Breakthrough tests on "mini cancers"

Scientists in London have discovered how to grow miniature copies of a patient’s cancer - these are called organoids - in a laboratory dish. These “mini-cancers” can then be used to screen over 50 different cancer drugs to find very quickly the combinations of treatments that will - or won’t - work for that person. Chris Smith spoke to study author Nicola Valeri from the Royal Marsden…

Nicola - the key issue here of when we face new drugs and new trials for patients is that we need to enrol a number of patients in clinical trials and then, many years later, we find out if the drug works or not. One the key issues we face in personalised medicine is to find the right drug for the patient in order to avoid toxicities to other patients that are unlikely to benefit from it.

Chris - Is it not also true that everyone is an individual and, therefore, just because a drug works in a population of patients, if you look at the individual, it may not works for them?

Nicola - That’s exactly the problem. That’s exactly what we face in the clinic even though you have positive data from clinical trials, sometimes the drug works, some other times it doesn’t and that’s what we’re trying to address.

Chris - Because, at the moment, say someone is enrolled into a course of treatment with an anticancer drug, we only find out months later in that person whether that drug’s doing them any good. So how can you improve on that?

Nicola - We need to develop tools that would enable us to define a kind of prediction whether the patient is going to benefit or not from this specific treatment. At the moment, in clinic, we have very to address this need. The real issue is that cancer is a very heterogeneous process; cancer changes over treatment so basically we need to have better models that we can manipulate in the lab and can tell us more about the way a single patient will respond to the treatment.

Chris - Is that what you’re now announcing, you’ve now found a way of doing this better?

Nicola - Yes. We can now take small biopsies from patients tumours. We focussed our attention on metastasis so when the cancer has spread outside the primary organ on bowel cancer and stomach cancer because these are two deadly diseases. Taking these biopsies from these patients we try to model what would happen in the clinic, in the lab.

Chris - You make it sound simple but I’m sure it’s not. How do you, from a biopsy, work out what would happen if you were to treat that particular patient from which that biopsy was collected with a certain drugs?

Nicola - What we have done, we have taken biopsies from liver, lung, or other organs affected by metastasis and we basically grew then as a 3D structure in the lab. Then we compare how this little tumour growing in my lab performs under treatment, response to anticancer agents as compared to the response observed in the patient.

Chris - I suppose one advantage of this is you can take that biopsy and divide it to grow lots of little mini identical cancers, so you can then experiment with lots of drugs in parallel and then work out what perhaps is the optimal drug, or combination of drugs for that person?

Nicola - Absolutely. That was one of the key points of our study. We grow tumours that are so small - like 200 micrometres - so we can grow as many as we want in a few weeks. Time is also another important bottleneck when we want to provide information to patients in real time, and you are correct, we can screen a number of drugs. In our case we screen 55 drugs and so we can, in a couple of months or even less, provide an answer about the right treatment.

Chris - That will be: does this person’s cancer respond to the drug we’re giving, is there evidence of side effects, is the cancer showing signs of becoming resistant, etc?

Nicola - Unfortunately, we cannot give information about the side effects, but what we do know from this test is if the organoid - the little tumour - in the lab does not respond to the drug, in 100% of cases the patient will not respond in the clinic. On the contrary, in 90% of the cases where we see a response in the lab, we will see a response in the patient as well.

Chris - If you were to translate this paper in Science this week to the clinic, what sort of an impact would it immediately begin to make for patients?

Nicola - First of all we might be able to rule out drugs that wouldn’t work, and then we might be able to make a quite accurate prediction of what is going to work. Of course, there are a number of other factors in the patient; the microenvironment for the tissues around the tumour they also play a role in cancer, and also the metabolism of the drug. At the moment, we are still working on these factors because we cannot account for these factors in our current system.

Chris - What’s next?

Nicola - For us, the next step is to allocate patients to treatment based on what we see on our organoids. We are developing protocols so that we can collect biopsies from a patient who’s coming to the Royal Marsden, test a number of drugs prospectively, and then allocate the patient a specific trial based on the findings we observed in the lab.