Abstract
Background
Historically, more than one-third of patients with DCIS treated with breast-conserving surgery (BCS) underwent additional surgery. The SSO-ASTRO guidelines advise 2mm margins for patients with DCIS having BCS and whole-breast radiation (WBRT). Here we examine guideline impact on additional surgery and factors associated with re-excision.
Methods
Patients treated with BCS for pure DCIS from 8/2015–1/2018 were identified. Guidelines were adopted on 9/1/2016, and all patients had separately submitted cavity-shave margins. Clinicopathologic characteristics, margin status, and rates of additional surgery were examined.
Results
Among 650 patients with DCIS who attempted BCS, 50 (8%) converted to mastectomy. Of 600 who had BCS as final surgery, 336(56%) received WBRT and comprised our study group. One hundred twenty-eight (38%) were treated pre-guideline and 208(62%) post-guideline. Characteristics and margin status were similar between groups. Re-excision rate was 38% pre-guideline and 29% post-guideline adoption (p=0.09), with 91% having only 1 re-excision. Re-excision for ≥2mm margins was uncommon (6% pre- vs 5% post-guideline). On multivariate analysis, younger age (OR 0.97,95%CI 0.94–0.99,p=0.02) and larger DCIS size (OR 1.43,95%CI 1.2–1.8,p<0.001) were predictive of re-excision; guideline era was not. Younger age (OR 0.93,95%CI 0.9–0.97,p<0.001) and larger size (OR 1.64,95%CI 1.3–2.1,p<0.001) were predictive of conversion to mastectomy, but residual tumor burden was low.
Conclusions
The SSO-ASTRO guidelines did not significantly change re-excision rates for DCIS in our practice, likely since re-excision for margins ≥2mm was uncommon even prior to guideline adoption, dissimilar to historically observed variations in surgeon practices. Younger age and larger DCIS size were associated with additional surgery.
Keywords: ductal carcinoma in situ, margins, breast-conserving surgery, whole-breast radiotherapy
As many as one-third of patients undergoing breast-conserving surgery (BCS) for ductal carcinoma in situ (DCIS) will require at least one re-excision for close or positive margins.1 This may be attributed to the multifocal, skip-pattern of growth that is observed in 70% of well-differentiated DCIS lesions as well as the inability to palpate the majority of DCIS.2 While the early randomized trials of RT demonstrated a clear association of positive margins with a higher risk of local recurrence (LR) after BCS for DCIS, most of these trials did not quantify negative margin distance and lacked standardized pathologic evaluation, limiting ability to study the impact of margin distance.3–6 As a result, there has historically been a lack of consensus among surgeons regarding the optimal margin width for DCIS.7
In 2016 the Society of Surgical Oncology (SSO), the American Society for Radiation Oncology (ASTRO), and the American Society of Clinical Oncology (ASCO) established a consensus statement based upon results of a meta-analysis including 7883 women from 20 studies.8 They concluded that among women undergoing BCS and WBRT for DCIS, a margin threshold of 2 mm minimized risk of LR as compared to more narrow margins (odds ratio [OR] 0.51, 95% confidence interval [CI] 0.31–0.85), while a margin threshold wider than 2 mm offered no greater benefit among patients receiving WBRT (OR 0.99, 95% CI 0.61–1.64).9
The 2014 dissemination of consensus guidelines for invasive breast cancer undergoing BCS and WBRT has resulted in a decline in re-excisions for invasive cancers reported in multiple single-center and population-based studies10–12, with a 35% reduction in odds of re-excision observed in a meta-analysis.13 We sought to examine the impact of the margin consensus guidelines on rates of additional surgery after initial BCS for DCIS, and to identify factors predictive of additional surgery.
METHODS
Following Institutional Review Board approval, we identified all women treated with initial BCS for pure DCIS at Memorial Sloan Kettering Cancer Center between August 2015 and January 2018. Patients with any microinvasive component or bilateral disease, and those with prior or concurrent invasive cancers were excluded from the study. Throughout the study period, all surgeons at our institution routinely used seed localization for non-palpable lesions. A cavity-shave margin technique was uniformly employed, with resection of the index lesion followed by separate submission of individual margins with a metal clip or suture designating the final margin surface. A single-view intraoperative radiograph of the index specimen was performed to confirm retrieval of the target lesion. Each margin was examined individually, and when tumor was present in a separately submitted margin, a millimeter measurement was provided for the distance of tumor cells from the inked surface. For analysis, margin status was grouped as tumor on ink, < 2 mm from the inked margin, or ≥ 2 mm from the inked margin.
