Chapters Transcript Video Surgical Considerations in DCIS and IDC Back to Symposium All right, so good evening and thanks to doctors Jane and Seidman, uh, for the invitation to discuss some of the surgical trials that were presented at San Antonio this year. This is my disclosure, which is not relevant to any of the talks. So we'll first, uh, we've got 5 abstracts in 20 minutes, so it's going to be quick. Uh, we, we're first going to look at LB 101, which is immediate breast surgery versus deferral of any surgery in women 70 and older. As we all know, the population is aging and the tumor biology is often more favorable in these older patients, uh, and that has permitted us, uh, to study de-escalation among them. So admission of axillary surgery and adjuvant radiation is fairly commonplace, but the question around primary endocrine therapy and whether immediate surgery at the primary site improves outcomes is still unknown. And so this group looked at the randomized controlled trials comparing tamoxifen versus surgery and tamoxifen in women 70 and older among those who had no chemotherapy and no radiation, and there are three such trials, all of which began enrollment in the 80s into the 90s, CRC in the UK, Greta in Italy, and Nottingham EPS 2 in the UK. Uh, you can see the number of patients that were enrolled in each of these trials, uh, and also the proportion of patients in these trials which are hormone receptor positive, which becomes important when you're investigating tamoxifen. Uh, it was not reported in the CRC UK study. 82% were hormone receptor positive in the Greta trial and all were hormone receptor positive in Nottingham. Uh, because EPS-1 was similar to the other trials where it was a mixed bag and so they refined their criteria to being all hormone receptor positive for EPS 2. A total of a little over 1000 patients were enrolled and randomized in all three of these trials, about half to tamoxifen in surgery and half to tamoxifen alone. In the tamoxifen alone group, surgery was deferred until progression was seen. This was a patient level meta-analysis, so they had all of the data for all patients as opposed to just abstracting data from the specific manuscripts. They looked to time the local regional failure to distant recurrence and to breast cancer mortality. The median age in all three trials, 76, 63% had T2 or greater tumors. In the surgery arm, a little under half underwent breast conserving surgery and of those who were randomized to surgery, 7% underwent no surgery. And we have about 7 years of follow up. So we looked at both local regional uh recurrence in the patients who had surgery and then those who were in the tamoxifen alone arm progression, which was defined as greater than or equal to 25% of an increase in tumor diameter. They first uh subdivided the population into node negative and node positive disease, and there's no surprise to anyone that in the tamoxifen alone group there were more progressions or what we would refer to as recurrences than the tamoxifen plus surgery group. What was maybe a little bit surprising is that there was not a huge difference between the node negative and node positive group. Looking at the forest plot, uh, specifically subdividing, uh, all three of the randomized trials and then by tumor size, uh, all of those subcategories of the patients favored tamoxifen plus surgery, uh, as opposed to tamoxifen alone. They then looked at some other breast cancer outcomes, distant recurrence as well as breast cancer mortality, and we can see that of those who underwent tamoxifen plus surgery they had improved rates of distant recurrence as well as breast cancer mortality. However, these were not evident in. Until years 3 and 4 respectively, uh, which then leads the authors to suggest what we really need to think about a patient's longevity, especially in a group of older patients in order to reap some of these benefits because it takes us a while to see a difference. So in the absence of radiotherapy, the authors conclude that immediate surgery offer or reduced local progression and recurrence when compared to deferred surgery. There were moderate reductions in distant breast cancer, distant recurrence in breast cancer mortality. Um, and there was no evidence of any effect on non-breast cancer mortality, so patients who underwent tamoxifen versus surgery didn't die any more frequently from other things. The limitations of this trial really are the mixed bag of hormone receptor positive and hormone receptor negative patients, though the majority were hormone receptor positive. They were on tamoxifen and in common practice we're using AIs in these patients at this point, so we don't. Have that data, nor do we have data on the comorbidities or quality of life from these patients, which would be really interesting to know. Uh, does a sicker patient actually benefit from surgery as much as a non-sick patient and so it leaves us that individualized discussion is still necessary. Surgery certainly improves outcomes in patients, but some patients may still not require surgery. Which brings us to GS 108, which is a study of young BRCA carriers with breast cancer looking at the benefit of risk reducing surgery. Uh, so what we all in the room know is the rates of breast cancer for BRCA1 and BRCA2 carriers being much higher than that, uh, in the, uh, general population, but contralateral