Active surveillance in the management of Ductal Carcinoma In Situ: the radiologist’s point of view

(1) Background: Considering highly selected patients with ductal carcinoma in situ (DCIS), active surveillance is a valid alternative to surgery. Our study is aimed at showing the reliability of post-biopsy complete lesion removal, documented by mammogram, as additional criterion to select these patients. (2) Methods: 2173 Vacuum Assisted Breast Biopsies (VABB) documented as DCIS have been reviewed. Surgery has been performed in all cases. We retrospectively collected the reports of post-VABB complete lesion removal and the histological results of the biopsy and surgery. We calculated the rate of upgrade of DCIS identified on VABB upon excision for patients with post-biopsy complete lesion removal and for those showing residual lesion. (3) Results: We observed 2173 cases of DCIS: 408 classified as low grade; 1262 as intermediate grade; 503 as high grade. The overall upgrading rate to invasive carcinoma was 15.2% (330/2173). The upgrade rate was significantly lower (8.2%) when considering patients showing mammographically documented complete removal of the lesion. (4) Conclusion: The absence of mammographically documented residual lesion following VABB is associated to a lower upgrading rate of DCIS to invasive carcinoma on surgical excision and should be considered when deciding the proper management DCIS diagnosis.


Introduction
Breast cancer is the most common cancer diagnosed in the female population, accounting for approximately 15.2%-30% of all new cancer cases among women [1].
Ductal carcinoma in situ (DCIS) of the breast represents a heterogeneous group of neoplastic lesions confined to the breast ducts and lobules, without showing invasive features nor metastatic potential [2].
About 25% of all breast cancer cases are ductal carcinoma in situ [3], thus their diagnostic and therapeutic management represents an important health challenge with fundamental public health implications. DCIS is usually diagnosed by imaging because it is often clinically occult [4]. Its incidence has rapidly increased from 1980 considering the dramatic improvement in diagnosis and screening imaging tools, mostly mammography which plays a central role, as a majority of DCIS is diagnosed by microcalcifications.
Nowadays, approximately 98% of patients with DCIS undergo surgery, often associated with radiotherapy [5]. However, it is now clear that most of them rarely progress spontaneously to invasive cancer, indeed the mortality rate is as low as 4% [6]. The risk of progression seems to be related the grade of the disease, being high grade tumour associated with a worse prognosis [7,8]. Besides, according to some studies, a higher aggressiveness is due to multifocality as well as to aberrant branching and lobularization, defined as neoductgenesis [9,10]. So, the identification of these patterns at imaging and histology could help in distinguishing intrinsic aggressiveness and tailoring the therapy accordingly.
Therefore, we can assume that aggressive treatment of DCIS, especially in patients with comorbidities, can be considered a form of overtreatment, as well as unjustified public health costs, particularly during the COVID 19 pandemic [11].
Four prospective international study protocols (LORIS, COMET, LORD and LORETTA) are currently in place to evaluate non-invasive treatment strategies for DCIS [12][13][14][15][16]. The main proposals consist in including selected patients in protocols based on active surveillance with or without associated endocrine therapy according to specific inclusion criteria, such as the nuclear grade of DCIS and biopsy technique (Table 1). The effectiveness of active surveillance can be improved by reducing biopsy-related diagnostic underestimation: presurgical biopsy proven DCIS may be upgraded to invasive carcinoma on submitted surgical specimens.
However, data regarding DCIS diagnostic underestimation rate are quite controversial: according to an important meta-analysis performed by Brennan et al, up to 26% of patients with biopsy proven DCIS reveal a synchronous invasive carcinoma on surgical specimens [17].
The primary purpose of our observational multicentre retrospective study is to determine the rate of upgrade of DCIS identified on Vacuum Assisted Breast Biopsy (VABB) upon excision.
The secondary objective of the study is to prove a correlation, if it exists, between the post-biopsy complete removal of the lesion and the lower likelihood to find an invasive carcinoma on subsequent surgical specimen.
In order to do this, the residual tumour rate found on surgical specimen was compared with imaging of mammogram performed post VABB but before subsequent surgery.

