Fig. 1.1
Schematic sections of breast duct in normal condition and in atypical hyperplasia. The normal duct (on the left) is composed of a single epithelial layer within the duct, which is bound by the basement membrane and a rim of myoepithelial cells on the lumen side of the basement membrane. Outside the basement membrane, the breast stroma contains the extracellular matrix (ECM ): blood and lymphatic vessels, stromal cells, immune cells and fat cells. In the atypical hyperplasia (on the right), an overgrowth of normal cells is observed which over time continue accumulating and develop abnormalities [6] mod
DCIS lesion contains multiple layers of cells that accumulate inwards into the lumen resulting in a stressful microenvironment, which may promote genetic instability. Cellular elements taking part in the carcinogenesis process are progenitor cells and malignant progenitor cells (also called committed progenitor cells).
Progenitor cells, like stem cells, have a tendency to differentiate into a specific type of cell, but are already more specific than a stem cell and are pushed to differentiate into its “target” cell. The most important difference between stem cells and progenitor cells is that stem cells are multipotent and can replicate indefinitely, whereas progenitor cells are uni- or oligo-potent and can divide only a limited number of times. Controversy about the exact definition remains and the concept is still evolving. In the normal duct progenitor cells :
Maintain the ductal-lobular architecture of the breast parenchyma.
Sustain the myoepithelial cell layer and basement membrane surrounding the parenchyma.
Balance the relationship of the parenchyma to the stroma.
Malignant (or committed) progenitor cells undergo complete malignant transformation. They replace the progenitor cells, take over and only in part retain their functions.
At the highest level of retained functions ( low-grade DCIS ), the malignant progenitors are able to maintain almost all of these functions and are able to renew the ductal and lobular structures.
At the lowest level of retained functions ( high-grade DCIS ), depending on the severity of the genetic alterations acquired, malignant progenitors may not be able to preserve the terminal units. Under their action:
New duct-like structures appear in close proximity to each other (ductal neogenesis).
Myoepithelium is formed in a defective way and focally disappears.
Periductal stroma undergoes remodelling and becomes infiltrated by lymphocytes.
The stressful microenvironment of the intraductal lumen may promote genetic instability. All of the components of extracellular matrix (ECM ) participate in the carcinogenic process by promoting or suppressing malignant progenitor cells that could arise within the mass of cells accumulating in the duct (Fig. 1.2).
Fig. 1.2
Some schematic features of process of intraductal cell proliferation at the highest level of retained functions (low-grade DCIS, on the left) and at the lowest level of retained functions (high-grade DCIS, on the right) [6] mod
Pluripotent malignant progenitor (stem) cells must adapt to survive for some time. However, in a later stage, process promotes the suppression of apoptosis in the face of genetic instability and could lead to the generation of genetically abnormal malignant progenitor cells before the onset of invasion. Invasion is associated with the loss of myoepithelial cells, periductal angiogenesis, fragmentation of the basement membrane and chemotaxis radially outwards (Fig. 1.3). Incidentally, understanding how genetically abnormal DCIS cells arise and survive within the high-stress microenvironment may provide targets for chemoprevention.
Fig. 1.3
Initial adaptation and subsequent evolution of stem cells with their associate phenomena. [6] mod
1.2.2 Outcome of DCIS Lesions
DCIS become invasive in an unknown proportion of cases, in different ways and at various times. Three hypothetical models for the possible outcomes of a DCIS lesion [2] are outlined in Fig. 1.4.
Fig. 1.4
Roughly schematic models for the different outcomes of a DCIS lesion. A: DCIS with prolonged intraductal phase. B: DCIS with short intraductal phase. C: Regressive or not evolving DCIS [2] mod
DCIS with prolonged intraductal phase. Tumour malignant cells spread along duct invasion with a persistent but slow phase, which may include a dormant phase with little cellular proliferation, which could last many years or decades. Possible specific factors are mutation or altered expression in critical gene, but also a simple basement membrane degradation. This last factor may be influenced by stromal signalling, hormonal milieu, age or other unknown environmental and dietary factors that may precipitate invasion of the basement membrane. Unlike what can be supposed, in this phase DCIS may be both low grade/ER positive and high grade/ER negative.
DCIS with short intraductal phase. DCIS arises within the ductal epithelial cells and may, in an unknown proportion of cases, rapidly become invasive after a short DCIS phase. The first possible factor is early critical mutation during the initiation phase of the tumour. Compared to BRCA2, BRCA1 tumours are more rarely associated with extensive intraductal neoplasia and may be an example of this type of progression. Other possible factors might be stromal factors, defect of basement membrane and/or the hormonal triggers at time of tumour formation. Usually DCIS are high grade with necrosis and ER negative.
