Magazine Article | October 30, 2014

GlaxoSmithKline Refocusing, Reshaping Oncology R&D

Source: Life Science Leader
wayne koberstein

By Wayne Koberstein, Executive Editor, Life Science Leader magazine
Follow Me On Twitter @WayneKoberstein

Soon after my first conversations with GlaxoSmithKline about interviewing its head of pharma R&D, the news hit. In a mega-swap of product lines with another global pharma giant, GSK gave up its entire commercial oncology portfolio to Novartis in exchange for the Swiss company’s vaccine line; at the same time, the two companies combined their consumer health units into a single business with GSK as the majority owner. Oncology was no small part of its drug division at the time, accounting for about 4 percent (about $1.5 billion) of pharmaceutical revenues before the Novartis deal.

Early on, it took reading past the headlines to grasp GSK’s full intentions. Essentially the company is refocusing its oncology R&D, all the while continuing to reshape the entire pharma R&D organization. The leader in charge of the transition is Dr. Patrick Vallance, president of Pharmaceuticals R&D. A former academic researcher, Vallance headed drug discovery before taking on the entire pharma R&D unit in 2012.

Thus he has witnessed the good, bad, and the less-than-pretty results coming from the organization during his eight years with the company. After a comparative dry spell, GSK staged a recovery in drug development in recent years, with six new drugs entering the market since 2012.

The company’s bonus program for discovery scientists, giving a small incentive at proof-of-concept and a bigger payment at approval, turned out to be a good surrogate marker for the new-drug surge. “Medicines can fail at any stage, so we didn’t want to give big rewards at the end of Phase 2a; we wanted to signal there was a big reward to be given at that point, which would come when the medicine becomes a medicine, and we paid up on that promise with the product approvals,” says Vallance.

Those successes, however, have not stopped the R&D transformation already under way — and, in fact, the new products are more reflective of the company’s past than predictive of its future direction in drug development. All of the recent approvals are in GSK’s established market areas, mirroring their relatively broad focus: COPD/asthma, diabetes, HIV, and cancer (melanoma), with the two molecular-pathway targeted drugs in the last category joining the exodus to Novartis. GSK’s current pipeline reveals that the company, not unlike most of its competitors, is heading into niche-drug territories, driven and guided by a careful reading of scientific opportunities and the reimbursement landscape.

At the same time, as heralded by Chairman Andrew Witty, GSK will seek to lead the industry in the trend toward sharply reduced R&D budgets, attendant cost-cutting, and collaborations with all sorts of research entities in the quest for biotech-like innovation — without, of course, incurring the usual, though oftignored high failure rates of the startup biotech sector. To reduce the risk of duplicating the darker side of entrepreneurial biopharma, the company aims to make the best possible use of its natural assets, the “platforms” of drug discovery, development, and commercialization that only a Big Pharma like GSK can maintain.

It is useful to know how and why Vallance joined the company, because his responsibility ultimately came to encompass both affirmation and optimization of the organization he would head. He was well along in an academic career at University College London, when he met GSK’s then head of R&D, Tadataka (Tachi) Yamada, and subsequently answered Yamada’s invitation to join the company in May 2006.

"We had to reintroduce personal accountability and individual leadership."

DR. PATRICK VALLANCE, President of Pharmaceuticals R&D, GlaxoSmithKline

“If you’d asked me the day before I decided to move to GSK, I would have said I absolutely had no intention of moving to industry,” he recalls. “I was a clinical academic. I saw patients. I ran a big department. I knew my next job would be running the medical school. I had been on the research advisory board for GSK for a couple of years, which was eye-opening for me both scientifically and in the way people thought about treating major medical problems.”

After an advisory board meeting in London, Yamada asked Vallance to head GSK’s drug discovery. Resisting the offer at first, Vallance later thought, “I could spend the rest of my career doing academic research that might lead to drugs or even, as I was doing at the time, making molecules that interfere with biological processes — or I could go and do the real thing, seeing molecules all the way to becoming medicines. This looked like a really big, interesting career change, and I decided to do it. It was a decision with no great planning or logic behind it and one that I have never regretted for a single second since then.”

