Ductal Carcinoma In Situ of Breast: From Molecular Etiology to Therapeutic Management (2024)

,

Article history

Abstract

Breast cancer is the most common cancer in women in several developed and developing countries; more than half of these cases are in developed countries. In the United States, the rate of breast cancer–related deaths is second only to lung cancer. According to the American Cancer Society, in 2022 an estimated number of 287 850 new cases of invasive breast cancer (IBC) are expected to be diagnosed in women in the United States. Moreover, 51 400 women are estimated to be diagnosed with ductal carcinoma in situ (DCIS) (1).

DCIS is stage 0 of breast cancer and a nonobligatory precursor to invasive ductal carcinoma (IDC) (Fig. 1). It is a noninvasive form of breast cancer, which is characterized by a neoplastic proliferation of epithelial cells that are surrounded by myoepithelial cells. In DCIS cases, the myoepithelial cells are confined by an intact basem*nt membrane, which separates these cells from the breast stroma, and prevents the neoplastic cells from metastasizing. DCIS cells have a morphology similar to tumor cells but are still confined by the ductal structure. When detected early, or treated appropriately, this type of breast cancer may be prevented from progressing into a more invasive cancer.

In fact, the intact basem*nt membrane is the key distinction between DCIS and IDC. DCIS cells can break through the membrane and invade surrounding breast tissue, progressing to IDC (2). IDC is the most common type of invasive breast cancer and accounts for approximately 80% of all invasive breast cancer cases. Although DCIS is noninvasive, affected women are at higher risk of developing IDC later in their lifetime. Currently, the standard treatment for DCIS is surgical intervention in the form of mastectomy or breast-conserving surgical resection followed by radiation therapy. However, the probability of recurrence within the same breast tissue is nearly 30% within 15 years after diagnosis (3). In most cases, DCIS is considered to be either over- or undertreated; thus, more time and effort must be devoted to fully understand the complexity of such disease to enable more efficient treatment measures.

Epidemiology of DCIS

DCIS makes up a relatively high proportion of newly diagnosed noninvasive breast cancer cases within North America. In screened populations, DCIS is typically detected as a small cluster of calcifications, but the area can be large, spanning several centimeters, or even occupying more than one quadrant of the breast. The widespread implementation of mammography screening has resulted in a dramatic increase of the incident rate of DCIS during the past few decades; however, these screening-detected DCIS lesions tend to be associated with a lower grade and smaller size (4). In fact, Surveillance, Epidemiology, and End Results (SEER) program data suggest that there was a 500% increase in DCIS cases from 1993 to 2003 among women older than 50 years, and a decline since 2003 (5), which could be explained by the decrease in use of postmenopausal hormonal therapies (6). The 20-year risk of breast cancer–related mortality is 3.3% among women with DCIS (7).

The development of DCIS lesions and IDC have similar risk factors, suggesting a common etiology for both diseases. Both carcinoma lesions have identical etiological relationships, including increasing age, family history, late age of first birth, late onset of menopause, elevated body mass index in postmenopausal women, and high breast density (8). Detailed studies of DCIS and existing IDC shows intratumoral genetic heterogeneity, cellular biology, and behavior (9).

Histological Grades of DCIS

Multiple systems have been used to identify histological grade of DCIS. The nuclear grade is an important part of overall tumor grade, which evaluates nuclear features of malignant cells. The presence or absence of comedonecrosis, or a buildup of dead (necrotic) cells within the tumor, is another key pathological component. There are 3 grades of DCIS: low grade (Grade I), intermediate grade (Grade 2), and high grade (Grade 3) (Table 1, Fig. 2); where increasing grades possess features ranging from normal or atypical ductal hyperplasia to pleomorphic carcinoma cells with irregular and large nuclei (Table 1). The low-grade DCIS resembles atypical ductal hyperplasia cells but can look more like normal breast cells. The intermediate-grade DCIS closely resembles a lower grade DCIS lesion but grows faster than normal breast cells. The high-grade DCIS tends to grow at a faster rate and looks abnormal. Patients with high-grade lesions have a higher risk of IDC and recurrence. Comedonecrosis is usually associated with high-grade DCIS.

