Tuesday 27 January 2009

A Brief History of Life

In order to reach the level of life that we have reached, we have had to progress from more primitive forms. We are vertebrates and so have a spinal cord. The earliest known occurence of something with a spinal cord is in a type of animal which has remained unchanged for millions of years: The sea-squirt. Adult sea-squirts are pretty basic tubular filter feeding creatures which remain stuck to the same place for their whole life, but the larval form of a sea squirt looks like a tadpole and possesses the crucial spinal cord. It is believed that around 550 million years ago, some of these sea-squirt tadpoles never found a rock to attach to, and instead remained in their fishy form, went on to evolve into fish and eventually into us.

Lets take it back even further. Before we evolved to the multicellular tadpole stage, we would have had to have been a single celled organism. Cells like ours with a nucleus that contains the genome, are known to exist as far back as 2.7 billion years ago. Lets go back further, before our cell can wrap its genome in a nucleus, you need a cell with loose genetic material, like modern bacteria. The oldest fossil record of bacteria is from 3.5 billion years ago. This is only a billion years after the earths crust cooled from boiling magma to solid rock.

What we have in common with bacteria is that we are vehicles for our DNA to replicate itself. We are a lot bigger and more complicated, but we do the same thing, we duplicate our DNA and we pass it on to a new generation. So where did this DNA come from? We make DNA with proteins and enzymes that are made with instructions from the DNA, so which came first? At this point its easier for me to just jump right to the start, its not that far away now.

Once the earths crust formed the oceans began to form, natural reactions began in the salty mineral rich water and organic chemicals began to form. These organic chemicals provide the well known 'primordial soup'. This soup would collect and become concentrated in the rock pools on the coasts, in here the first amino acids would have formed, as well as the first nucleotides. Amino acids are the basic building blocks of proteins. Nucleotides are the basic building blocks of RNA and DNA. These can assemble themselves and interact with eachother, eventually some proteins will randomly have developed a structure that enabled the nucleotide chains to be copied. Other proteins could use nucleotide chains to produce a new protein. Suddenly we have the first instance of replication.

As this chemical replication proceeded, in its very early form it would have been imprecise and many errors would occur. But errors are good in biology, errors produce variation, variation produces differences in performance. The best performing combinations of proteins and nucleotides would become the most common. The complexity of the combinations would have increased with time. The single strands of RNA formed complimentary double strands, and acquired functions that outperformed the primitive proteins. Some of these RNA functions still exist within our cells. For a while the world was dominated by the RNA chains, with help from proteins.

Eventually the RNA would have found a performance boost by being isolated within a bubble of oil, the products of its work being kept close instead of washing away into the ocean. This is essentially the cell membrane that us and bacteria possess. The complexity of the replication reaction has taken another step and inside the bubble complexity grew ever greater. Sometimes however, other more primitive RNA systems might get into the bubble, and take advantage of the resources there, we call them viruses today. Later proteins became more complex and outperformed RNA which was replaced with DNA and life started to look like something we would recognise.

Thats a heavily summarised account of the most accepted theory of how life developed. But at what point can we say life actually commenced? Do simple chemical reactions count as life? Is it the basic replication where life starts? The fact is that we are the product of a basic chemical reaction that began more than 4 billion years ago, and inside us the reaction continues, growing ever more complicated as long as it enables us to reproduce ourselves better. We are only aware of our environment because awareness helps us to find food, survive and reproduce. Every aspect of human nature can be related back to how it helps us ensure successful propagation of our DNA. With the global human population nearing 7 billion, we're certainly doing quite well, but not nearly as well as those bacteria, there are a hundred trillion of them in your gut alone, and bacteria will be around long after humans are forgotten.

Viruses in Gene Therapy

Since the publication of the human genome in 2003 great developments have been made in genetic technology. But one of the big challenges is developing reliable methods for the delivery of the desired gene to cells in a human body. This is where our old enemy the virus can help us, as I mentioned briefly in my post about targetting therapies.

