We were wrong.

Now, though, we have a new idea that flips the previous one on its head. Hi, I'm David, and this is MinuteEarth.

For a long time, doctors have known that there are lots of different genetic mutations that can cause cancer.

Even the same kind of cancer, say breast cancer, can be caused by very different mutations.

That means that a treatment that works well for one person with breast cancer doesn't necessarily work well for another person with breast cancer. This is exciting when, a few decades ago, genetic testing advanced enough that scientists could sample a patient's cancer, find the mutation that caused it, and then, once they figured out a treatment that worked for that one patient, they could essentially copy and paste that treatment for everyone else with the same cancer-causing mutation.

Which sounds great, and the initial results were super promising. For example, researchers copy-pasted a treatment that worked for one breast cancer patient with a particular mutation, and found that 80% of -- they pinpointed more mutations and developed more drugs, we'd soon get to the point where, whenever anyone got cancer, we could destroy it by copying and pasting a mutation-specific treatment.

But that obviously hasn't happened. First, there are a lot of different cancer drugs and a lot of different cancer-causing mutations out there, and scientists have only stumbled on a few specific combinations that actually work.

Second, even when scientists do find a combination that works in one patient, and they do the whole copy-paste thing, other than in a few select cases, the treatment doesn't work all that well in other people with the same mutation. It turns out that how most cancers respond to treatment isn't just a matter of the mutation that caused them, but also all sorts of other complicated stuff that can vary from person to person.

As a result, copy and pasted treatments, which scientists once thought might basically cure cancer, barely work better than traditional treatments like radiation and old-school chemo. But instead of giving up on the copy-paste idea, today scientists are tweaking it.

Instead of copying and pasting treatments, some scientists are copying and pasting the cancer itself.

Outside of the patient, of course.

By sampling, growing, and then divvying up a patient's cancer into a ton of different testing wells, doctors can pit hundreds of different drugs against it at the same time. They don't have to know exactly what mutation caused the cancer or what other complicated stuff is going on.

They can just see which drug works best for that particular cancer, and then give that drug to the patient.

Scientists have only done a few clinical trials of this new copying and pasting cancer approach, but it seems to be way more effective than copying and pasting treatments. Almost 40% of patients have gotten better results than they would have with radiation and old-school chemo alone.

It may turn out that this new breakthrough is underwhelming, too.

We're still a very long way from actually curing cancer.

But ironically, it may turn out that one of the best ways to fight cancer is to make more of it.

Perhaps a CTRL-C and CTRL-V. We need to do a lot more research to show the effectiveness of this approach.

This is why we were super excited to make this video in partnership with Florida International University, which has been a leader in this field, known as Functional Precision Medicine. Dr. Diana Assam, along with postdoc Dr. Arlette Aconda de la Rocha, and the rest of the team at the Robert Stemple College of Public Health and Social Work at FIU, are pioneering the next phase of clinical trials in both pediatric and adult cancer patients.

After taking tumor samples and growing them in the lab, they expose them to more than 120 different drugs to find the right fit.

This process takes less than a week. Their initial study results are incredibly promising.

Eighty-three percent of kids who received treatments guided by this approach showed improvement.

To learn more about FIU's clinical trials, visit www.fiu.edu.