The Memorial Sloan Kettering Cancer Center breast disease surgery team adopted the SSO-ASTRO consensus recommendation on September 1, 2016. Patients were categorized into two study periods: “pre-guideline”, defined as those treated before September 1, 2016; and “post-guideline”, defined as those treated after September 1, 2016. Patients treated during this time did not receive intraoperative radiation, partial-breast radiation, or brachytherapy.
Clinicopathologic characteristics, margin status, and rates of re-excision and conversation to mastectomy were examined. Patients in the two study periods were compared using the Wilcoxon rank-sum, Fisher’s exact, and chi-square tests as appropriate. Univariable and multivariable analyses were carried out using logistic regression. Categorical predictors were candidates for the multivariable models if each category had at least 10 patients who experienced the event of interest. Multivariable models were chosen using an automated search procedure that selected the models with the best Akaike Information Criterion (AIC). Two-sided p-values of less than 0.05 were considered statistically significant. All analyses were performed using R 3.6.2 (R Foundation for Statistical Computing, Vienna, Austria).14
RESULTS
Between August 2015 and January 2018, 650 consecutive patients with pure DCIS were treated with initial BCS, of whom 50 (8%) converted to mastectomy (Fig. 1). Among 600 patients who had BCS as final surgery, 336 (56%) received WBRT and comprised our primary study group: 128 (38%) were treated pre-guideline and 208 (62%) post-guideline. Clinicopathologic characteristics are summarized in Table 1. The median age was 55 years, median DCIS size was 1.2 cm, and 37% had high nuclear grade. Nearly all cases (98%) were clinically occult and detected by imaging, and most patients had heterogeneously dense or extremely dense breasts (67%). The majority of patients were estrogen receptor (ER) positive (91%), preoperative MRI was performed in 32%, and 40% of ER positive patients received adjuvant endocrine therapy. All clinicopathologic features examined were similar between the pre-guideline and post-guideline periods (Table 1).
Fig. 1.
Study population.
DCIS ductal carcinoma in situ
TABLE 1.
Clinicopathologic characteristics and margin status of patients treated with breast-conserving surgery and whole-breast radiotherapy, stratified by guideline era
| All n = 336 |
Pre-guideline n= 128 |
Post-guideline n = 208 |
P | |
|---|---|---|---|---|
| Age, years, median (range) | 55 (31–79) | 54 (39–79) | 56 (31–78) | >0.9 |
| Diagnosed by | 0.5 | |||
| Exam finding | 6 (1.8%) | 2 (1.6%) | 4 (1.9%) | |
| Mammogram/U S | 329 (97.9%) | 125 (97.7%) | 204 (98.1%) | |
| MRI | 1 (0.3%) | 1 (0.8%) | 0 (0%) | |
| Breast density | 0.7 | |||
| Fatty | 19 (5.7%) | 9 (7.0%) | 10 (4.8%) | |
| Scattered fibroglandular | 93 (27.7%) | 38 (29.7%) | 55 (26.4%) | |
| Heterogenous | 193 (57.4%) | 70 (54.7%) | 123 (59.1%) | |
