It seems that the latest buzz phrase to be going around the medical world is the notion of “personalized medicine”. Which sounds like a bit of a joke, because isn’t all medicine personal? I mean, I’m the only sick person here, and I’m the only one putting medicine inside of my body. Doesn’t that already feel quite personal? But, if this isn’t, what is personalized medicine?
we should not treat everybody the same way
Personalized medicine is the idea that we should not treat everyone the same way. No one is exactly the same as anyone else. You are your own unique height, weight, hair colour, shoe size and anything else. You are your very own snowflake.
But if we are all different, why should we treat everyone the same way? Well, we usually think of traditional medicine as a simple flowchart, a patient comes in, the patient is diagnosed with condition X, therefore, treatment with Y. The problem is that this is too much of a simplification. In truth, we discriminate all the time. Something as simple as whether you are a man or a woman can make a big difference to treatment.
When you go beyond the very most basic medical problems, there are a whole bunch of inherited and environmental influences that change how you approach treating sickness. Someone that has smoked 20 a day, every day since they were 14, might very well have a different lung cancer treatment to that athlete that never touched cigarettes.
As science has progressed, healthcare has evolved. Routinely with cancer, a small sample (known as a biopsy) is taken and looked at under a microscope by a technician. By looking at the colour, shape and how the cancer cells respond to certain dyes, allows the cancer to be typed. Knowing which subcategory of cancer your tumour belongs to is crucial to fighting it. You know how it is likely to progress, how aggressive it will be and crucially the best way to fight it.
But this still isn’t personalized medicine, this is simply the evolution of modern medicine.
All of this is already one step forward in what President Obama in his State of the Union address, and others, call “precision medicine” – the idea that a one-size-fits-all solution cannot be applied to healthcare. Obama therefore, has made a big push in funding the development of new diagnostic tests.
Personalized medicine is the idea of taking huge amount of data to make a profile and treatment as unique as you are.
But precision medicine is not the same as personalized medicine.
Personalized medicine is the idea of taking huge amount of data to make a profile and treatment as unique as you are.
Crucially, personalized medicine means applying huge datasets, complex statistics and big data approaches to healthcare and medicine.
Patient records today
Traditionally, every patient would have their own medical record – typically a paper folder containing basic information and a few sheets of doctors’ notes. Things like your age and blood group would be recorded. In countries with established healthcare systems, maybe there would be a questionnaire about your family history, the drugs you take, how much you drink and smoke and all those other weird questions you get asked.
This system has largely gone unchanged for the last 200 years. And it works great! But only if you see the same doctor for absolutely everything in your whole life. You’d better also hope that nothing ever happens to the place that stores those records, like a flood or fire!
Let’s imagine what happens if you are visiting a friend in a different city and you have an accident? You get rushed to a hospital, but that hospital has no idea what blood group you are. They have no idea that you are allergic to penicillin. They might not know about that metal hip or pacemaker and metal things become really bad news if they try and give you an MRI in a really big magnet.
The hospital, in most cases, can, of course, request this information – subject to data protection laws. If your family doctor is actually open and organised, then they might put get a copy of your medical history in the post the same day. Or if you’re really lucky that might be able to fax it over (remember faxes!).
Broken bones in a snapshot of time
Lucky for you, it was just a broken bone! To check out the extent of the damage, the doc sends you for an X-ray. Well, good news! You are using a technique that fundamentally has not changed since 1895. You use a “flash”, just like on a normal camera (but you use X-ray light instead of visible light) to light up your body against some photographic film.
You know all those TV shows when the doctors hold up an X-ray, here I’ll find you some stock photography. That is the medical equivalent of taking your 32 mm camera film down to the drugstore to be printed back in the 90s.
Philips (yes the same Philips as TVs and toothbrushes), are one of the biggest players in medical imaging, and they only really started selling “digital” X-ray machines in 2007. They even celebrate on their website the fact that they’ve installed 7,000 digital X-ray scanners. So that’s one of the world’s largest healthcare providers, selling one of the most universal instruments, to all the hospitals in the world, and they’re celebrating 7,000 sales! For perspective, in the US alone, there are 5686 hospitals (not including clinics and dentists).
