Smoke from cauterized tissues can ID tumor margins

by Jeffrey Perkel      

Sealing a head wound with heat (cautery) during the medieval era.

Many years ago, a close family member was diagnosed with breast cancer. The tumor was sufficiently small that she had options and chose a lumpectomy.

From the patient’s point of view, lumpectomy–removal of only the tumor–trumps mastectomy–complete breast removal–cosmetically, emotionally, and physically. But there’s one level where such a decision is more fraught. A surgeon who goes in to actually remove the tumor must be certain to get all of it. Otherwise, the cancer will simply return.

If there’s any doubt during the removal, the surgeon preserves the cancerous tissue either for immediate analysis while the patient is still on the operating table or for later examination. Both options are problematic. The time required to do the tissue analysis on the spot extends the length of the operation and thus the amount of time the patient is under anesthesia. If the tissue is evaluated post-op and the tumor margins aren’t “clean,” the patient must go under the knife again.

What surgeons really need is a way to determine, essentially in real time, where healthy tissue stops and the tumor begins, thus removing the analytical middle-man. But no such tool exists. Now, one is in development.

In the 17 July issue of Science Translational Medicine, Zoltán Takáts of Imperial College London, and colleagues in the UK and Hungary describe an in-development surgical device called the “intelligent knife,” or iKnife. It’s basically an electrosurgical tool coupled to a mass spectrometer, like a contractor’s stud-finder for the operating room. A mass spectrometer helps to identify the chemical content of a sample.

The REIMS-iKnife experimental setup. Image courtesy of Science Translational Medicine/AAAS.

Electrosurgery uses electrical current to cut and cauterize or seal tissue. That process generates smoke. As Takáts explains in the article,

Historically, surgical smoke has been considered as toxic and an irritant; therefore, it is often dispelled from the operative field. We hypothesized that this is a rich source of biological information and therefore used mass spectrometry to measure its metabolomic composition.

In other words, they thought that tissue going up in smoke might contain important molecules that mass spectrometry could identify. The iKnife samples that smoke via a process called REIMS (rapid evaporative ionization mass spectrometry), looks at its molecular content, and uses that to determine the nature of the tissue it just touched. The idea is that if the smoke arises from tumor tissue, the analysis of the smoke will distinguish that from smoke arising from non-tumor tissue, and a surgeon will know when the end of the tumor has been reached.

How does the machine do that? First, it needed to learn what constitutes healthy and diseased tissue. So, Takáts and his team built up a database of tissue profiles of 2,933 healthy, cancerous, and diseased but noncancerous tissues from 302 patients with gastric, colorectal, liver, breast, lung, or brain cancer.

The analysis looks specifically at lipids, the fatty molecules that make up cell membranes. Lipids are highly abundant, easily adaptable for mass spectrometry, and surprisingly telling molecular markers. But it wasn’t the specific identity of these molecules that mattered so much as their proportions. “What is really specific for a tissue is the distribution … the lipid fingerprint,” Takáts told me earlier this year, when I spoke to him for an article I was writing on mass spec imaging techniques.

The system interprets the chemical information – surgeons aren’t mass spec experts, of course – and gives the physician “histology level identification,” Takáts said.

To test the system, Takáts and his team applied it first to a series of surgically removed tumor samples. By looking at solid tumors in the lung and liver, they found the system could distinguish different types of tumors –adenocarcinoma and squamous cell carcinoma, for instance – as well as nearby healthy tissue. The system could also, in just a few seconds, differentiate lung and colon cancer metastases that had travelled to the brain from primary tumors that arose in the brain itself.

But all that was done “ex vivo,” outside of the patients. How would the iKnife fare during surgery itself? Superbly. As the authors write,

A total of 864 spectra were acquired from 81 patients and identified in all cases—that is, the spectra confirmed the result of postoperative histopathology. A sensitivity of 97.7% and a specificity of 96.5% were reached for binary classifications (cancer/healthy) of all cases. There was a low rate of both false-positive (3.5%) and false-negative (2.3%) results.

In a few cases, the technology flagged tissues that were thought to be cancerous but actually weren’t – a suspected case of colonic adenocarcinoma that turned out to be Crohn’s disease, for instance.

According to Takáts, the utility of the iKnife will vary from tumor to tumor. As he told me in March, it would be of less value in, say, the large intestine, where surgeons tend to cut out relatively large pieces of tissue and are limited by where blood vessels happen to be located. But in tissues like the kidney, pancreas, and brain, it could be a significant boon.

“We got the most positive reactions from brain surgeons,” he said, because of the difficulty they have in assessing how far into the healthy brain tissue the tumor infiltrates.

The iKnife system is being commercialized by Hungarian startup MediMass. According to the company’s web site, the company hopes to achieve regulatory approval “in all major markets” by the first quarter of 2015.

Should that happen, the decision of lumpectomy over mastectomy could potentially become significantly easier for patients to make.

PS: The iKnife isn’t the only system being developed for real-time histopathology in the OR, by the way. Researchers also are pursuing designs based on Raman imaging. But that’s a story for another day…

Image credit: Public domain image of smoke via Wikimedia Commons. Cautery image, public domain in the United States, also via Wikimedia Commons.