A groundbreaking initiative is underway to train dogs to detect the odour of colorectal (bowel) cancer in urine samples, harnessing their remarkable sense of smell.
The charity, Medical Detection Dogs, is exploring whether our canine companions could provide a precise, non-invasive, and highly sensitive method for early bowel cancer detection. This innovative approach could enhance screening uptake and improve health outcomes.
Bowel cancer is the fourth most common cancer in the UK, with a survival rate of around 60%. Many patients present with advanced disease, highlighting the challenges in early diagnosis.
The current colonoscopy screening process is invasive, leading to only about half of those offered the test actually taking it up. Less invasive faecal sampling lacks specificity, often still requiring a follow-up colonoscopy.
In a UK first, Medical Detection Dogs will use urine samples instead of faecal samples due to the reduced stigma associated with providing urine. This could make diagnostic tests more accessible, encouraging more people to participate in screening. Additionally, it may aid in monitoring treatment responses and help scientists pinpoint tumour-specific compounds that the dogs detect, which can then be targeted in lab tests.
This project is in collaboration with Humber Health Partnership, which will supply the charity with samples to train the dogs. Medical Detection Dogs will apply their proven methodology that has shown dogs can accurately detect the odours of prostate cancer and bladder cancer in peer-reviewed trials.
These dogs have also successfully identified the odour of other diseases, including Parkinson’s disease, Pseudomonas, and COVID-19.
Six dogs will be trained to detect the odour of bowel cancer in the charity’s training room by sniffing both positive and negative samples and indicating when they have found it. This project marks the first time Medical Detection Dogs will use interactive stands developed with colleagues at the Open University. These stands will collect additional data on how the dogs search, providing unbiased insights into the dogs’ interactions with the samples and any emotional changes in their responses.
The initial canine team includes Willow, a Flat-coated Retriever; Mango, Callie (main photo), and Dotty (pic above), Cocker Spaniels; Hetty, a Fox Red Labrador; and Rosie, a Black Labrador.
Claire Guest, Co-Founder and CEO of Medical Detection Dogs, says: “We believe that information learned from our dogs about the odour of bowel cancer could help deliver an accurate, rapid, and non-invasive test for early diagnosis that would be offered to clinicians to use alongside existing diagnostic methods or for post-treatment monitoring.
“The team has just started in the training room and early signs suggest that once again, the world’s most powerful bio-sensor, a dog’s nose, will have no trouble recognising the odour.”
Andy Hunter, Colorectal Surgeon at Humber Health Partnership, adds: “My team is dedicated to reducing the side effects associated with conventional treatments for rectal cancer while maintaining high cure rates.
“I became interested in methods to detect cancer at its earliest stages, and during a conversation with Dr Guest, it became evident that we shared the same vision. With MDD’s support, I swiftly gained the commitment of Hull University Teaching Hospital’s Trust to conduct a trial assessing the utility of canine olfactory detection in cancer diagnosis and follow-up. We began collecting urine and stool samples from our willing and enthusiastic patients. These samples, now combined with relevant clinical data, will provide the MDD team with the necessary tools to train our canine collaborators. I eagerly anticipate seeing the dogs in action. Ultimately, I hope this work can be translated into clinical settings, particularly in the realm of early rectal cancer treatment and follow-up.”
Dr Clara Mancini, Professor of Animal-Computer Interaction at The Open University, says: “Medical Detection Dogs do amazing work, leveraging dogs’ extraordinary olfactory intelligence to advance the early detection of life-threatening diseases, such as colorectal cancer. At the Open University’s Animal-Computer Interaction Lab, we are very excited to collaborate with this pioneering charity. Our first-of-its-kind interactive technology uses sensors to capture the dogs’ spontaneous response to the biological samples they sniff, producing data that can be analysed using machine learning to offer key insights into the dogs’ knowledge about the odours they examine. This will effectively enable the dogs to share with us information that could relate to the stage or aggressiveness of a disease.”