When Is Sentinel Node Biopsy Indicated in High-Risk Ductal Carcinoma in situ? Four Hundred Sixty-Eight Cases from Three Institutions

Abstract

Introduction

Sentinel lymph node biopsy (SLNB) in ductal carcinoma in situ (DCIS) is not indicated. However, in certain cases (size >3 cm, high grade, mass effect on mammography, or palpable mass), it may be possible to find incidental invasive carcinoma (IC) that requires an SLNB. We studied the correlation of the aforesaid factors with the probability of finding IC in the surgical specimen.

Methods

Data was collected from 3 different institutions between 2010 and 2016, recording characteristics such as, but not limited to: high grade, size >3 cm, mass effect on mammography, and palpable mass.

Results

On the whole, 468 “high-risk” DCIS cases were identified, 139 (29%) of which had IC. When the DCIS was high grade or the size was >3 cm, there was no significant difference in the probability of finding IC in the surgical specimen (OR = 1.13; 95% CI 0.84–1.51; OR = 1.2; 95% CI 0.85–1.40). Nevertheless, when a high grade and size (>3 cm) were combined, IC was more likely to exist (72.7 vs. 27.3%; p = 0.001). In addition, mass effect and palpation were independently associated with a significantly greater degree of IC (OR = 12.76; 95% CI 6.93–23.52).

Conclusions

The results suggest that high-grade DCIS or DCIS with a size >3 cm, independently, does not require SLNB. Nonetheless, in the event that both factors are found in the same case, SLNB may be indicated. Additionally, SLNB is advisable for DCIS cases that are palpable or show a mass effect on mammography.

Introduction

The introduction of screening mammographies in the last 25 years has led to a decrease in the number of aggressive treatments and to improvements in both detection at an early stage and the survival rate.

The performance of opportunistic mammographies has also improved the diagnosis of ductal carcinoma in situ (DCIS), a preinvasive lesion [1, 2] that can lead to invasive carcinoma (IC) in 40% of the cases. This type of lesion currently represents 20% of cases [1]. The treatment of DCIS consists of excision of the lesion either conservatively with a lumpectomy or through a mastectomy in lesions that are extensive. Radiotherapy is performed in the case of conservative surgery, creating a boost in the area where the tumor was.

The role of sentinel lymph node biopsy (SLNB) in DCIS cases is controversial. DCIS is a type of lesion that does not have the capacity to disseminate, and thus the probability of it leading to a lymph node metastasis [3, 4, 5, 6] is very low. Therefore, SLNB is not indicated according to the majority of the clinical guides [7, 8, 9, 10]. Currently, there are studies evaluating the outcomes of not performing SLNB in early breast cancer, with no axillary lymph nodes affected within ultrasonography studies.

However, recent studies have recommend SLNB in patients undergoing breast-conserving surgery [11]. SLNB is a safe technique, but complications can occur, especially in elderly patients [12, 13, 14].

The scientific literature outlines that 20% of DCIS cases diagnosed by core needle biopsy are invasive cancers (occult or incidental carcinoma), in the final surgical pieces [15, 16, 17, 18, 19, 20, 21], needing SLNB in a second surgery. However, we know that involvement of the sentinel lymph node is unlikely. Likewise, if it were to be affected, there would be low-volume disease (isolated tumor cells or micrometastasis) [18]. Hence, the significance of SLNB would still be questionable.

The probability of finding a hidden or incidental IC is very low, but when DCIS cases present some characteristics such as a palpable mass, a mass effect on the mammography, extensive lesions (>3–5 cm), or a high grade (grade 3, poorly differentiated) the probability of finding an infiltrating carcinoma is significantly higher.

We studied the correlation of these factors with the probability of finding an IC in the final surgical piece that leads to the performance of an SLNB.

Methods

This is a retrospective study in which all “high-risk” DCIS cases filed between 2010 and 2016 as a result of biannual mammographic screening (according Spanish public health care) or an opportunistic mammography, from 3 different institutions (Hospital Universitari General de Catalunya, Valld'Hebron University Hospital, and Germans Trias I Pujol Hospital).

High-risk DCIS cases are defined as those that show a mass effect on the mammography and/or are a palpable mass and/or are high grade (grade 3) and/or have a size >3 cm. Moreover, according to Spanish clinical guidelines, SLNB is indicated in high-risk DCIS.

The breast cancer TNM staging was defined using the American Joint Committee on Cancer (7th edition) manual. We also collected the age of the patients, mammography imaging, the type of surgery performed, and hormonal receptor expression. We excluded patients who showed suspicious axillary lymph nodes or had ipsilateral infiltrating carcinoma.

All biopsies were performed through core needle biopsy either with stereotaxis in the case of microcalcifications (vacuum-assisted biopsy) or by ultrasound if a lesion was visible. The samples were studied by pathologists specialized in breast pathology from the breast cancer units of the 3 hospitals.