| Extremely dense | 31 (9.2%) | 11 (8.6%) | 20 (9.6%) | |
| Preoperative MRI | 0.5 | |||
| Yes | 106 (31.5%) | 44 (34.4%) | 62 (29.8%) | |
| No | 230 (68.5%) | 84 (65.6%) | 146 (70.2%) | |
| Pathologic DCIS size, cm, median (range)* | 1.2 (0.1–8.5) | 1.2 (0.1–8.5) | 1.3 (0.1–6.0) | 0.4 |
| Nuclear grade | 0.8 | |||
| Low | 21 (6.3%) | 8 (6.3%) | 13 (6.3%) | |
| Intermediate | 191 (56.8%) | 70 (54.7%) | 121 (58.2%) | |
| High | 124 (36.9%) | 50 (39.1%) | 74 (35.6%) | |
| ER status | 0.5 | |||
| Positive | 307 (91.4%) | 119 (93.0%) | 188 (90.4%) | |
| Negative | 29 (8.6%) | 9 (7.0%) | 20 (9.6%) | |
| Necrosis | 0.4 | |||
| Yes | 231 (68.8%) | 92 (71.9%) | 139 (66.8%) | |
| No | 105 (31.3%) | 36 (28.1%) | 69 (33.2%) | |
| Endocrine therapy | >0.9 | |||
| Yes | 123 (36.6%) | 46 (35.9%) | 77 (37.0%) | |
| No | 213 (63.4%) | 82 (64.1%) | 131 (63.0%) | |
| Closest margin width at index surgery | 0 2 | |||
| Tumor on ink | 38 (11.3%) | 19 (14.8%) | 19 (9.1%) | |
| < 2 mm | 99 (29.5%) | 40 (31.3%) | 59 (28.4%) | |
| ≥ 2 mm | 199 (59.2%) | 69 (53.9%) | 130 (62.5%) | |
| Number of < 2 mm margins at index surgery | (n = 137) | (n = 59) | (n = 78) | > 0.9 |
| 1 | 90 (65.7%) | 38 (66.4%) | 52 (66.7%) | |
| 2 | 30 (21.9%) | 13 (22.0%) | 17 (21.8%) | |
| ≥ 3 | 17 (12.4%) | 8 (13.6%) | 9 (11.5%) | |
| Re-excision performed | 109 (32.4%) | 49 (38.3%) | 60 (28.8%) | 0.09 |
Represents n = 235 cases; n = 73 had slides counted and n = 28 had DCIS present with no measurement of extent
US ultrasound, MRI magnetic resonance imaging, DCIS ductal carcinoma in situ, ER estrogen receptor
The pre- and post-guideline eras did not differ in closest margin width at index surgery (p = 0.2) or in the number of < 2 mm margins at index surgery (p > 0.9) (Table 1). Overall, the rate of re-excision was 32.4%. Re-excisions declined from 38.3% in the pre-guideline era to 28.8% in the post-guideline era, although this did not reach statistical significance (p = 0.09). Among 109 patients who underwent re-excision, 99 (90.8%) required only 1 re-excision (85.7% pre-guideline vs 95.0% post-guideline, p = 0.11). Nearly all patients with tumor on ink underwent re-excision during both eras (100% pre-guideline vs 89.5% post-guideline). Re-excision for negative but < 2 mm margins was similar (65.0% pre-guideline vs 62.7% post-guideline), while re-excision despite having ≥ 2 mm margins was uncommon throughout the study period and was similar between the two eras (5.8% pre-guideline vs 4.6% post-guideline), respectively.
On univariable analysis, larger DCIS size (OR 1.34, 95% CI 1.10–1.65, p = 0.004) was associated with re-excision, while guideline era, breast density, preoperative MRI, receptor status, and grade were not associated. On multivariable analysis, younger age (OR 0.97, 95% CI 0.94–0.99, p = 0.02) and larger DCIS (OR 1.43, 95% CI 1.16–1.77, p < 0.001) were predictive of re-excision, while guideline era was not (Table 2).
TABLE 2.