Viruses are the smallest form of life, essentially just parasitic packets of genetic material. They have adapted to infect a a variety of tissues, using a variety of different methods to get their various forms of genetic material into the cell. All this variation makes viruses more diverse than all the other forms of life put together, as viruses have adapted to use them all and for each species there are a whole collection of associated viruses. This variation also provides us with a potential toolbox which we can use to achieve our own objectives.

Having small and well understood genomes, viruses are easily modified to carry genes of human interest, and there you immediately have a highly efficient gene delivery mechanism. This method has produced an interesting range of viral therapies for a range of diseases. One company that is pursuing this technology in a broad range of diseases is Oxford Biomedica, which has a range of viral based gene therapies for diseases including Parkinson's disease, age-related and diabetic sight-loss, and in early development; motorneurone disease, AIDS, spinal cord injury; haemophilia. These developing therapies all utilise a modified horse Lentivirus to deliver therapeutic genes to a specific tissue.

Parkinson's disease involves a depletion in the brain of the critical neurotransmitter dopamine which adversely affects the brains normal functioning. Oxford Biomedica's viral therapy carries three genes into the brain tissue, which encode enzymes that produce dopamine. This new dopamine production increases the level to a point where normal brain function can resume, as shown in animal models and currently looking very promising in early human trials. The sight-loss therapy works in a similar way; The virus is modified to target only the desired retinal cells and delivers genes that halt the uncontrolled growth of blood vessels on the retina that occurs in certain eye diseases.

Hereditary conditions have been successfully treated, as shown by experiments by University of Pennsylvania Medical School and University College London with a rare form of hereditary blindness called Leber congenital amaurosis. In this therapy they inject into the eye a tamed strain of Adenovirus carrying a working copy of the mutated gene that causes the blindness. Vision improved enough for the patients to sucessfully navigate an obstacle course in dim light, a task that would previously have proved very difficult for them. There are six genes involved in the disease so further improvements may be made to the treatment by including more of these genes. The teams also believe the treatment may cause more improvement in children, as their retinas will have degenerated less than adults.

As the understanding of genetic causes of diseases, both acquired and hereditary, are being developed faster than ever by geneticists, the opportunities for gene therapies such as those described become ever more numerous. Viruses are going to be instrumental in delivering these therapies to the cells that need them.

Sunday 18 January 2009

Immunotherapy - Switching Anti-Cancer Immunity Back On

The immune system is a powerful tool that protects us from being infected by the many bacteria, fungi and viruses that would otherwise find a warm wet nutrient rich wound an ideal place to colonise. But it also has a role in protecting us from our own cells. Immune surveillance is the body's way of keeping a check on cells that begin to divide uncontrollably. Such cells have mutated somehow and will be recognised as defective and destroyed by the immune system. Generally this works very well for most of our lives; unfortunately it applies the 'survival of the fittest' rule to cancer cells. The 'fittest' cancer cells being those that acquire the mutations enabling them to mutate faster than the immune system can recognise them, or those that acquire a mutation that causes some kind of immunosupression. Eventually the cancer evolves to a form that completely evades the immune system, usually by secreting a cocktail of chemical messengers that stop immune cells recognising them.

Many approaches are being tried in an effort to make the immune system recognise the cancer again and begin killing it. Cocktails of lab-grown tumor cell vaccines have had mixed success in trials, ongoing are Phase II trials by Onyvax whose therapy consists of three types of lab grown cancer cells, which are injected into patients with prostate cancer. The immune system will recognise these cells as totally foreign, and in the process should notice some cancer associated proteins which it had previously missed due to the immunosuppression by the cancer. Although this approach has shown some effectiveness in previous trials by extending patients lives by around 6 months, it only slows the cancer and never causes remission. This could be because this approach doesn't do anything to overcome the immune suppression by the cancer, there may very well be a well armed immune response ready to attack the cancer, but every time it gets close it is prevented from attacking.

Dendreon is another company with a prostate immunotherapy, this one uses lab-primed immune cells called Antigen Presenting Cells (APC) and injects them into patients, once in the patient the APCs 'teach' the patients immune system to attack the cancer. This approach is similar to the Onyvax approach, but is further in development, unusually it is in its 3rd Phase III clinical trial, ideally those should only happen once. It is clearly having an interesting effect on cancer, enough to continue investigations, but its not significant enough to bring it to market yet. Again, this therapy doesn't address the issue of immune suppression.