The performed surgery was a lumpectomy or mastectomy depending on the size of the lesion or on the possibility of performing radiotherapy. According to Spanish guidelines SLNB is indicated for high-risk DCIS. Henceforth, bearing in mind that all of our collected cases were considered high risk, SLNB was performed in each case.

A logistic regression model was designed to evaluate the correlation of the aforementioned variables with the odds of finding IC in DCIS cases. Other collected variables, such as age, tumor size or the existence of hormone receptors, were taken into account. OR and 95% CI were calculated. The model was constructed taking into account those variables that resulted statistically significant in the regression analysis. A receiving operator curve (ROC) was carried out in order to evaluate the accuracy of the resulting model. The level of significance was set at 95%. The following software was used: pEpiR, ggplot, caret and pROC packages from R Studio (R Core Team, 2017).

Results

Four hundred sixty-eight high-risk DCIS cases were collected, 139 (29%) of which were IC in the final pathology specimen.

The average age was 55 years, mastectomy was performed in 120 patients (26%), and there were 348 lumpectomies (74%).

The size of the invasive component was <10 mm in 84 cases (60%) and >10 mm in 55 cases (40%). The SLNB was positive in 50 patients (10%).

Regarding age, no significant differences were observed between infiltrating tumors and noninfiltrating tumors (mean age: 51 vs. 53 years; p = 0.73; Fig. 1).

Fig. 1.

Age and probability of infiltration.

With regard to the hormonal receptor condition, no significant differences were found in the probability of finding infiltration within DCIS cases (OR = 0.96; 95% CI 0.64–1.44).

Concerning the histological grade or the presence of necrosis, 322 cases were low grade, 230 (71%) of which were noninfiltrating and 92 (28%) of which were invasive. One hundred forty-six cases were high grade, 99 (67%) of which were noninfiltrating and 47 (32%) of which were infiltrating (Fig. 2). Therefore, the results suggest that DCIS cases that were high grade or presented necrosis showed no significant difference in the odds of finding an IC.

Fig. 2.

Distribution of cases depending on the histological grade.

As for size, 365 cases were <30 mm, 259 (70%) of which were noninfiltrating and 106 (29%) of which were invasive. One hundred three cases were >30 mm, 67 (70%) of which were noninfiltrating and 33 (32%) of which were infiltrating. With relation to extensive lesions (>3 cm), there was no greater probability of finding an IC (OR = 1.39; 95% CI 0.89–2.17; Fig. 3, 4).

Fig. 3.

Distribution of cases regarding size.

Fig. 4.

Distribution of cases depending on size.

A composite variable including size >30 mm and high histological grade was analyzed. When the 2 conditions were not present (n = 183), 44 cases (24%) were invasive in the final piece, while 139 (76%) were noninvasive. Likewise, when a size >3 cm and a high histological grade (n = 33) were combined, 24 cases (72%) were invasive, and 9 (27%) were noninvasive in the final pathologic study (72.7 vs. 27.3%; p = 0.001; Figure 5).

Fig. 5.

Distribution of cases when size and grade were combined.

With regard to the mass effect on mammography, 200 cases had no mass effect, 187 (93%) of which were noninfiltrating and 13 (6%) of which were invasive. Two hundred sixty-eight cases had a mass effect on mammography, 142 (52%) of which were noninfiltrating and 126 (47%) of which were infiltrating. Hence, the results argue that a mass effect on mammography is associated with a significantly greater probability of infiltration (OR = 12.73; 95% CI 6.93–23.52; p < 0.001).

Moreover, those DCIS cases that were palpable resulted in a significantly higher probability of being infiltrating (OR = 21.82; 95% CI 9.38–55.30; p < 0.001). Hence, those tumors that were palpable or had a mass effect on the mammography were infiltrating in 47.5% of the cases compared to 6.5% of those that were not.

Table 1 summarizes the probability of each factor (size, grade, mass effect, and palpable) in finding an incidental or occult carcinoma.

Table 1.

Probability (odds ratio) of preoperative factors and the presence of IC in surgical specimens

	OR	IC (2.75%)	IC (97.5%)	p value
Hormone receptor	0.96	0.64	1.44	>0.05
Size >30 mm	0.63	0.33	1.19	0.12
High grade	1.36	0.74	2.54	0.31
Size >30 mm + high grade	4.42	2.32	8.43	0.001
Mass effect	15.48	7.05	36.10	0.00
Palpable	21.82	9.38	55.30	0.00

A logistic model assessed all variables of interest adjusting for age. Only 2 variables were found to be significant, i.e., mass effect and histological grade. The final model was constructed taking into account those 2 variables. The ROC curve showed an AUC of 77% (95% CI 73–81%) and a negative predictive value of 93.5%. In this model, the presence of a mass effect was sufficient for the case to be classified as positive. When a high-grade tumor was also observed, the probability of having an IC increased by 19%.