Univariable and multivariable regression analysis of risk factors for re-excision among patients with pure DCIS treated with breast-conserving surgery and whole breast radiotherapy
| Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| OR | 95% Cl | P | OR | 95% Cl | p | |
| Age, years (continuous) | 0.98 | 0.96–1.0 | 0.11 | 0.97 | 0.94–0.99 | 0.021 |
| DCIS size, cm (continuous) | 1.34 | 1.1–1.65 | 0.004 | 1.43 | 1.16–1.77 | <0.001 |
| Breast density | ||||||
| Fatty or SF | REF | |||||
| Heterogeneous | 1.59 | 0.96–2.69 | 0.08 | |||
| Extremely dense | 1.57 | 0.66–3.64 | 0.3 | |||
| MRI performed | ||||||
| No | REF | |||||
| Yes | 1.11 | 0.67–1.80 | 0.7 | |||
| Nuclear grade | ||||||
| Low | REF | |||||
| Intermediate | 0.73 | 0.29–1.92 | ||||
| High | 0.83 | 0.32–2.25 | 0.7 | |||
| ER status | ||||||
| Negative | REF | |||||
| Positive | 0.90 | 0.41–2.09 | 0.8 | |||
| Guideline era | ||||||
| Pre-guideline | REF | |||||
| Post-guideline | 0.65 | 0.41–1.04 | 0.08 | 0.65 | 0.35–1.20 | 0.2 |
DCIS ductal carcinoma in situ, OR odds ratio, CI confidence interval, SF scattered fibroglandular, ER estrogen receptor
In a separate analysis, patients with pure DCIS who converted to mastectomy (n = 50) were compared to the remainder of the population who underwent BCS alone (n = 600). Patients who converted to mastectomy were younger (median age 49 vs 58 years, p < 0.001), had larger DCIS size (median 1.75 cm vs 0.9 cm, p < 0.001), and more often had high nuclear grade (40% vs 24.5%, p = 0.02), dense breasts (86% vs 61.5%, p = 0.003), and preoperative MRI (46% vs 28.7%, p = 0.02) (Table 3). On univariable analysis, younger age (OR 0.94, 95% CI 0.91–0.97, p < 0.001), larger DCIS size (OR 1.45, 95% CI 1.14–1.82, p = 0.002), extremely dense breasts (OR 4.98, 95% CI 1.72–14.8, p = 0.003), preoperative MRI (OR 2.12, 95% CI 1.17–3.8, p = 0.01), and high nuclear grade (OR 6.05, 95% CI 1.71–38.5, p = 0.02) were associated with conversion to mastectomy, while guideline era was not. On multivariable analysis, younger age (OR 0.93, 95% CI 0.9–0.97, p < 0.001) and larger DCIS size (OR 1.64, 95% CI 1.3–2.1, p < 0.001) remained predictive of conversion to mastectomy (Table 4). Prior to conversion to mastectomy, 32 (64.0%) had no re-excisions attempted, 14 (28.0%) had 1 re-excision attempt, and 4 (8.0%) had 2 re-excision attempts. Residual tumor burden at mastectomy was low, with 30% of patients having had no residual tumor identified, and with a median size of 0.5 cm (range 0.15–1.8 cm) among those with measured residual disease.
TABLE 3.
Clinicopathologic characteristics of patients who converted to mastectomy compared to those underwent breast-conserving surgery alone
| All (n = 650) |
Converted to mastectomy (n = 50) |
No conversion to mastectomy (n = 600) |
P | |
|---|---|---|---|---|
| Age, years, median (range) | 58 (30–91) | 49 (33–79) | 58 (30–91) | <0.001 |
| Diagnosed by | 0.02 | |||
| Exam finding | 11 (1.7%) | 2 (4.0%) | 9 (1.5%) | |
| Mammogram/US | 634 (97.5%) | 46 (92.0%) | 588 (98.0%) | |
| MRI | 5 (0.8%) | 2 (4.0%) | 3 (0.5%) | |
| Breast density | 0.003 | |||
| Fatty | 38 (5.8%) | 1 (2.0%) | 37 (6.2%) | |
| Scattered fibroglandular | 200 (30.8%) | 6 (12.0%) | 194 (32.3%) | |
| Heterogenous | 351 (54.0%) | 35 (70.0%) | 316 (52.7%) | |
