AFG Venture Group Dispatches

Corporate advisory and consultancy in Australia, South East Asia and India.

Making A Difference

Ian Dixon
Founder and CEO, Genscreen Group

Every day I seek to use my own skills – and those around me – to make a tangible difference to the lives of others – and make some money when we are successful. My chosen vocation is the development of human therapeutics (and diagnostics) to solve serious unmet medical needs – potentially one of the most challenging fields of endeavour a 54 year old lapsed-engineer could select.

Let me explain the challenges, in a way that I hope gives you some insights . . .


Every year, the FDA approves around 20 (21 in 2010, 25 in 2009 and 21 in 2008) truly new “drug” treatments under its New Drug Application (NDA) / Biologic License Application (BLA) system – an amazingly small number when you think about it. The USA is the world’s largest market for drugs, so drugs addressing significant needs and also sold in Europe (under EMEA) and other markets will tend to come to the USA.

Human afflictions

My colleague Dr Ian Smart (an immunologist) explained to me many years ago that there are three types of diseases. The first type is named after the cause, where the cause is well understood (e.g. Influenza). The second type of disease is where it is named after the symptoms – because the cause is not known (e.g. arthritis, multiple sclerosis, cancer). The third type of disease is named after the first person to identify the disease or a person of note who suffered (e.g. Alzheimer’s Disease, Meniere’s Disease, Lou Gehrig’s Disease). In the third type of disease we are missing clarity on a few key things – what causes it and a firm set of symptoms. Every now and again a disease shifts from third-type to the second (Lou Gehrig’s Disease has become amyotrophic lateral sclerosis, or ALS) but that may not signal a curative success. By now you will have concluded that having a disease named after a person is not a good plan. Unfortunately, even the diseases named after the cause – e.g. HIV-AIDS – do not always quickly yield to the efforts of medical science.

Most diseases are not simple

There are actually very few diseases that are simple – and most of those we have stopped talking about years ago because the answer is on the supermarket/pharmacist’s shelf (e.g. headaches). There are around 1400 listed diseases with genetic linkages but often the genetic association is loose (e.g. Schizophrenia) and science struggles to explain why some people are afflicted whilst genetically similar people are not. Some notable genetic diseases have a simple origin that we can explain well – e.g. Down syndrome is associated with duplication of chromosome-21 – but often the actual symptoms, key mechanisms and target proteins involved in the affliction are still unclear. We all know of aged smokers who live to a ripe old-age seemingly impervious to the self-inflicted dangers. We also know that only a small percentage of those exposed to asbestos suffer the fatal consequences of mesothelioma.

Human biology is complex

Without attempting to explain human biology in this forum, it is important to understand that our human body is made up of around 100 trillion (1014) cells of around 200 types, around 20,000 protein-coding genes, 206 bones, 600 muscles, and 22 internal organs. Once we start to look at the surface of the cells or the inside of the cells we see complexity and highly-interactive factors that include cytokines, enzymes, kinases, cytoskeleton, receptors, ligands, DNA, messaging RNA and more and more. The key point here is that biology is a complex dynamical system – full of redundant complexity and interactions (edges) that provide feed-back and feed-forward interactions that we are yet to fully understand.

Complex – dynamical systems

Obviously, the complex dynamical system named Angelina Jolie has got many things right. A feature of complex dynamical systems is that they are very robust and able to handle big changes (e.g. onset of viral infection) through the involvement of multiple modes of response (e.g. humans deal with viral infection through a complex and interactive cascade of Compliment, innate immune response, T-cell adaptive immune response and B-cell adaptive immune response made up of multiple cell-types, cytokines etc.).

The flip-side of complex dynamical systems is that they are remarkably difficult to manipulate – their powerful features of stability and flexibility becomes a problem when biotech people (like me) want to change the status of the system (from what has become the disease-state norm to a new non-disease-state) through an intervention (e.g. a drug or a medical device). Many diseases take decades to become the norm (e.g. obesity, diabetes type 2 and cancer) and so have become the “attractor” (or set-point) of the person’s complex biology.

As a rule of thumb to shift the attractor (normal state) of a complex dynamical system requires a shift affecting around 10 – 20% of the edges that affect that attractor. This means that an intervention must either be (i) very broad acting (e.g. a “dirty” vaccine) or (ii) target a pivotal “edge” in the complex dynamical system (often easier said that done).

Where treatments come from – history, luck or rational drug design

Back in history many cures were found by trial and error or luck (or desperation). Salicylic acid (aka aspirin) comes from the bark of a tree and has been used since around 400BC to treat pain and headaches – well before University degrees and white coats.