Cell Genesys had a vaccine very similar to Onyvax's, except it included an extra immune stimulatory factor in an attempt to overcome the immunosuppression with an even more powerful immune response. This attempt failed, their PhaseIII trial was terminated and the therapy abandoned. Things are beginning to look a bit bleak, I shall bring us onto some more promising therapies.

NovaRX have a therapy similar to that of Onyvax and Cell Genesys but for advanced lung cancer, it uses four sorts of cancer cell in its vaccine; crucially it also blocks a signal protein called TGF-β, which is an immunosuppressor secreted by the tumor cell. This approach seems to be highly effective, extending patients lives by years in a phase II study and is now in a PhaseIII study.

These types of therapies are important, because once a working therapy is produced for one cancer type, the same approach can be used for nearly all other types of cancer. In all of these trials however, the vaccines usually only work for a certain subpopulation of those tested. I wrote previously about the great variation in cancer between people and the need to personalise treatments using biomarkers. These companies will be monitoring patients and discovering biomarkers that will help optimise future treatments with their products.

One way to overcome the problem of treating unique cancers with general vaccines, is to make the vaccine unique to the patient. Antigenics have a vaccine which is tailored to each patient by taking a biopsy of the tumor, taking it to the lab and creating a personalised vaccine from it. This vaccine has been effective in prolonging survival in kidney cancer and melanoma has been approved for treating kidney cancer in Russia and has been submitted for approval in Europe.

I have presented a small selection of therapies here, a search of 'cancer vaccine' on clinicaltrials.gov provides a list of 303 clinical studies that are seeking patients, hundreds more have completed recruitment and are ongoing. So there is much more information to be gained in the not too distant future. Using this information the next generation of cancer vaccines are likely to incorporate several modes of action including blocking the immunosuppressive activity of the tumor, priming the immune system to attack, and boosting that attack with immunostimulators. Perhaps by doing this, we will be able to extend lives by decades rather than months and years.

Saturday 17 January 2009

Targeting Therapies

Getting your drug therapy to the tissue that you want to treat is easy, the body's circulatory system is perfect for that. The problem occurs when the drug gets into other tissues and causes side effects, a long list of which can be found on the information sheet supplied with any drug.

There are a multitude of different ways of specifically targeting certain tissues or cells, and many new ways being developed with the use of new nanotechnology. One such method is being developed at the City University of New York. Here the drug is attached to a mesh of fatty acids, making it inactive. This mesh will disperse around the whole body like any other drug, but could be designed so that the drug can be detached from the mesh by an enzyme that is only present in the tissue being targeted. In this way the drug is only released in its active form at the desired location, thus limiting the chances of the drug getting into other tissues and causing side effects. This is in very early development and has yet to be proven outside of bench-top experiments, there is undoubtedly still a lot of work to be done to make this method work, but it shows us the kind of thinking going on in this area at the moment.

Here's some background to a different problem. DNA encodes the 'blueprints' for all the proteins your cells need to do their business, it is like the master copy. When the cell needs to make a protein it uses a slightly different chemical called RNA to make a copy of the gene, the cell then uses that copy to construct the protein. Many copies are made and transmit the message of how to construct the protein to the cellular machinery. When a cell is making a protein that it isn't supposed to, it can often cause disease. In the lab it is possible to block the RNA message by designing small segments of interfering RNA that stick to the RNA message. As the cellular machinery works its way along the message, making the protein as it goes, it reaches this interfering RNA segment and can't read the message anymore because it is blocked out and so the protein is never completed. Inject this interfering RNA into the body however, and you'll find it is destroyed pretty quickly in the blood before it ever reaches where it supposed to.

