The future of early cancer detection?

Along with a crew of technologists and scientists, Jorge Soto is developing a simple, noninvasive, open-source test that looks for early signs of multiple forms of cancer. Onstage at TEDGlobal 2014, he demonstrates a working prototype of the mobile platform for the first time.

Almost a year ago,

my aunt started suffering back pains.

She went to see the doctor

and they told her it was a normal injury

for someone who had been playing tennis

for almost 30 years.

They recommended that she do some therapy,

but after a while she wasn't feeling better,

so the doctors decided to do further tests.

They did an x-ray

and discovered an injury in her lungs,

and at the time they thought that the injury

was a strain in the muscles and tendons

between her ribs,

but after a few weeks of treatment,

again her health wasn't getting any better.

So finally, they decided to do a biopsy,

and two weeks later,

the results of the biopsy came back.

It was stage 3 lung cancer.

Her lifestyle was almost free of risk.

She never smoked a cigarette,

she never drank alcohol,

and she had been playing sports

for almost half her life.

Perhaps, that is why it took them almost six months

to get her properly diagnosed.

My story might be, unfortunately,

familiar to most of you.

One out of three people sitting in this audience

will be diagnosed with some type of cancer,

and one out of four

will die because of it.

Not only did that cancer diagnosis

change the life of our family,

but that process of going

back and forth with new tests,

different doctors describing symptoms,

discarding diseases over and over,

was stressful and frustrating,

especially for my aunt.

And that is the way cancer diagnosis has been done

since the beginning of history.

We have 21st-century medical treatments and drugs

to treat cancer,

but we still have 20th-century procedures

and processes for diagnosis, if any.

Today, most of us have to wait for symptoms

to indicate that something is wrong.

Today, the majority of people still don't have access

to early cancer detection methods,

even though we know

that catching cancer early

is basically the closest thing we have

to a silver bullet cure against it.

We know that we can change this in our lifetime,

and that is why my team and I

have decided to begin this journey,

this journey to try to make cancer detection

at the early stages

and monitoring the appropriate
response at the molecular level

easier, cheaper, smarter

and more accessible than ever before.

The context, of course,

is that we're living at a time

where technology is disrupting our present

at exponential rates,

and the biological realm is no exception.

It is said today that biotech is advancing

at least six times faster than the growth rate

of the processing power of computers.

But progress in biotech

is not only being accelerated,

it is also being democratized.

Just as personal computers or the Internet

or smartphones leveled the playing field

for entrepreneurship, politics or education,

recent advances have leveled it
up for biotech progress as well,

and that is allowing

multidisciplinary teams like ours

to try to tackle and look at these problems

with new approaches.

We are a team of scientists and technologists

from Chile, Panama,

Mexico, Israel and Greece,

and based on recent scientific discoveries,

we believe that we have found

a reliable and accurate way

of detecting several types of cancer

at the very early stages through a blood sample.

We do it by detecting a set of very small molecules

that circulate freely in our blood

called microRNAs.

To explain what microRNAs are

and their important role in cancer,

I need to start with proteins,

because when cancer is present in our body,

protein modification is observed

in all cancerous cells.

As you might know,

proteins are large biological molecules

that perform different functions within our body,

like catalyzing metabolic reactions

or responding to stimuli

or replicating DNA,

but before a protein is expressed or produced,

relevant parts of its genetic code

present in the DNA

are copied into the messenger RNA,

so this messenger RNA

has instructions on how to build a specific protein,

and potentially it can build hundreds of proteins,

but the one that tells them when to build them

and how many to build

are microRNAs.

So microRNAs are small molecules

that regulate gene expression.

Unlike DNA, which is mainly fixed,

microRNAs can vary depending on internal

and environmental conditions at any given time,

telling us which genes are actively
expressed at that particular moment.

And that is what makes microRNAs

such a promising biomarker for cancer,

because as you know,

cancer is a disease of altered gene expression.

It is the uncontrolled regulation of genes.

Another important thing to consider

is that no two cancers are the same,

but at the microRNA level, there are patterns.

