the personal views of a doctor in industry

Posts Tagged ‘interpreting data

A Walk in the Valley of Death

with one comment

The NIH has announced quite loudly that it is taking a bold step in helping to find cures for rare and neglected diseases (NIH Announces New Program to Develop Therapeutics for Rare and Neglected Diseases, NIH News). So what do the numbers look like?


According to a recent NIH press release, the atrition rate in the pre-clinical phase of development is so bad that researchers call it the “valley of death”! They think 80-90% of compounds fail to make it to human testing; some inside industry think that even this is low-balling it, and that number is closer to 90% (The NIH Takes the Plunge, In the Pipeline).


Some more numbers from the NIH press release:

  • “NIH estimates that, in total, more than 6,800 rare diseases afflict more than 25 million Americans. However, effective pharmacologic treatments exist for only about 200 of these illnesses.”

  • “Studies suggest that it currently takes more than a dozen years … to take a potential drug from discovery to the marketplace”

  • “it takes two to four years of work and $10 million, on average, to move a potential medicine though this preclinical process”

The WSJ reports that, “Stephen Groft, the director of the rare diseases office, said the program is starting with $24 million in funding this year with expectation of receiving the same amount each year until 2013”

  • USD24 million for 4 years is an investment of USD92 million;
  • Each drug costs USD10 million “on average” to get through pre-clinical development, and so there is enough money to develop about 10 candidates;
  • 90% of these candidates will fail, leaving just one successful compound;
  • With the timelines given, this compound would come to market in 2020;
  • According to Steven Paul, President at Lilly Research Laboratories the chance of getting from phase 2 trials to market is 1 in 8 (Blockbuster Drugs and Innovation) – I cannot find the ph1 attrition rate;


This program has about a 12% chance of resulting in one effective treatment in 2020 for one out of the 6600 untreated rare diseases.


It is a step in the right direction, but let’s not get carried away.



Written by Pillhead

May 22, 2009 at 3:33 pm

Sales of New Drugs – no more blockbusters?

leave a comment »

An article in Forbes (The Value Of New Drugs Is Dropping) cast an eye over the initial sales of new drugs launched in the last decade. The article looked at data from IMS, and found that drugs launched in 2008 sold only a tenth as much in their first 6 months on the market as did drugs launched in the banner year of 1999. Despite the good news that the FDA approved more drugs in 2008 than in 2007, “new medicines contributed much less than 0.5% to the growth of the global market, another way in which the first half of 2008 constituted six of the weakest months in a decade”

At first sight, this article appears to fit our gut feeling that the “low hanging fruit” are taken and few blockbuster drugs remain. It also fits and confirms the results of the drive espoused by several firms for more drugs of middle revenue.

But looking at the way this article presents data looks a little like cherry picking to me. The period of study is 1998 to 2008, but one of the big comparisons is made with 1999 a “banner year”. And using “six” as a statistical cutoff is odd: “2008 constituted six of the weakest months in a decade”?

How bad is it really looking for us from this data?

first 6 month sales for new drugs

first 6 month sales for new drugs

If there is a trend here, and I do not see a clear one, then it is equally valid to say that the last three years have seen a return to form. Never since the hayday of the late 90’s, when pharma launched Singulair, Vioxx, Viagra and Avandia, have we seen sales at these levels. 

I do not think this would be an accurate refection, but it is important not to see what we want to see in the data. 

What I do see in the graph is a remarkably stable picture of first 6 month sales in the region of 5-10 million USD. It would be interesting to look deeper to see why. 

There is certainly no strong indication here that the drugs launched more recently are any different from those 10 years ago. For that conclusion, we need more data.

Written by Pillhead

February 23, 2009 at 1:22 pm

Science Education, East vs West

with 2 comments

I have an unexamined belief that I am a good product from the best educational system in the world. I was educated in the West, specifically in England, and I confess to being a UK education snob seeing the Western system as the best way to prepare a person to solve problems at work, and to be creative. My view is largely supported by personal experience and anecdotal stories (see American Creativity vs. Chinese Skills in Education Week – boy those educationalists can really discuss an issue!)

It was refreshing, therefore, to get cold water poured over my preconceptions in a paper by Lei Bao in the Jan 2009 issue of Science (Learning and Scientific Reasoning). The study tested nearly 6000 freshmen about to commence a physics degree in China or USA to see whose education system was better at preparing them up to this point.

The Chinese system is rigourous and knowledge focussed, while the US system is broader and less focussed on the acquisition of knowledge. Which students would perform better on tests of physics knowledge and problem solving skills?

The results, in graphs below, we striking. The Chinese students trounced the Americans in terms of knowledge (eg 66% vs 27% on the Brief Electricity and Magnetism Assessment, as Bao noted the Americans scored little better than expected by chance). However, the problem solving results were almost identical (the graphs are almost superimposed below).

results graphs

results graphs

The paper alone is fascinating enough, but the most remarkable thing is the way in which so many commentators are so quickly using the results to support wildly different opinions:

“Our study shows that, contrary to what many people would expect, even when students are rigorously taught the facts, they don’t necessarily develop the reasoning skills they need to succeed” – author, Lei Bao, in Ohio State U Research Communications

“… modern science education (at least in the UK) focuses too much on the knowledge and too little on the method” – Ed Yong at Not Exactly Rocket Science

“… this finding might actually apply to evolutionary studies. Maybe just teaching fact, Fact, FACT! isn’t necessarily the best thing for science students” – Thinking Christian

People will always use scientific evidence to support a variety of positions. But the more selfish perspectives should be kept on the fringe. In this case, the centre is being drowned out by the egos of the ones reviewing the data. It has gone almost unnoticed that deep within his commentry, Bao states that, “We need to think of a new strategy, perhaps one that blends the best of both worlds”.

If one removes oneself from the systems under review here, and our assumptions already built around these systems, what conclusion would one draw from the results? The results show that the American and Chinese systems both similarly prepare students with problem solving skills, but that the Chinese students also get a truck-load of knowledge at the same time, which the American students do not. On the basis of this paper, the Chinese system is superior.

So, why is this not discussed at all, by anyone? 

I believe the answer is not savoury. I believe that it goes to our prejudices in favour of what we feel comfortable with because it is what we grew up with. It is not scientific, and goes against the scientific method that we hold dear.

The reason we do not champion the Chinese system comes from outside the paper itself. It is captured in a comment made by one of the co-authors, Jing Han, who says, “To do my own research, I need to be able to plan what I’m going to investigate and how to do it. I can’t just ask my professor or look up the answer in a book”. This is the sort of thing that the net is full of in discussion threads on the subject of Chinese vs American students. It is not pretty.

I think that if we get off our hobbyhorses then perhaps we will be able to move this discussion along a little. We should begin by defining more carefully what it is that we are looking for. If it is graduate students who can come up with their own topics for research then say so. But, I do not think that this is an important criteria for anything of any real importance, however annoying it is.

Perhaps we should leave the last word to Peng Guohui, principal of Jindao Middle School in Guangzhou, China, who was quoted in the Education Week article. He suggests that, “Knowledge acquisition is the basis for creativity …”

Written by Pillhead

February 8, 2009 at 4:48 am