His initial revelations at the company were mainly positive: “One of the things that struck me was the outstanding quality of the science that goes on and the attention to data reproducibility and integrity, a huge issue in many academic sectors and in the published literature. The dedication to more reproducible assays was very impressive. It was important for me to get to grips with what it means to have robust and reproducible data at this scale of R&D.”

Scale itself — the industrial scale of discovery and development activities at the company — also impressed Vallance, and at the same time led him to ponder an inherent conundrum in the large-pharma model. “The number of projects is big, and of course, that’s good in one sense. But it leads to a question: What is the depth of understanding underpinning every project?” That thought became his mission: finding a way to increase scientific understanding of drug mechanisms and drug interaction with the human body, so the company might focus on fewer projects in greater depth.

Vallance saw a potential opportunity in the industrial setting to create an advantage that eluded the academic world. At its best, GSK allowed all the different R&D disciplines — chemistry, biology, biopharmaceuticals, and so on — to interact in “an incredibly fluid way,” he says. “Getting all of those elements together to work as an integrated, multidisciplinary team is frankly the dream of a lot of people in academia, but it only seldom works.” Again, over time he looked for a way to make the concept even more effective in practice.

During the six years following his move to GSK, Vallance rose from head of drug discovery to senior vice president of medicines discovery and development, helping plan and implement the first major R&D reforms driven by Witty and Moncef Slaoui, now chairman, Global R&D & Vaccines. Those reforms began in discovery, involving the creation in 2008 of about 40 teams, consisting of 40 to 60 members, called Discovery Performance Units (DPUs), out of the six former disease- area units. The DPUs were designed to integrate disciplines even more effectively than did the preexisting company culture. They were also to echo a new “open-research” philosophy of external collaboration, by re-creating start-up-style innovation internally.

When Vallance was appointed head of Pharmaceuticals R&D in January 2012, he continued to lead reforms reaching past discovery through all the stages of drug development. But in addition to the internal restructuring, GSK also began a campaign to shift much of the R&D burden and risk to a massive external network of companies and institutions. Overall, Vallance believes the organizational changes have reinforced a more bottom-up, scientifically driven approach to new drug development, moving the company away from the old top-down, market-driven paradigm that held sway for so long in the industry.

“We now have the freedom to discover on the basis of science, which makes it possible to make surprising new discoveries rather than the ones we wished and expected to discover. You have to build new medicines from what is scientifically credible, rather than go where you see simply a big market. Of course, the scientific line must be linked to a vision of patient need, but if you are completely focused on the market from day one, you close things down — and if you start by aiming at where you see the biggest dollar sign, you will end up in a weird place scientifically and medically.”

As the mechanics of GSK’s R&D reorganization continued to evolve, all the related efforts conformed to the two key principles employed by the DPUs, according to Vallance — personal accountability and openness to external collaboration. Both principles were reflected in progressive structural changes.

“We had to reintroduce personal accountability and individual leadership so that we didn’t end up with everything being a total team effort and nobody really taking accountability for projects,” he says. Personal accountability called for smaller, integrated teams. It was not always as painless as it sounds. Teams that underperformed were cut, with their funds, and often their personnel shifted to new teams. Some long-term people prospered; others found themselves demoted or worse.

With the startup of the DPUs, the company eliminated some therapeutic areas in R&D, such as some late-stage neuroscience work, and reportedly made around 3,000 R&D-related job cuts worldwide. Besides respiratory, the company’s largest market area, GSK’s current pharmaceutical pipeline lists candidates in oncology, respiratory, cardiovascular/metabolic, immuno-inflammation, infectious diseases, ophthalmology, neuroscience, and various rare diseases. DPUs are also subject to periodic pruning. In 2012, following a review of all DPUs, the company cut three of the units, reduced funding on five, and raised funding on six.

Thus, it is apparent DPUs compete for budgets and must manage their finances like an independent company. They are judged on their outputs by a panel that includes external advisers as well as internal experts from R&D. “The change to DPUs has been an extremely successful path for us,” says Vallance. “It has created what I call integrated drug discovery. When I joined GSK, most discovery people were either running a chemistry line, a biology line, or a clinical development project. What the DPUs have done is create individuals who run virtual biotechs. The units have been very successful in the integration of disciplines and progression of projects in a focused way, rather than a reliance on the volume or number of projects.”