| Extremely dense | 61 (9.4%) | 8 (16.0%) | 53 (8.8%) | |
| MRI performed | 195 (30.0%) | 23 (46.0%) | 172 (28.7%) | 0.02 |
| Pathologic DCIS size, cm, median (range)* | 1.0 (0.05–8.5) | 1.75 (0.3–5.6) | 0.9 (0.05–8.5) | <0.001 |
| Nuclear grade | 0.02 | |||
| Low | 91 (14.0%) | 2 (4.0%) | 89 (14.8%) | |
| Intermediate | 392 (60.3%) | 28 (56.0%) | 364 (60.7%) | |
| High | 167 (25.7%) | 20 (40.0%) | 147 (24.5%) | |
| ER status | >0.9 | |||
| Positive | 609 (93.7%) | 47 (94.0%) | 562 (93.7%) | |
| Negative | 38 (5.8%) | 3 (6.0%) | 35 (5.8%) | |
| Unknown | 3 (0.5%) | 0 (0%) | 3 (0.5%) | |
| Necrosis | 0.4 | |||
| Yes | 373 (57.4%) | 32 (64.0%) | 341 (56.8%) | |
| No | 277 (42.6%) | 18 (36.0%) | 259 (43.2%) | |
| Radiotherapy | <0.001 | |||
| Yes | 336 (51.7%) | 0 (0%) | 336 (56.0%) | |
| No | 313 (48.2%) | 50 (100%) | 263 (43.8%) | |
| Unknown | 1 (0.2%) | 0 (0%) | 1 (0.2%) | |
| Endocrine therapy | 0.001 | |||
| Yes | 203 (31.2%) | 5 (10.0%) | 198 (33.0%) | |
| No | 446 (68.6%) | 45 (90.0%) | 401 (66.8%) | |
| Unknown | 1 (0.2%) | 0 (0%) | 1 (0.2%) | |
| Guideline era | >0.9 | |||
| Pre-guideline | 276 (42.5%) | 22 (44.0%) | 254 (42.3%) | |
| Post-guideline | 374 (57.5%) | 28 (56.0%) | 346 (57.7%) |
Represents n = 435 cases in entire population; n = 122 had slides counted and n = 93 had DCIS present with no measurement of extent
US ultrasound, MRI magnetic resonance imaging, DCIS ductal carcinoma in situ, ER estrogen receptor
TABLE 4.
Univariable and multivariable regression analysis of risk factors for conversion to mastectomy among patients with pure DCIS treated with initial breast-conserving surgery
| Univariable analysis | Multivariable analysis | |||||
|---|---|---|---|---|---|---|
| OR | 95% Cl | P | OR | 95% Cl | P | |
| Age, years (continuous) | 0.94 | 0.91–0.97 | <0.001 | 0.93 | 0.90–0.97 | <0.001 |
| DCIS size, cm (continuous) | 1.45 | 1.14–1.82 | 0.002 | 1.64 | 1.27–2.11 | <0.001 |
| Breast density | ||||||
| Fatty or SF | REF | |||||
| Heterogeneous | 3.66 | 1.69–9.11 | 0.002 | |||
| Extremely dense | 4.98 | 1.72–14.8 | 0.003 | |||
| MRI performed | ||||||
| No | REF | REF | ||||
| Yes | 2.12 | 1.17–3.8 | 0.01 | 2.14 | 0.93–4.91 | 0.07 |
| Nuclear grade | ||||||
| Low | REF | |||||
| Intermediate | 3.42 | 1.0–21.4 | 0.1 | |||
| High | 6.05 | 1.71–38.5 | 0.02 | |||
| ER status | ||||||
| Negative | REF | |||||
| Positive | 1.06 | 0.37–4.05 | >0.9 | |||
| Guideline era | ||||||
| Pre-guideline | REF | |||||
| Post-guideline | 0.93 | 0.52–1.69 | 0.8 | 0.54 | 0.24–1.18 | 0.12 |
DCIS ductal carcinoma in situ, OR odds ratio, CI confidence interval, SF scattered fibroglandular, ER estrogen receptor
DISCUSSION
In this large, consecutive cohort of patients with pure DCIS treated with BCS and WBRT from August 2015 to January 2018, we found that rates of re-excision declined from 38% to 29% after adoption of the SSO-ASTRO consensus guidelines, but this was not statistically significant. Re-excision for ≥ 2 mm margins was uncommon throughout the study period, while re-excision of margins < 2 mm was frequent, potentially explaining the lack of a significant change in the post-guideline period since the margins advocated by the guideline were already widely adopted. Young age and larger DCIS size were independently predictive of both re-excision and conversion to mastectomy.