Some DCIS lesions can have microinvasions (MI) (Fig. 2) or be known as DCIS-MI (8). DCIS-MI is a rare diagnosis, comprising 10% to 20% of DCIS cases and approximately 1% of all breast cancer cases. The American Joint Committee on Cancer (AJCC) staging system defined MI as cancer cells breaching through the wall of the duct and entering the adjacent stroma with invasive focus not more than 1 millimeter in greatest dimension (10). Compared with pure DCIS, DCIS-MI have both a noninvasive DCIS component and a small component of invasive disease. Although the invasive component is relatively small, prognosis in DCIS-MI is more similar to IDC than pure DCIS (11). DCIS-MI is the first step of invasion into the nearby stroma, or tissue past the basem*nt membrane, which further illustrates the importance of the complexity of this disease.

Interestingly, most substantial cases of DCIS show multiple grades within the same lesions (8). This knowledge might help fully understand the evolution of this carcinoma, but most importantly, what causes the progression of DCIS to become more invasive. It’s important to understand that each lesion can consist of one or more nuclear grades, since that information would direct the treatment plan.

Most DCIS lesions have an intertwining makeup of several different nuclear grades, each aided by multiple cells within the microenvironment (Fig. 2). Despite the prevalence of DCIS, there is no uniform accepted classification system for diagnosing this disease. There has been a growing opinion as to the importance of grade identification over its phenotypic morphology (8).

Emergence of diversity during breast cancer evolution could help explain the reasoning behind multiple histologic nuclear grade lesions (Fig. 3). It is believed that DCIS influences the progression to IDC. There are 3 main theories behind progression of DCIS to IDC, understood through the idea of evolution: Independent Evolution, Evolutionary Bottleneck, and Multiclonal Invasion (12). Both Evolutionary Bottleneck and Multiclonal Invasion are a direct lineage from a single clonal ancestry cell. Allred and colleagues analyzed 120 cases of pure DCIS to understand the nuclear grade makeup, as well as evaluate the biological diversity among different grades of DCIS to further understand its role within breast cancer evolution. According to the study, 46% of cases showed the presence of multiple histologic nuclear grades, primarily between Grade 1 and Grade 2, and 33% showed diversity of several important biomarkers (13).

Molecular Mechanisms/Characteristics of DCIS Progression to IDC

The molecular characteristics of DCIS and mechanisms of its progression to IDC are still poorly investigated. The heterogeneity of DCIS lesions and variation in clinicopathological characteristics, such as estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2/ERBB2) status, create challenges in identifying unique markers that is associated with high risk of progression to IDC.

Nonetheless, previous studies identified specific genomic alterations for different grades of DCIS lesions and for different stages of breast cancer progression from ductal hyperplasia to IDC (14). For example, amplification of FGFR1 and MYC are more frequent compared to DCIS (14-16). The variation within these genomic features between DCIS and IDC could be related to the tumor microenvironment (17), cancer stem-like cell (CSC) populations, myoepithelial differentiation, acquisition of mesenchymal phenotype, changes in expression of cancer-related genes, and some epigenetic changes that can contribute to breast cancer progression (18). It was shown that DCIS cells and tumors contain high cancer stem-like cell subpopulations (ALDH1 + and CD44+/CD49f+/CD24-) with increased self-renewal, treatment resistance, and tumor development capabilities (19).

The recent study using in vitro 3D culture model of DCIS showed that reduction of Rap1Gap, a negative regulator of the small GTPase Rap1, resulted in ERK/MAPK activation, cytoskeletal reorganization, epithelial-mesenchymal transition (EMT) and invasion (20). Ectopic expression of Rap1Gap reversed the invasive phenotype. Immunohistochemistry (IHC) analysis of DCIS human samples compared with normal breast tissue and IDCs shows an increased Rap1Gap expression levels (20). This study suggested that Rap1Gap can function as a switch from DCIS to IDC.