breast cancer at 10 years being 12 and 9% in BRCA 1 and 2 respectively compared to 5% in the general population. And the associated risk of ovarian cancer certainly being higher among BRCA1 carriers and of those with cancer treatment plans really need to take into account not only these, uh, the risk of relapse of the primary tumor, but also the risks of secondary malignancy, which is why I have the contralateral tumor data here. And age is likely a factor, mostly because younger women have a longer time to live and therefore have a longer time to have a contralateral breast cancer. And it may also be because of estrogen exposure, that being they may still have their ovaries in place. And so the BCY collaborative asked whether risk reducing surgery impacted survival among breast BRCA carriers with breast cancer, uh, and they looked at about 5300 patients who were younger or at 40 years of age. Stage 1 to 3 treated from 2000 to 2020. You can see this is a multi-site, uh, study. They did an analysis specifically of risk reducing mastectomy in about half of patients in the retrospective analysis underwent risk reducing mastectomy, and they also did an analysis of risk reducing salpingo oophorectomy and similarly about 50% of patients had a salpingo oophorectomy. When looking at the combination, about 1/3 of patients had both, and so that was a cohort that the authors were very interested in. So who were the patients overall young because that's what they were selected for 35, uh, predominantly BRCA1 carriers with a mixed bag of tumor stage of nodal stage. Uh, about a third had luminal-like cancers, about half triple negative breast cancer, and 7% HER2 positive breast cancer, and the vast majority had chemotherapy. Their primary outcome was overall survival, and this is the uh Kappa Mayer curve for the risk reducing mastectomy group, and you can certainly see that risk reducing mastectomy improved overall survival with a hazard ratio of 0.65. They did subgroup analysis. For overall survival and found that there was no difference uh according to the BRCA gene, according to age, according to time of testing, meaning whether or not these patients knew at the time of their breast cancer diagnosis that they were a carrier for BRCA or whether this information uh came about afterwards or their pathology um of their breast cancer. There was certainly a similar uh finding of a benefit to risk reducing mastectomy for disease-free survival and a breast cancer free interval. When looking at the salpingo oophorectomy group, there was a similar improvement in overall survival among patients who underwent salpingo oophorectomy. And unlike in risk reducing mastectomy, there was a seen benefit, uh, in BRCA1 when compared to BRCA2 and in triple negative breast cancer when compared to the other subtypes, and this is likely because BRCA1 patients are more likely to present with triple negative breast cancer. And a similar finding for disease-free survival, uh, and breast cancer free interval. They then asked a question about that 3 of patients who underwent both and whether or not there was an interaction between risk reducing mastectomy and risk reducing salpingo oophorectomy, and they found no interaction with overall survival, but there was a significant interaction for both disease-free survival and breast cancer disease free interval. And so the authors suggest exploring that uh a bit more. Uh, so risk reducing mastectomy and oral pingo oophorectomy improved overall survival irrespective of the BRCA pathogenic variant that the patient harbored, but risk reducing salpingo-oophorectomy was more beneficial for BRCA1 patients, and that's counter and to. Thinking that maybe BRCA patients who have uh a larger preponderance for hormone receptor positive uh disease might actually uh derive a greater benefit from salpingo oophorectomy, but this is likely due to the ovarian cancer risk reduction. And so these data do help to improve counseling on risk and management strategies for young, so less than 40. BRCA patients with breast cancer. What the authors stress is that these recommendations are not for older patients or for those contemplating risk reducing surgery who do not have a breast cancer diagnosis. That's not the population that was studied. So this information, you know, you've got a 38 year old woman with breast cancer in front of you, this is when you can use these data for, um, discussion. Highly anticipated abstract, uh, was the initial uh Comet trial, uh, analysis, and Doctor Shelly Huang from Duke presented, uh, the cancer outcomes from Comet and Doctor Anne Partridge, uh, presented the patient reported outcomes and we'll review them both. So the question is, are we overtreating DCIS? Since the advent of screening mammogram in the 1980s, we see a huge uptick in the cases of DCIS that we're diagnosing and probably greater than 50,000 expected in this country this year. But we have not seen a corresponding decrease in the local and regional, um, as well as distant disease, uh, when we diagnose this DCIS. So is there some part of DCIS or some form of DCIS that we're sort of overtreating where we can leave it alone and we don't need to treat it? And that's sort of what the comment asked Is active monitoring or watchful waiting for those who treat prostate cancer an appropriate strategy for low risk DCIS? They enrolled 957 patients with hormone