Optimizing local control in patients with DCIS undergoing BCS is of great importance, as half of all LRs are invasive and confer a subsequent increased risk of disease-specific mortality.5 Historically, there has been controversy and lack of consensus regarding the ideal margin width in this setting. In a population-based surgeon sample from the Surveillance, Epidemiology and End Results (SEER) registry for patients treated from 2005 to 2007, 42% of surgeons preferred a margin width of > 1–2 mm, while 48% preferred a margin width of > 5 mm among patients with DCIS having BCS with planned radiation.7 The 2016 SSO-ASTRO consensus guidelines established a threshold of 2 mm as minimizing risk of LR among patients undergoing BCS and WBRT for DCIS, based upon review of available literature, including results from a study-level meta-analysis of 7883 patients from 20 studies wherein no greater benefit was seen with wider margins than 2 mm (OR 0.99, 95% CI 0.61–1.64).8,9 However, favorable long-term rates of local control have previously been demonstrated in early trials which required a margin of only “no ink on tumor”.5 Accordingly, for women with negative but < 2 mm margins, the guidelines advise use of clinical judgment for selective re-excision with consideration of factors such as individual risk of LR, cosmesis, and life expectancy.8
The earlier dissemination of consensus guidelines endorsing “no ink on tumor” as a negative margin for stage 1 and 2 invasive cancers undergoing BCS and WBRT has resulted in a significant decline in re-excisions, reported in multiple single-institution studies and population-based samples.10–12,15 In a population-based study of 3729 women with stage 1 and 2 breast cancer treated with initial BCS from 2013 to 2015 identified from the Georgia and Los Angeles County SEER registries, a significant decrease in additional surgery was observed after dissemination of the clinical guidelines, with final rates of BCS increasing from 52% to 65%.15 A meta-analysis including 16,282 pooled pre-guideline and 15,900 post-guideline patient data points from 7 studies of patients with stage 1 or 2 cancers similarly reported a reduction in re-excisions from 22% to 14% after guideline publication, with an overall 35% reduction in the odds of re-excision.13 However, the impact of the more recent guidelines for DCIS remains unclear and, to our knowledge, has not yet been studied in any large patient samples. In a survey of 767 responding members of the American Society of Breast Surgeons, over 90% were familiar with the recent guidelines for margins in DCIS, and were more likely to avoid re-excision to attain margins ≥ 2 mm (82% pre-guideline vs 88% post-guideline, p = 0.002) and more likely to undertake re-excision for a margin < 2 mm (26% pre-guideline vs 37% post-guideline, p < 0.001).16 Our current study showed an absolute 9% decline in re-excisions after guideline adoption in our practice; however, this was not statistically significant. We attribute this to consistent practice patterns during both periods at our institution, with re-excision for ≥ 2 mm margins being similarly infrequent before and after guideline adoption (5.8% pre-guideline vs 4.6% post-guideline), respectively.
The less-pronounced impact of the margin guidelines for DCIS as compared to invasive cancer may be, in part, due to the individual physician- and patient-level decision making advised for patients with < 2 mm margins. In our current study, a moderate frequency of re-excision was observed for such patients throughout the study period: 65.0% in the pre-guideline era vs 62.7% in the post-guideline era. An accurate estimation of individual risk of LR is a key component of this decision making. LR risk among DCIS patients treated with BCS is known to be modified by risk factors such as age, volume of disease, treatment period, and receipt of radiation and/or endocrine therapy. In a contemporary cohort of 1491 patients with DCIS treated with BCS with or without adjuvant radiotherapy, Tadros et al. found the association between margin width and locoregional recurrence to vary by receipt of radiotherapy (interaction p = 0.02). While there was no difference in 10-year locoregional recurrence rates among those with < 2 mm and ≥ 2 mm margins receiving RT (HR 0.8, 95% CI 0.2–3.2, p = 0.7), there was a significantly higher likelihood of recurrence among those with < 2 mm margins compared to ≥ 2 mm margins when foregoing RT (HR 5.5, 95% CI 1.8–16.8, p = 0.003).17 These findings emphasize the need for a nuanced multidisciplinary discussion for such patients with negative but < 2 mm margins to accurately estimate individual risk and optimally select further treatments, whether by means of adjuvant therapies and/or additional surgery.
Conversion to mastectomy after initial BCS for DCIS is an uncommon event. In our study, only 8% of all patients ultimately underwent mastectomy, similar to previously reported rates ranging from 9% to 11%.15,18,19 However, in this study, we found that 64% of such patients did not attempt any re-excisions prior to completion mastectomy, and that residual tumor burden at mastectomy was very low. The retrospective nature of our study does not allow us to delve into patient decision making regarding decisions to convert to mastectomy. However, the observed minimal residual disease supports consideration of further intervention in decision making at this juncture, as it is possible that many patients may have achieved adequate margins with re-excision and could have avoided mastectomy.