Receiving the nascent medicines from the DPUs are Medicines Development Teams (MDTs), which take responsibility for getting the surviving product candidates through late-stage development. Even smaller than the DPUs, the MDTs each own an individual project in the latestage pipeline with “a lot of accountability, a lot of ability to do things the way they want to do them, and a lot of team responsibility for results,” Vallance says.

On an even wider scale than the internal reorganization, the externalization of research has brought extensive restructuring affecting several thousand more jobs and careers. The company saw the strategic changes as necessary to break the ice jam holding back pharma R&D productivity for GSK specifically and the industry in general — in any case, forcing the large companies to obtain most of their winning new drugs from the entrepreneurial sector. “Underpinning everything are strategies to diversify through business development and increasingly through links in academia,” says Vallance.

In Vallance’s view, the final leaf in the table, ace in the hole, winning play, or what-have-you is the leverage of using the unique platforms a Big Pharma brings to the table — small molecule and/or biopharmaceutical production, clinical trials management in multiple countries, a highquality regulatory organization, reimbursement planning, and other resources tasked with supporting the small R&D units. “The platforms exist to support the projects and not the other way around,” he says with emphasis, highlighting a role reversal from the industry’s traditional platforms-drive-research paradigm.

Six years into GSK’s R&D restructuring, another earthshaking change now comes to challenge, or perhaps liberate, the organization. As proposed, the Novartis deal redraws the landscape of therapeutic areas and destinations in GSK’s pipeline. As of press time, the regulatory process is still unfolding and it is too soon to describe the agreement as final, but the general outlines are clear: GSK’s vaccine business, separate from the pharma unit, will of course grow larger, but the entire commercial line of cancer drugs will go to the other company.

Aside from the corporate impacts on critical mass and market portfolios, the deal’s effects on R&D may be dramatic. In oncology, the organization will no longer focus on postmarket development but must realign to an early development, pioneering mode in epigenetics and immunooncology. Why does the company want to abandon even a nominally successful product line to chase after therapeutic approaches still as unproven as they are far-reaching?

Vallance explains that, despite having a good track record in oncology discovery, GSK ranked only about 15 in cancer drug sales, and though it might aspire to the top 10, it would likely never reach 1 or 2. Moreover, the drugs the company had commercialized would continue to demand more and more support with expensive postmarketing studies.

“The deal allows us to go back and focus on the earlier areas that have excited us, epigenetics and immuno-oncology, to make sure that we really invest in those properly. If all goes well, they will become anchor areas for the next wave of products in the pipeline, which at some point will take us back into oncology commercially or make us partner of choice for Novartis or another company.”

He says the oncology situation illustrates another principle that Andrew Witty has advocated: “We will discover what we discover, and we will develop the medicines we want to develop. We won’t always be the right company to commercialize a product, but we must be sure to apply our focus and our size appropriately.” Such a flexible strategy seems well-suited to the small-team, entrepreneurial approach of DPUs and MDTs, where risk may be more quickly recognized and, hopefully, mitigated.

So is GSK starting over in a totally different way with oncology? “That is exactly what we are doing,” Vallance replies. “Immuno-oncology and epigenetics are likely to require greatly different development tools and a different way of thinking about development, and exploring the new areas gives us an opportunity to do just that. Increasingly, one gets drawn into more and more combinations of therapeutics with a variation of MOAs (mechanisms of action). This gives us a chance to rethink things. No one knows yet exactly how to develop an immunotherapy or epigenetics drug. We don’t even know yet how we should be thinking about the ways to deliver the drugs, what the safety profiles should be, and so on. All of it will require a whole new look at the organization.”

Some of the products in GSK’s pipeline may initially target rare diseases or narrow indications, yet turn out to be applicable to much broader treatments. In immuno-oncology, certain MOAs now in testing seem to work equally well in many different kinds of common tumors as conventionally defined, by organ. (See also Part 3 of "Combination Cancer Immunotherapy — A Virtual Roundtable,” on page 36.)

Vallance gives an example in another area, an ex-vivo stem-cell/gene therapy in Phase 3 for a rare, fatal childhood disease called adenosine deaminase severe combined immunodeficiency (ADA-SCID). Although the disease is extremely rare, the therapeutic mode has far-reaching implications, he says. “Cell/gene therapy is where we need to be because it allows us to really understand how to apply a new technology. We may ultimately have broader applications to much more common diseases. That’s exactly the question going on at the moment in oncology, and it places an emphasis on epigenetics — whether immunotherapies will also become fragmented as we realize that patient populations respond very differently depending on their epigenetic profile.”

