When surgeons can't determine the edges of a tumor, it's a problem. Cut too much, and they risk hurting the patient. Cut too little, and they may leave stray cancer cells behind. Now, researchers have developed a surgical knife that can sniff the smoke made as it cuts tissue, almost instantly detecting whether cells are cancerous or healthy.
The souped-up scalpel works by analyzing lipids, the fatty molecules that make up much of the cell membrane. In the past few years, chemists have shown that ratios of certain lipids can be used to identify various biological tissues, including tumors. But this requires first removing and preparing the tissue for a technique called mass spectrometry, which analyzes the mass and structure of charged molecules.
Hungarian chemist Zoltán Takáts wondered if he could speed things up by directly analyzing the smoke created by the electrosurgical knives that surgeons use to cut and cauterize blood vessels. Although the smoke is "a very nasty" tarry mixture, Takáts says, he realized that one component is a vapor containing ionized molecules—just what mass spectrometry needs. His team has shown that the lipid profiles identified by piping this vapor from an electrosurgical knife to a mass spectrometer correspond to different tissue types from animals.
Now, his group at Imperial College London, together with Jeremy Nicholson, a biochemist who heads Imperial College's department of surgery and cancer, has tested what they've dubbed the "intelligent knife," or iKnife, in the operating room. The team collected nearly 3000 tissue samples from about 300 cancer patients' surgeries, had pathologists identify if a sample was healthy tissue or a type of cancer, then matched up each result with the lipid profile they got by touching the iKnife to the same sample. They showed that the iKnife could distinguish normal and tumor tissues from different organs, such as breast, liver, and brain, and could even identify the origin of a tumor that was a metastasis, a secondary growth seeded by a primary tumor elsewhere in the body.
The researchers next tried out the iKnife during 81 actual cancer surgeries using the 3000-sample database as a reference. The iKnife results matched pathology lab results after the surgery for cancerous and normal tissues for nearly all patients, the researchers report online today in Science Translational Medicine. With only a 1- to 3-second delay for an iKnife readout, "it's real-time information" Takáts says. (Waiting for pathologists to analyze a sample can take up to 30 minutes.) That feedback could minimize the time a patient is under anesthesia and allow surgeons to work faster and more effectively. The team's next step is to conduct clinical trials to find out if using the iKnife helps patients develop fewer recurring tumors and live longer.
Although researchers are working on other ways to detect tumor margins in real time, those ways often require injecting a patient with a special dye to highlight the tumor. The iKnife is simpler because the process "is no different from what [physicians] normally do," Nicholson says. Surgeons will monitor a traffic light-like video display as they cut, he explains, with red indicating tumor cells, green healthy tissue, and yellow something in between.
The iKnife is "cool" because it is uses "a byproduct of surgery," says biomedical engineer Nimmi Ramanujam of Duke University in Durham, North Carolina, who uses optical imaging to detect tumor margins in tissue samples. She wonders, however, how useful the tool will be to surgeons, who would ideally have a complete image of the tumor margins before they cut into tissue, instead of having to feel out the tumor's edges by measuring single points with the iKnife. "The edge of a tumor is more challenging to detect than cancer versus healthy tissue due to a mix of different tissue types, and it is the edge that surgeons want to know about," Ramanujam says.