Understanding factors predictive of conversion to mastectomy is of relevance in an era of increasing choice of upfront mastectomy for DCIS, particularly among young women.20,21 In our study, we found young age and larger DCIS size to be predictive of both conversion to mastectomy and re-excision, consistent with prior reports. In a population-based sample of 316,114 patients with stage 0 to 2 breast cancers who underwent initial BCS from 2004 to 2010, younger age and larger DCIS size were significantly associated with a greater likelihood of repeat surgery, and these remained predictive on subset analysis of patients with DCIS alone.19 Cognizance of these factors is important for appropriate preoperative discussion and counseling.
Limitations of our study include its retrospective nature and inability to examine physician- and patient-level decision making. The population size may also limit ability to further discern differences between the two eras. However, its strengths included examination of a large, consecutive population of patients treated at a high-volume center with consistent surgical and pathologic techniques. Awareness of risk factors and implementation of the evidence-based recommendations for margin width may lead to continued improvements in tailoring and optimizing surgical management for patients with DCIS.
Conclusions
In this consecutive population of pure DCIS patients treated with initial BCS and WBRT, re-excisions declined from 38% to 29% after guideline adoption, but this was not statistically significant, likely as our institutional re-excision practices for ≥ 2 mm margins were consistently low throughout the study period. This is in contrast to the wide variation in surgeon practices that was observed prior to consensus margin definitions, with many surgeons favoring wider margins. Conversion to mastectomy was uncommon. Young age and larger DCIS size are predictive of additional surgery after initial BCS. Our data support further study regarding patient-level decision making at the juncture of conversion to mastectomy after initial BCS, particularly in young patients.
Synopsis:
Here we examine SSO-ASTRO guideline impact on additional surgery and factors associated with additional surgery after lumpectomy for DCIS. The guidelines did not significantly change re-excision rates in our practice, and younger age/larger DCIS size were associated with additional surgery.
ACKNOWLEDGEMENTS
The preparation of this manuscript was funded in part by NIH/NCI Cancer Center Support Grant No. P30 CA008748 to Memorial Sloan Kettering Cancer Center, and this study was accepted for presentation in ePoster format at the 2020 Society of Surgical Oncology International Conference on Surgical Cancer Care. Dr. Monica Morrow has received honoraria from Genomic Health, and all other authors have no conflicts of interest to disclose.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
REFERENCES
- 1.Morrow M, Jagsi R, Alderman AK, et al. Surgeon recommendations and receipt of mastectomy for treatment of breast cancer. Jama October142009;302(14):1551–1556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Faverly DR, Burgers L, Bult P, Holland R. Three dimensional imaging of mammary ductal carcinoma in situ: clinical implications. Semin Diagn Pathol. August1994;11(3):193–198. [PubMed] [Google Scholar]
- 3.Warnberg F, Garmo H, Emdin S, et al. Effect of radiotherapy after breast-conserving surgery for ductal carcinoma in situ: 20 years follow-up in the randomized SweDCIS Trial. J Clin Oncol. November102014;32(32):3613–3618. [DOI] [PubMed] [Google Scholar]
- 4.Donker M, Litiere S, Werutsky G, et al. Breast-conserving treatment with or without radiotherapy in ductal carcinoma In Situ: 15-year recurrence rates and outcome after a recurrence, from the EORTC 10853 randomized phase III trial. J Clin Oncol. November102013;31(32):4054–4059. [DOI] [PubMed] [Google Scholar]
