Introduction
Microglandular adenosis (MGA), defined by the World Health Organisation as a benign, proliferative condition, presenting as a haphazard proliferation of small round glands consisting of single layers of epithelial cells with no myoepithelial components [1], is a very rare lesion of the breast, identified in up to 0.1% of breast biopsies, and affecting patients mainly at the age of 29–77 with an average age of 46 [2].
Microglandular adenosis can co-exist within the mammary gland with invasive breast carcinoma (IBCa), mostly of the triple-negative type (TNBC), but the reports of such a co-incidence are relatively rare, accounting for approximately 27% of all MGA cases [3]. The significance of the MGA component in the MGA-TNBC lesion is unclear. It remains uncertain whether MGA is merely a harmless bystander or a nonobligate precursor to IBCa. The latter is supported by the morphologic, immunophenotypic and molecular evidence, which fits to the model of IBCa evolution via the MGA→IBCa transformation [2, 4]. Microglandular adenosis shares morphological features with IBCa such as the infiltrative growth pattern and the lack of the myoepithelial cell layer [5–8]. Some cases of MGA, so-called atypical MGA (aMGA), show architectural complexity and cytological atypia [9]. Microglandular adenosis cells do not express either hormone (estrogen receptor – ER, progesterone receptor – PR) receptors or HER2, which is consistent with the classification of a triple-negative (TN) lesion [1, 5, 10–12]. Moreover, there are reports of the similarities between the patterns of gene copy number alteration in MGAs, aMGAs and associated TNBCs [3, 13, 14]. The information on the genetics of MGA is, however, limited as only a small number of molecular studies on MGA have been carried out to date.
Here, we describe a case of MGA with an atypical TNBC component, whose histological features, highly suggestive of the MGA origin, advocate for the MGA→IBCa path of disease evolution.
Case report
The 66-year-old female was admitted to the hospital with a palpable mass in the left breast. At admission, physical examination did not reveal any significant abnormalities, and laboratory tests detected only slightly decreased eosinophil (0.03 × 103/ul) and lymphocyte (0.03 × 103/ul) counts. Mammography revealed two pathological masses: a 7-mm calcified nodule in the upper pole of the right breast (Fig. 1 A) and a 20-mm calcified nodule in the retroalveolar area of the left breast (Fig. 1 B), classified as a second (BI-RADS 2) and a fifth category (BI-RADS 5) in the Breast Imaging-Reporting and Data System, respectively [6]. A subsequent fine-needle biopsy of the lesion in the left breast revealed atypical cells.
Fig. 1
Mammography showing a 7-mm nodular density in the right upper pole (BI-RADS 2) (A), mammography of the left breast showing a 20-mm mass in the retroalveolar area of the left breast (BI-RADS 5) (B)
The patient underwent breast-conserving surgery combined with a biopsy of the left sentinel lymph node (guided by lymphoscintigraphy with 99mTc nanocolloid). The resected specimen of 9.0 cm × 4.0 cm × 3.0 cm contained a visible tumour measuring 1.5 cm × 1.5 cm × 1.0 cm. The microscopic appearance of the tumor was equivocal. It consisted of glands infiltrating adipose tissue; however, these glands were well-formed, and large areas lacked a desmoplastic response (Fig. 2 A). Some glands were uniform and small, while others were more complex, forming cribriform structures (Fig. 2 B). Most glands were lined by a single layer of cuboidal epithelial cells, periodic acid-schiff (PAS) stain – positive and in certain areas, they contained colloid-like, eosinophilic, PAS-positive secretions. Cellular atypia was variable, ranging from mild to severe, with areas of marked atypia suggestive of malignancy (Fig. 2 C). A focal desmoplastic response indicated a possible transition towards malignancy. Immunohistochemistry showed negativity for ER, PR, and HER2 (Fig. 2 D, Table 1), an elevated K-i67 proliferation index (Fig. 2 E), and heterogeneous positivity for S100 and SOX10 (Figs. 2 F, G). There was also low positivity for calponin and p63 in some glands (Figs. 2 H, I). The myoepithelial layer was absent around most glands. Considering the high atypia, areas of desmoplastic response, and the mixed immunohistochemical profile, a diagnosis of triple-negative, basal-like invasive breast carcinoma associated with MGA was made, following World Health Organization criteria.
Fig. 2
Microscopic findings. In hematoxylin and eosin staining, the tumour exhibited tubular and cribriform morphology, with myo epithelial cells infiltrating the adipose tissue of the breast. Typical microglandular adenosis areas consisted of round glands lined by a single layer of flat to cuboidal epithelial cells, lacking a myoepi thelial layer. The tumour displayed a high level of atypia, including abnormal mitoses, with the presence of myoepithelial cells. Mag nification at 10×, 100×, and 400×, respectively (A–C), immunohis tochemistry demonstrated negativity for oestrogen (100× magnifi cation) (D), slightly elevated Ki-67 index (100× magnification) (E), and positivity for SOX10 and S100 (respectively, 100× magnification) (F, G), staining for calponin and p63 confirmed the absence of myo epithelial cells (respectively, 100× magnification) (H, I)
Table 1
Immunohistochemical profile of the tumour
| Stain | Positive | Negative | Equivocal |
|---|---|---|---|
| E-cadherin | + | ||
| Cytokeratin 5/6 | – | ||
| HER2 | – | ||
| Estrogen receptor | – | ||
| Progesterone receptor | – | ||
| S-100 | + | ||
| 34B12 | +/– | ||
| Calponin | +/– | ||
| Cytokeratin 18 | +/– | ||
| p63 | +/– | ||
| SMA | +/– |
The patient underwent four cycles of docetaxel and cyclophosphamide adjuvant chemotherapy, according to the standard regimen. For 30 months of follow-up no recurrence has been detected.