Discussion

The results suggest that when DCIS had presence of high-grade/necrosis and a size >3 cm, the probability of finding IC was 4 times greater than when the 2 factors were independently used. Thus, SNLB may be indicated when these 2 features are present.

Additionally, a mass effect in a mammography or palpable masses had greater odds of IC, and therefore SLNB would be advisable.

On the other hand, in cases that were low grade and showed no mass effect on mammography, the probability of being non-IC was 93% (negative predictive value). Thus, in these cases, SLNB is not indicated.

When DCIS requires mastectomy, SLNB is mandatory [22, 23] on the grounds that if there happens to be an invasive component in the surgical specimen there will be no possibility of performing a SLNB in the second surgery.

Since the type of surgery is not a predictor of infiltration, the mastectomies that were performed for high-risk DCIS were included in this study in order to have more cases [9].

The scientific literature suggests that SLNB should be indicated in DCIS cases with characteristics such as: age <55 years, smaller core needle biopsy, solid mass on imaging, size >25 mm, high grade [18, 24, 25]. Nevertheless, our results did not find age to be a risk factor for IC. Regarding the biopsy, all of our patients were diagnosed through a core needle biopsy or a vacuum-assisted biopsy (stereotactic cases).

Nowadays, all patients with an infiltrating breast tumor undergo SLNB for assessment of the axillary nodal status; however, the majority of patients do not need this surgical procedure given that SLNB tends to be negative in most patients in early stages of breast cancer. In our study only 10% had a positive SLNB.

Currently there are 2 clinical trials (SOUN and INSEMA) evaluating the overall survival rate, the disease-free survival rate, and quality of life in women with early breast cancer randomized into 2 groups, i.e., women who underwent SLNB and women who had clinically and sonographically negative lymph nodes who did not undergo SLNB [26, 27].

There are several studies assessing the role of SLNB in DCIS [28, 29] and its clinical significance depending on whether it is positive or negative. Broekhuizen et al. [30] suggested in their study that the survival rate of patients with DCIS and micrometastasis in the SLNB is not affected. It is clear that the presence of positive lymph nodes can modify the treatment that the patient will receive later. van Roozental et al. [9] demonstrated, in a series of 1,251 patients in whom SLNBV was not performed, that the indication for systemic therapy would be missed in 1.9% of population.

In our study we did not place the spotlight on the result of the SLNB, focusing rather on predicting the probability of IC to decide whether to perform an SLNB. The possibility of carrying out an SLNB in all DCIS cases is not a viable option due to the possible side effects or comorbidity associated with this technique, especially in elderly patients, and because of the cost of the procedure. Additionally, it is necessary to consider the stress that having a second surgical procedure may cause the patient. Besides, we must take into account that drainage of the lymphatic node can be affected especially if an extensive lumpectomy or oncoplastic techniques have been performed. Thus, it seems reasonable to perform SLNB in selected cases, following Dutch breast cancer guidelines [22] which recommend performing it in the following cases: young age, size >30 mm, high grade, mass effect on the mammography, or palpable mass as confirmed in our study.

Our results also suggest that age, histological grade, and size, independently, do not seem to have a significant correlation with the probability of finding IC; therefore, SLNB should not be indicated. Nonetheless, when DCIS is a palpable mass or has a mass effect in the mammography, the odds of infiltration are significantly higher, and hence SLNB should be highly recommended.

SLNB in DCIS cases that are diagnosed through a core needle biopsy can be omitted during the first surgery, since it can be performed in a second surgery if IC is detected, avoiding the morbidity associated with surgery and medical costs.

However, within those DCIS cases that are palpable and show a mass effect on mammography, the odds of finding an invasive component are significantly high; thus, it would be reasonable to indicate am SLNB in the same surgical procedure in order to avoid the discomfort of a second surgery for the patient. Likewise, when a DCIS is high grade and has a size >3 cm, it seems reasonable to include the patient in the decision making. Nonetheless, we will continue to expand our database in the future given that broader literature is needed, for the surgeon and for the patient, to make the most beneficial, optimized, and effective decision in each case.

It is worth bearing in mind that our study does have certain limitations, since it is a multicentric and retrospective analysis. Nevertheless, it is still a large study utilizing real-world data that can help us in our daily clinical practice and decision making process to correctly counsel patients with high-risk DCIS about whether or not an SLNB should be carried out.

Statement of Ethics

This study was approved by the Ethics Committee of the Hospital Universitari General of Catalunya. Written informed consent was obtained from all of the patients.

Conflict of Interest Statement

The authors certify that they have no affiliations with or involvement in any organization or entity with financial or nonfinancial interest in the subject matter discussed in this paper.

Funding Sources

No funding was obtained for this work.

Author Contributions

T.C.: study conception. T.C., P.A., J.J., and M.M.: study design. T.C., R.P., M.G., and X.G.-F.: drafting of this work. J.A., A.G.-M., M.L., V.P., and M.X.: statistics.