Regressive DCIS. It is also possible, only theoretically and without any direct evidence, that DCIS undergoes regression in some cases. Regressive DCIS , or otherwise not evolving DCIS, usually low grade, are occasionally found in re-examination of histological samples. It remains unknown if DCIS remains dormant or could it undergo regression along with ductal epithelium following post-lactational settlement and postmenopausal involution.
Post-lactational settlement has been well described, but there is no description of the reconstitution of the ductal system for the next child. Is the pattern the same developed from residual stem cells lining a path through the stroma? Or completely different generated from rudimentary ducts behind the nipple? This remains to be explored.
As regards postmenopausal involution, in a latest study carried out among women with multiple biopsies, a significant association of higher BC risk among those with involution stasis, as compared with those with involution progression, has been observed [7].
1.2.3 What We Know About Natural History
Since the current standard of care of DCIS is surgical removal of the lesion, the natural history of DCIS is poorly understood because it cannot be directly observed. It is still disputable whether DCIS is an obligate precursor of invasive BC and, even if non-obligate, what proportion of in situ breast lesions progress to invasive cancer. Yet, an estimation of the proportion of DCIS that may not progress to invasive cancer and factors that can predict progression is therefore of clinical importance.
Available evidences, which strongly suggest DCIS is a precursor of most invasive BC, are:
Some invasive cancers have an adjacent DCIS component, and molecular studies show marked similarity in the genetic profile of the two components supporting origin of invasive cancer from DCIS.
Invasive cancers occurring in the excision site of a previous DCIS lesion show similarities to the primary DCIS suggesting that they have arisen from residual DCIS.
Molecular studies show the presence of shared identical genetic abnormalities between genetic changes in DCIS lesions and in recurrent lesions. These strong similarities are proved also in synchronous invasive BC when present, giving a further demonstration of a clonal relationship.
Epidemiological risk factors are also largely similar between DCIS and invasive BC.
Evidences also suggest not all DCIS will progress to invasive cancer in the medium term, but precise estimates of progression are not possible given the limitations of the data. It is unknown whether all BCs through a prolonged, potentially detectable intraductal phase or whether some of them rapidly invade surrounding stroma.
Actually there are several sources of evidence that shed light on the natural history of DCIS; however, none can provide a definitive answer on the proportion of DCIS that will progress to invasive cancer. The main studies are related to misdiagnosed DCIS, invasive recurrence and autopsy series.
DCIS initially misdiagnosed as benign and treated by biopsy alone. Studies give the most direct evidence regarding the progression of DCIS to invasive BC. These studies suggest that between 14 and 53% of DCIS may progress to invasive cancer over a period of 10 or more years [2].
Recurrence of DCIS as invasive cancer. Long-term series show that up to 40% of women treated with breast-conserving surgery will develop recurrent disease in the ipsilateral breast and about half of these recurrences will be as invasive cancer. The risk of recurrence has been shown to depend on patient characteristics such as history of breast cancer in a first degree, relative younger age at diagnosis and tumour factors such as histology, presence of necrosis, nuclear grade, size and architectural patterns.
Diagnosis of DCIS in autopsy studies. The reported incidence of DCIS has been used to suggest a larger reservoir of DCIS may exist in the population. Curiously, but conceivably, a highest reported autopsy studies on women who had not been diagnosed with breast cancer during their lifetime vary by study design, with a highest number of DCIS found in forensic rather than hospital autopsy.
Other evidences come from mathematical and animal models, as well as from epidemiology. Among these, mathematical models are promising, even if limited by broad assumptions and lack of complexity. However, modelling of various scenarios of progression, together with studies of genetic factors involved in progression, may shed further light on the natural history of DCIS. A summary of evidences on natural history of DCIS is presented in Table 1.1 [2].
