Cancer organoids personalise therapy

Balls of tumour tissue grown from a patient's cancer cane be used to optimise and personalise tumour treatments...
11 May 2015

Interview with 

Aine McCarthy, Cancer Research UK


Cancer is not just one disease; there are hundreds of different types, from breast to brain and prostate to pancreatic. Making matters more complicated, it's also becoming increasingly clear that cancer is as individual as each patient and needs treating in a much more personalised way. Secondary tumor deposits in the liver from a primary cancer of the pancreasBut how do doctors find out what drugs an individual's cancer is most susceptible to, and the likely prognosis? Now, scientists in the Netherlands have developed a way to grow in the lab tiny balls, known as organoids, of a patient's cancer, which can be used to test treatments before they go anywhere near the patient. Aine McCarthy, Science Information Officer at Cancer Research UK, took Kat through the study...

Aine - At the moment, the main techniques we use to evaluate how well tumours will respond to drugs are cell cultures and something that are called xenografts, where they transplant cells from patients into mice and watch how they grow. These two techniques are fundamental to all basic science research but they have their limitations and their flaws. With cell cultures, the problem is, we don't necessarily get to see the whole tumour or look at all the different types of cells within the tumour. So, it's a very, almost, one-dimensional look at what's going on and what's happening in the cancer cells.

Kat - What have the researchers done now that's different?

Aine - What they've done in this really interesting paper is they've taken cells and samples from patients with colorectal cancer and they've taken both cancer samples and samples of normal tissue. They've grown what are called "tumour organoids". The best way to think of these is almost as a 3D model and it's a way to look at lots of different aspects of the tumour that will give us a more, maybe, real-time look at how drugs respond and what's happening.

Kat - So, how many patients were they looking at?

Aine - They looked at 20 patients. They got normal samples from 19 of these patients and interestingly, they also had actually 22 tumour samples because in 2 of the patients that had cancer, they had 2 separate tumours. So altogether, they looked at 22 tumour samples and 19 normal samples.

Kat - So, they're growing these little 'organoids' in the lab and then what were they doing with them?

Aine - So they looked first to see how long they could grow them for and whether they could grow them, and then the next thing that they wanted to look at was, can they use this organoid model to figure out how well tumour cells will respond to drugs. They're taking the organoids - so the tumour samples and the normal samples - and they're doing what's called DNA sequencing. So, they're reading the entire DNA or genetic code of them. They're also looking at another molecule, which is the RNA expression, which allows them to look at the gene level. And then they're using this information to see which genes and which proteins are mutated or faulty and then they'll base their drug screen on this to see if specific drugs work properly to target these.

Kat - Because that would tell you what drugs might work for this patient.

Aine - Exactly, and the ultimate aim that they have is to give a more personalised, more targeted treatment and therapy to the patients. Also, what they want to do is create a biobank where they'll have these samples stored so they can be used in the future to test further drugs as they come along.

Kat - So, in terms of what they found in this paper, did it work for them?

Aine - It did work and they found that there was consensus among their findings with other literatures. So, they found similar things to other people and it does show that by using these organoids, you can identify what they call "molecular marks" or "molecular signatures" that will tell you how well a tumour will respond to certain drugs.

Kat - Now, the key thing with treating cancer is you need to do it quickly. You need to find out what's wrong with someone, what sort of cancer they have, which drugs to give them. Is this the kind of research that will be in a meaningful timescale to get those answers for doctors?

Aine - In its current state, it's not clinically used but also not able to be used in the clinic. But the main aim will be to hone the technique and to get it working in a more quick way, in a quicker manner. Basically, they acknowledge themselves that it takes a bit of awhile for the organoids to grow so ideally, they'd speed up this process and they'd make the whole thing a bit more clinically applicable. That's the best phrase I can use.

Kat - This has just been used for colorectal or bowel cancer. Do you think it could work for other types of cancer too?

Aine - In theory, I think it absolutely could. I think there's no reason to say it couldn't. But it's important to be aware that the trials and the studies will have to be done to make sure that it can accurately represent and depict what's happening in all different cancer types. But I don't see why, if tested and used properly, it couldn't give us the same information about different types of cancers.

Kat - One of the things about this organoids technique is that it means that you don't have to use mouse models, animal models at all. Could something like this completely replace the use of animal models in cancer research?

Aine - No. I don't think it could completely replace the use of animals, but it could absolutely be used in parallel. It could give us some other vital information but animal models are a necessity because they allow us to look at the whole system - the lymphatic system, the blood supply, absolutely everything. So, we could glean some more tailored, more targeted information from the organoids that we can then use in animal models, but they won't be eradicated completely.


Add a comment