Webinar: Transforming oncology surgery with label-free technologies
In a recent webinar, Rob Morgan, Sagentia Managing Partner along with Rachel Smith and Thore Bücking from the Applied Science’s Team explored the potential role of label-free sensing in the smart surgery phenomenon.
Exempt from the regulatory and clinical burdens affecting biomarkers, these technologies could unlock new possibilities for oncology surgery. The webinar discusses the latest research, technical challenges and future potential in Raman, tissue autofluorescence and mass spectroscopy.
We would like to thank all who attended, and those who joined in by asking questions. Due to time restrictions, we were unable to answer all of them on the day, but we have published a selection of questions and answers here.
A: Fluorescence detection systems are already commercially available in the OR so auto-fluorescence can be used today. However, advances are needed to improve the discrimination of tissue, for example using time resolved fluorescence, necessitating the introduction of more advanced systems. Raman spectroscopy is also at an advanced stage as it has already been used in pioneering intra-operative studies and we are aware of one commercial system aiming for launch in 2021.
A: Several studies have used this technology to detect tumour margins in a variety of ex vivo samples. Although ex vivo analysis is helpful, because it provides the basis for development of in-vivo procedures, the main goal should be in vivo analysis during surgery in the OR. There is one group that is getting close to this at the University of California, and they have integrated tissue autofluorescence into the Da Vinci surgical robot. This has allowed real-time guidance during surgery of oral cancers of swine and human patients. It is an exciting development in this area and a step in the right direction for in vivo analysis during surgery.
A: It does depend on the surgical system that is being used, it’s probably easiest to integrate this data into the immersive displays found in some robotic surgery systems and in surgical microscopes. It’s interesting when you look at conventional laparoscopic surgery and how you might integrate this data with the screens which are used in lab surgery. Heads-up displays are an interesting area and there's other new technology emerging in the consumer electronics world that we will see applied in surgical environments.
A: Robotic surgery brings many benefits to surgery, in common with laparoscopic and minimally invasive surgery. However, the surgeon is one step removed from the surgical site and cannot use the sense of touch: many surgeons use palpate tissue during open surgery and get information that way. The ability to sense, characterize tissue and look at the spectra from the tissues is therefore particularly important in robotics because it makes up for the surgeon being one step removed from the surgical site.
A: In terms of resolution, the PIRL scalpel has a greater level of resolution in comparison to the iKnife. The resolution of the PIRL scalpel is around 200 μm compared to 4 mm with the iKnife. The higher resolution of the PIRL scalpel comes at the expense of sampling time. The sampling time of the PIRL scalpel is 5-10 s per spot compared to less than 2 s per spot for the iKnife. This would need to be resolved before the PIRL scalpel is used during surgery.
A: In terms of Raman spectroscopy, speed is an interesting area as it depends on how it is being sampled. It can be fast: the sampling rate for the point probes is about five hertz, which is a fraction of a second for every datapoint. That is only for the data acquisition; after this it needs to be analysed and due to advances in machine learning this is quick as well. While it takes a long time to train a neural network, using a trained network to make a decision for an input can be negligible in time when in comparison to the sampling speed. A powerful computer might be needed to do the calculations but that wouldn’t be an insurmountable hurdle from a technological point of view.
A: At the start of the presentation, we talked about the use of real-time data in many different applications in surgery. One of those was the identification and location of critical structures, for example major blood vessels that you don't want to hit during surgery, and nerves that you don't want to damage or compress. Nerves are a challenge as it isn’t easy to visualize nerve structures. There are already ways of sensing nerves using electrical signals during surgery but an enhanced ability to sense critical structures would be useful in in all sorts of different surgical applications and surgical specialties. It is an interesting area, it will be good to see how Raman spectroscopy, advanced auto-fluorescence spectroscopy and mass spectroscopy might be used in that application.
A: Raman spectroscopy imaging is commonly done but not in the OR. The benefit of using a point probe is that it is much faster to analyze because only one point is sampled at a time. By raster scanning, imaging is possible, however traditionally, something like that would take hours which is not practical during surgery. A lot of work has gone into making the imaging systems much faster and in recent years we've seen the scanning time go down drastically. One research group has achieved imaging speeds for Raman spectroscopy as high as about 10 minutes per sample. While this is not instant, a ten-minute wait time could be acceptable under some circumstances, such as analyzing resected tissue in the OR next to the patient.
A: Both sampling tools have similar accuracy in determining the tumour margins, but the MasSpec Pen is slightly more accurate; around 10% more accurate than DESI-MS. When comparing the DESI-MS and the MasSpec Pen, the main difference is the sampling time. The MasSpec Pen takes a few seconds, whereas the DESI-MS takes about three minutes so there is a significant difference there in a surgical setting. At the moment the MasSpec Pen is slightly more accurate and faster as well.