If by some circumstance the Novartis deal does not play out as planned, GSK’s intentions in oncology will be nonetheless revealed. It will, by one measure or another, pivot its R&D focus from older cancer drugs and MOAs to the cutting edge of cancer therapy, based on the solid, but always shifting, ground of scientific understanding.

One aspect of R&D the Novartis deal does not change, at least in direction, is GSK’s externalization strategy. To the extent that the pharma R&D organization feeds the Rx-to-OTC pipeline, it will gain global scale and strength from the ex-Novartis consumer business. But the deal will not impede the growth of GSK’s worldwide collaborations with companies and academics. Other initiatives, such as the Oncology Clinical and Translational Consortium (OCTC), an international, collaborative research network of six major cancer centers, will also continue unabated. (See “Lynn Marks: Accelerating GSK’s R&D Innovation” on page 27.)

Summed up, the combination of entrepreneurial internal teams and external partners builds on a distributed research model. “What we won’t do is build more brick-and-mortar innovation centers across the globe. But we are forging strong alliances with biotech and, all importantly, with academia worldwide. We have some interesting ways of attracting academics leading research in our areas of interest.”

Discovery Partnerships in Academia (DPAc) is a global program that builds virtual companies around selected ideas submitted by academic scientists or labs based on preclinical or early clinical results. “We place a senior drug discoverer from GSK with the academic, agree that together we will make a medicine, and the whole thing is milestone-driven. They get the keys to GSK, they get a topclass team on their project, and they can publish everything.” If a project must end because of technical roadblocks, the academic partners can keep the data and continue to work on the idea with another academic or company partner.

“The DPAc program has formed a series of virtual biotechs across the globe, the most advanced of which is now in Phase 3, and it has worked really well as a system,” Vallance attests. “We now have an office in San Diego where we’re forming more such alliances. We’ve done partnerships with groups like Avalon Ventures to create new venture partnerships with academics, and we have opened a new space in Boston. The program is about making links with academics rather than putting up big new laboratories. In fact, we believe that the brick-and-mortar facilities tend to end up being fortresses rather than places where true collaborations occur.”

GSK has now traveled quite a distance in reforming the R&D organization — far enough, in fact, to have tracked and measured the payoff in improved performance. Every year, the group publishes its internal rate of return, comparing it to a target rate of 14 percent. The rate has climbed about one percentage point per year since 2010 to reach 13 percent in 2013. The period fairly well conforms to the time when most of the reforms took effect while sales fell, so, objectively, the evidence suggests the reorganization is working.

"Medicines can fail at any stage, so we didn’t want to give big rewards at the end of Phase 2a; we wanted to signal there was a big reward to be given at that point, which would come when the medicine becomes a medicine."

DR. PATRICK VALLANCE, President of Pharmaceuticals R&D, GlaxoSmithKline

The company appears to be alone in revealing its return on R&D. In an industry dominated these days by real-time stock trading, the multiyear measurement alone may set GSK apart from its market-jittery peers.

“One of the principles that is critical for R&D is to stay true to an approach long term,” says Vallance. “The tendency to change bets every couple of years in a business where it takes 10 years or 15 years to develop a drug is terribly counterproductive, so we are very happy with the overall structure we have built and with the way the returns are going, looking long term. I believe we’ll stay with that strategic approach.”

Many years ago, a young writer/editor waited for hours in a hot room outside the chairman’s London office. The company was Wellcome; the chairman, Sir Alfred Shepperd. On another day, the same journalist sat in SmithKline Beecham’s London-suburb headquarters speaking with CEO Jan Leschly, and on another, in North Carolina’s Research Triangle Park with Ernst Mario, CEO of Glaxo. All of those companies and more became the components melded into GlaxoSmithKline over a long journey of many steps.

GSK faces another long road ahead as it continues to tweak and put its new R&D structure to the test. Execution will prove more important than theory. At every level of implementation, from the DPUs to the partners and suppliers, the returns will rise or fall on the choices the company and its leaders make with every step.


