Jack Andraka – Detecting Pancreatic Cancer... at 15
I'm here to talk about the really devastating disease. It's created a huge impact on basically everyone's lives. It's killed some of the most well-known celebrities, and most notably, Steve Jobs, recently. It's called pancreatic cancer, and pancreatic cancer is a devastating disease. Only 5.5% of people who are initially diagnosed will survive after 5 years. And one of the main reasons due to this is that, usually once it's diagnosed, the cancer is spread around the body, when the patient has a very small chance of survival. And in order to prevent this late detection, we need a method for early detection, meaning a simple, rapid, sensitive, versatile, and minimally invasive sensor for the early detection of pancreatic cancer.
And, so, there are a variety of proteins that are found in the blood of patients who had pancreatic cancer that are at elevated levels. I show society one called mesothelin. Mesothelin-- it's highly overexpressed by pancreatic cancer, but not by healthy tissue. But, in addition, it's highly overexpressed by these things we called non-invasive precursor lesions, meaning if you can detect this protein, you can detect pancreatic cancer before it becomes invasive, possibly making the survival rate jump above 50% rather than a measly 5.5%.
So, in order to detect these ultra-low concentrations of mesothelin in your blood, I employed these carbon nanotubes, atom-thick tubes of carbon. And they're really the superheroes of science and material science, because they have these broad applications which lead to inexpensive biosensors. In my case, I was looking at how the distance between neighboring carbon nanotubes highly impacts how electricity is transported along that.
And, so, I'm really fascinated by carbon nanotubes. I was reading this really interesting paper in my biology class, and all of the sudden, we were learning about these new things called antibodies: organic molecules that actually bind specifically to one protein; in my case, it would be to mesothelin. Kind of like a lock-in key.
So then I though, in my biology class, I was just sitting there behind my desk looking at this little paper, I thought, "What if I put this antibody in a network of carbon nanotubes?" just wildly, on a whim. And then it hit me. Amazing. I was very very happy. My biology teacher wasn't as happy when she found me reading a paper instead of writing an essay on biology class.
So then, what had to happen is a substrate that you can basically support this network of antibodies and carbon nanotubes. And I thought it had to be porous to take up this aqueous solution. But also, it had to inexpensive and readily available. Filter paper-- it fulfills all of these. And basically, all you have to do is create a ? dispersion of these antibodies of carbon nanotubes that are just floating around, chilling out. And then you basically just dip-coat your sensor, making it really easy to produce.
And so, here is a visual depiction of the sensor being created. And how you test it, basically, it only requires 1/6 of a drop. And you just put it on, and then you just measure the change in electrical potential-- just an electric parameter.
So then, in order to detect pancreatic cancer from the detection of this protein, basically what you have to do is you have to establish what we call a cut-off level, where anything above that level of mesothelin is considered positive; anything below we consider negative. And so, usually it's 10 nanograms per mL for mesothelin, so that's what I chose. And so then I ran an in-vitro study just to make sure it was working pretty well, and it worked spectacularly, actually. And then, we went into actually testing human samples. We had patients who had pancreatic cancer, and then also the non-invasive pre-case. You can see that they all fall in this range of 18 to 28 nanograms per mL. That's well above the
10 nanogram per mL cutoff. All of these would be found positive by my sensor, meaning that it had a 100% sensitivity towards pancreatic cancer.
Then we had patients who did not have pancreatic cancer, that were either perfectly healthy and skipping around, or they had chronic pancreatitis, meaning an inflammation in the pancreas, the disease most often mistaken in diagnosis of pancreatic cancer. All these, you can see, are below 10 nanograms per mL, meaning that they would all be found negative on my sensor, meaning that it had a 100% selectivity towards pancreatic cancer. So it was giving a near perfect diagnosis of pancreatic cancer. That's revolutionizing how we look at cancer diagnosis.
So, then what we had-- we were looking at another application of the sensor. We wanted to see if it could effectively look at how well a treatment was doing on a patient. What we had is 5 chemotherapeutic drugs that were given to patients over 5 weeks, and you can see for each of the 5 chemotherapeutic drugs that they were distinctive drops. So it could reliably monitor how effective a cancer treatment was.
And so, what happens is, the sensor-- it's really revolutionizing how we look at cancer treatment and diagnosis, because, compared to the current gold standard of protein detection in cancer, it's 168 times faster, over 26,000 times less expensive at 3 cents, and over 400 times more sensitive. And also, it has limitless applications. So you switch off the antibody, and you can detect an entirely new protein, meaning you can look at other forms of cancer, but also you can look at cancer treatments effectively. You can look at cancer drug resistance, because these are caused by unique mutations in the DNA, producing a unique protein, which it can pick up with my sensor. Also, you can look at the effectiveness of a cancer treatment, meaning that we are revolutionizing how we look at cancer treatment, personalizing a treatment at a patient so you don't have that conventional guess-and-check method. Also you could look at all their forms of disease, such as in food and water safety: ecoli, salmonella, rotavirus. And then also, you could look at other blood pathogens, such as AIDS, STD's, and all of those.
And so it could operate in any setting, such as at a regular check-up, but also in developing nations, revolutionizing and possibly saving thousands of lives in cancer diagnosis treatment in the biomedical field.
Edit news description to add:
- Historical context: how the event or text affects the world and history
- An explanation of the work's overall story (example: "Here, President Obama confirms the legality of drone strikes...")
- The work's impact on current issues