receptor positive DCIS. It could be either or either ER positive, PR positive, or both. Uh, low to intermediate grade, so no high grade, and these patients had to be over the age of 40. If there was a span of calcifications greater than or equal to 4 centimeters, two biopsies were required to confirm that this was just DCIS and there was no evidence of an invasive component, but there was no limit on the span of calcifications for enrollment in the trial. Uh, you can see the randomized stratification by age, by, uh, the, um, grade as well, uh, as, um, specifically the, the two arms here, guideline concordant care, which is half of patients, and active monitoring. And what guideline concordant care was was upfront surgery, uh, radiation if recommended, and endocrine therapy was permitted. And for active monitoring, this was a mammogram every 6 months and a biopsy if a change on the mammogram had been found and again, endocrine therapy was permitted. Interestingly, however, the active monitoring arm seemed to be pretty enticing to the participants because you can see that over half of the patients who were randomized to guideline concordant care chose to undergo active monitoring instead, and a smaller proportion of patients who were randomized to active monitoring decided to go undergo guideline concordant care instead. Uh, these patients represented about a third of the cohort who did not undergo their assigned or their randomized arm, which the authors, um, say that they a priori, um, assumed that that's what, uh, they would find and so, uh, this is well powered for their outcome, which is 2 year ipsilateral invasive cancer. Uh, you can see the difference between the arms. They're very well balanced in terms of age, grade, and, and comorbidities, aside from the fact that greater comorbidities were seen in the guideline concordant care, but these were all sort of minor comorbidities. And when they did analysis in four cohorts, meaning assigned to guideline concordant care and underwent guideline concordant care, assigned to guideline concordant care, but underwent active monitoring, and so on, uh, those 4 cohorts were also well balanced. So the slide that everybody's looking for is, is active monitoring inferior to guideline concordant care and you can see the intention to treat analysis on the left and then the per protocol analysis on the right and we can see that active monitoring in fact was not inferior to. To guideline concordant care and probably what we assume right the the uptick in the breast cancers in blue, which is the guideline concordant care are those patients who upstage at the time of their excision of their DCIS um and then those who, uh, have, um, have invasive cancer in the active monitoring arm or those who have a change on their mammogram over time. Interestingly, they looked at then what happened after randomization and if they underwent any sort of treatment, the mastectomy rate and the intention to treat arm, uh, 5.5% versus 3.7% with a higher proportion in the guideline concordant arm, which you would expect, but in the as treated protocol you can see 10% versus 1%. Uh, and that must mean that of those patients who are undergoing active monitoring, they were already in their mindset of doing a less is more type of a paradigm and therefore, uh, chose breast conservation rather than a mastectomy. Uh, high rates of endocrine therapy, about 2/3 up to 3/4, uh, radiation more in the active monitoring arm, probably because, uh, they had more breast conservation, uh, and a low rate of chemotherapy, 1%. From a patient reported outcomes, there were many surveys that they did. Here are some of the, um, some of the surveys. I won't, um, talk about each of them, but a high response rate, about 100% completed at least one survey and greater than 83% response rate at each time point, which is really pretty good for a PRO type study. Uh, and then when they looked, we're going to look at the intention to treat, but when they did sensitivity analysis and per protocol, it was the same. So as you can see, no significant difference in general health or mental component of the PROs. There is a significant but non-clinical non-clinically significant, uh, difference, uh, in physical functioning and probably the biggest question is, well, what about the worry of DCIS? Is this an all consuming, uh, feature of this patient's life if they're on active monitoring and there was clearly no difference between the groups and worry about DCIS. So in the short term, active monitoring had a non-inferior rate of invasive cancer. Uh, patient reported outcomes didn't differ, and active monitoring did little to affect the well-being of these patients, but this is probably not ready for prime time. This is just 2 years. You probably need a longer follow up in order to really put this into into practice, um, and also more data on the effect of endocrine therapy. There was no presentation about data of how about the patients who didn't take endocrine therapy are those the patients who ultimately um had an invasive cancer. So more to come on comet, but really good early data to suggest that we probably have some patients for whom this active monitoring strategy is acceptable. And our last abstract, EEA. Does axillary surgery improve outcomes is a question that surgeons will debate until they can't debate any longer, and there's been continued