Castle Biosciences’ TissueCypher® test can offer patients with Barrett’s esophagus a clearer picture of their risk of developing esophageal cancer.
Esophageal cancer is a killer. It’s the eighth most common form of cancer worldwide and the sixth leading cause of cancer death. The five-year survival rate is around 20%.
One reason for those grim numbers is that most esophageal cancers are diagnosed too late. The first symptom many patients develop is difficulty swallowing, and at that stage the tumor is not easily treatable in most cases.
Cadman Leggett, M.D.
Gastroenterologist, Mayo Clinic
“That’s essentially what happened to my father,” said Cadman Leggett, M.D., an expert in esophageal diseases and gastroenterologist with the Mayo Clinic. “My dad was in his early 50s when he developed esophageal cancer. He underwent chemotherapy, radiation therapy, and surgery, but unfortunately those measures were not enough to cure his disease. This is the reason why I decided to dedicate my career to fighting esophageal cancer.”
The tragedy of esophageal cancer is that it can be a very treatable disease if it’s caught early. Many people first develop a condition called Barrett’s esophagus (“BE” or “Barrett’s”), which is the only known precursor condition to esophageal cancer. A small percentage of patients with this condition may progress to cancer. However, if patients with BE at the highest risk for developing cancer are identified and treated early, they can often avoid esophageal cancer entirely.
As a physician scientist who has dedicated his career to the care of patients with BE and esophageal cancer, Dr. Leggett notes that today’s minimally invasive treatment for BE through endoscopy is highly effective in preventing advanced esophageal cancer. However, Dr. Leggett emphasized the importance of identifying patients at risk of progression and acting within the window of opportunity.
Barrett’s esophagus
BE is increasingly prevalent around the world. “A normal esophagus has a type of lining that is very similar to our skin,” Dr. Leggett explained. “With BE, that lining is injured by chronic reflux — basically, acid. The lining heals but is replaced by a type of lining that looks more like the lining found in your stomach, but with a distinct composition of cells.”
While BE is not itself malignant, it can progress to cancer over time. The term “dysplasia” is used to describe specific, pre-cancerous changes in the Barrett’s tissue — if there are no signs of these dysplastic changes, the condition is described as non-dysplastic BE (NDBE). Most people with Barrett’s are non-dysplastic. But in a subset of NDBE patients, BE can become dysplastic and may eventually become esophageal adenocarcinoma, an invasive form of esophageal cancer.
That makes screening for BE crucial. “If a patient has experienced chronic reflux symptoms at any point in their life, they should talk to their healthcare provider about being screened for BE,” Dr. Leggett said.
However, relying on chronic reflux (heartburn) symptoms in conjunction with other risk factors (like being older, being obese, or being a smoker) to trigger BE screenings is not completely effective. It’s possible to have BE and not know it because one can still have Barrett’s but not manifest any heartburn symptoms. That’s why everyone with any history of reflux should discuss screening with their healthcare provider regardless of whether they are currently experiencing heartburn.
Emmanuel Gorospe, M.D., M.P.H.
Gastroenterology Medical Director, Castle Biosciences
“BE can be a silent disease that may go on for years and put people at risk for cancer progression,” explained Emmanuel Gorospe, M.D., M.P.H., gastroenterology medical director at Castle Biosciences.
Dr. Gorospe calls BE a “paradox” because the treatments available to remove precancerous cells are usually incredibly effective — the biggest challenge arises in finding the highest-risk patients to apply them to.
Risk and limitation
Patients with BE are diagnosed and monitored through a procedure called endoscopy. “We use a flexible tube with a camera,” Dr. Leggett explained. “We examine the esophagus, the stomach, and the small bowel and focus on the Barrett’s segment. We target subtle abnormalities — in texture and vasculature — and obtain a tissue biopsy or perform a technique called resection. A high-quality endoscopic examination is extremely important in detecting these subtle areas of dysplasia.”
Those biopsies (esophageal tissue samples) are sent to a pathology lab where a pathologist will confirm whether BE is present and assign it one of four grades: non-dysplastic (NDBE), indefinite for dysplasia, low-grade dysplasia, or high-grade dysplasia. The patient’s gastroenterologist will use the pathology grade to determine the patient’s risk of progression and discuss next steps, which may include routine surveillance or discussion of BE treatment with endoscopic eradication therapy (EET).