ER, PR, and HER2

Most DCIS lesions are positive for ER and PR, while also being HER2-negative (21-23). These lesions can be closely classified as being Grade 1 or 2 (Table 1). Comparison analysis of DCIS and IDC lesions have not shown significant differences in the expression of ER and PR (22). The HER2 is amplified in 20% of IDCs, associated with increased proliferation abnormalities, which promotes cell motility that contributes to tumor cells metastasizing (22). Regarding DCIS lesions, HER2 is overexpressed in higher grades, such as Grade 3 (Table 1) (23, 24). Interestingly, lesions with HER2 amplification were found with higher frequency in DCIS than in IDC tumors (22).

The St. Gallen system can be used to classify DCIS into surrogate molecular subtypes (25). Five surrogate molecular subtypes were used to classify 381 DCIS cases: Luminal A (ER + and/or PR+, HER2−, Ki67 < 14%), Luminal B/HER2− (ER and/or PR+, HER2−, Ki67 ≥ 14%), Luminal B/HER2+ (ER+ and/or PR+, HER2+), HER2+/ER− (nonluminal) (ER−, PR−, HER2+), and Triple negative (ductal) (ER−, PR−, HER2−) (26). Identifying these subtypes within DCIS cases further confirms that these lesions are precursors of the different subtypes of IDC carcinomas. The ability to correlate these subtypes raises the possibility of preventing DCIS lesions from progressing, similar to that of IDC.

Tumor Suppressor Genes

Surprisingly, the genetic comparison between DCIS and IDC shows identical genetic abnormalities when it comes to loss of tumor suppressors and amplifications. Like most cancer derived cells, DCIS cells show a faster proliferation rate than most cells surrounding it, which contributes to their positive growth imbalance. The progressive growth of DCIS comes from the alterations of normal growth-regulating mechanisms and microenvironment.

The tumor suppressor gene TP53 is located on chromosome 17p13 and encodes 53-kDa nuclear phosphoprotein, which regulates a cell cycle and a programed cell death (27). Loss of tumor suppressor gene TP53 is thought to be an early driver of breast cancer and to occur before the development of IDC, in particularly in DCIS of nuclear grade 2 or 3 (23, 28, 29). TP53 mutation was not found in normal breast epithelium, hyperplasia areas, or DCIS of nuclear grade 1 (29).

In normal breast tissue, TP63, a member of the tumor suppressor p53 family of transcriptional factors, is associated with differentiated myoepithelial-specific genes and is essential for development and maintenance of epithelial stem cells. In cancer, loss of TP63 expression promotes migration, invasion, and distant metastasis. DCIS cells showed a significant decrease in TP63 + population and impaired differentiation of myoepithelial cells (30).

Myoepithelial Differentiation

Myoepithelial cells play essential roles in normal mammary gland development by contributing to the differentiation, proliferation, and polarity of luminal epithelial cells, branching and elongation of ducts, and milk production (31-33). The layer of myoepithelial cells and the basem*nt membrane outline the duct lumen, form a structural barrier, and serve as a limiting factor of DCIS to IDC progression (30). The intact continuous myoepithelial cell layer and basem*nt membrane inhibit penetration by malignant cells into surrounding tissue and progression to invasive disease. Immunohistochemical analysis of myoepithelial markers including smooth muscle actin (SMA), calponin, p63, smooth muscle myosin heavy chain (SMMHC), cytokeratins (CKs) 5/6, P-cadherin, and CD10 can be used to detect the presence or absence of an intact myoepithelial cell layer and differentiate between DCIS and IDC (30, 34).

P-cadherin is a cell adhesion molecule within myoepithelial cells that helps with intercellular connection, cellular migration, and growth (35). P-cadherin is frequently overexpressed in high-grade IDC and is an enhancer of migration, which can be correlated with tumor aggressiveness (36). When comparing P-cadherin levels across DCIS and IDC cases, the majority of lesions were positive for P-cadherin in the luminal cells of the DCIS (35).

Microenvironment

In DCIS, the proliferative epithelial cells are enclosed in a continuous myoepithelial basem*nt membrane along with enrichment of immune cells and fibroblasts (Fig. 2) (37). The progression of DCIS to IDC happens when the continuity of myoepithelial basem*nt membrane is disrupted, and the tumor cells invade into the stroma. The primarily collagenous stroma of a normal duct contains fibroblasts and immune cells.