- 5.Wapnir IL, Dignam JJ, Fisher B, et al. Long-term outcomes of invasive ipsilateral breast tumor recurrences after lumpectomy in NSABP B-17 and B-24 randomized clinical trials for DCIS. J Natl Cancer Inst. March162011;103(6):478–488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Correa C, McGale P, Taylor C, et al. Overview of the randomized trials of radiotherapy in ductal carcinoma in situ of the breast. J Natl Cancer Inst Monogr. 2010;2010(41):162–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Azu M, Abrahamse P, Katz SJ, Jagsi R, Morrow M. What is an adequate margin for breast-conserving surgery? Surgeon attitudes and correlates. Ann Surg Oncol. February2010;17(2):558–563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Morrow M, Van Zee KJ, Solin LJ, et al. Society of Surgical Oncology-American Society for Radiation Oncology-American Society of Clinical Oncology Consensus Guideline on Margins for Breast-Conserving Surgery With Whole-Breast Irradiation in Ductal Carcinoma In Situ. J Clin Oncol. November202016;34(33):4040–4046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Marinovich ML, Azizi L, Macaskill P, et al. The Association of Surgical Margins and Local Recurrence in Women with Ductal Carcinoma In Situ Treated with Breast-Conserving Therapy: A Meta-Analysis. Ann Surg Oncol. November2016;23(12):3811–3821. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rosenberger LH, Mamtani A, Fuzesi S, et al. Early Adoption of the SSO-ASTRO Consensus Guidelines on Margins for Breast-Conserving Surgery with Whole-Breast Irradiation in Stage I and II Invasive Breast Cancer: Initial Experience from Memorial Sloan Kettering Cancer Center. Ann Surg Oncol. October2016;23(10):3239–3246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chung A, Gangi A, Amersi F, Bose S, Zhang X, Giuliano A. Impact of Consensus Guidelines by the Society of Surgical Oncology and the American Society for Radiation Oncology on Margins for Breast-Conserving Surgery in Stages 1 and 2 Invasive Breast Cancer. Ann Surg Oncol. December2015;22Suppl 3:S422–427. [DOI] [PubMed] [Google Scholar]
- 12.Bhutiani N, Mercer MK, Bachman KC, et al. Evaluating the Effect of Margin Consensus Guideline Publication on Operative Patterns and Financial Impact of Breast Cancer Operation. J Am Coll Surg. July2018;227(1):6–11. [DOI] [PubMed] [Google Scholar]
- 13.Havel L, Naik H, Ramirez L, Morrow M, Landercasper J. Impact of the SSO-ASTRO Margin Guideline on Rates of Re-excision After Lumpectomy for Breast Cancer: A Meta-analysis. Ann Surg Oncol. May2019;26(5):1238–1244. [DOI] [PubMed] [Google Scholar]
- 14.R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ (Accessed March 3, 2020). [Google Scholar]
- 15.Morrow M, Abrahamse P, Hofer TP, et al. Trends in Reoperation After Initial Lumpectomy for Breast Cancer: Addressing Overtreatment in Surgical Management. JAMA Oncol. October12017;3(10):1352–1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.DeSnyder SM, Hunt KK, Dong W, et al. American Society of Breast Surgeons’ Practice Patterns After Publication of the SSO-ASTRO-ASCO DCIS Consensus Guideline on Margins for Breast-Conserving Surgery With Whole-Breast Irradiation. Ann Surg Oncol. October2018;25(10):2965–2974. [DOI] [PubMed] [Google Scholar]
- 17.Tadros AB, Smith BD, Shen Y, et al. Ductal Carcinoma In Situ and Margins <2 mm: Contemporary Outcomes With Breast Conservation. Ann Surg. January2019;269(1):150–157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Langhans L, Jensen MB, Talman MM, Vejborg I, Kroman N, Tvedskov TF. Reoperation Rates in Ductal Carcinoma In Situ vs Invasive Breast Cancer After Wire-Guided Breast-Conserving Surgery. JAMA Surg. April12017;152(4):378–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wilke LG, Czechura T, Wang C, et al. Repeat surgery after breast conservation for the treatment of stage 0 to II breast carcinoma: a report from the National Cancer Data Base, 2004–2010. JAMA Surg. December2014;149(12):1296–1305. [DOI] [PubMed] [Google Scholar]
- 20.Rutter CE, Park HS, Killelea BK, Evans SB. Growing Use of Mastectomy for Ductal Carcinoma-In Situ of the Breast Among Young Women in the United States. Ann Surg Oncol. July2015;22(7):2378–2386. [DOI] [PubMed] [Google Scholar]
- 21.Park HL, Chang J, Lal G, Lal K, Ziogas A, Anton-Culver H. Trends in Treatment Patterns and Clinical Outcomes in Young Women Diagnosed With Ductal Carcinoma In Situ. Clin Breast Cancer. April2018;18(2):e179–e185. [DOI] [PMC free article] [PubMed] [Google Scholar]