“There are limitations in using dysplasia as a marker for risk,” Dr. Leggett noted. “We are only sampling a very small percentage of the esophagus — less than 5% — so it is prone to sampling error, and there is a chance of missing dysplasia. In addition, there can be a lot of discrepancy in the interpretation by the pathologist.”
In other words, all physicians have a margin for error when it comes to spotting existing dysplasia—and even if they do believe that they have spotted dysplasia, the pathologist receiving the resulting tissue samples may disagree or provide a subjective diagnosis.
Once someone is diagnosed with BE, performing surveillance of the BE tissue for signs of progression can be yet another challenge. Current guidelines outline disease management and treatment recommendations for patients with Barrett’s according to population-based risk data. As an example, the guidelines recommend that patients with non-dysplastic Barrett’s be rechecked for signs of dysplasia every three to five years. We know, however, that some patients will progress faster than expected, which is virtually impossible for physicians to predict without additional information.
Castle Biosciences, however, has introduced a “game-changer” to the process: TissueCypher.
Click here to learn more about TissueCypher
A fuller picture
“When you take a biopsy, it’s like looking up in the sky and seeing that it’s all clear at that moment,” Dr. Gorospe said. “But the question is, will it get cloudy tonight? Or tomorrow morning?”
Biopsies provide a snapshot of what the patient’s cells look like at a single point in time. But a pathology lab’s traditional analysis of those “snapshots” provides little information about the risk of future disease progression. That is a critical detail needed to prevent cancer, and physicians are often left to make educated guesses about when to schedule the next surveillance endoscopy. They can be cautious and schedule surveillance every year (with added costs), or they can schedule the next follow-up in three years, guessing that the BE won’t progress before then.
This is the game that TissueCypher is changing. Rather than have physicians make educated guesses about an individual patient’s risk of progression, this AI-driven test looks more deeply at the biopsies to predict the patient’s specific risk over the next five years.
“The TissueCypher test works by identifying nine biomarkers and the structure of the tissue,” Dr. Gorospe explained. “It looks at those together to produce a risk class for each patient — either low, intermediate, or high — as well as a score; a numerical figure that has a corresponding probability of cancer progression.”
The process looks beyond the visual clues traditionally used by pathologists and delves into the “unseen world” of molecular changes — changes that are impossible for the human eye to detect. TissueCypher employs a validated, AI-driven algorithm that removes human subjectivity from the analysis.
Human cells are incredibly complex; there are millions of data points in a typical tissue sample, well beyond the capability of humans to analyze efficiently. The AI platform used in TissueCypher was trained by analyzing these data points in both patients who progressed to cancer as well as those who didn’t progress. That resulted in identifying a “signature” that was consistent with progression. The intensity of that signature in a sample can be translated into a risk score.
“TissueCypher allows you to better inform the patient of their individual risk,” Dr. Leggett said. “When I get a TissueCypher test and it’s a low-risk score, then the patient often feels reassured. Or, if it’s an intermediate or high-risk score, it can empower the patient to be proactive about their own care.”
The larger medical community is recognizing that changes are needed in our approach to BE and esophageal cancer risk. The American Gastroenterological Association recently published clinical practice guidelines related to the use of EET in the treatment of patients with BE. The guidelines recognize that a subset of patients with NDBE may have a higher risk of developing esophageal cancer than traditionally assumed — and specifically acknowledged the potential benefit that tests like TissueCypher can have in identifying these patients.
Both Dr. Gorospe and Dr. Legget believe TissueCypher is a “game changer.” “Having an objective, personalized, and evidence-based prognostic test for BE can certainly change the way we care for patients,” Dr. Leggett said.
“Today’s biggest challenge in preventing esophageal cancer is not the availability of treatment, it’s the ability to identify patients at risk,” Dr. Gorospe said. “Using TissueCypher on a patient’s endoscopic biopsy tissue can enable clinicians and patients to have a more personalized discussion about therapy or surveillance aligned to the patient’s individual risk of developing cancer.”
Click here to learn more about TissueCypher