For complex diseases, the more recent approach is called “rational drug design”, based upon the idea that one can derive a useful and safe therapy by progressing through the rational process of (i) understand the cause of the symptoms (ii) identify the errant “messenger” (e.g. receptor, ligand, gene) (iii) find a way to modulate the errant “messenger” (e.g. small molecule inhibitor of ligand-receptor binding) (iv) develop the agent into a safe drug or device (v) conduct clinical trials to show safety and efficacy and (vi) put the treatment into commercial use.

MOA (Mechanism of action)

Rational drug design certainly appeals to scientists and engineers – but it isn’t plain-sailing for aspiring Nobel Prize winners.

Mechanism of action (MOA) is a buzz-word used in many scientific meetings. MOA is really understanding the cause of the symptoms, identification of the errant “messenger” (e.g. receptor, ligand, gene) and then find a way to modulate the errant “messenger” (e.g. small molecule inhibitor of ligand-receptor binding).

As we saw earlier, many diseases have vague or ill-defined symptom sets – often with person-to-person variation. In the case of cancer the symptoms are relatively clear – uncontrolled growth of cells (of some type) and then spread of nodes of uncontrolled growth of cells – yet the MOA to chase after is less clear.

Unfortunately, finding “A” MOA and “THE” MOA can be two different things.

As an example, stomach ulcers were “known” to be caused by excess stomach acid and “known” to be treated by antacids – a large over-the-counter drug class. Then along came Drs. Barry Marshall and Robin Warren who challenged the accepted MOA, proposed an alternative MOA (that stomach ulcers are caused by stomach infection with bacteria Helicobacter pylori), struggled for years against the establishment, proved that antibiotics cured both stomach infection with bacteria Helicobacter pylori AND stomach ulcers – and eventually won the 2005 Nobel Prize in Physiology or Medicine.

The story of the beta amyloid hypothesis (MOA) and Alzheimer’s Disease – and the enduring efforts to cure Alzheimer’s Disease with inhibitors of beta amyloid is now losing favour, and may be replaced by a new direction – i.e. a new MOA for that disease.

In cancer, there are so many potential MOAs to go after, yet our success in treating most cancers is still poor. Read Clifton Leaf’s article in Fortune March 22nd 2004 entitled “why we are losing the war on cancer – and how to win it” if you want a view from an Industry-insider on how poorly we are doing.

Despite around $30b p.a. spent on anticancer research and anticancer drug discovery many patients are still treated with severe side effects and poor post-treatment prognosis. If we have a clear MOA for cancer then the world would be a different place!

Reductionism versus holism

In scientific endeavour there is often a tussle between the reductionist-approach (where one tries to dissect the problem into small and smaller parts to arrive at the Eureka-moment) compared with the holistic-approach (which seeks useful new understanding from looking at the problem as a whole).

If you accepted the proposition that human biology is a complex dynamical system – then you will have sympathy with the scope for a reductionist-approach in better understanding diseases like cancer, neurodegeneration, aging, HIV-infection and autoimmune diseases – yet you may also guess that breakthroughs will come from a holistic-approach.

In our company, we like to have the benefit of the fine and painstaking reductionist work of others (which they publish in great detail in electronic journals) whilst having a very holistic view on innovative MOA-selection.

Interestingly, many of mankind’s biggest medical successes against fearful diseases have used a holistic approach. The smallpox vaccine arose from the 1796 observation of Jenner that milkmaids (exposed to benign cowpox) were protected against devastating smallpox. Along the same lines, Fleming identified the antibiotic substance penicillin from the mold Penicillium notatum in 1928 through a fortunate series of events, then the discovery was made practical by further hard work (and some considerable luck) by Florey, Chain and Heatley.

Back to cancer – many researchers claim that cancer is not one disease and that there will be many treatments required to address the different forms. Alternatively, maybe there is a MOA that pin-points a pivotal target for the treatment of all cancers.

How many drugable targets are there left for us to conquer

Journals on the subject of medicine development do ponder how many “novel” targets remain for biotech people to discover and then conquer. A recent publication proposes that the world’s present drugs address around 300 “targets” (proteins, receptors, kinases etc.) and guesses that there are around a further 300 useful drugable targets yet to address. If that estimate is correct, then each biotech that can identify a novel and useful disease target is the equivalent of Columbus finding the Americas.

Unfortunately, Big Pharma tends not to get fond feelings about a biotech venture until the biotech can show (i) that they have a novel and useful disease target AND (ii) that they have discovered an intervention that works (e.g. a drug) AND (iii) such intervention is safe (i.e. has been tested in clinical trials. Getting (i) AND (ii) AND (iii) is the challenge of commonly under-funded biotech – regardless of their merit and stamina.

