Molecular Interception: Discussing the Future of Surgery in Breast Cancer

The surgical landscape is undergoing a fundamental transformation, moving from a scalpel-first tradition to a biology-driven strategy. In a recent dialogue with CancerNetwork®, Patrick Borgen, MD, discussed how neoadjuvant treatments are rewriting the clinical rulebook. By treating the whole body first, surgeons can now leverage advanced antibody-drug conjugates (ADCs) and checkpoint inhibitors that were previously restricted to later stages of care.

One of the most profound shifts involves minimal residual disease (MRD) and ctDNA. Borgen described “molecular interception”, the ability to detect cancer DNA fragments in the blood long before symptoms appear. This shift from calendar-based dosing to biological monitoring allows for surgical de-escalation and reduced toxicity.

Additionally, despite practicing in a global financial hub like Brooklyn, Borgen highlighted the stark reality of “surgical deserts.” He advocated for hub-and-spoke models to ensure underserved populations gain access to the same sophisticated testing and immuno-oncology found at major academic centers, ensuring that biology dictates a patient’s survival rather than geography.

Borgen is a chair in the Department of Surgery at Maimonides Medical Center.

CancerNetwork: Considering the emergence of total neoadjuvant surgery, how are you navigating the conversation with patients regarding the “Watch and Wait” protocol, and what are the most reliable clinical or imaging markers you use to ensure you aren’t missing a window for curative resection?

Borgen: I would break [this question] into 2 parts. For most of my career, surgery always came first: we took the mass out, we took the lymph nodes out, and then we [gave adjuvant therapy] systemically. Overwhelmingly, we’re in a new era where most of our patients will get neoadjuvant [treatment]. It might be chemotherapy; it might be endocrine [or] ablative therapy. The legitimate question that comes up for the patient is, “Well, isn’t it dangerous not to take the lump out a priori?”. The answer is no; it makes much more sense to treat the whole body first than to focus on the breast.

We come up against patients who push back. The problem that we face with sequencing of treatments, whether we do surgery or medical oncology first, is that on a lot of protocols, if you do the surgery first, you don’t have access to the ADC or the checkpoint inhibitor. We, the surgeons, can get it wrong. Sequencing used to be an art. It’s not that way anymore. It’s almost black and white, that there’s a right and a wrong way to sequence these treatments.

The second part is, what do we do after very successful neoadjuvant therapy where we’ve had, for example, a radiologic complete response? Do we still need to do surgery? If so, what imaging do we trust? There’s not a single [type of] imaging out there that will tell us whether there’s been a pathologic complete response. The best imaging that we have in the breast is MRI with enhancement, but it’s far from perfect.

The next question is, what about multiple core biopsies? What about image-guided assessment using core biopsies? There are lots of groups that are doing that. It makes perfect sense. Some of the extreme examples have talked about 25, 30, or 35 core biopsies. A core biopsy needle is 8 gauge. It’s almost as big as a pencil, so this is not a minimally invasive procedure. This has led researchers like Henry Kuerer, MD, from MD Anderson Cancer Center, to say, “If we have a complete response [on] imaging, why don’t we go to radiation? Why are we doing any surgery at all?” That’s the right question.

It’s not quite ready for prime time. There is a differential response from radiation. Some tumors respond beautifully, some don’t. Going straight away, assuming a pathological CR, is not quite ready for prime time. I believe it will be in subsets of tumors.

How do you envision MRD monitoring changing the timing of extensive surgery in the next 5 years, given the emergence of liquid biopsies and ctDNA? Could a negative ctDNA result eventually lead to omitting standard-of-care lymphadenectomy in certain disease states?

There probably has not been a more exciting development than the [MRD] or the ctDNA space. Measuring a person’s blood for fragments of DNA that do not belong is absolutely a technical tour de force and something that, 20 years ago, we would have thought was science fiction. Unfortunately, the technology arrived before our ability to interpret and act on it. Those trials are underway, and we will close that gap quickly. At the moment, we’re still struggling with when, where, and how to use MRD and ctDNA.

For decades, we’ve treated patients with breast cancer based on risk. This changes that. This treats them based on biology. We also used to treat patients based on a calendar; dose-dense doxorubicin [Adriamycin], cyclophosphamide, and paclitaxel [AC-T] was 4 doses of AC for over 8 weeks, followed by 10 weeks of a taxane. That doesn’t make a lot of sense if we’re using MRD and we see a complete disappearance of the ctDNA. It’s likely that it won’t be necessary to use the calendar to complete those treatment courses.

It’s easy to imagine avoiding CDK4/6 inhibitors in estrogen receptor [ER]–positive disease. It’s easy to imagine not having extended hormone ablative therapy. We’re talking about 10 years in most patients with [stage II] ER-positive breast cancer. It’s easy to understand that we may shorten or avoid chemotherapy based on MRD. It’s completely logical that it will de-escalate the surgery as well. I think the answer is yes, it will change what we do surgically.

