Robotics, AI and Other Colonoscopy Enhancements

When it comes to screening and diagnosing colorectal cancer, colonoscopies remain the gold-standard method of detecting colorectal polyps and tumors. However, like many other screening methods, it is not perfect and has its limitations, including how it:

• May not detect all polyps and tumors depending on location and size
• Can cause temporary discomfort and pain to the patient
• Can have complications such as bleeding and rarely colonic perforation
• Has sedation-related risks

While new screening methods such as liquid biopsy are developed to help overcome some of these shortcomings, this doesn’t mean that colonoscopy procedures remain at status quo. The outcome of the procedure — the successful detection of polyps and tumors in the colon and rectum — can be enhanced when a colonoscopy is paired with other technologies.

In this article, we explore the different ways in which the colonoscopy procedure can be enhanced, and if these have been implemented in clinical practice.

One of the limitations of a colonoscopy is that small polyps and tumors can sometimes be overlooked. This is especially true for sessile polyps, or polyps that grow flat against the colonic lining instead of protruding or growing from a stalk (pedunculated polyp). While most polyps are not cancerous, that sessile polyps tend to be easily missed or overlooked means that the risk of a cancerous growth going undetected is also higher.

Thankfully, it is possible to enhance the visibility of these hard-to-spot polyps. A blue dye known as methylene blue can be used to selectively stain “subtle and non-polypoid lesions, both adenomas and sessile serrated adenomas”, according to Dr Alessandro Repici, the primary investigator of a study that investigated the blue dye’s efficacy.

The use of blue dye during colonoscopies is not new. In the past, the dye had to be prepared right before the colonoscopy, and then sprayed onto the colon lining during the procedure itself. This had limited efficacy, as the dye would be localized to the sprayed area, and the process was also time-consuming and imprecise.

In their phase 3 clinical trial, Dr Repici and his team investigated the efficacy of delivering the dye to the colon through a delayed-release oral tablet, such that the dye is released into the colon in time for the colonoscopy. The team studied 1,205 patients who were scheduled for colonoscopies at 20 centers worldwide, and each patient was assigned to one of three test groups:

• A full dose (200 mg) of the methylene blue dye
• Half a dose of methylene blue dye
• Placebo group (no dye)

Note: The group that received half a dose of dye was not analyzed for efficacy, and were instead included to mask which groups the patients were assigned to, to prevent physician bias.

Compared to the placebo group, patients who received the full dose achieved a higher detection rate of adenomas, polyps or carcinomas. On top of that, more flat and small polyps and lesions smaller than 5 mm were identified in patients who received the full dose. Another encouraging outcome of the study was the minimal side effects of ingesting the dye tablet — outside of the expected blue feces and urine discoloration.

Despite the promising outcomes, the blue dye tablet has not been released for commercial use as of 2018. According to Cosmo Pharmaceuticals, the company behind the tablet, the US Food and Drug Administration (FDA) has recommended the researchers “provide confirmation of effectiveness with a second phase III trial”, citing that the outcomes were not “sufficiently robust” despite the statistical significance.

AI has slowly but surely become integrated into daily life, and the same is also true for cancer care. One of the ways AI has been implemented, specifically for colorectal cancer, is in screening colonoscopies. The expectation? Computer-aided detection, or CADe, would improve colonoscopy outcomes by increasing the detection rate of polyps and growths. The assumption here is that an increased detection rate would enable more polyps to be removed, thereby lowering the risk of cancer developing from these growths.

And, for the most part, CADe does indeed increase the detection rate of polyps. In a clinical trial of over 3,000 patients, more adenomas were detected overall when CADe was incorporated into the screening procedure. However, an important caveat that cannot be overlooked is that much of the detected adenomas were small and the least likely to develop into cancer. In the same vein, CADe did not increase the detection rate of advanced adenomas, which have a higher likelihood of becoming cancerous.

The benefits of detecting small, benign polyps have also been called into question. It is standard procedure to remove the polyps during the colonoscopy, but Dr Rodrigo Jover, who led the clinical trial, wrote that “removing them does pose small but real risks”, alluding to the risk of colon injury from having multiple polypectomies.

Aside from polyp detection, the use of AI in automated polyp diagnosis, or computer-aided diagnosis (CADx) has also been studied. Early studies in 2009 and 2011 reported that polyps could be diagnosed with an accuracy of 85.3% to 93.1%, though at the time, diagnosis by observers was reported to be superior still.

More recently, one study utilizing real-time CADx systems with ultramagnifying colonoscopes reported a negative predictive value of greater than 90% in real-time CADx of diminutive (smaller than 5 mm) polyps. The negative predictive value, which is a way of measuring how specific a test is, indicates the likelihood that person with a negative test result indeed does not have the disease. Therefore, CADx can recognize polyps that do not require polypectomy, thereby minimizing the number of unnecessary procedures.

At present, while AI does some benefits to the existing colonoscopy procedure, and experts believe it may be a few years yet before AI plays a critical, key role in colorectal cancer screening, owing to the lack of larger-scale randomization studies to weigh the risks and benefits of incorporating AI.

Another limitation of the colonoscopy procedure that is a big deterrent to timely screening is the pain, discomfort and injury to the colon it can cause. In order to increase patient compliance and encourage timely colonoscopies, Dr Keith Obstein of the Vanderbilt University Medical Center, Nashville, TN, hopes that his team’s “capsule robot” can make the procedure easier for both the patient and clinician conducting the scope test.

Traditional colonoscopes are inserted into the colon from the rectum, and are guided through the colon by the back end of the scope, which can result in “tissue stress”, as Dr Piotr Slawinski, one of the study authors, describes. The magnetized capsule colonoscope, on the other hand, can navigate the colon guided by an external magnet attached to a robotic arm that “pulls” the scope forward from the front.

In addition, the colonoscope is also able to complete retroflexion, which involves maneuvers that bends the scope backwards to allow technicians a reverse view of the colon lining. This can be done autonomously with the push of a button. To top it off, the capsule colonoscope could also perform the standard therapeutic procedures such as tissue biopsies and polypectomies.

For their study, Dr Obstein and his team successfully maneuvered through the colon of a pig 30 times. The capsule was also able to complete 30 retroflexions at an average of 12 seconds per retroflexion, which was within their expectations.

As of October 2023, the team has begun phase 1 clinical trials to investigate the ability of the capsule colonoscope — referred to as a magnetic flexible endoscope (MFE) in the study — to navigate the human colon from the rectum to the cecum (the beginning of the large intestine).

While efforts are being directed towards lowering the risks and barriers of the colonoscopy procedure, the clear benefit that timely colorectal cancer screening cannot be understated. Though being put off by the potential pain and discomfort will stop many from getting screened, at-risk individuals are nonetheless encouraged to prioritize their health and well-being by going through with the potentially lifesaving examination.