Reporting to Patrick Vallance, GSK’s head of pharma R&D, Lynn Marks is a senior vice-president in charge of a group called Projects Clinical Platforms and Sciences (PCPS). His group’s responsibilities cover the conduct of all of the Pharma R&D Phase 1 through Phase 4 clinical trials, with rare exceptions. PCPS has a staff in more than 40 countries and is one of the company’s “platform” organizations. In addition to clinical trial operations, its functions include business IT support provided by a PCPS subgroup, Business Planning and Performance. In parallel to PCPS, a closely related nonclinical platform group covers clinical trials drug formulation and supply. Marks also sponsors the company’s “Simplifying Clinical Development Change Initiative” and is corporate secretary for TransCelerate Biopharma, the trans-company collaboration to improve efficiency and productivity in clinical development.

“All of the companies in this industry realized five to 10 years ago that they were spending way too much money for the deliverables of a new medicine launched into the world,” says Marks. “All the change initiatives that companies did — we called ours Simplifying Clinical Development — was to look at ways to reduce costs, increase quality, boost efficiency, and simplify the processes across clinical development. So we put our program in place about five years ago, and we have tracked the magnitude of savings and monitor the quality of our trials.”

Having accomplished its mission, with projected savings of “hundreds of millions of pounds across that time frame,” the initiative comes to an end later this year. But Marks says initiatives such as Simplifying Clinical Development also helped spark the thinking that led Global R&D Chief Moncef Slaoui, along with Patrick Vallance and other industry R&D leaders, to begin building what turned out to be TransCelerate Biopharma — and attract 18 other, mainly midsize and large pharma companies into the not-for-profit collaboration to date. Marks worked with R&D executives from other companies to get TransCelerate up and running, initially chaired the operations committee, and now serves as secretary of the organization.

“Dr. Slaoui was very keen on the idea that we could work in collaboration in a precompetitive way on how to increase quality, decrease costs, and increase efficiency in the clinical development space. So I was brought in very early in those conversations in regard to how my organization could participate,” he says.

One area TransCelerate took on was data standards, with CDISC, a dictionary of terms associated with therapeutic standards that all member companies use when they report their information into regulatory agencies. The member companies have since put additional resources into CDISC to accelerate the development of therapeutic standards, according to Vallance. “Our goal was ‘55 in Five,’ or introducing 55 new therapeutic area standards in five years. We didn’t want to re-create something that was already in place but rather get our resources aligned and make clinical development move faster and more efficiently.”

The obvious and oft-asked question about TransCelerate is why collaborate with a set of companies that are otherwise competitors in the industry? Marks describes the dynamics of collaboration. “We are fiercely competitive, and we have to be ever-conscious of anti-trust issues, so we make sure we’re clear of all that by having active involvement from our colleagues in Legal. We look for areas where we want to advance something and believe, as a group, if we work together, we can actually advance it faster. Good clinical practice training was one example in which we could make the lives of our investigators around the world simpler. We decided to train them with an agreed set of fundamentals on good clinical practices, which each of us would have done independently and redundantly in the past.” TransCelerate is also working to build an electronic portal through which all 19 companies would communicate with investigators globally.

Comparator drugs for clinical research is another area the group took on — to the surprise of doubters, Marks adds. “In the old world, if we wanted to run a clinical trial using another company’s marketed product, we would often go through a third party that would buy it in various countries and ship it to us. Now we have multiple agreements among companies to buy directly from each other, so we know the pedigree and the characteristics of the product and can ensure a high-quality, reliable source of the clinical trial material. I am particularly proud of the team for this because many people thought we couldn’t do it.”

This industry “League of Nations” gives new meaning to the popular term “partnering.” No longer is a partner just a license holder or trusted supplier; now the term may apply to a lengthening list of collaborations, including the “precompetitive” tide that lifts all boats.

“The space of where we can go is only limited by our imagination and our appetite for precompetitive collaboration as an industry,” Marks says. As an infectious disease specialist, he sees antibacterials as a key example that such cooperation can have a big impact. “It is one of those areas in which we will have to call on the collective leadership of government, academia, and industry, and we will have to break down the barriers across companies to improve our response to the growing threat of antimicrobial resistance. How can we advance that kind of collaboration across the industry collectively?”