Calando Pharmaceuticals in California are testing in humans a kind of Trojan-horse system where the interfering RNA is packaged inside a nanoparticle studded with a molecule called transferrin. They chose this molecule because cancer cells are abnormally rich in receptors for that molecule, and when they detect it on the particle they will take the whole particle inside the cell. The acidity inside the cell is different to the blood, and this causes the particle to burst, releasing the interfering RNA into the cell where it can do its job. This technique is very promising as it can relatively easily be modified to target any receptors and deliver interfering RNA to all sorts of cells, not just cancer cells, and could potentially be adapted to deliver regular drugs.

A third approach is to make the cells produce the drug themselves. Viruses exist by infecting cells and making them produce all the proteins it needs, the virus just brings along the appropriate genes and the cell does all the work. Viruses are also very specific about the cells they infect, which is half the work already done for us. They are already being used selectively infect cancer cells, thereby killing them. Companies such as Oncolytics, Genelux and others are carrying out trials of this method. Viruses or artificial virus-like particles can be designed to deliver a gene for a specific enzyme that makes an active drug out of an inactive 'pro-drug' that is injected normally. This means the infected cell becomes a kind of drug factory at the precise location the drug is required, minimising the exposure of the rest of the body to that drug.

In ways such as these the treatments of the future will have a much reduced range of side effects, while at the same time being more effective and improving the quality of life of people suffering from chronic diseases.

Sunday 11 January 2009

Book Review - The Time Traveller: One Man's Mission to Make Time Travel a Reality

The story of Ronald L. Mallett's life has essentially been defined by one tragic moment in his childhood. In 1955 Ronald's 33 year old father died suddenly and unexpectedly of a heart attack and 10 year old Ronald lost the man who was the centre of his universe. After this event he receded from his friends and his life, instead developing a passion for reading. A year after his fathers death he discovered a comic book version of H.G. Wells' The Time Machine, and this lead him to dream of the possibility of travelling back in time to warn his father of the heart attack and thereby prevent his untimely death.

While most of our childhood dreams fade with time, Ronald never forgot his, indeed it became a secret obsession that he would see his father again and he became determined to build a time machine. Using knowledge of electronics he had learnt by helping his father repair television sets, he secretly built a replica of the machine depicted in The Time Machine. Of course it failed, and he realised he needed to learn more if he was to make it work. His autobiography details how he overcame the hurdles of poverty and racism in order to gain himself an education in theoretical physics, eventually receiving a PhD from Penn State University in 1973.

Ronald knew he was unable to openly admit his goal was to build a time machine as he would not be taken seriously and would be ridiculed, thus far the only person he had told of his plan was his wife. So in 1975 he joined the University of Connecticut as an assistant professor, and studied the only thing that was known to manipulate time: Black holes. He pursued an academic career there becoming more and more despondent as the years passed and he seemingly got no closer to his goal. His marriage failed, as did his health, himself suffering from heart trouble and having to take time out from his career. Realising he was running out of time he reignited his research, and eventually produced a 4 page paper with an equation that predicted that light, as well as gravity is able to manipulate time and built a small experimental model of a time machine to demonstrate it.

He published the paper, and after 40 years of secret work, found that there was a great interest in his work, and the physics community took him completely seriously. He revealed the driver behind his life's work after a question and answer session at a presentation of his theory in Washington DC. In response to this story, Bryce DeWitt, who proved Einstein's theory of relativity, said in front of the conference audience that he didn't know if Ronald would ever see his father again, but did know he would be proud of him. That statement provided the validation of his life that Ronald had always needed, but one part stuck with him, DeWitt had said he wasn't sure Ronald would see his father again. Ronald went back to the equations and soon realised that even if he did build a time machine, the farthest back in time he would ever be able to travel to would be the moment the machine was first switched on. But that didn't matter anymore because he already knew his father would be proud of him, and didn't need to see him to know that.

The book is an inspiring life story as well as an introduction to the various theories of time manipulation. The physics of Ronald's work is presented in easily readable metaphors and although you might not understand them completely (I certainly didn't), you get the general idea and that's all you need to know to follow the story in this book, it really is more about a man missing his father than it is about physics. I certainly enjoyed reading it and believe it is worth a few hours of antibody's time to read this story.

Friday 9 January 2009

A Little Help from Man's Best Friend

My last post suggested that some new drugs don't make it to market because the existing trials process doesn't provide critical information that can be used to optimise the trial design and reveal the true potential of the drugs being tested.

