The Currently Incurable Scientist: Your Own Personal Immunotherapy
“Each patient’s cancer is unique and personalized – why shouldn’t their treatment be too?”
My latest The Currently Incurable Scientist column for Fight Colorectal Cancer has published and it ended with the above quote. The column discusses an area of cutting edge oncology research that I find absolutely fascinating: a type of experimental immunotherapy called a “personalized cancer vaccine”. I already mentioned personalized immunotherapies briefly in my post “My New Year’s Vision of a Colorectal Cancer Immunotherapy Future” and the basics were covered in the Fight CRC column – but as usual in this blog I wanted to dig a bit deeper into a Director’s Cut… I hope after reading this post you will find the concept of personalized immunotherapies as fascinating and exciting as I do!
What I describe may sound like such disruptive technology that it is far away from widespread use. Although it is still unproven & highly experimental – the future may be closer than you think.
The Original “Model T” Strategy of Drug Therapy
Aspirin was first produced and sold by Bayer worldwide in 1899. It is the ultimate mass-produced medicine. 40,000 TONS of it are manufactured in factories and sold around the world each year! It is relatively safe for most people and it has broad activity against a gazillion sorts of aches & pains, prevents heart attacks (and more! Including potentially cancer but that is the subject of another post!). I look at it as just about the ultimate non-personalized medicine. It is probably hard to find almost any adult, in most countries of the world, who hasn’t been treated with aspirin at least once in their lifetime.
Traditionally, oncology drug discovery was quite similar. Mass-produced drugs meant to be taken by ~all patients based upon average effects seen in large clinical trials. A specific example of this in colorectal cancer (CRC) is the chemo cocktail I am currently taking: FOLFIRI. In clinical trials, about 35% of patients see significant tumor shrinkage (including both tumor shrinkage and stable disease, it has a “response rate” of about 75%). Adding bevacizumab (Avastin) increases the response rate a bit further. Who takes it though? Essentially 100% of patients do. Why? Because it is completely non-personalized and oncologists can’t predict beforehand who it will work for. All that they know is that “most patients” will benefit from taking it but they also know… that some of their patients will endure the harsh toxicity of this chemotherapy without being helped. At the end of the day, you just have to use the best tools you have available. Thankfully research has been busy trying to improve the tools!
Personalized Medicine (v 1.0) – Choosing the Paint Color
Most of you have probably heard of “personalized medicine” which first became a buzzphrase starting in the late 1990’s. This referred to the first use of medicines based upon genetic information.
Was it a BIG step forward? DEFINITELY. Is it really “personalized medicine”? Kind of.
What the first generation of “personalized medicine” meant was essentially “grouping patients together” based upon a genetic test. Don’t get me wrong – this is very powerful and beneficial… I am just pointing out that it is still treating groups of genetically similar patients with a particular drug based upon average response. An example from CRC is treatment with anti-EGFR antibodies such as cetuximab (which I took & responded to with very good quality of life last year). Not all CRC patients are treated with anti-EGFR antibodies. Although this is surely too simplistic of grouping, currently genetic tests are run and only patients genetically identified as “KRAS-wild type” are treated with drugs like cetuximab because it is known that the opposite group of patients (“KRAS-mutant”) will not have any benefit from taking these drugs.
Personalized Medicine (v 2.0): Building a Custom Race Car
This brings us to the point of this post (about time!): truly personalized immunotherapies. Starting from the Model T of traditional chemo, through the custom paint jobs of targeted agents – we’re finally now starting to get into the world of custom-built race cars!
What makes these therapies “fully personalized”? They are based upon the genetic sequencing of an individual patient’s tumor. Although there may be shared common mutations between some CRC patients (e.g. KRAS G12D), the exact list of all mutations within a patient’s tumors are as completely unique as his or her fingerprints. No two patients’ tumors are exactly alike.
Scientists have known this for a long time but there were very real practical/scientific/engineering issues to obtaining the exact genetic mutations for patients’ tumors – at least on a scale beyond a very small number of patients. But there have been HUGE strides made in recent years making individualized tumor genetic profiling start to approach an efficiency & cost-structure feasible for a relatively large numbers of patients.
Commercial companies now offer “Whole Exome Sequencing (WES)” which determines all of the mutant genes in a tumor compared to the rest of your body – but focused only on genes which make proteins (the things which are actually potentially noticeable by the immune system). An additional technique “RNA-seq” takes the process a step further and looks into which mutated DNA genes are making it to the level of RNA, one step closer to protein production.