Generating “drugs” that inhibit a target

The good news is that the options for developing inhibitors of molecular targets have never been more numerous. Companies like us can choose from small molecules, peptides, monoclonal antibodies, aptamers, silencing RNA, gene therapies and more. We can use High Throughput Screens (HTS), in silico (denovo design computer-based) modelling, X-Ray crystallography and other techniques. We can immunise mice, sharks and lamas to generate target-specific ligands that can then be humanised.

If one has the target, money and willingness to embrace innovation – then there is little excuse not to find an inhibitor or two.

Safety – safety – safety – does it work?

Getting a drug to market is a task of some magnitude and sensitivity. Typically the type of drug development (including cell-based therapies) we do, involves the making of something genuinely new, and putting it into people – after testing in animals. Unfortunately, animals (even non-human primates) are not perfect mimics of humans – and first-in-man testing of novel new chemical entities is a delicate activity which is highly regulated and subject to rigorous controls.

Around 23% of drugs fail because they do not have efficacy – only 1 in 4 fail because they don’t work!

Most drugs fail because they were not safe in animals, or they failed to be safe in humans at doses that work – the therapeutic window problem. Safety and adverse reactions are the main focus of the drug industry and regulators.

It is easier to have a safe drug that doesn’t work particularly well than a drug that works really well for some people but has a poor safety/adverse-reaction record.

Myth – the cost to develop a drug

It is a common expectation that it costs over $1 billion to get a new drug on the market. Recent publications have used industry data to show that the actual real cost for a biotech to do the job is a median of $55m (average is $75m). This fits with costs we know and understand. Typical costs for a cancer drug might be (i) drug discovery (e.g. HTS) $2m (ii) drug screening and optimisation $3 – 5m (iii) preclinical testing $2m (iv) IND preparation $2 – 3m (v) Phase I/IIa clinical study $5m (vi) Phase IIb clinical study $10 – 20m  (vii) Phase III study $20-30m.

How to be a sure thing in biotech? Allocate a budget of $500m to progress 10 good programs and get at least 1 (and probably 2 or 3) through.

Potential partners – the biotech model

The life of a biotech is something like being the male praying mantis – with success being associated the fragile endeavour by being in a dance with a Big Pharma partner.

A partnership with “Big Pharma” (a pharmaceutical company with sales > $3b p.a.) is generally sought by biotechs to take their drug into the market. Pfizer is the worlds largest “Big Pharma” with 110,000 employees, sales of $68b p.a. a market cap of $155b on PE of 20. Clearly a biotech will struggle to dance with Big Pharma – but that is part of the challenge.

Those people involved in a new drug or medical device can hope to see their product part of the portfolio of the Big Pharma partner – and to receive the royalty cheques every quarter.

The size and sort of prizes for winners

If a biotech gets a project through into the clinic and licensed to a Big Pharma, the payoff can be very large and rewarding.

Numbers vary, but milestone fees paid to the biotech might be $200m to $500m, with royalties on sales around 10%. A blockbuster drug (sales of over $1b p.a.) might deliver $1b or more to the biotech over a 10-year period.

Based on a cost to the biotech of less than $50m for a single drug, that is a potential 20x return for investors.

How should investors select their biotech investment?

As an insider, here are my tips for picking a winner.

Firstly, look for people and organisation that are keeping their costs to a minimum. A biotech does not need to raise $150m to get a drug to the point of being licensed.

Secondly, look for the mechanism of action (MOA) that makes sense and is novel. Novelty is not enough on its own. Most MOA’s will seem sensible – many will be wrong.

Thirdly, support biodiversity. The world does NOT need another vascular disrupting agent to treat cancer. There is already doubt that vascular disrupting therapy actually extends life in a meaningful manner.

Finally, worry more about safety of the drugs under development. If the target or the drug-series, is showing signals of having safety concerns, that will be the main cause of commercial failure.

What does a biotech CEO need from investors?

Financial support and patience.

Getting a drug development project to the point of recognition-of-value requires at least $25m – ideally $50m plus – and around 5 years.

For me, heaven would be a war chest of $250m in available funds over the next 5 years to take 10 programs as far as they can get. Based on industry metrics, that should give around 2 significant successes and get the company cash flow positive.

About the Author

Ian Dixon is founder and CEO of Genscreen Group (, a privately-held Melbourne-based biotechnology accelerator with a focus on cancer therapeutics and diagnostics. Ian is also non-executive Director of Cell Therapies Pty Ltd ( (of which Genscreen owns 22%), a leader in the field of cell-based therapies for cancer vaccines and regenerative medicine. Since 2004 Ian has concentrated on biotechnology innovation and human therapeutics. Before that, Ian worked in the medical devices and diagnostics/pathology instrumentation industry.

Ian has qualifications in engineering (mechanical and electronics), an MBA and studies in biomedical engineering at Monash University. Contact details ph +613 9894 4555