T cells are rarely detected inside the duct (38). In DCIS, an intact myoepithelial cell layer and basem*nt membrane serve as a barrier between immune system in stroma and the intraductal tumor cells. During transition from DCIS to IDC, the status of the immune microenvironment changes to become more immunosuppressive (38, 39). In DCIS, there is an activated immune environment surrounding the ductal structure, with enrichment of cytotoxic CD8 + T cells (38, 39). Compared with DCIS, IDCs contained fewer activated MKi67+CD8+ and GZMB+CD8+ T cells (38). In IDC, there is a suppressed immune environment, with enrichment of Treg cells (38, 39). DCIS and IDC also have significant differences in expression of several immune checkpoint proteins which may contribute to immune escape during breast cancer progression (38, 39). In particular, DCIS contained more TIGIT+ T cells, whereas IDCs had higher expression of PD-L1 and CTLA4 (38, 39).

Besides promoting DCIS invasion, fibroblasts can help aid the rapid development of most breast cancer lesions. Specifically, breast cancer–associated fibroblasts (CAFs) release growth factors, angiogenic factors, and proteases into the tumor microenvironment that support tumor growth, angiogenesis, metastasis, and therapy resistance (40).

In IDC, the fibroblasts transform into CAFs, which express PDGFRα/β, αSMA, FAP, and FSP1 (41). The CAFs assist tumor cells invading into the surrounding stroma. Fibroblasts produce proteases, like matrix metalloproteinases (MMPs) that degrade the extracellular matrix, which results in the stromal reorganization and releasing of growth factors (42).

Several studies have been conducted to understand how fibroblasts are capable of contributing to DCIS invasion. Sung et al showed that co-culture with human mammary fibroblasts induced invasion of MCF10DCIS cells through collagen reorganization (43). Using a 3D model, Hu and colleagues co-cultured fibroblasts with MCF10DCIS cells within a recombinant basem*nt membrane (44). From this study, it was shown that there was an increased invasive branching within the Matrigel and higher levels of MMP-9 and MMP-14 with the presence of fibroblasts (44). MMP-9 promotes cancer cell invasion and disease progression whereas MMP-14 activates other MMP family members to promote cancer (45). Using the mammary architecture and microenvironment engineering 3D model, Osuala and colleagues demonstrated that CAF-derived IL-6 contributed to invasion of MCF10DCIS cells (46). Another study identified that CAF-derived chemokine CXCL1 induced DCIS invasion in the MMTV-PyMT mouse model (47). Moreover, Brummer et al conducted a study using a mammary intraductal injection DCIS model, demonstrating that CCL2/CCR2 chemokine signaling promoted invasion of DCIS cells through fibroblast-dependent mechanisms (48).

Diagnosis and Treatments for DCIS

Since the implementation of mammography screening, there has been a dramatic increase of the incident rate of DCIS. Mammograms, an x-ray imaging of the breast, are the most efficient way to detect breast cancer cases early on its progressive state. Mammographic screening accounts for majority of diagnosed DCIS cases. DCIS presents on a mammography as microcalcifications, or bright white clusters, that have irregular sizes and shapes (49). Further evaluation with core breast needle biopsy is essential for determining the diagnosis and grade of DCIS.

There are no clear guidelines to predict which DCIS cases would progress into IDC and require aggressive treatment as opposed to cases that would remain indolent during a women’s lifetime. As a result, many women with low-risk DCIS are overtreated with surgery (lumpectomy or mastectomy), radiation, and hormone therapy (50). Between 1991 and 2010, approximately 70% of DCIS cases were treated with breast-conserving surgery with or without radiation, and approximately 25% of DCIS patients underwent mastectomy (50). Only a very few cases of mastectomy are followed by breast reconstruction. The appropriate treatment plan option depends on a variety of clinical-pathological factors like age and the extent of the disease (8). Most breast-conserving surgical cases are associated with minor excision of the ductal structure affected.

