January 10, 2021

Real-time identification of cancer cells during oncological surgical procedures

 

Real-time identification of cancer cells during oncological surgical procedures

Intra-abdominal malignancies often result in intraperitoneal free cancer cells (IPCCs), which increase the chances of cancer spreading to distal organs. Peritoneal carcinomatosis is a very common form of metastatic spread in gastric and colorectal cancer patients. Curative treatments for peritoneal carcinomatosis, such as cytoreductive surgery and intraperitoneal chemotherapy, have been shown to be effective, especially in malignancies of colorectal origin, thus increasing interest in free malignant cell detection. Various therapeutic strategies have been proposed to address the presence of IPCCs, currently detected with techniques such as pathological examination, immunocytochemistry (ICC), and polymerase chain reaction (PCR). However, the time between IPCC sampling and detection takes days to weeks. Consequently, treatment cannot be performed during the surgical intervention (i.e. detection in real time), a critical stage where immediate adjuvant treatment could significantly decrease a patient’s chances of relapse and metastasis. The aim of this research was to establish an ultra-rapid method for the detection of cancer cells in abdominal fluid during surgical procedures. Such a tool could be used to increase the efficiency of both surgical as well as post-operative treatments, thereby abating the recurrence and metastasis of intra-abdominal cancers.

For that purpose Prof. Guy Lahat and Dr. Shelly Loewenstein from the surgical division research laboratory collaborated with Prof. Noam Shomron, head of the functional genomics laboratory at the Sackler faculty of medicine in Tel Aviv University. Together they developed an ultra-rapid method combining deep Learning with Nanopore DNA sequencing technology for the detection of cancer cells in abdominal fluid during surgical procedures. The MinION (Oxford Nanopore Technologies) – the first handheld genetic sequencer – is capable of reading long stretches of DNA in a real-time. They used deep learning to identify and differentiate digital signatures of non-cancerous DNA (healthy DNA from patient mouth swabs or blood) and DNA from cancer cells (collected from intraperitoneal fluid) that have undergone Nanopore sequencing. These samples contain heterogonous populations of healthy and cancerous cells. Thus, deep reading of the DNA allows identification of small quantities of cancer-derived DNA. Preliminary results obtained from several gastric cancer patients’ samples have validated their approach demonstrating the utility of real time DNA sequencing and the use of deep learning to compare non-cancerous DNA to mutated cancer DNA. Combining DNA sequencing (as a ‘digital signature’) with deep learning will afford surgeons the opportunity to quickly identify IPCCs during surgical procedures, thereby allowing immediate treatment and decreasing the need for future intervention.