My point is, chances are, you go for an X-ray, CLICK, wait a few minutes and you’re holding an X-ray. The image you hold in your hands is probably the only thing in existence (except accounting) that proves it actually happened. Later you’ll give it to a nurse, who will add it to your patient folder and that’s it. Easily lost, easily misplaced, never to be seen again.
Electronic medical records
Personalized medicine becomes possible from the same principles as big data. As data storage has become cheaper and cheaper, simultaneously data processing has gotten faster and faster. Meaning more and more data is now available, on-demand, than has ever been accessible in the history of humanity.
Healthcare providers are starting to see the value in electronic medical records. A single, central database of patient records sounds, in principle, a great idea for patients. Obviously, assuming that concerns about privacy and security can be met.
In the UK, the National Health Service (NHS) was an earlier mover towards the hopes of having a nationalised database. The project, however, was abandoned in 2013 after going over budget its £4 billion budget, BY £6.4 BILLION.
Enter big data
Companies like 23andme now exist that offer kits in the post where you swab your cheek, post it back to them and they genetically test it for you. This simple test, for around $200, gives you your genetic susceptibility to around 100 common diseases.
If you think about that as $2 per disease, to know whether you’re likely to get some disease as you age – it almost seems stupid not to. Think about it. If you know that you are highly likely to have a heart attack, you can start making lifestyle choices now to minimise your risk.
Thanks to advances in genetic sequencing, known as next generation sequencing (NGS)(inventive name huh?), you can have your entire genome sequenced for around $1000.
That is 3.2 billion bits of DNA.
The entire blueprint to make you, who you are, can be yours in about 2 weeks.
In raw data format, your genome summarises to about 700 MB of raw data. A single CD-ROM.
Every genetic disease. Every genetic susceptibility is available to you with a blood test and $1000. And this price is only going to come down with more time. The human genome project took almost a decade and billions of dollars to complete the first genome. It is already down to $1000 and 2 weeks, think where it will be in another decade. Pretty routine.
The potential for this information, even in of itself, is immense. But the real power comes from the ability to compare yourself against yourself.
Copy of a copy
Almost every cell in your body is, at best, is 10 years old. This is because our cells are constantly dividing to replace themselves as they get old and die. Aging is essentially a result of a photocopier problem.
If you think back to high school, we all had that one teacher that used the same handouts, year after year. The problem is that when you make a copy, you always lose a small amount of detail, of quality. If you photocopy, the photocopy then you lose even more quality. Of course, you can still mostly read the page and guess (error correct) some of the missing letters, but at a certain point, you’ve lost so much quality in your copy of a copy of a copy, that you can’t read the page anymore. Now imagine that that page of words are the instructions to keeping you alive. Aging is just less than perfect copying, amplified.
Cancer, almost by definition is when your own cells losing control of themselves. Cancer becomes like a photocopier where the big green “Copy” button is stuck. The cells have lost control and everything just keeps copying.
Cancer genomics is personalized medicine
The power of big data, in this case, comes in comparison. Imagine that along with your corrupted photocopied page, you had access to the original. Then you could do a side-by-side comparison and see where the errors are.
In medicine, a doctor when taking a biopsy for analysis could also take a cell or two from the neighbouring, but normal, tissue. You could literally compare “normal” you and “cancer” you. In the lab, this technology already exists, they are called microarrays (great tutorial here) and make for pretty diagrams where things that change are green or red, and stuff that goes unchanged is yellow.
This gives your doctor a genetic map of exactly what is wrong in your cancer cells.
When you know exactly what has gone wrong, then it is much easier to come up with a plan to combat it. This is personalized medicine.
You are not alone
Big data also allows for other innovations. When you receive that terrible news that you never wanted to hear, that it looks like cancer, you can take some comfort, knowing that you are not alone. You are not the first person to ever suffer this horrible disease. Big data is now allowing us to use the information in those other patients. Thanks to projects like The Cancer Genome Atlas project (TCGA), thousands of patients are having their cancer tumours genetically sequenced, and the results are being freely made available to the public.