In order to identify the most suitable treatment for every single DCIS, the Van Nuys Prognostic Index (VNPI) was established based on characterizing the most important predictive factors and combining them in an algorithm (Table 1). VNPI scores can be associated with predicting the 10-year local recurrence-free survival in treated patients with DCIS, the percent of those that would remain cancer-free after excision, and which excision is more appropriate for the given lesions (23). Those with a low score (VNPI 4-6) account for 33% of patients, have a 97% local recurrence-free survival, and wide-local excision (WLE) would be the treatment plan. Those with an intermediate score (VNPI 7-9) account for 57% of patients, have a 73% local recurrence-free survival, and the appropriate treatment would be WLE with radiotherapy (RT). Those with a high score (VNPI 10-12) account for 11% of patients, have a 34% local recurrence-free survival, and the appropriate treatment would be a full mastectomy. RT after DCIS reduced the risk of local recurrence by 50% but did not reduce the risk of dying of breast cancer (51). Based on lesion size, margin width, histological features, and patient’s age, the recommended treatment could be breast-conserving surgery alone, lumpectomy followed by adjuvant radiotherapy, or mastectomy (Table 1) (52). Other scoring methods have been used to determine both DCIS recurrence and to predict benefits of RT. OncotypeDX DCIS scoring was the first clinically validated genomic test to determine a 10-year risk of local recurrence using a 12-gene assay and was shown to complement traditional clinical or pathologic factors (53, 54). Another scoring method developed is DCISionRT (PreludeDX), which analyzes the expression of specific markers and provides information associated with the benefits of RT (54). With this scoring method, physicians can determine if a patient who has undergone breast-conserving surgery would benefit overall from RT, therefore limiting unnecessary radiational exposure (55).

There has been very little progress to identify innovative targeted treatments for DCIS, simply because of its intertwined subtypes, with different biological potential (56). Each subtype of DCIS is composed of different biological potential; thus, the only successful targeted treatment for DCIS is tamoxifen, which is used for Grade I or II ER-positive lesions. For 21 years, tamoxifen has been known for treating IDC. Essentially, tamoxifen binds to ER on the cancer cells and prevents estrogen from binding in return prohibiting cell proliferation. Tamoxifen is used for estrogen and progesterone receptor-positive diseases, consistent with the biologic mechanism of action of the drug (57).

To confirm the significant benefits of tamoxifen, Allred and colleagues used a clinical trial to evaluate ER and PR relationships in DCIS after lumpectomy and radiation, followed by adjuvant tamoxifen (58). This study was undertaken to evaluate the relationship between adjuvant tamoxifen and the receptor status among DCIS lesions. With prolonged follow-up, the results showed that the adjuvant tamoxifen had a significant reduction of ipsilateral breast cancer in patients with ER-positive DCIS lesions; similarly, tamoxifen reduced contralateral breast cancer in patients with ER-positive and ER-negative DCIS lesions. However, in ER-negative only DCIS lesions, there was no ipsilateral benefit with tamoxifen. This emphasized that tamoxifen must bind to functional ER to be effective on preexisting tumor cells (57, 58).

A similar trial was conducted to demonstrate the impact and effect when combining the use of tamoxifen and lumpectomy with radiation. The results showed an additional 37% reduction in relative risk of local recurrence and a decrease in contralateral breast cancer of comparable magnitude (59). The RTOG 9804 trial showed that RT significantly decreases ipsilateral recurrence in the good-risk subset of DCIS patients (60). These studies potentially offer patients and physicians an additional option that is not only therapeutic but more efficient.

Wesseling and colleagues published a meta-analysis summary of the current knowledge and prognostic factors for invasive disease after a diagnosis of DCIS. Most of the factors pointed to a higher relative risk of subsequent IDC for patients with DCIS, but the effects were generally small. Six prognostic factors had a statistically significant pooled estimate (61). The 6 factors were race, premenopausal status, detection by palpation, involved surgical margin, high histologic grade, and high p16 expression, the majority of which can be biologically explained. A recent study used immunofluorescence multiplexing and single cell analysis to study co-expression of breast cancer associated cellular biomarkers in limited formalin-fixed, paraffin-embedded DCIS tissue microarrays (62). Using the developed multi-step comprehensive analysis, the investigators demonstrated that high HER2 and low ER and PR in DCIS samples were associated with breast cancer events. In contrast, Ki67 and COX2 did not show significant correlation with breast cancer events. A large cohort study is needed for further validation of this prognostic signature (62).