The second part of the equation is true. For the patients who have metastatic disease absolutely sterilized by immunotherapy, systemic therapy, and checkpoint inhibitors who still have residual disease in the breast, we used to almost never operate on those patients. This could change that. We could be compelled to obtain locoregional control in someone whose systemic disease has been sterilized and is left with the [mass] in the breast. That will be another paradigm shift based on MRD. It is fun to be part of this right now as the story evolves.

As a leader in the field, what systemic changes are most critical to ensure that patients in underserved “surgical deserts” have access to the same subspecialized oncologic care found at major academic centers?

I practice in Brooklyn, New York, after having been at Memorial Sloan Kettering Cancer Center [MSKCC] for 17 or 18 years. Our cancer center is 6 miles from MSKCC. It’s 3 miles from Wall Street, the financial epicenter of the world, and it is a desert. Brooklyn [has] 3 million people. If it was a state, we would have more breast cancer [cases] than 15 US states. But the number and level of centers that are providing sophisticated testing and sophisticated surgery are very limited. This has to do with a lot of variables, including things like having a large uninsured and underinsured population. [They are] not always welcome at some of the academic medical centers; resources are not always there.

In Brooklyn, New York, I’m happy to report that we happen to have them, but a lot of patients don’t. Brooklyn has about 14 hospitals. Three of the 14 have the resources to do this. What we are trying to do is to create hub-and-spoke models, so that patients from other centers can be referred to the centers of excellence that have MRD testing, next-generation sequencing, sophisticated surgical techniques, [and] immune-oncology. We’re making some progress in that. In the next 5 years, we’ll see a major shift in urban centers like Brooklyn, New York, on not only where these patients are treated, but how they’re treated.

Technologies like fluorescence-guided surgery are helping surgeons identify occult nodes and positive margins in real time. Which disease fields do you believe stand to benefit most from this technology, and how close is this to becoming a standard tool in every oncology operating room?

If I step outside of breast cancer for a moment, we’re using fluorescent technologies in surgically resecting gliomas and other brain tumors, where it makes a tremendous difference between identifying tumor cells and brain cells. This is the most robust use I’ve seen today of these fluorescent technologies. In breast cancer, we’ve been tracking things like intraoperative margin assessment for years, and there are 2 problems.

One is that the technologies at the bedside are expensive. In a lot of payer mixes like ours, it’s difficult to come up with a business model that justifies that. With limited resources, that’s a bit of a luxury. Now, the technology will certainly evolve, but I’m not currently aware of a best-of-breed technology in breast or intraoperative margin assessment or sentinel node assessment. We’ll get there, but the better systemic therapies are, the less important these types of assays will be.

We talked earlier about not doing surgery in patients who’ve had successful runs of neoadjuvant therapy. There’s the phase 3 COMET trial [NCT02926911] in ductal carcinoma in situ [DCIS] looking at what [would happen] if we don’t take the DCIS out. That’s not ready for prime time, but that would certainly indicate that margin assessment may not be the critical thing. In breast cancer, it’s got ways to go. It’s ready for prime time in central nervous system tumors, and like everything else, it’s exciting.

What advances do you believe will have the most profound impact on surgical oncology over the next 5 years?

I’m going to expand the question, if I might, to oncology at large, and I’m going to focus on what I see happening in breast cancer. I’m very territorial about the Miami Breast Cancer Conference. Our motto for 4 decades has been, “Hear it on Friday and use it on Monday.” Even that’s evolving, because part of what’s going to happen this year is “I came in doing it on Friday and abandoned it on Monday.”

You’ve got these legacy defaults [like] ctDNA, for example. We used to wait for symptoms or radiologic progression to start treating progressive disease. Today, we can act on ctDNA immediately in things like looking for ESR1 mutations in previously treated ER-positive breast cancer, before scans show anything and before patients develop a symptom. In hormone receptor (HR)–positive disease, we’ve been treating it as a single entity, and it’s not. It’s a big family of diseases. We saw data this year about patients with a PIK3CA mutation in HR-positive disease. It’s a whole different subset of breast cancer.

The default for most of my career was that chemotherapy was our biggest gun. When we needed results––downsizing tumors, treating metastases––we used chemotherapy. That’s being replaced. ADCs are going from a salvage therapy to the backbone of our therapy, and that’s an exciting change. The bottom line is that we’re letting tumor biology drive our decisions about surgery, medical oncology, [and] radiation.

We’re going to use the biology of what’s there to select the appropriate therapy. I love the concept of molecular interception coming off the Super Bowl. We are replacing watching and waiting for symptoms to develop with molecular interception; the tumor tells us it’s coming back. Those are the paradigmatic changes that are going to lead breast cancer into the future.

Reference

Hwang ES, Hyslop T, Lynch T, et al. The COMET (Comparison of Operative versus Monitoring and Endocrine Therapy) trial: a phase III randomised controlled clinical trial for low-risk ductal carcinoma in situ (DCIS). BMJ Open. 2019;9:e026797. doi:10.1136/bmjopen-2018-026797