When huge datasets are made freely available to the public, it allows clever people the ability to do some amazing things.
GenomeDx in Vancouver is one such company. Thanks to a partnership with the National Research Council of Canada, GenomeDx could take huge amounts of genetic information from cancer patients, apply some complex statistics and pull out significance. The analogy of finding a needle in a haystack does not do these guys justice. They are looking for only one or two changes in a sequence 3.2 billion characters long. A sequence that naturally has huge sections of junk, that varies wildly and changes over time.
GenomeDx focussed their efforts initially on prostate cancer, a cancer that has a lot of samples available, but also a cancer where subtle differences are the difference between doing nothing and huge, life-changing surgery. Thanks to their efforts and analysis they uncovered 22 unique genetic markers for the aggressiveness of prostate cancer. This is huge! Previously, prostate cancer typing was almost entirely based on the Gleeson scale – i.e. what does it look like under a microscope. GenomeDx not happy with just coming up with some markers, wanted to make a clinical difference, making a dedicated iPad app that pushes test results straight into the hands of the specialist clinician.
In their first study, GenomeDx reclassified 60 % of high-risk patients to low-risk thanks to this extra genetic information. That’s the difference between a few pills or no action at all, versus painful surgery with a long recovery time.
Their trials were so successful, they went back through previous patients outside the trial. With the new information, some patients that were fine by the old microscope and pee tests, and not scheduled for follow-ups, were called back to the clinic in the nick of time, ultimately saving lives.
GenomeDx is just one company, currently focussing on one disease. Around the world, there are hundreds of similar startups looking to take advantage of all this new data.
Personalized medicine is not all about you
Human genomics is only one of several new fields that are flooding the world with data. Fields like “metabolomics” – the study of what the body breaks molecules like food down into or “proteomics” – the study of all the proteins in your body are exploding. Perhaps the “-omic” that holds the most promise is that of microbiomics – the study of the unique cocktail of bacteria, viruses and fungi inside and on you, that keep you alive.
You have 10x more bacteria cells in you, than you actually have of yourself.
The human genome is a code comprising of 3.2 billion characters, these in turn code for 22,000 genes (bits of genetic data that code something). When the Human Microbiome Project (HMP) started, they took small samples from the bodies of 242 people. From these 242 people, the scientists found 100 TRILLION unique bacteria. That’s 10x more types of bacteria than you have cells in your body. All of those bacteria, combined had 8 million unique genes.
The bacteria inside and on you accounts for 360x more genetic information than you have.
Think about that for a second.
The ecosystem, just the bacteria that you need to keep you alive, ignoring fungi and viruses, totals 1 trillion bits of information.
We coexist in a huge balancing act, and getting it wrong is so easy, as seen by almost all of post-surgical infections come from our own bacteria being in the wrong place.
The answer is accessibility
Of course, having all of the data means nothing if it is not accessible.
But great things only happen when great minds have access to data
Across the globe, there are countless libraries with millions of books that will never be read, simply because people are too lazy to research an answer. In this internet age, almost all of the collective knowledge of humanity is instantly accessible and searchable, and we have started to do great things with that.
But great things only happen when great minds have access to data, to do intelligent things and build information, drawn from their own conclusions.
Consider the success of Google. A huge library of knowledge is accessible in 0.3 seconds. But imagine a world where that same Google search took an hour to return an answer. Would you care what other films this actor was in if you had to wait an hour? What if Google took a week, a month or even a year to come back with an answer? Remember that the answer to the Ultimate Question of life, the universe and everything is meaningless after 7.5 million years.
When information is not accessible in a timely fashion it loses its value.
A doctor needs results now. A doctor cannot wait weeks to make a decision.
Big data provides the power that makes personalized medicine possible. The ability to make complex queries, taking a patient’s history, their current prescriptions, their genetic makeup, their cancer’s genome and compare them to clinical trials and international statistics in real time will change the world. Medicine will, by definition, become personalized medicine. Custom, tailor-made solutions for you.