Table 1.1
Summary of evidences on natural history of DCIS
Type of evidence | Conclusions | Limitations |
|---|---|---|
DCIS initially misdiagnosed as benign lesion | Studies suggest 14–53% may progress to invasive cancer over 10–15 years | Higher-grade lesions less likely to have been misdiagnosed. Follow-up likely to be more complete for women subsequently diagnosed with cancer |
Recurrence of DCIS as invasive cancer | Overall recurrence rate is founded between 1.45 and 22.5% of cases, about half of these showing an invasive BC | Recurrence is strongly dependent on excision margins and moreover may not reflect situation in absence of surgery |
Autopsy studies | Large reservoir of undetected DCIS in the population, thus not all DCIS progress to invasive cancer | Modelling predicts such a reservoir would be expected due to differing growth rates of tumours |
Mathematical models | Large reservoir of DCIS relative to invasive cancer predicted even if all DCIS progress | Broad assumptions and models lack complexity |
Epidemiology | Risk factors similar between DCIS and invasive cancer | Does not give estimate of progression rates, only that DCIS is likely to be a precursor for invasive cancer |
Animal models | Some useful models for studying genetic alterations associated with progression | Genetic background and hormonal milieu differ between models and applicability to humans questionable |
All of these studies are materially affected by:
Inadequate data on age-specific recurrence incidence, as it is possible that the proportion of DCIS that progresses to invasive BC may differ by age
Unpredictability of progression for low-grade DCIS, when the boundaries between benign or proliferative lesions, with or without atypia, and DCIS are very weak (see also Sect. 1.4)
Since the natural history of DCIS remains unobservable, it is essential that new methods are developed to estimate the progression of DCIS to invasive cancer. Mathematical models may provide important insights into growth and progression of DCIS. Identification of crucial molecular changes associated with progression of DCIS in animal models or in recurrent DCIS following surgery gives important incomes to identify what DCIS lesions are likely to progress or remain clinically benign on the basis of the presence of such key genetic alterations.
As DCIS progression to invasive BNC may never occur or take decades in some cases, the concern regarding the potential for overdiagnosis and overtreatment has fostered recently novel randomised trials investigating active surveillance versus active management. Together with multigene expression assays on vacuum biopsy, it seems possible to stratify patients at increased risk for progression to invasive BC even in the absence of surgery.
1.3 Space: A Revisited Anatomy
Clinical Practice Points
While knowledge of the microscopic anatomy of the mammary ducts is well established, less account is taken of its large-scale anatomy.
Subgross anatomy should refer to new anatomical models, which partly confirm results of past Cooper’s achievements.
There is noticeable variation in the extent of different lobes, and it could be assumed that some lobes develop earlier and larger than other ones.
To some degree the branches of lobes intertwine with each other, mainly in periphery but also in central part.
The boundary between multifocal and multicentric cancers may be elusive, as demonstrated by potential skips observed with ductal endoscopy and ductal echography.
1.3.1 Understanding the Misunderstood Subgross Anatomy
Since the 1970s, starting from the studies of Wellings and others [8], it has been widely accepted that all breast cancer begins at the junction of the duct and lobule or the terminal ductal-lobular unit (TDLU) . For this reason many believe BC is multifocal and/or multicentric in origin and could not begin in a ductal tree.
As regards multicentricity, the presence of more foci varies considerably depending on the criteria adopted, leading to not a little confusion and inappropriate adjudications. Nevertheless, at much the same time, researchers who have used techniques of whole breast sectioning have concluded that the in situ component is most often located in a single ductal tree or lobe.
In the 1990s, Tibor Tot, on whole-mount specimens, has hypothesised that DCIS and consequently BC in general are lobar diseases with simultaneous or asynchronous and often multiple in situ tumour foci localised within a single lobe.
As a result of these studies, the great importance of subgross anatomy behind multifocality became clear. New studies were launched and almost all had come to the same conclusions made by the first great anatomist Sir Astley Paston Cooper. In dealing with the anatomy of ducts and glandules, Cooper [3] puts together three illustrations of the individual lobes (Fig. 1.5), injected separately with different coloured waxes. In the first anatomical preparation, main ducts show their radiated direction, maintained also in the next preparation. In the last preparation, in which also appears the glandular component, the whole structure appears much more convoluted and indefinite.
Fig. 1.5
Anatomy of ducts and glandule, in Plate VI of Cooper’s text [3]. In the plate there are three images displaying, as Cooper’s caption read, “Lactiferous tubes or ducts injected with (coloured) wax, showing their radiated direction, and, in some places, their inter-ramification”. In the first image, not shown here, the main and segmental ducts having a rather regular course in the radial direction are showed. In the second (left), injected subsegmental ducts are intermixed and there are some discrepancies of volume of the lobes. In the third (right), “Ducts (are) injected more minutely (…) at the lower part of the preparation the separate ducts are seen passing above and beneath each other, to render the breast a cushion; whilst at the upper part the ducts are single”
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