A change in the existing trial process would have serious legal and ethical issues to contend with, and so is unlikely to occur. It can however be supplemented in such a way that streamlines the process, more accurately determines the effect of a cancer drug and can reduce the time to market. That solution is dogs. These are not your typical animal experiments, such as the rat or mouse models of cancer, in which the animal often lacks a fully functional immune system and the cancer itself is a cross-species 'xenograft' implant. Its fairly easy to see how these models do not accurately represent the type of cancer that occurs naturally in the body, and that's where the dogs come in.

In the USA up to 6 million pet dogs are diagnosed with cancer every year. Canine cancer is surprisingly similar to human cancer. Dogs get the same types of cancers as humans, they are genetically similar to humans and crucially, large scale genomic analyses of canine tumors have shown that there are no differences in the genetic mechanisms of the cancer. The other similarities in canines include their size and the fact that they possess fully functioning immune systems. The similarity between dogs and humans is so close that most existing drugs can be used to treat equivalent diseases in both species.

Naturally dog owners are keen to pursue any therapies that may prevent the death of their pet. Fortunately they can, in the USA the National Cancer Institute operates a network of animal hospitals, fully equipped with state of the art imaging technologies to accurately diagnose and monitor canine cancer. Similar work is carried out at the Roslin Institute in Scotland. These centres are used to test new therapies on the plentiful supply of canine patients following Good Clinical Practice guidelines and central reporting of dangerous side effects not too dissimilar to those in place to protect human patients.

The canine trials can start providing detailed information on the mechanisms of the drugs action before even the Phase I trial in humans starts. Information such as genetic profiles for which the drug is ineffective or blood markers that can be used to determine successful responses in advance of significant tumor reduction. These can all be translated to human systems and used to better design the human trials. These types of biomarkers are not often discovered until around Phase II of human trials, and are usually validated during phase III. Having them in place for Phase I is very useful as they can be validated in the early phases and used to optimise the later more expensive and time consuming phases. For example selective enrolment of patients who have the genetic or biochemical profile compatible with the drug would make the trial more decisive, while freeing other patients to pursue other therapies with more likelihood of success for them. Another benefit is instead of waiting years to determine the survival of the patient and therefore whether the drug was effective, biomarkers that provide advanced indicators of survival could provide that information in months, drastically reducing the length of the trial and therefore the costs.

Comparative Oncology such as this has the potential to give us a much more detailed understanding of the drugs we are testing and should help bring more new drugs to market and quicker. Cancer is a complex and diverse disease that will not be overcome by a single therapy alone, we will need to use combinations of therapies specifically targetted to a patients personal disease to attack it from several fronts. Therefore understanding each cancer type and each drug as much as is technically possible will be critical in determining potential drug synergies and creating successful recipes for treatment.

Wednesday 7 January 2009

The Problem with Clinical Trials

All new medicines undergo a rigorous series of controlled studies to establish safety and efficacy before they are licensed. The need to test drug safety on a small scale before allowing it to be prescribed was underlined in the early 1960's by the well known Thalidomide Tragedy, a situation where the drug thalidomide was prescribed to pregnant women in Europe and Canada as a treatment for morning sickness. Unfortunately the drug had only been tested in animals, and nobody foresaw the severe birth defects that would be inflicted on the children of these women. In 1962 the US Food and Drug Administration (FDA) put a system in place to ensure all new drugs would be rigorously tested before coming to market.

This has resulted in the clinical trial system we have today which typically consists of three phases. Phase I begins if laboratory and animal experiments have shown convincing evidence that a new drug is effective and safe. At this stage the primary concern is drug safety and only the minimum number of patients will be treated, initially with a very low dose of the drug, increasing gradually as the trial proceeds. If the trial goes well and no patients were harmed by the drug, then it may enter phase II. In Phase II the objective is often to establish an effective treatment regime, different dose levels and frequencies are likely to be tried in order to establish how to make the drug most effective. At completion of this phase the data will be studied to determine if there is a benefit associated with use of the drug. If there is, then phaseIII will begin, often with hundreds of patients in a large scale placebo controlled study across many sites to establish beyond doubt whether the drug is truly beneficial.

