Improving analysis of functional imaging

Bookmark and Share
Published: 23 Nov 2016
Views: 2022
Prof Nandita deSouza - The Institute of Cancer Research, London, UK

Prof deSouza speaks with ecancertv at NCRI 2016 about the role of functional imaging in guiding patient treatment choices.

She summarises presentations at the conference on whole body diffusion weighted MRI, molecular diagnostics with radionuclides for enhanced targeting, and the quantitative value of image analysis.

Prof deSouza also considers computer standardisation of data analysis, the need for national guidelines, and the work of the Cancer Research UK Imaging Centre Network.

The session is on functional imaging for influencing management. Functional imaging is a new way of looking at images that doesn’t tell you just about the anatomy but tells you something about how they might function or how tumours behave. So what we’re trying to do with this session is to understand where functional imaging fits in the pathway of decision making in order to treat patients. I’ve got three excellent speakers lined up, one is going to talk about decision making and where functional imaging might fit in the pathway. She’s going to particularly highlight whole body diffusion weighted MRI which is a new technique that does not involve ionising radiation because it uses magnetic resonance imaging and can get information from the whole body in one hit, as it were, and so would be an important imaging pathway when you’re making decisions for cancer patients. The second speaker is going to talk about molecular diagnostics, identifying targets with imaging and being able to specifically locate those targets within the body. That’s with a nuclear medicine technique using targeted radionuclides. Then the third speaker is going to talk about quantitation because images are all very pretty to look at and we can qualitatively look at them and make assessments but actually functional imaging is so much more than that, it’s about getting numbers out but getting them out meaningfully so that we can follow tumours through treatment. But we have to be able to standardise those numbers and understand what they mean so we can interpret them correctly so the whole process of image analysis and quantification is really crucial. Hopefully we will address the breadth of functional imaging with those three talks.

Are there any computer programmes to standardise this analysis?

There are two aspects to it, one is the standardisation at the point of data acquisition and that is actually quite complicated because there are so many different equipment manufacturers out there, so many different hardware and software issues that you have to grapple with and that has to be done at a level of the technologists acquiring the imaging data at source. Then there is the problem of standardising the analysis which, as you rightly point out, is to be done by computer programmes largely but there needs to be human intervention in order to make sure that it’s done correctly. There need to be processes of quality assurance and quality control built in to that.

Are you aiming for country-wide guidelines?

Absolutely, that’s going to be necessary if we used these techniques in a quantitative way for decision making. That doesn’t exist yet but we need to get there.

What are the financial considerations and practicalities?

Imaging is costly, it is very costly though, like with everything, costs come down - the more of it you do, the cheaper it gets, it’s just because of the bulk demand. Nevertheless the hardware is costly, radioisotopes are costly and the process of doing it is costly. But it will get cheaper with time so unless you are doing new and very challenging techniques it’s unlikely to get more expensive but we can bring these costs down with time if we’re more efficient and more time efficient, especially if we know exactly what we want to do so we don’t waste time on the unnecessary measurements.

What are the key messages?

That functional imaging is here to stay but that it has to be interpreted with caution and with knowledge. That the quantitation requires quite a lot more work and that we, as a scientific community, have got to put that work in to make it meaningful so that we can introduce imaging into the treatment pathway.

Could you tell us about your other work?

We do a lot of studies, in fact we’re part of a Cancer Research UK imaging centres network with other centres at Manchester, Cambridge, Oxford and University College London. One of the main focusses of this imaging centre network, and it’s driven by CRUK, is to introduce functional imaging into clinical trials and to do them in a multi-centre way to be able to roll them out nationally. So there are a lot of trials that we are involved with together with these other imaging centres.

Is there any work that particularly stands out?

A personal one for me is one that we’ve been doing in ovarian cancer, it’s called Discover, looking at response assessment with diffusion weighted imaging in epithelial ovarian cancer in patients either that are treatment naïve and having chemotherapy prior to debulking surgery or in patients who have recurrent or relapsed disease who are then on second and third line treatment, looking at diffusion related imaging as a quantitative response biomarker.