Materials and Methods
We reviewed all cases of breast biopsies with DICS diagnosed on VABB at our Departments of Pathology from January 1st, 2000 to December 31st, 2018 and subsequent surgical excision performed in two centres.
Since VABB provides a better diagnostic performance than core needle biopsy [18], we selected patients submitted to this procedure using a 12G needle.
During the considered period of this study, we have been using a subcategorization of DCIS according to the so-called DIN system (Ductal Intraepithelial Neoplasia) as previously published [19]. Briefly, DIN1C corresponds to low grade DCIS, DIN2 to intermediate, DIN3 to high grade, according to nuclear morphologic features of the neoplastic cells [20,21].
Patients younger than 40 years of age, those with concomitant invasive carcinoma or past personal history of breast cancer and those showing DCIS with comedonecrosis were excluded from the study.
All these data have been retrospectively collected. By using mammogram before surgery, we recorded the absence or the presence of post-VABB residual lesion and we compared the outcomes of these two groups of patients.
The upstage rate of DCIS to invasive carcinoma following surgical excision was always recorded.

Statistics
The collected data were compared using the compare proportions test. A p-value <0.05 was used to determine statistically significant differences.

Results
A total number of 2173 vacuum assisted breast biopsies were performed under stereotactic guidance showing DCIS: 408 cases were low grade (DIN1C); 1262 cases were intermediate grade (DIN2), and 503 cases were high grade (DIN3). The mean age of the patients was 62 years (range 32-84 years). The overall mean diameter of the lesions was 20 mm. Table 2 summarizes clinicopathologic characteristics of the patients. These data led to the first observation: patients showing complete removal of the lesion experienced a significantly lower upgrade rate compared to those showing mammographically detectable residual tumour after VABB (p-value < 0.01).
Data considering the three diagnostic categories (DIN1C, DIN2 and DIN3) are summarized in Table 3.

Discussion
Considering that most of DCIS will never progress to invasive breast cancer during a patient's lifetime, surgical therapy and radiotherapy of DCIS, especially in patients with comorbidities, can be considered a form of overtreatment, without taking into account unjustified health and social care costs.
Surveillance, Epidemiology, and End Results (SEER) data shows that the 20-year breast cancer-specific mortality rate in patients with DCIS is as low as 3.3% [11,22]. On the other side, according to a significant meta-analysis, 25.9% (18.6-37.2%) of presurgical cases diagnosed as DCIS have been upgraded to invasive carcinoma upon excision [17].
In this study, we tried to minimize the risk of diagnostic underestimation by applying strict inclusion criteria. In particular, in our series, all cases were biopsied by VABB with at least a 12G needle; patients younger than 40, or patients with previous history of breast cancer were excluded.
Our overall upgrading rate of 15.2% was in line with other previous studies [23][24][25] that have reported upgrading rates in the range 11-25% (Table 4). Furthermore, in order to reduce the diagnostic underestimation rate as much as possible, we took into account additional parameters such as the post-biopsy complete removal of the lesion [26] information not reported in other studies and the diameter of the lesion, information evaluated only in Loretta trial (diameter <25mm) [16].
The results of our study show that if the lesion is completely removed during biopsy, the overall diagnostic underestimation rate is significantly lower. Indeed, DCIS patients showing complete removal of the lesion experienced a significantly lower upgrade rate to invasive cancer compared to those showing mammographically detectable residual tumour after VABB (respectively 8.2% vs 15.2%).
Therefore, we strongly believe that this last parameter should be considered as a possible selection criterion to enrol DCIS patients in surveillance protocols.
As far as we know, this study has one of the largest number of biopsies considered in a single retrospective study, but the main limitation is its retrospective nature. Moreover, involved patients do not perfectly match inclusion criteria of LORIS, LORETTA, COMET and LORD protocols.
The absence of mammographically documented residual lesion following VABB is associated to a lower upgrading rate of DCIS to invasive carcinoma on surgical specimens and should be taken into account when deciding the proper management of patients with ductal carcinoma in situ diagnosis. Data Availability Statement: Data sharing not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.