In this post, we interview Rebecca Fitzgerald, chief investigator for the BEST3 trial. Her work at the MRC Cancer Unit focuses on the early detection and treatment of cancer of the oesophagus. Oesophageal cancers are the 8th most common cancer worldwide and the 6th most common cause of cancer death with only 15% surviving 5 years. The two main subtypes of oesophageal cancer: adenocarcinoma and squamous cell carcinoma (OSCC) have remarkably improved survival when diagnosed at an early stage.
What did you work on before the Cytosponge and the BEST trials?
Back when I was training to be a doctor, I was allocated to a gastro firm. I had several patients with various cancers who died, including oesophageal cancer, and couldn’t make up my mind if I wanted to be an oncologist or gastroenterologist. After getting married and following my husband and his job to the US, I took the opportunity to do some research and talked to various people about possible projects. There was a gastroenterologist working on infectious diseases that I couldn’t get very excited about, but the doctor who was working on cancer of the oesophagus was very charismatic and encouraging. His research appealed to me because it combined the gastro and the oncology I had been working on.
The first challenge that he set me was working out how Barrett’s oesophagus progress – what makes it so bad? There was already information about acid reflux being a key risk factor, so I set about doing a lab project to expose cells to acid and bile, to see how it affected cell behaviour. I’ve worked on Barrett’s and cancer of the oesophagus ever since. It really captured my imagination, because I was already convinced that there was an unmet clinical need and the lab worked seemed like one day it’d be clinically useful. For quite a long time I was interested in exploring acid reduction to reduce cancer risk.
Throughout my whole career I’ve been working on Barrett’s and cancer of the oesophagus in one way or another. But increasingly, I got interested in detecting it early.
Where did you get the inspiration for the Cytosponge?
When I was in London I was working at Bart’s and the London Hospital, training to do an endoscopy. I increasingly realised that endoscopy was a barrier to diagnosis because most people with heartburn don’t get an endoscopy. I remember chatting to my boss at the time, Professor Mike Farthing who worked on inflammatory bowel disease- so wasn’t an oesophageal person at all, when he said, “What you need is a bottle brush for the oesophagus”. That got me thinking.
I moved to Cambridge and thought I need to devise something that could enable us to collect cells from the oesophagus without an endoscopy, ideally something that could be done in the GP surgery. I went to the engineering labs in the laboratory of molecular biology- the famous Nobel prize lab in Cambridge- and got them to make me a bottle brush. We made several of these prototypes- no one ever swallowed it! I spent a lot of time thinking about how to make it more user friendly. So I looked around to see what had been done in the past, particularly in parts of the world where they have a lot of oesophageal cancer – not the type that is connected with Barrett’s and heartburn, but squamous cancer in Japan and Iran. In Japan, in the past, they have done a lot of work on cell collection devices in the oesophagus with balloons that you inflate on a catheter.
It gradually evolved from a rigid catheter to something soft that you swallow- like a string. I then realised that the benefit of having a rigid bottle brush meant that you can target the area of the oesophagus- but a sponge on a string could scrape the entire length of the oesophagus. This meant that you’ve got all the cells you might be interested in, and it doesn’t require any operator skill. You also don’t need to know or feel the pressure at the end of the oesophagus to know when & where to inflate a balloon. You don’t need to look and see where it is to inflate, you just sample the oesophagus in the same operator-independent manner every time.
I then realised you need to couple the device with lab tests because the proportion of the cells that are Barrett cells are only a fraction of all tests sampled. So, I started developing biomarkers that we could use as part of the test. I collaborated on the device side of it and got a small company who made medical devices to manufacture it. I spent a lot of time, and still am, working on making the lab test as effective as possible- that’s how we developed the trefoil factor 3 test (antibody test) to diagnose Barrett’s.
What were peoples first reactions to the Cytosponge?
I swallowed the first one we made. I hadn’t thought about the fact I should starve for a few hours beforehand – I’d had a chicken sandwich for lunch, so it came up looking quite disgusting! The amazing thing is that research patients are willing to try something new. If we explain why and how it’s better than an endoscopy- which isn’t particularly nice or convenient and means taking a day off work- then the reaction is generally good.
What were the biggest challenges & opportunities with BEST3?
We ran a 13,000-patient randomized trial- that’s big. That’s overwhelming, and once you start you feel like you might never finish. So, the scale of it was challenging, but it also involved primary and secondary care. The test was done in primary care, and then we had to engage109 different hospitals so that we could perform the confirmatory endoscopies, so the logistics were quite a challenge too.
In terms of opportunities, I wasn’t expecting the results to be as striking as they were with ten times more cases of Barrett’s detected. The cancer detection aspect also wasn’t a main endpoint of the trial- so to see patients getting treatment for dysplasia and early cancer as a result of Cytosponge testing was amazing. The fact that it finished just before COVID, and that we are now seeing capacity problems with endoscopies has given us some extraordinary opportunities to elevate the implementation.
What’s the timeline for implementation of the Cytosponge? Has COVID affected your implementation?
The BEST3 trial was for patients who weren’t actively seeking their GP because they had symptoms- they were on medication for heartburn, and on repeat prescriptions when contacted.
Currently, implementing the Cytosponge is looking more feasible in patients with symptoms who are referred to their GP for an endoscopy, and people with known Barrett’s who are being monitored. It is in this population that implementation will be accelerated due to COVID.
In Scotland they’re planning to commission Cytosponge clinically predominantly in symptomatic referrals with heartburn and the plan is to have several thousand cases- an exciting opportunity indeed.
Just before COVID we got a grant from INNOVATE UK for project DELTA, which looks at implementation. Project DELTA will have a primary and secondary care arm. The primary care arm is on hold because of COVID, but the secondary care arm is about to start and will look at patients referred to secondary care with Barrett’s symptoms and surveillance. As a result, we’ll look at all the aspects of implementation- where it fits in the pathway, appointment scheduling, training, processing samples at scale and results delivery for example. The primary care arm will be widened to more of a screening context. We want to know whether a GP organising a repeat prescription could get an alert indicating the patient has not had an endoscopy. If the complaint is predominantly heartburn, then they should recommend the Cytosponge. We’ll also be able to see in this next stage what the uptake is. In BEST3 around ¼ of people offered it took it up. However, in BEST3 they only got a letter from their GP, whereas in the implementation stage the offer will be in a clinical context which we hope will increase uptake.
Further Reading:
Barrett’s ESophagus Trial 3 (BEST3)
A ‘sponge on a string’ test to detect oesophageal cancer earlier
A trial using the Cytosponge test in GP surgeries for people with heartburn symptoms (BEST3)
BBC Look East coverage of BEST3 trial results
The views expressed are those of the author. Posting of the blog does not signify that the Cancer Prevention Group endorse those views or opinions.