Several scientific studies have shown

that abnormal microRNA expression levels

varies and creates a unique, specific pattern

for each type of cancer,

even at the early stages,

reflecting the progression of the disease,

and whether it's responding to medication

or in remission,

making microRNAs a perfect,

highly sensitive biomarker.

However, the problem with microRNAs

is that we cannot use existing DNA-based technology

to detect them in a reliable way,

because they are very short sequences of nucleotides,

much smaller than DNA.

And also, all microRNAs are
very similar to each other,

with just tiny differences.

So imagine trying to differentiate two molecules,

extremely similar, extremely small.

We believe that we have found a way to do so,

and this is the first time that we've shown it in public.

Let me do a demonstration.

Imagine that next time you go to your doctor

and do your next standard blood test,

a lab technician extracts a total RNA,

which is quite simple today,

and puts it in a standard
96-well plate like this one.

Each well of these plates

has specific biochemistry that we assign,

that is looking for a specific microRNA,

acting like a trap that closes

only when the microRNA is present in the sample,

and when it does, it will shine with green color.

To run the reaction,

you put the plate inside a device like this one,

and then you can put your smartphone on top of it.

If we can have a camera here

so you can see my screen.

A smartphone is a connected computer

and it's also a camera,

good enough for our purpose.

The smartphone is taking pictures,

and when the reaction is over,

it will send the pictures

to our online database for processing

and interpretation.

This entire process lasts around 60 minutes,

but when the process is over,

wells that shine are matched
with the specific microRNAs

and analyzed in terms of how much and how fast

they shine.

And then, when this entire process is over,

this is what happens.

This chart is showing the specific microRNAs

present in this sample

and how they reacted over time.

Then, if we take this specific pattern of microRNA

of this person's samples

and compare it with existing scientific documentation

that correlates microRNA patterns

with a specific presence of a disease,

this is how pancreatic cancer looks like.

This inside is a real sample

where we just detected pancreatic cancer.

(Applause)

Another important aspect of this approach

is the gathering and mining of data in the cloud,

so we can get results in real time

and analyze them with our contextual information.

If we want to better understand

and decode diseases like cancer,

we need to stop treating them

as acute, isolated episodes,

and consider and measure everything

that affects our health on a permanent basis.

This entire platform is a working prototype.

It uses state-of-the-art molecular biology,

a low-cost, 3D-printed device,

and data science

to try to tackle one of humanity's
toughest challenges.

Since we believe early cancer detection

should really be democratized,

this entire solution costs at least 50 times less

than current available methods,

and we know that the community can help us

accelerate this even more,

so we're making the design of the device

open-source.

(Applause)

Let me say very clearly

that we are at the very early stages,

but so far, we have been able

to successfully identify the microRNA pattern

of pancreatic cancer, lung cancer,

breast cancer and hepatic cancer.

And currently, we're doing a clinical trial

in collaboration with the
German Cancer Research Center

with 200 women for breast cancer.

(Applause)

This is the single non-invasive,

accurate and affordable test

that has the potential to dramatically change

how cancer procedures and diagnostics

have been done.

Since we're looking for the microRNA patterns

in your blood at any given time,

you don't need to know
which cancer you're looking for.

You don't need to have any symptoms.

You only need one milliliter of blood

and a relatively simple array of tools.

Today, cancer detection happens mainly

when symptoms appear.

That is, at stage 3 or 4,

and I believe that is too late.

It is too expensive for our families.

It is too expensive for humanity.

We cannot lose the war against cancer.

It not only costs us billions of dollars,

but it also costs us the people we love.

Today, my aunt, she's fighting bravely

and going through this process
with a very positive attitude.

However, I want fights like this

to become very rare.

I want to see the day

when cancer is treated easily

because it can be routinely diagnosed

at the very early stages,

and I'm certain

that in the very near future,

because of this

and other breakthroughs that we are seeing

every day in the life sciences,

the way we see cancer

will radically change.

It will give us the chance of detecting it early,

understanding it better,

and finding a cure.

Thank you very much.

(Applause)