The use of artificial intelligence may help determine how breast cancer tumors develop resistance to certain therapies, notes Rakhshanda Rahman, MD, FRCS, FACS.
It is important for providers to establish a “network” among themselves and their patients to keep up to date with all the latest technological advancements in oncology and continue to push the envelope further for breast cancer care, according to Rakhshanda Rahman, MD, FRCS, FACS.
Rahman, a professor of Breast Surgical Oncology at Texas Tech University Health Sciences Center and the medical director of UMC Cancer Center, spoke with CancerNetwork® about the latest breast cancer surgical treatment strategies and technologies she has adopted in her clinic, including improvements in cryoablation and use of magnetic surgical devices.
She also discussed the importance of implementing early breast cancer screenings to bolster the efficacy of systemic therapies.
“We need to keep increasing the advocacy towards screening for patients. There are still a lot of misgivings in people’s minds,” Rahman said. “They may be scared of radiation or do not have time; there are lots of reasons why people choose to not have a screening. Nothing makes a bigger difference to mortality from breast cancer like screening does. All these advancements are great, but if patients do one thing, they should get screened.”
Additionally, Rahman spoke about the importance of the shared decision-making process, especially when determining potential surgical strategies such as lumpectomy vs mastectomy. Considering whether the patient would prefer a better cosmetic outcome or the peace of mind that may come with a mastectomy is valuable, she said.
Rahman: The whole point of screening is to detect cancers at a very early stage. We do know that the curability rate is much higher for a screen-detected cancer, which is north of 95%, compared with a symptomatic patient. Secondly, the smaller the bulk of the tumor, the more likely the success of systemic therapies. All the therapies we use have some toxicities, but the risk/benefit ratio becomes much better if you have a smaller bulk to deal with. Then we can choose appropriate therapies depending on the type of cancer.
It’s basically a lot of shared decision-making. The first issue is whether we can save the breast. The size of the tumor vs the size of the breast is very important. Even with relatively larger tumors, we can use a lot of oncoplastic techniques to make a cosmetically appropriate outcome, even though a large volume of the breast was removed. Those are some technical things that we assess when we look at the patient. But then it comes to the question of shared decision making. Generally speaking, when we save the breast, we need to follow up with radiation therapy to the breast.
Radiation therapy is a time commitment on behalf of the patient and the health care system. Because that’s part of saving the breast, that is on the table to be worked into the decision-making process. In other words, there are patients who sometimes choose not to save the breasts—even if the tumor was small—because they live too far away and thus cannot follow through with radiation. Sometimes they are not willing to experience the potential adverse effects [AEs] of radiation, as well as work-related issues. There are many things that go into that decision making. Cosmetic outcome is so important to some patients that they want to do everything possible to save their breasts. Some people are too scared about having to keep up with follow-up mammograms that they want both breasts gone. You must meet the patient where they are and entertain all the options to land on a feasible decision.
Within the last few years, we have come to understand the tumor biology much better. For example, I do genomic profiling with 150 gene tests of the tumor to figure out exactly what type it is. Based on these test results, we can predict who would benefit from chemotherapy and who would benefit from endocrine therapy. If the test shows that someone would benefit from chemotherapy, then I work towards trying to use the chemotherapy first. If there are more responses up front, the surgical task becomes easier.
The other important thing is that with more advancements in screening technologies in terms of quality of image, we’re able to find very tiny tumors, which is good news for the patient. It is also a little bit of difficult for the surgeon because now I must find this tiny tumor and remove it. When the tumor is very tiny, we have to figure out the best possible way to find exactly where to go and get it. For many decades, we had been using a very thin wire. The patient will go to the imaging center, they’ll put in a wire, the wire will stick out of the breast, and then the patient will be transported to the operating room where the wire will be followed to take the tumor out. [There are many] logistics issues associated with that procedure.
Nowadays, we have magnetic devices like MOLLI®, which is a little magnet that is implanted in the breast with imaging. However, nothing sticks out of the breast, which means it doesn’t have to be placed the morning of surgery; it doesn’t bother the patient or anybody else. When the patient comes to the operating room, I can use a detecting wand that will find the magnet and give me the exact distance so that I can plan my surgery and do it as cosmetically as possible.