Discussion
The presented case demonstrates co-existence of MGA and TNBC lesions. Consistent with existing reports, despite being recognized as a benign lesion, MGA manifested several morphological characteristics of IBCa [15–17]. Furthermore, as described in previous studies, the MGA/aMGA and IBCa shared the same immunophenotype – negative for ER, PR, HER2 and CK5/6 with expression of S100 and basal-like markers such as – CD8/18 and epidermal growth factor receptors [8, 18, 19]. To date, only five cases of MGA-IBCa with an immunophenotype other than TN have been described [11, 16, 19, 20].
Morphologically, histological subtypes of MGA-associated IBCa reported thus far include: not otherwise specified ductal carcinoma, metaplastic matrix producing carcinoma, carcinoma with metaplastic squamous differentiation, acinic cell carcinoma, spindle cell carcinoma, and adenoid cystic carcinoma and, as in our case, salivary type carcinomas [8, 21, 22]. Curiously, invasive tumours often replicate the morphology of the associated MGA and aMGA components, such as clear cell features, cytoplasmic granules resembling acinic cells, and secretory activity [14].
As demonstrated by several studies, genetically, MGA is a heterogeneous group of lesions. Shin et al. identified significant genetic instability in MGA, marked by chromosomal aberrations across multiple chromosomal arms. The most common alterations in MGA included gains in 2q and 8q, and losses of 5q and 14q. Atypical MGA exhibited additional changes, such as a gain in 1q and losses in 14q and 15q [3]. Moreover, available results of genetic analyses show that MGA/aMGA and concurrent IBCa share the same molecular alterations. Using high-resolution microarray-based comparative genomic hybridization, Geyer et al. found that MGA/aMGA and concurrent IBCa exhibited similar genomic profiles. This demonstrated that, in each case, the MGA/aMGA/IBCa components were clonally related, pointing to the MGA as a likely precursor of the invasive carcinoma [13, 14]. Moreover, Guerini-Rocco et al. proved that MGA/aMGA arising with TNBC display a genomic landscape similar to TNBC. This includes recurrent TP53 mutations, somatic mutations in other cancer-related genes (PI3K pathway-related genes e.g. PTEN, PIK3CA, INPP4B and tyrosine kinase receptor signalling-related genes e.g. ERBB3 and FGFR2), and a complex pattern of copy number alterations. In contrast, pure MGAs did not exhibit clonal non-synonymous somatic mutations and showed only limited copy number alterations [23]. Additionally, Radner et al. found that MGA and adjacent IBCa shared mutations in genes such as TP53, CTNNB1, FGFR2 and PIK3CA as well as amplification of the ERBB4 gene. This study also demonstrated that gain of chromosome 2q was the most common copy number alteration in MGA, which was also found in adjacent IBCa. Interestingly, Radner et al. identified hypermethylation of GATA3 CpG sites in MGA, leading to reduction of GATA3 protein expression that is a key factor for luminal differentiation and ER expression [24]. Notably, the molecular landscape of MGA/aMGA differs from that of the well-known precursors of oestrogen-positive IBCa (such as atypical ductal hyperplasia, flat epithelial atypia) [8]. Taken together, available evidence supports the hypothesized MGA→TNBC transformation pointing at MGA/aMGA as a true precursor of molecularly atypical TNBC.
Although in the presented case, no genetic analyses were carried out, results of histological and immunohistochemical characterization are consistent with the postulated path of disease evolution.
However, it is important to realise that despite morphological features indicative of poor prognosis, in most cases of MGA-IBCa, including ours, the outcome is significantly more favourable than in the patients with pure TNBC [5]. Being classified as a benign condition, there are not data on MGA survival rates. In cases with documented progression to invasive carcinoma, 5-year survival reaches the level of approximately 77–90% [7]. In contrast, a 5-year survival rate for IBCa is about 34–42% [7]. This points to the need to reconsider MGA-TNBC treatment protocols towards potentially less aggressive therapies [1, 2, 5, 13, 16, 17, 21, 25–27]. It also raises the question of whether pure MGA should continue to be treated as a benign lesion or as a precursor to TNBC with advanced follow-up care.
Conclusions
Our case emphasizes the importance of recognizing MGA as a potential precursor to invasive breast carcinoma, especially the TN subtype. Given that IBCa-MGA cases have better outcomes than IBCa alone, it is important to revaluate MGA treatment protocols to avoid unnecessary overtreatment. Increasing awareness and applying comprehensive diagnostic methods are crucial for improving patient outcomes and effectively managing the risks associated with MGA. Further genetic research is essential to understanding the molecular mechanisms driving this progression to resolve this clinical conundrum.
