Search This Blog

Wednesday, December 29, 2010

AstraZeneca vs the CCS

In a drunkards walk there was an example of decisions based on improper data sets. A movie producer was hailed as a success after a string of money making films. Then there was a string of less successful films and she was fired. If the data set was taken as one set, the leadership would have realized that the producer was par for the course over of her tenure.

AstraZeneca is going down the same path. The CEO believes in R&D.
Brennan’s commitment to research has its origins at Merck & Co., when the U.S. drugmaker dominated global drug discovery and blockbusters were easier to find. He joined Whitehouse Station, New Jersey-based Merck as a 21-year-old salesman in 1975, and was there when the company’s labs pioneered new ways to treat hypertension and cholesterol.

“This is what David saw happening,” says P. Roy Vagelos, who ran Merck’s R&D and later, when he became CEO, promoted Brennan to run its collaboration with Swedish drugmaker Astra AB, a predecessor to AstraZeneca.


David, the CEO, saw something happening at Merck while working as a salesman. Now he is a CEO and he is going to save the science culture at AstraZeneca. He finds others who have succeeded, so he thinks. Let us refer to the Cargo Cult Science speach once again. The natives observed the Allied forces building and operating their airport. They took notes and replicated the work. How is David saving R&D? By hiring people he thinks have succeeded.

Menelas Pangalos helped build an industry-leading pipeline of experimental drugs at Wyeth, helping persuade Pfizer Inc. to buy the company in 2009 for $68 billion. He was lauded by New York-based Pfizer’s then-CEO Jeffrey Kindler for his “incomparable” expertise.


At 43 Panalos has not had a history of scientific success. He was in the right place at the right time when some "experimental drugs" were put into a pipeline. Where are those drugs now? What exactly did Panalos do?

Like the movie producer, people in positions such as Panalos live and die on the decisions of others. Statistics tell us that someone has to emerge in the position Panalos was in that so impressed David. Together they have gone over the highly unsuccessful R&D branch (the airport) and they have honed it into what they think will work. It sounds as though they've sat in an office and worked out a plan to change the shape of the Cargo Cult Airport Controllers' "two wooden headphones and bars of bamboo sticking out like antennas".

But wait, they brought in a scientist. Martin Mackay is president of R&D. Now we are going to get closer to the problem. In a recent setback the U.K. drugmaker failed to win U.S. approval for a new blood thinner, Brilanta, to rival Plavix, the world’s second-best selling drug. Regulators didn’t ask for new clinical trials for the drug. and “Our highest priority is to provide the requested Plato analyses to the FDA,” Martin Mackay, AstraZeneca’s president of research and development, said in the company’s statement.

So Martins job is to articulate what the company is doing scientifically. The people he communicates with are indeed part of the problem. The FDA, the board of directors... All of these people are convinced that they are the judge and jury of scienticif merit and progress. As we have seen, even their success stories lead to death and/or a lower quality of life. The formula AstraZeneca has come up with puts new symbols into a formula that never calculated anything. It is the logic that is missing.

I would suggest taking one project at AstraZeneca and subject it to the CCS Manual For Discovery. What seems trivial to the big picture guys is actually the core of scientific discovery. Building a solid foundation for a project is non-negotiable. What was once considered trivial must now be written about by the highest members of the R&D staff. It must be clear to the laboratory workers what the leadership thinks is happening. Then give the laboratory staff the opportunity to "bend over backwards to prove them wrong". Most importantly, document everything. In general, I suggest flipping the project upside down for a change. Let the guys in the lab judge the words of the executives. After all, they have scientific backgrounds and laboratories to help them make their points. People in offices have their words and their arrogance. That is the old formula. Nothing has changed at AstraZeneca. The new leadership is focused on dealing with issues left behind by the old leadership. The laboratory scientists are left in their labs to figure out the new structure of discovery while those who created it try to get the FDA to approve Brilinta.

Tuesday, December 28, 2010

The Next Step

Version one of your companies manual has been written. There are various chapters describing all of the groups, their roles and the roles of their members. There are details about communication and who has what responsibility. Now lets focus on what matter most to the company, the product or service. In the case of our fictional company, the product is a service where we help biotech companies stray from the status quo and start making real discoveries by following a path. This path quickly discards things that do not work. The path has the ability to identify things that are working and will work in the real life scenario.