As for the third new option, there are some very low-risk cancers that do not require chemotherapy or radiation therapy either if the patients are elderly. In that case, ablating the tumor while it is still in the breast—having to remove the lump and give a scar—is not really needed. I prefer cryoablation; I think the benefits are higher and the risks are lower with cryoablation. It uses a needle that can skewer and freeze kill the tumor. It takes about 20 minutes and then the patient can go home.
If you really see what AI is, it’s basically the ability to compute things that the human brain cannot compute. Ever since we had computers, many things have already been using AI when we do things like data mining. [The most recent forms of AI] have a new twist, not because the AI is new, but because the way it’s delivered is new. There are some misgivings I have about it. But from a scientific perspective, if we stick to what was always [understood as] AI in terms of being able to compute large swaths of data beyond the capacity of a human brain, then our learning gets that much farther.
Right now, for example, we’re seeing some great work, which I think will majorly change the face of breast cancer. Some companies are looking at the entire human genome on the tumor cell. Without AI and computational capabilities, you don’t know which gene is the culprit [for the disease] and what to attack. With a blood test, we were able to find those genes, but we didn’t know what to do with it. Now that all those data can be computed with the use of AI, you can draw these blood tests to figure out what changed in the tumor that taught it to learn resistance, and what to attack to kill it. That’s where AI is going to be most helpful.
We also heard about the use of the imaging side of AI. There are lots of little pixels that can be fed into some algorithms, which can compute for you what is more likely to be cancer than not. There is some positivity to it. The impact will be a little bit smaller [than the genome application] because we are already very good at detection. The room for improvement is not that significant with all the technological development that has already been out there. However, it could help from a fiscal responsibility perspective.
For example, there could be an algorithm that might say that if you are in a certain group based on an image, you [may not require testing] for a couple years instead of needing testing every year, which may have an impact on health care costs. [This tool] is in infancy at this point; I don’t know which direction it would go. I don’t think the detection alone would make a huge impact, but if we can do it in a more cost-effective way [with AI], that might be a more fruitful impact.
One of the things I’ll caution, though, is that with all these advancements, people should not misunderstand the use of AI to replace the good old-fashioned regular [human] intelligence. That’s what my worry is. Talking to your patient, understanding where they’re coming from, what the realities are, and making the best treatment plan; that’s the human connection. There are also benefits with telemedicine, and I’m all for it in certain angles. But some algorithms are being promoted to replace a doctor/patient relationship. I don’t think that’s ever going to be a good idea.
We were able to document that our operating time was much faster when we use MOLLI compared with when we used wires. We’re going to work on that paper; I’ve submitted an abstract, so we’ll see where that goes. We are also trying to move cryoablation forward. I am offering cryoablation to low-risk patients with cancer. On the research side and some of the animal model work that I’ve done, we’ve shown that there is some immune response to cryoablation, as well. We feel that even some of the higher-risk cancers could benefit from cryoablation in addition to regular treatment; that’s what we’re investigating right now.
When you bring those questions to the human trials, there are lots of ethical concerns that we have to figure out to make sure that people still get the best possible treatment while you’re answering your scientific questions. That’s the second area I’m working on.
The third area we’re working on is leveraging this human genome profiling of tumors that I already talked about. We are looking at a couple of genes that we feel are less used in the clinical realm but may have something to do with why cancer becomes more aggressive. We’re trying to figure that part out to see if that will materialize into some new targets and new drug development.
Keeping up with stuff is very important; there are so many papers every month. We used to be generalists; we would manage the hernia one day, the gallbladder another day, and then cancer on another. Things have gotten very specialized because of the technological advancement. It is impossible to keep up with all that’s going on. I feel that the takeaway for providers is to find your niche in the specialty and keep up with all the newer technologies. They should have that network because when we try new things—as is the nature of science—they are not going to be perfect.
There will be good ideas, but sometimes you figure out something didn’t go well. That network is very important; when people are trying new things, we should actually help each other out. We must keep meeting and talking to each other to keep pushing the envelope further and further. From the patient perspective, that is important for them to ask the right questions. They can push us into delivering better and better care that’s more suitable for them.
In the middle of this, the elephant in the room becomes the access. Not every patient living in every rural county is going to have access to the latest and greatest advancements. We also need to be cognizant to focus on scalability of these technologies. How can we not just develop stuff that’s available in Manhattan, New York but make sure that someone in a [small rural county] has access to the same quality of care with these technological advancements?
MOLLI: precision surgery made simpler.MOLLI Surgical Inc. Accessed November 9, 2023. https://shorturl.at/mvCM6