So I’m presenting the results of the new clinical study at this meeting which describes a new way of detecting a precursor for cancer of the oesophagus. The test is a device which collects cells on it called the cytosponge without needing an endoscopy, which is a normal part of it, and it’s coupled with a molecular test. It’s a simple test that’s applicable to routine labs, could be used in the NHS, which finds a protein which is specific to the precancerous cell. So just to tell you a bit more about the device, it’s a capsule on a string and compressed within the capsule is a sponge. You swallow the capsule with some water, it pops down into the top of the stomach, you hold onto the string so it doesn’t disappear off, you leave it literally for three or four minutes and the capsule dissolves to let the sponge expand. Then a nurse simply pulls the sponge out and as it travels up the oesophagus it collects cells, plenty of them, about half a million cells on the surface and within the mesh of the sponge, the device. You simply put that into a standard preservative pot and the cells can be spun off and tested for this protein which is specific for the precancerous condition. The precancerous condition is called Barrett’s Oesophagus and Barrett’s Oesophagus occurs particularly in people who have heartburn symptoms, indigestion symptoms, who are at risk for cancer of the oesophagus.
So what we did in this study, this follows up from previous work where we showed that this test was feasible, safe, straightforward and in this study we wanted to get more information about the accuracy of this test for diagnosing Barrett’s Oesophagus. So it was a case controlled study design so we took patients without Barrett’s Oesophagus who got heartburn symptoms who had been referred by their GP for an endoscopy and we also took patients with known Barrett’s Oesophagus. The study overall was over 1,000 patients. It was performed in eleven hospitals around the UK and 27 nurses were trained to give the device, just to give you an idea of how generalised it is as a test. Patients came to the hospitals, they’d been referred for endoscopy anyway, the nurse administered the cytosponge and then we tested it for presence of Barrett’s Oesophagus and compared the results to the endoscopy which they had on the same day.
The results of the study showed that, first of all, it was very safe. We didn’t have any serious adverse events due to the sponge itself. Secondly, patients found it quite acceptable; we asked them to rate it on a score of 1 to 10 and compare it with endoscopy. You get a whole range of values – patients thinking it’s really quite pleasurable to patients thinking it unpleasant but the proportion of patients who found the cytosponge unpleasant was smaller than the proportion of patients who found endoscopy unpleasant. So it’s really very simple and pretty quick and easy and not too bad as a test.
In terms of how good it is, the sensitivity overall was around 80% for detecting Barrett’s Oesophagus and that increases with the length, which isn’t surprising. So Barrett’s can be a very variable length, from literally a centimetre or two up to ten or even fifteen centimetres. The longer the length the more likely we were to pick it up and the sensitivity increased to around 87% when we’d got a longer length present. We know also that the length correlates with the risk of cancer so the more cancer prone Barrett’s we’re more likely to detect. A proportion of our patients swallowed it more than once and if you swallowed a second time the sensitivity increases to nearly 90%. In terms of how specific it was, it’s 92% specific.
So we found these results very encouraging in terms of safety, the acceptability and the accuracy of the test.
So this should impact the clinic?
That’s what we hope. We think that this is a completely novel paradigm compared to endoscopy. GPs are discouraged from endoscoping every patient with a bit of heartburn because endoscopy is quite expensive. There’s a lot of pressure on NHS endoscopy units, it’s quite invasive for the patient. So we see this as a test which is very applicable for primary care, GPs can do it and indeed our very first study was in GP surgeries, as a prelude to deciding who needs an endoscopy so we can really focus endoscopy on the high risk patients.
It isn’t very expensive. Our patients that were asked to swallow it a second time were quite willing to swallow it a second time. Some tests, screening tests, like the fecal occult blood test, for example, you actually do three specimens. So it’s all a trade-off between balancing the sensitivity and the specificity but if you really wanted to maximise the sensitivity one could do it twice.
What will you do next?
Two things are next. One thing is we want to not only diagnose Barrett’s but identify the patients at greatest risk who are already on the road to progressing to cancer. So you can do a lot of different things with one specimen, you don’t just have to get the patient to swallow it again but as well as testing for the TFF3 protein you can test the mutations. So we’re interested in, and we’ve shown for example, that you can find p53 mutations which is the commonest mutation for this cancer. So now we’re working out what second tier of tests you would do to identify the people with Barrett’s who are at greatest risk of cancer and that way you don’t have to triage everyone that you find Barrett’s in towards endoscopy, you could just focus more specifically on the high risk.
So we’re now analysing that data for this same cohort and the second thing is we want to really go back into the primary care setting and do a much bigger study as a prelude to really getting this accepted in clinical practice. So we’re planning a third study and discussing this with CRUK which would be in primary care, about 10,000 patients randomised to have standard of care or to be investigated with endoscopy versus cytosponge. So that’s what we’re planning, to try and get this out into the general practice.