As vice-president of Worldwide Business Development, Damien McDevitt is also site director rolling out GSK’s new U.S. west coast office in San Diego, planned as a center for company partnering, collaboration, and funding for innovative ideas in the life sciences. He speaks to us here about how the new office will operate, some of the implications and advantages of the west-coast location, and GSK’s collaboration with Avalon Ventures to tap into the region’s scientific and entrepreneurial environment through direct financial and practical support for startups.


MCDEVITT: Yes, I’ve got a number of roles. I head up business development for some of the areas within R&D and also for R&D Asia, and I combine that with my new role to help set up and run this west coast R&D site.


I’ve been in business development for 10 years doing deals with companies on the west coast and I’ve always wondered, why don’t we have a physical presence on the west coast, given the tremendous amount of innovation here? Just over a year ago, when we did our deal with Avalon Ventures to build 10 small biotech companies together in San Diego, it was the catalyst for saying, we need to have a presence on the west coast and that presence should be based in San Diego, where we can work closely with Avalon Ventures and our new, innovative deal model: a pharma company and venture group building companies together. If they work out, GSK will be able to buy the companies at agreed rates.


Clearly, partnering with biotech and academia is a core part of our strategy, and a chief reason for our success. Many of the programs that we have in development and products we recently launched have come through the partnerships, and we believe the next generation of medicines will come from partnerships. We now have more than two dozen partnerships on the west coast, and our next task is to identify and develop new relationships. We are looking for new technologies for our manufacturing and clinical support platforms, and new therapeutics for the pipeline as well. We are focused in and around preclinical programs right now and would like to add some more to the pipeline. The Avalon Venture partnership is going very well, and we’ve announced three startup companies: Thyritope Biosciences, Silarus Therapeutics, and Sitari Pharmaceuticals.  By the end of the year, hopefully we will have one to two more companies to add to the three we have set up already.


The innovation is typically in the university setting, and we work with the university to build a small company around that innovation. We and Avalon both need to like the idea before we start the company, and then Avalon and GSK between us will commit $10 million worth of Series A investments into the company and set up the collaboration. All of us will work together to develop a candidate, and GSK has a built-in, exclusive option to buy the company at a pre-negotiated price. We house the company in the Avalon incubator here at Torrey Pines, and then if it’s successful in producing development candidates, GSK will buy the IP from the company and bring it into the R&D pipeline.


Yes, we are very close to SR-One and are actually now in a discussion with them on a joint project. They obviously have a network out here on the west coast as well, with the office in San Francisco.


Most of the innovation in the Bay Area, San Diego, and Seattle, has been in early stage biotech, but the local industry has now matured significantly. Like other companies before, we would always do a GSK west coast trip with our execs once a year, visit a bunch of institutes and companies, and do our deals. But now the big companies are willing to make significant investments in facilities and establish a presence. GSK, unlike J&J and Pfizer, is not building large infrastructure; we’re much more interested in building relationships, but we are just bringing it to a new level here, being on the ground, being closer, developing stronger relationships, and so on. People treat you differently when you’re local.


There is a collision now of medicine, engineering, and technology in California, which is an advantage to our bioelectronics programs, for example. The engineering innovation going on in Silicon Valley and in San Diego will help inform how we think about making medicines in the future. If you look at manufacturing in many large pharmaceutical companies, a lot of the technology is actually quite outdated — it could be 30, 40, 50 years old. But at GSK, we have an initiative to augment and improve the technology in our manufacturing processes, so we also are looking at innovative companies out here on the west coast that can help us bring our manufacturing technology up to 21st century standards.


Everyone was delighted that GSK’s in town. They want to meet us, to hear about the Avalon Ventures deal, and to hear about what we have taken an interest in today. It is a relatively small, collegial community, and everybody seems to be rooting for each other. There is a lot of innovation happening in San Diego and in San Francisco, and there’s a lot of opportunity out there, probably even more than I expected.


We have been active in partnering there for 30 years with scientists at places like UCSD, Scrips, and Sanford‑Burnham, as well as up the coast to Stanford, UC Berkeley, and so on. We have been very successful working with the technology transfer offices in these universities, and so that’s really helped GSK here. The university and institutions are even more sophisticated now, I would say. They are beginning to understand that it’s still a long, long way to go with their innovation from discovery to practice, so they are a bit more reasonable from a business perspective in how to establish relationships, be productive and deliver medicines to patients.