The status quo tells us that PHds are the leaders and those who fell short of the PHd is the followers. However, our company employs purists. Our University system is not a group of vocational schools. If in fact the world were fair, leadership roles would employ the best leaders. If the University system selected for the best of the best, their PHds would compete for and win most of the leadership roles in a biotech company. As it is, a PHd is required for leadership positions. This precludes any chance of a newly minted PHd earning a leadership role in an environment very different from the one they previously succeeded in (college). The PHd is simply given a leadership role instead of being given a chance at earning it.

Let's consider a PHd who is hired to run a protein purification group. Developing methods and using cutting edge systems and leading others are not things taught in graduate school. Quite frankly, most University laboratories can't afford the same technology as industry. What is needed is a specialist. The medical field has long had specialized doctors who require a very different education from other MDs. A cardiac surgeon differs from a Gynecologist. The protein purification group requires a leader who also has a specialized education. There is little chance that a PHd will come out of the university properly educated to perform the tasks required by industry. Therefore the PHd must be taught. The manual will provide a path the new leader must take. He/she must learn how to develop, how the equipment and the operators work, and how to lead.

True leadership will not hire and hand over responsibility. The leaders will hire someone to fill the purification leadership role. They will then provide the education that is needed.

Finally, leadership requires something that most people just assume PHds do well. We assume PHds know how to communicate. We assume they speak and write in full sentences. Give this a test. Secretly record a conversation you have with a friend. Type out the words that are spoken and see how many proper sentences are spoken. The chances are there will be a correlation between education level and percentage of full sentences. But there will be many outliers who must be dealt with. The non-PHd who communicates well and understands methods must be promoted. The PHd who is the opposite must be demoted. This is a business model, not a PHd club.

To conclude todays topic, the University system is not a vocational system. Don't assume the answer to scientific discovery is the PHd. It's what the PHd does that leads to success not what they did in the past. You must provide them with the platform on which they will work. It's your company, not theirs. A biotech company shouldn't be started and left to succeed or fail by people who did not participate in the founding ideology of the company. They are just looking for a job like everyone else. It is the job of leaders to define the job and give the workers every chance at succeeding.

Sunday, December 26, 2010

A Manual for Discovery

How arrogant must I be to propose that I could help anyone discover the truth about anything? But of course, this is just a thought experiment. I'm doing it on a blog that I assume no one reads. So I can do it. I can invent a biotechnology company that does succeed.

We are talking about business, of which I know nothing. But since biotechs are run by people who know nothing of science then I will put together a business plan. I will assume the executive and administrative are simple. I will search for people with the proper qualifications and looks. There will be no board since this company is run on my own vast fortune. Therefore, the executive staff will answer to me. They will all be paid a cut below the scientific staff. A new paradigm for sure. Our mission will be to create a company that has a platform of discovery that can be sold to others. That is our product, a platform for discovery.

Scientific progress is not done by hiring smart people to figure out that which you do not know. You are the ones who start your business. You must hire smart people to follow your plan. If your plan is to tell other people to discover a cure for cancer and the only thing you give them to go on is RNAi technology or monoclonal antibody technology, you are going to fail. If my company is hired, we will run a scenario, start to finish, of your plan. How do you go from a basic concept of using technology X to cure disease Y?

But we are here to help. You will be writing a manual for The Company with our help. Like any act of writing, you can't just start with a pencil and a piece of paper. We'll give you a template. When any employee in the corporate world is told to evaluate themselves, they are given a template. What were your goals? Did you acheive all of your goals? What could you have done better? So our template would help you evaluate your company discovery platform. Fill in your introduction and mission statement. Lay out your justifications and methods. Explain the kind of people who you need and what the scope of their work will be. Timelines? Fill in the data where we have provided the space in our template. All you will need will be the answers to the questions.

It's easier for the board and the execs and all of the potential partners to not have such a template. But the high rate of failure has provided a good reason to rethink the old business model. So step one is to acknowledge that you need a platform of discovery. It needs to be put into writing in a specific way.

Thursday, December 23, 2010

Epiphanies of 2010

I was sitting in the break room when I came across a publication called Pharmaceutical Manufacturing. It was volume 9, issue 1. The article was 'Tech Transfer: Do It Right'. Sounds boring I know but the following excerpt explained the latest debacle of my career. My latest job was abysmal.

There are two groups in biopharma today: those who "get" tech transfer and those who don't. For those in the first group, technology transfer is a mature discipline that follows a structured approach, with predictable outcomes. For those in the other, tech transfer is a perennially new frontier with surprises at every turn.


