Pharma Strategy Blog

Commentary on Pharma & Biotech Oncology / Hematology New Product Development

The title of this post today is inspired by one of my bioinformatician science buddies on Friendfeed, Neil Saunders, who has a great blog that’s worth checking out.

Here’s a wonderful simplified picture of many of the pathways thought to be involved with different types of cancer and was shown by Dr Wafik El Deiry of Penn at the recent AACR meeting on colorectal cancer in Philly:

Source: AbCam (pdf download)

Imagine all those pathways that are overexpressed in any given cancer, some may well be mutated, most will be passengers, a very few will be actual drivers.  Now imagine that all of them are lit up like a Christmas tree.

Not so easy to see the wood from the trees now, is it?

Think about drug development… how many people rush off and pick off one target and take a targeted therapy, either a monoclonal antibody or kinase inhibitor, combine it with standard chemotherapy and think it possibly might work, let’s suck it and see.

Ummm, no.

What are the chances of such a random unscientific approach actually working?  Pretty low.  Then we sit back and realise that the old models (animal, clinical etc) aren’t working any more and such an approach is no longer sustainable.  The drain on resources, whether time, money or people is too high.  The phase II/III attrition rate with that throw the mud at the wall approach is horrendous in oncology.

There is another way.

The smart researchers and companies are now using more modern, highly evolved animal models, doing more extensive preclinical research and thinking differently using a more holistic systems biology approach.  They’re trying to figure out what the logical drivers are, which might suggest some logical combinations (think two unapproved targeted agents with an approved one perhaps) based on the constitutively activated  or mutated targets, cross-talk, feedback loops and going into phase I research later with a more solid rationale.

The winners in this will be the companies with the most useful and broad pipeline who can mix and match more easily in this strategic pathway rather than tumour approach, with their own compounds than someone else’s.  Collaborations on the pharma side are still relatively few and far between.  They are often also a nightmare to manage, no matter what the original intentions were.

Change is already happening judging by the many enlightening conversations I’ve had this year with academic and pharma researchers, clinicians and commercial clients alike.  This is great news and it’s driven by a greater understanding of basic research, better animal models, a panopoly of potential druggable targets and a broad, deep pipeline across the oncology companies as a whole.  We just need to start putting the jigsaw together now and maybe we’ll see a difference in outcomes in the not too distant future.

The future may not be so far off as we think.  The future is now.

10 Responses to “What you’re doing in cancer research is rather desperate”

  1. Claudia Donnet

    Wonderful post. Glad to know that the concepts of complexity and holistic systems biology are slowly trickling down throughout the pharmaceutical industry. Mathematical modeling using the information we already have about the cancer-related signaling network also becomes a powerful tool to test these multi-point attack strategies before even trying on preclinical testing.

  2. Alexey Bersenev

    Great post Sally!
    I believe that this pathways map is simplified. it gets more messy in reality. And we are rather desperate and lost. That’s why I don’t like approaches targeting particular pathways. Can you imagine if in one particular cancer 30 of them on and off and this whole machine is evolving in progression? What should we target?
    One of approach I like is non-specific: Let immune system to do the job – eliminate mutant cells efficiently.

    Btw, what new smart animal models are you talking about Sally?

    • maverickny

      Hi Alexey, Thanks for your kind comments.

      Regarding the animal models, I was referring to, for example, ones from Tyler Jacks’ group at MIT or Jeffrey Engelman’s PI3K model.

  3. Irfan Alvi

    For my perspective on the cancer problem, see:

    Also, I recently did a PubMed search and found that the systems biology approach to cancer first appeared in the literature in a substantial way around 2006, and the number of papers has been gradually increasing every year since then. Focus more specifically on the complexity of cancer has been more recent, more like the past year or two.

    • maverickny

      Good points, Irfan. I’ve noticed the shift over the last two years, too. You can see the trend emerging from AACR as they organise their sessions and posters around pathways and targets whereas ASCO and ASH (sadly) still do sessions by tumour type.

Comments are closed.

error: Content is protected !!