Optimizing Surgery With New AI Technologies in Breast and Ovarian Cancer

CancerNetwork® spoke with Anant Madabhushi, PhD, executive director of the Emory Empathetic AI for Health Institute and faculty in the Department of Radiology and Imaging Sciences and the Department of Biomedical Informatics and Pathology, and researcher of Cancer Immunology at Winship Cancer Institute of Emory University; and Farzad Fereidouni, PhD, an associate professor in the Emory University School of Medicine, about MarginCall, the AI-driven oncology surgery tool they are developing, at the Winship Cancer Institute of Emory University.

Currently, during breast and ovarian cancer surgeries, surgeons have to rely on the frozen section procedure, which requires them to freeze a piece of tissue from the surgery that a pathologist will evaluate for tumor tissue.² If the frozen specimen is positive for tumor tissue, the surgeon must go back to the patient and continue the surgery, and if it is negative for tumor tissue, the surgeon knows they are finished with the surgery. This process requires 2 specialists, and is time-consuming, often forcing the surgeon to spend 30 minutes in the operating room doing, essentially, nothing, according to Madabhushi.

They are developing the MarginCall tool in hopes that it will be able to optimize this process by greatly reducing the amount of time it takes, ultimately getting confirmation about whether there is positive or negative tumor tissue in minutes.

Additionally, Fereidouni added that the frozen section is a “terrible thing” and that it is destructive, ruining the tissue and affecting downstream molecular analysis.

Transcript:

Madabhushi:

One of the big challenges, right now, that we have in pathology is being able to provide rapid information and feedback to a surgeon in the operating room when they’re operating on [a patient’s] cancer. [Another] one of the big challenges that we have in cancer surgeries today is knowing exactly where the margin is, and what surgeons tend to do is try to be somewhat conservative, because you don’t want to take out too much normal tissue. You want to take out the tumor, obviously, but you don’t want to take out too much outside the tumor, because the more tissue taken out, obviously, [is] going to compromise the normal function of the patient. We all appreciate that there’s going to be some spread of the tumor outside the margin.

What typically tends to happen is that the surgeon will take out the tumor and some of the margin, and essentially, that piece of tissue gets frozen, and that frozen specimen is then looked at by a pathologist… If that’s completely negative [for tumor tissue], then the surgeon knows that she or he is good to go and there’s a high likelihood they got all the tumor. If it’s all positive, then that tells the surgeon that they probably need to take out more tissue, because there’s a spread of the tumor, and they need to make sure that they…don’t leave any tumor behind.

The problem with this current paradigm is that it’s somewhat labor-intensive and time-consuming because the surgeon is still in the [operating room], and there’s a time lag because that specimen then has to be frozen. It’s got to be taken over to the pathologist, or the pathologist to come down, [to] do the reading. It takes time. This could [last] from 30 minutes to an hour. It takes time, and at that same time, you have, essentially, the surgeon twiddling her or his thumbs…just waiting for that diagnosis to come back.

The question that we’re trying to address here is: “Could we make this far more rapid?” Could we accelerate the time in which we get the information about margin positivity or the lack thereof back to the surgeon so that instead of wasting precious minutes in the operating room, the information gets fed back to the surgeon, in maybe even 2 or 3 minutes. Essentially, providing rapid on-site diagnosis of the surgical margins so that the surgeon can then decide right there and then [know] whether they need to expand the margin or are [be fine] with what they took out.

Farzad:

Frozen section is a terrible thing, but we also don’t do it in many organs. Breast is 1 of them. We cannot do even a frozen section on them, so the situation is even worse than that. The frozen section is destructive. It ruins the tissue, it consumes the tissue, and it affects downstream molecular analysis.

References

Emory researchers awarded up to $17.6M from ARPA-H to innovate cancer surgery, improve outcomes. Emory Winship Cancer Institute. January 6, 2025. Accessed January 22, 2025. https://tinyurl.com/y3dpmtxr