The problem with this system is that it is very expensive and is a very long process, taking up to a decade or more to complete. The race to get the drug to market means companies often try to complete the first two phases as soon as possible, they see promising data from these trials and dive headfirst into PhaseIII to save as much time and money as possible. This means that detailed studies into the method of action of the drug are often overlooked, as it isn't considered worth spending the money on that until you know the drug is safe and effective, its as if nobody cares how the drug works, they just want to find out if it is effective at treating the disease. This is thought to be one large factor in why so few drugs make it through PhaseIII to market. If you don't know how the drug works, then you don't know why its failing. Within a trial the drug may work well for some patients, and have no effect on others, frequently the benefit is seen in so few patients that the drug is considered ineffective and dropped from the process. Any benefits that were seen are easily forgotten once the trial is branded a failure. There was however a potential to learn a lot from these studies, information which when used correctly could have meant the drug could have been brought to market.

The trial system is certainly protecting the public from potential dangers of new drugs, but it may also be indirectly harming them by being such an expensive and time consuming process that drugs that only seem to have a small benefit, or benefit only a small selection of patients, are never brought to market. In my next post I will explain how a surprising addition to the current system is already giving us the information we need to optimise the trial process and bring more drugs to market.

Monday 5 January 2009

The Increasing Use of Biomarkers in Cancer Research

Hello and welcome to my first 'SciBite'. Whilst looking for a subject to write my first post about I came across an article about prostate cancer and a particular type of androgen (male hormone) receptor. This article details the discovery of several variants of androgen receptor, and how the expression of one particular variant seems to be responsible for enabling prostate cells to grow even without the androgens they usually require.

This is important because current standard treatment for prostate cancer often involves disrupting this androgen signal, either by castration, or more commonly in the richer countries of the world, hormone therapy. Both of these methods are initially highly successful, often large decreases in the size of the tumor are observed, but frequently the cancer develops independence from the androgens that used to dictate its growth, and a new phase of the disease is entered.

The discovery of this variant androgen receptor sheds some light on the mechanism of the cancers evolution to this phase, but usefully it also provides a 'biomarker' to help consultants decide the best course of therapy to place their patient on. It is possible to test a biopsy of the tumor to establish if this variant receptor is present, if it is already present at a high level then hormone therapy may not be a very successful option and the patient may benefit from an alternative treatment. For patients who are undergoing hormone therapy, it may provide a useful guide as to how far along the tumor has got to becoming androgen independent, and enable a switch to another treatment before tumor growth becomes completely uncontrolled.

This is one good example of a current trend in cancer research and treatment: Identification and utilisation of biomarkers to better diagnose, monitor and treat cancer. Previously cancers have been categorised according to the organ or tissue of origin, and all cancers of that origin tended to be treated the same way. However, more recently new technologies such as gene expression profiling have shown that the genetics of tumors can differ greatly between patients, even if the tumors are in the same tissue. Furthermore it has exposed the fact that there can be large genetic differences even within a patients own cancer, primary tumors being significantly different to secondary tumors. The differences extend beyond the genomic level, as all changes at that level are translated into cellular changes, the up or downregulation of tumor specific proteins may be detectable in the blood as is the case with PSA in prostate cancer.

These differences go some way to explaining why cancer treatment has until now been so hit-and-miss. If cancers of the same tissue can be caused by completely different mechanisms, then it is unreasonable to expect one treatment to be successful with them all.

The advantage we gain from the genetic analyses available to us now is that we can see easily which genes are undergoing changes in the development of a cancer, and investigate them specifically. This will inevitably provide many new cancer biomarkers to help us understand the processes occurring within cancers, give us a clue about the prognosis for the patient and enable us to more precisely target specific processes in the tumor with proven drug combinations.

I recently heard a speaker at a lecture say something along these lines: "We don't need to develop any new cancer drugs, we already have a lot of those, we just don't fully understand how they work or who they will work for, and that is where we need to focus our efforts".