Taken all-together this information gives you a genetic snapshot of exactly one person’s tumor. No one else’s! Just his/hers. It can’t get any more personalized than that!
Once you have a genetic snapshot of a tumor – you have a “catalog” of everything that is different about the tumor (at the genetic level) compared to the normal parts of the body. This is very powerful information because differences between normal tissue and an “invader” (in this case the unwanted tumor) are exactly what the immune system was created to deal with. Unfortunately, not all mutations are recognizable by the immune system – but fortunately assays exist which can tell you which ones are. In that way, scientists can generate a catalog of not only tumor-specific mutations but also a catalog of the mutations recognizable by the patient’s immune system.
But what can you do with this information? It turns out that you can do a lot… including things potentially leading to personalized treatments.
The problem with tumors is that although they are genetically different from normal tissue, they set up all sort of defenses to “hide from the immune system”. This is where science and engineering can come into play – to use techniques to help the immune system to take notice.
Therapeutic Cancer Vaccines
Therapeutic Cancer Vaccines have been studied for years. Like childhood infectious disease vaccines, they are intended to teach the immune system that something in the body shouldn’t be there. To teach the immune system that it is dangerous. To teach the immune system to attack & destroy it. Unlike childhood vaccines, they are intended to teach the immune system to attack existing tumors, not just new tumors attempting to enter the body.
Even though traditional therapeutic cancer vaccines have steadily improved (e.g. some very promising results in a form of brain cancer!), historically most of them have struggled to generate strong anti-tumor efficacy in the clinic, probably for a number of reasons. In addition to immunosuppressive tricks done by tumors, I think it also boils down to them not being personalized/specific enough. These vaccines have tended to be based upon proteins that are either merely at higher levels in tumors compared to normal tissues or based upon mutant proteins shared across patients without regard to how that mutant protein exactly relates to that patient on an individual genetic & immunological level. These factors do not align themselves with generating the most robust immune response possible. For example, there are a number of safeguards to prevent the immune system from attacking proteins seen in normal tissue (to avoid autoimmune diseases!) but this same system gives “tumor associated antigens” an uphill battle to fight against since they are not 100% tumor specific. Likewise shared antigens may be more efficient from a logistical point of view (think Henry Ford & his Model T factory) but one person’s immune system may be less of a fan of buying a Model T than another’s…
Enter the Personalized Therapeutic Cancer Vaccine
You can see how the two sides of my story are now coming together!
- A belief that therapeutic cancer vaccines need to become more personalized to increase efficacy
- The recently feasible generation of personalized tumor data of which genes are unique to the tumor & which of their protein products activate the immune system.
A recent immunotherapy chocolate & peanut butter match made in Heaven!
The first fully personalized cancer vaccines have been generated and tested against preclinical models in recent years – including this one with very high activity in an animal model for CRC. The key question now becomes: can this strategy and preclinical data be translated to successful clinical trials?
Signs of immune system activation are already being seen in melanoma patients. Cancer Centers, biotechs and big pharma (e.g. BioNTech, Neon Therapeutics/BMS, Gritstone Oncology) are slowly ramping up personalized vaccine therapy trials against a number of cancer types, including recently CRC (e.g. NCT02600949). The trials will start out relatively small but any success will quickly feed growth.
I believe the full power of this strategy may be seen when it is combined with immune activating drugs such as the deservedly famous PD-1 checkpoint inhibitors. As I said in the Fight CRC column “This combination is like teaching the immune system how to drive (the vaccine) and then filling its car up with gas (the immunotherapy drug)!”
In addition to proving efficacy in clinical trials, as you can imagine logistics & production costs for fully personalized therapies are much more complex than the pharmaceutical equivalent to a Model T factory. But I look at this as an engineering issue more than a scientific one. And I am a big believer that when engineering (and not basic science) is a bottleneck – the Human Race accomplishes amazing things. If the possibility of a solid tumor cure is dependent on it – what could be more incentive to engineers to figure things out than that? Disruptive technologies are always hard but when they offer something previously unobtainable (and desperately sought after) – they lead and the world follows.
I can’t accurately predict if a personalized cancer therapeutic vaccine will eventually cure anyone’s cancer – since by definition, science is all about pushing the boundaries of knowledge & attempting the currently unknown. What I can say though is that many scientific and engineering stars are aligning for the very first time to make a strong shot on that goal using this strategy. As a scientist-patient-advocate that is what I am looking for – for myself & others. Strong shots on goal. I need it, my fellow cancer patient community needs it, and the world needs it.