Hey there, folks. Today, we’re diving into some research about esophageal and stomach cancer, and let me tell you, it’s a concerning situation. Over the past few decades, the rates of esophageal adenocarcinoma (that’s EAC for short) and gastric cardia adenocarcinoma (GCA) have been skyrocketing in the United States and other western countries. And yeah, these types of cancer are no joke, they’re highly fatal.
But hold on, we’ve got some good news too. Dr. Joel Rubenstein, a research scientist and professor of internal medicine at Michigan Medicine, believes that early detection and screening can be a real game-changer. He says that by screening patients, we can catch pre-cancerous changes in the esophagus, known as Barrett’s esophagus. Now, this condition is often found in folks who have long-term gastroesophageal reflux disease, or GERD. And when we catch it early, we can take additional steps to prevent cancer. How about them apples?
The thing is, even though guidelines already recommend screening for high-risk patients, a lot of healthcare providers are still in the dark about this. It’s crazy, but true. So many people who develop these types of cancer never had a proper screening to begin with. But fear not, my friends, because there’s a new tool on the block that’s here to bridge the gap between provider awareness and patient risk. It’s an automated tool that’s embedded in the electronic health record system. Fancy stuff, I know.
Dr. Rubenstein and his team of researchers used some artificial intelligence magic to crunch the numbers. They examined data from over 10 million U.S. veterans who were at risk of developing EAC or GCA. And you know what they found? The results were published in Gastroenterology and they’re pretty darn cool.
They developed this tool called K-ECAN (I guess they like their acronyms) that can predict an individual’s risk of developing esophageal adenocarcinoma and gastric cardia adenocarcinoma. And get this, the tool uses basic information that’s already there in the electronic health record – stuff like patient demographics, weight, previous diagnoses, and routine lab results. It’s like magic, folks.
I should mention that they had another tool before called M-BERET, but it required measuring hip and waist circumferences, and it was a bit cumbersome to use. So, they came up with K-ECAN to make things easier for healthcare providers. They wanted to use the data that’s already there in the record and present the patient’s risk at just the right moments. Like when they’re due for a colorectal screening or when they’re refilling their acid reducing prescription meds. Genius, right?
And here’s the kicker, my friends. K-ECAN is more accurate than the current guidelines or any previous prediction tools. It can actually predict cancer at least three years before a diagnosis. I mean, that’s mind-blowing. And what’s cool about it is that it doesn’t just rely on GERD symptoms. See, GERD is a risk factor for esophageal adenocarcinoma, but not everyone with GERD ends up getting cancer. And get this, about half of the folks who do get cancer never had GERD symptoms to begin with. So, K-ECAN can identify those who are at risk, regardless of whether they have symptoms or not. That’s real game-changer stuff right there.
Dr. Akbar Waljee, who’s a professor at the University of Michigan, says that this research wouldn’t have been possible without a real collaborative effort. It took a whole team of dedicated folks to make it happen. They used data from millions of U.S. veterans, and it really shows the power of working together and using data and machine learning to improve cancer prevention.
Now, imagine incorporating this fancy AI tool into the electronic health record. It could give healthcare providers an automated notification about which patients are at an increased risk of developing esophageal adenocarcinoma and gastric cardia adenocarcinoma. That’s some next-level stuff right there. And you know what that means? It means we can significantly decrease the burden of these cancers. I mean, imagine the lives that we could potentially save.
Dr. Rubenstein and his team are thrilled about this new tool. They used some seriously sophisticated machine learning to make it happen, and they’re excited about the potential for increased screening and a decrease in preventable deaths. They’re even planning to do more work to validate K-ECAN for use outside of the VA. That’s just awesome.
So, my friends, let’s keep our eyes on the future of cancer prevention. With tools like K-ECAN, we’re getting closer to a world where we can catch these deadly diseases early and save lives. And that’s something we can all raise a glass to. Cheers.
– Michigan Medicine – University of Michigan
– Journal reference: Rubenstein, J. H., et al. (2023). Predicting Incident Adenocarcinoma of the Esophagus or Gastric Cardia Using Machine Learning of Electronic Health Records. Gastroenterology. doi.org/10.1053/j.gastro.2023.08.011.