question the concept of building a portfolio that will deliver products by targeting appallingly low industry attrition rates. Finally, it would be fair to question whether our investment in understanding disease biology in support of our selecting the best targets has been sufficient.

However, it is also fair to say that the diseases and targets we are pursuing today represent formidable challenges, in terms of their late age of onset, chronic nature, polygenic origin, and combination therapy requirements. Innovation is the key driver for success in our industry and it is not lacking. However, we can drive greater...



In order to spur innovation, you must create a culture and an environment that fosters it. To do this at Roche we reinvented our R&D organization. We created a drug discovery culture and mindset that, while initially focused on generating clinical candidates (measuring quantity), is now focused on progressing programs into pivotal proof of concept Phase II clinical trials (measuring quality).

This evolution required a long-term approach, which included having credible leadership, hiring good people, and providing them with the right tools and a clear target. We needed to let our people succeed, measure their efforts, and challenge them to improve. We also created a globally integrated informatics component that allows us to collect and query data, while creating new knowledge. Finally, we radically modified our management model, which reqiured the implementation of clear decision rules that heavily focus on quality first and also reward innovation.

Our long-term approach began in 1990, when Roche acquired a majority stake in Genentech because we anticipated that a component of the future of healthcare would be protein therapeutics. Now biologics represent 40% of Roche's revenues-a substantial return on investment. However, in 1998, we took a hard look at our pipeline and processes, and developed a plan for our organization. In 2002, we obtained a majority stake in Chugai following a merger with our Japanese affiliate. While this relationship is different from Genentech, it still focuses on a long-term growth strategy aimed at capturing synergies and creating value, and provides Roche with an advantage in the rapidly growing Asian market.

Now, every scientist at our five global research facilities has access to all the information being generated at our company, and can make comparisons based on common assays and standards. Our research centers have their own culture and, by preserving these differences, we have avoided 'group think,' which can cripple an organization and stifle innovation. Rather than creating a centralized research focus, we maintain a critical mass at each of our global R&D centers. Internal capabilities are complemented by our majority-ownership interests in Genentech and Chugai, which further expands our access to innovation. We refer to this as our "hub and spoke" model, which is the concept of majority-ownership coupled with strategic alliances.

Collectively, this approach has been successful, allowing us to create an environment that fosters innovation. We also expanded our scope of innovation and significantly enhanced the size and quality of our pipeline.1 Our pipeline now includes 107 research projects across seven therapeutic areas; 64 new molecular entities, including six with opt-in opportunities; and 27 regulatory filings planned for the next five years, eight of which are for lead indications.

Allowing a strategy to reach a point at which it can be evaluated takes patience. While strategies may require modifications or tactical changes to evolve, it's best to let them mature before making drastic changes. All too often in our industry, good strategies are violated by changes in management and/or mergers and acquisitions.


In recent years, the business model for biotech has evolved from creating enabling technologies to drug discovery. At the same time, consolidation in pharma has caused many good scientists with drug discovery experience to move into the biotech sector, adding to the credibility and likelihood of success for this model. Therefore, partnering has to be one of the main strategic levers to drive innovation in pharma.

For alliances to work, natural synergies must exist. For biotechs, pharma provides credibility, which can equate to higher valuations, important operating capital, and value-chain expertise (i.e., scientific expertise, regulatory guidance, and help in building commercialization capabilities). For pharma, we can supplement our internal productivity, gain access into areas where we do not invest internally, and realize true value via equity investment.

When considering a potential partner, it is equally important to consider the experience and management style of their leadership, along with their assets. Partnerships should be measured on both sides for their value. At Roche, we look for partners with long-term potential because the reality is that only a small percentage of compounds make it to market or have a return on investment.

Roche has been a leading force in alliances in recent years, entering into them with a wide range of companies in various areas. In 2003, Roche led the industry with 10 product deals.2 As a result, we now have alliances with over 50 companies around the world and are viewed as a distinctive partner with a flexible, collaborative approach.

As the biotech sector matures, opportunities for partnering will continue to grow and be a critical component of a company's strategy. The key is in finding the best companies and working in partnership to achieve mutual goals.


In the future, as new healthcare models evolve, biomarkers will play an increasingly significant role in drug discovery and development, helping with target evaluation, and establishing proof of concept/mechanism and variable drug response. While there is a need to identify, validate, and utilize bio-markers in clinical development and-in some cases-the market, it is important that we intelligently generate and test hypotheses and appropriately invest to deliver the highest value medicines (sometimes linked with a diagnostic) to the market.

In the past we thought we knew what the profile was for a type II diabetic patient based upon a clinical differential diagnosis. However, we're learning that our definition of the disease no longer applies and is evolving to a more molecular-based diagnosis. All diabetic patients do not respond in the same way to medicines and there are many forms of the same disease, causing patients to react differently to the same drug. Advances in genetics and genomics are making this increasingly clear, and we're finding that these are helping us identify patient subpopulations with common genetic factors, which when understood, enable us to develop treatments with higher success rates.

There is another important factor in the perception, or more accurately the misperception, that healthcare innovation is somehow lacking or declining. High-profile advances, such as the elucidation of the sequence of the human genome, have spawned the usual bold claims about a "revolution" in healthcare. Such overstated claims have plagued every era of medical progress. The fact is, some of these claims will never be realized. But the more gradual revolution we are witnessing will deliver long-lasting improvements in the ability of doctors to predict, manage, and treat diseases. In the end, good science will prevail and we will reap the value from a better understanding of human biology.

As our understanding of the molecular basis of disease advances, it will be increasingly possible to identify and treat a range of subclasses, and the most value from biomarkers will be derived from making better decisions earlier and ensuring that new markers are fully validated. Also, as global regulatory authorities better define their requirements for inclusion of pharmacogenomic data in clinical trials, the industry should continue to incorporate bio-markers in their clinical programs.3

In developing an innovation strategy, companies should focus on quality, quantity and value, and on building target profiles based upon what the organization is capable of delivering. Clear strategies need to be set to achieve these goals. For Roche, our visionary investment in biotech in the early 1990s set the framework for our innovation strategy. We continue to explore ways to maximize value. While we have adjusted our strategy based upon real-time assessments, we remain committed to carrying it out and ensuring that we succeed.

We know that success can only be achieved by hiring the best people, setting clear goals, and giving them the tools they need to be innovative and successful. And, in the end, we must measure, challenge, and improve on the course they chart. If companies do this more effectively, I believe we will see more innovation, and stronger and healthier pipelines resulting in new, innovative treatments in line with the 21st century healthcare vision.

Lee E. Babiss is vice president, Preclinical Research & Development for Roche in Nutley, NJ, where he is responsible for developing and directing oncology, metabolic diseases, and inflammation research strategy. He is a member of the International Research Management Team, which develops global research strategy, the Roche Biomarker Leadership Team, and chairs the Roche R&D Center China Scientific Board.

He can be contacted at lee.babiss@roche.com.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?