de-escalation of surgery in the axilla from axillary lymph node dissection for all, which we haven't been doing for a long time, omitting axillary lymph node dissection in pathologically node negative disease on sentinel node biopsy, or for those with limited no nodal disease like Z11. However, there's now a push for how about the patients who are clinically node negative, can we omit axillary surgery in these patients as well? And that's what the EEA trial asked. Asked specifically can surgical axillary staging be entirely omitted as part of breast conserving therapy without compromising outcomes and they they randomized uh about 5000 patients who were T1 to T2 and 0. That's clinical, but I have an asterisk asterisk here because they coined this IN 0, which is imaging node negative because all patients had to have a negative preoperative axillary ultrasound. Uh, all of them went underwent breast conserving surgery and whole breast irradiation. Uh, there are about 5000 patients treated per protocol, and they were randomized in a 1 to 4 fashion. Uh, no sentinel node biopsy, which is the 1 and sentinel node biopsy, which is the 4. Uh, they looked to see, well, what do we find when we do the sentinel node biopsy and you can see, you know, not a huge amount of nodal disease, but they wanted to take this a little bit, a little step further and specifically for those with 1 to 3 macro metastasis, they further randomized them to axillary dissection or not, and those results have not yet been presented, so we're just focused on the the sentinel node biopsy portion of this. Uh, the groups were well balanced, um, as, as usual, but some things to point out is median age 62 and only about 10% were less than 50%, so an older group. Uh, most were T1, um, unlike sound where they were all T1, uh, here we have 10% which were T2. Uh, from a grade perspective, only about 3% being grade 3, and the vast majority being hormone receptor positive, HER2 negative, as expected. From a chemotherapy standpoint, we can actually see there was chemotherapy more frequently utilized in those who underwent sentinel node biopsy. So if you start to think, well, maybe we medically treated those who didn't undergo axillary surgery and that's why their outcomes are the same. That wasn't true. Um, those were equivalent in sound but higher here, um, and a high rate of endocrine therapy and. Uh, we're sort of used to these types of Kaplan Meyer curves in the de-escalation, uh, uh, studies, uh, as it comes to breast surgery. There was absolutely no difference uh between the arms as it relates to, uh, disease with a 5 year, uh, with good 5 year follow up in the group. Uh, when they looked at the intention to treat, it was the same. Uh, they not to sort of belabor the point forest plot, looking at all the subgroups that we might be interested in, T2 and so on, and again, uh, no statistically significant difference between any of the groups. So great, you know, we can sort of prove that they didn't have any difference, but what about the rates in the axilla? Do do these patients all of a sudden manifest axillary disease? Well, similar to sound, uh, the local regional recurrence rate was only less than 2%, 1% axillary recurrence in the no sentinel node biopsy arm, and 0.3% in the sentinel node biopsy arm. Uh, 1% is greater than 0.3%, but 1% is really low, and we probably shouldn't, uh, be doing a bunch of surgery for 1%. Uh, no distance, uh, no difference in distant relapse. So they conclude omission of axillary surgery associated with no improvement in local regional or distant control. No improvement in overall survival and a lower use of adjuvant chemotherapy, but they don't actually give us genomic data and genomic data are probably going to be what guide our systemic therapy uh in especially our postmenopausal patients if they have the one node, it's still going to guide your systemic therapy. Like sound, all patients had a negative preoperative axillary ultrasound. The primary outcome in sound was not, uh, was not, uh, ipsilateral, uh, recurrence. It was distant disease-free survival, but their local regional recurrence was very similar. So they conclude that de-escalation using this paradigm of a negative preoperative axillary ultrasound is suitable for patients who are clinical T1 and 0, age 50 or greater, with grade 1 to 2 breast cancer that is hormone receptor positive, HER2 negative, and again this is specifically in those undergoing breast conserving therapy. No patients underwent a mastectomy in this trial. So in summary, while de-escalation continues to be a prevailing theme in breast surgical oncology, an important role remains, meaning we're not putting ourselves out of a job for risk reducing surgery. We saw that in the BRCA, uh, abstract for primary site surgery, we saw that in our patient's 70 and older abstract and for axillary. Surgery in those who are not who don't fit the criteria that we just discussed and certainly as always multidisciplinary discussion and patient preference, which is important, remains critical to identifying patients who are suitable for these de-escalation strategies and with that we almost did it 5 abstracts in 20 minutes. Created by Related Presenters Austin Williams, MD Assistant Professor, Department of Surgery, Fox Chase Cancer Center Assistant Professor, Department of Surgery, Fox Chase Cancer Center View full profile