Conclusion

Why study DCIS? The core reason for studying this disease is to understand its etiological factors. Some patients diagnosed with DCIS have the possibility of progressing to IDC if left untreated or treated incorrectly. Currently, the most used treatment for DCIS is a lumpectomy followed by RT. The treatment for DCIS patients could be improved based on comprehension of which lesions have the potential to become invasive. The problem is that all lesions lack different capacities for progression under a microscope. Since only 20% to 50% of DCIS cases progress to IDC if left untreated and since the patients with DCIS after surgical resection with or without radiation have an overall good prognosis, DCIS is often overlooked. The primary focus is on understanding the invasive disease and its treatment strategies, rather than on DCIS. Thus, there is a lot unknown about DCIS. This unknown aspect leads to the misperception of whether this disease can be determined as invasive.

To fully grasp the progression and explanation of the mechanisms of DCIS, one must uncover the molecular pathways involved and the risk factors associated with each lesion on DCIS that allows invasive properties to evolve. After a patient has DCIS lesions, it takes approximately 10 years for these lesions to achieve an IDC diagnosis. Because of this, some scientists and doctors are not convinced of the urgency to further investigate DCIS lesions or search for better treatment strategies. If the gap of knowledge was breached, or a better comprehension of DCIS lesions on a molecular level was understood, then more manageable measures would be taken that could nearly eliminate the current mortality rate associated with this diagnosis.

Breast cancer makes up an entire field of researchers, professors, and doctors, who share the same goal to help the 300 000 women affected yearly by this cancer. Breast cancer research is a multibillion-dollar industry. However, most of these funds support research in invasive breast cancer rather than DCIS. There is little to no research conducted to elucidate the progression of DCIS to IDC, or the mechanism to explain why certain lesions progress beyond the basem*nt membrane into its surrounding tissues. Members of the National Institutes of Health DCIS conference proposed that the word carcinoma should be removed from the term ductal carcinoma in situ because DCIS is noninvasive and has a favorable prognosis. The term indolent lesions of epithelial origin (IDLE) has been proposed for use instead of DCIS (54). However, experimental studies of human and mouse DCIS lesions are showing the opposite: carcinoma precursor cells exist in these lesions, and the aggressive phenotype of breast cancer is predetermined early at the premalignant stage (2).

By identifying the mechanism that inhibits the progression of DCIS to IDC, a major unknown about DCIS would become known. This discovery alone will broaden the understanding of breast cancer, as well as develop a more concrete explanation of why this carcinoma might be the initiator of IDC. This would create a big impact within the field and change the outlook of DCIS.

Abbreviations

Abbreviations

• CAF

  cancer-associated fibroblast

• DCIS

  ductal carcinoma in situ

• EMT

  epithelial-mesenchymal transition

• ER

  estrogen receptor

• HER2

  human epidermal growth factor receptor 2

• IDC

  invasive ductal carcinoma

• MI

  microinvasive (microinvasions)

• MMP

  matrix metalloproteinase

• PR

  progesterone receptor

• RT

  radiotherapy

• VNPI

  Van Nuys Prognostic Index

• WLE

  wide-local excision

Acknowledgments

This work was supported by Emory startup and the Northwestern University Zell scholar fund.

Financial Support

Emory startup. NIH grant: R01CA202948.

Conflict of Interest Statement

The authors declare that there are no conflicts of interest.

Data Availability

Data sharing is not applicable to this article as no new data have been included in this review.

References

Author notes

Shelby Lynn Hophan and Olena Odnokoz share co-first authorship and have an equal contribution in preparation of this manuscript.

© The Author(s) 2022. Published by Oxford University Press on behalf of the Endocrine Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Email alerts

Citing articles via

More from Oxford Academic