There seems to be a pattern in biotechnology of people who don't "get it". We all study science but we don't all see the structures that lead to real progress. Most work is begun at the top levels. The executives present the board with ideas for research projects. Drug targets like TNF alpha and Amyloid beta are popular due to the massive profits that can be made off of people growing old and falling apart. TNF alpha is a molecule targeted by several products already on the market. It can be prescribed for aches and pains as well as for cancer. Amyloid beta is the protein found in plaques that form in alzheimer's patients. Executives and board members keep thinking that a molecule can be found that will bind to a certain region of the protein and prevent the plaques from forming. A simple minded approach to a complex problem. Once the executives have made their decisions it's handed off to the next level down, the scientists.

The scientists read up on the literature and go off into the perennially new frontier with surprises at every turn. Junior personnel without PHds generally do the laboratory work and end up taking the brunt of the criticism for ideas that don't pan out. This leads me to the second epiphany I had in 2010.

The Handbook of Process Chromatography is a manual that describes ways in which to develop purification methods. It depicts methods for developing methods. It occurred to me that there is no such book for conducting scientific research. As Feynman said, we assume that we are teaching people how to arrange things so that they get some
wealth in their system. The Handbook does just that. It teaches people how to arrange their process development efforts so that they can get work done as quickly as possible. Each company should have a handbook that describes their process for developing drugs. But there isn't one. They call it creative minds at work. But that is where process chromatography was 40 years ago. They showed up and winged it. Somewhere along the way a small percentage of forward thinkers make progress and set up systems. We have yet to articulate the system of translating ideas from the boardroom to the market.

Wednesday, December 22, 2010

Non Religious Women Have Better Sex Lives

This post is not about religion or sex. I thought of a "research study" that I thought would get peoples attention. Religion and sex are attention grabbing topics. Why not put them together and add in my own bias towards a blissful non-religious world? If I could fund the research, I could find people to do the work, let them know what my expected outcome is, only accept outcomes that fit the title and this headline would make the news.

It takes on the notion that women who are atheists are rewarded with better sexual satisfaction. Religion and sex may have nothing to do with each other however. In statistics we have the example that home break-ins increase in the summer months. Thus one can make the case that as ice cream consumption increases, so does home invasion. We haven't actually said that eating ice cream makes people want to break into a house. The truth is that the weather leads to an increase in ice cream consumption, home invasions and other things like days spent on the beach. Correlation is not causation.

The title does not state that atheism causes a better sex life. It states a possible correlation. It simply says that, as a population, atheist women may enjoy sex more than their religious counterparts. If you put this up on the Huffington Post you will get a spate of responses that have little to do with defending the science behind the study. People will respond to the title of the study. Religious folk who are open minded about sex will shout it down by sharing their own sexual outlook. Religious folk who are not open minded about sex may condemn the study as being biased. Atheists will say that they knew it all along. The scientist however, will respond to what they feel the data is depicting. Maybe the study was flawed and the religious group of women were all catholic nuns and the atheist women came from the local strip club.

If Dr. Ioannidis is correct and roughly 90% of the medical research that doctors rely on is misleading, exaggerated, or flat-out-wrong, then the question is why. Why do we as scientists select for research that is interesting as opposed to being true? We know that what is true should trump what is more interesting. It should be just as interesting to the scientific community to find out that there is no correlation between religion and sex.

Feynman touches on this with the example of the scientist who tested the maze used in mouse studies. The scientist tested the test (the maze) and found that there was an alternative explanation why mice do what they do inside a maze. The sound or texture of the floor below them led them to the door where food was previously found. The science is in telling the mouse testing world that they have to control for extraneous signals that will confound their research. That should be of interest to scientists.

Getting someone to listen to you is what anyone who desires to move up the ranks of their field needs to do. The trick in science is to get people interested in simple truths. If a study is interesting enough to get funding, it's interesting enough to be published. Imagine a science journal that publishes experimental design then later publishes the outcome. Reviewers of the work have their reviews published along with study. Review the proposed study. Review the work. Review the conclusions. Then review the reviews. Give researchers attention before during and after their work and they'll go back to pursuing what is true versus what will produce most eye catching headlines.

Monday, December 06, 2010

Five Years In the News

2010 Nice review
2009 BOOM
2008 ZING SON OF A...
2007 BLA-ZAMM
2006 OUCH
2005 POW

Okay, I'll stop at five years gone by. You get the idea.

A new decade is soon upon us. What will come? In the last decade we had a net loss of about 3500 jobs and an approximate 60% drop in market cap valuation. How shall we measure success from 2011 to 2020?