ICC is a rare invasive breast carcinoma with an excellent prognosis. The incidence of ICC was reported to range from 0.3% to 3.5% [1, 6–8]. While the WHO earlier definition (2003) and Page’s description classified ICC into classical and mixed subtypes [1, 9], the WHO new edition (2012) seems to weaken the concept of the classical ICC . In any case, both the old and new WHO definitions propose that pure ICC should be defined as breast carcinoma with infiltrating components presenting an almost entirely (>90%) invasive cribriform pattern. Venable  reported 62 cases of ICC from 1,087 patients with primary breast carcinoma and found a five-year survival rate of 93.1% for all ICC cases (all died of ICC with < 50% invasive cribiform pattern), and 100% survival for ICC with 50% to 99% invasive cribiform pattern (including pure ICC). Page et al.  reported 51 cases of ICC (35 classical and 16 mixed ICC) over an average follow-up time of 12.5 years. They found 10-year survival rates of 97% for classical ICC and 50% for mixed ICC. One patient with classical ICC died from a second and contralateral carcinoma, not the primary ICC. Ellis et al.  compared 10-year survival rates for different histological types in 1,621 cases of primary breast cancer. Ductal carcinoma in situ, cribriform and TC had excellent prognoses, with 10-year survival rates of 92%, 91% and 90%, respectively, compared with 80% for mucinous carcinoma, 51% for medullary carcinoma, 54% for lobular carcinoma and 47% for invasive ductal carcinoma-not other specified (IDC-NOS). The frequency of metastasis in the axillary lymph nodes was lower in pure ICC than in mixed ICC (14.3% vs. 25%, respectively) . Venable found that the maximal number of positive lymph nodes in pure ICC was three . To date, evidence of distant metastasis in ICC have been limited, especially for pure ICC.
In this paper, we report a case of pure ICC with bone matastasis that had been left untreated for 13 years. The patient’s pathology revealed that over 90% of the invasive tumor cells grew in a cribriform pattern, thus, the patient was considered to have pure ICC. The tumor cells that had locally invaded the nipple, skin and muscle were still arranged in a cribriform pattern (as indicated by Figure 3). Histology verified only one obvious metastatic lymph node and one micro-infiltrating lymph node, possibly due to the effect of two cycles of neoadjuvant chemotherapy received by the patient. Although we detected multiple nodules in both lungs during CT imaging on initial examination, we could not ascertain whether or not the patient had lung metastasis because she did not show any pulmonary symptoms at her first diagnosis and subsequent follow-up. Surprisingly, the patient was alive seven years after neoadjuvant chemotherapy, radical mastectomy, and postoperative systemic therapy. Taking the 13 years prior to treatment into account, this patient had survived for at least 20 years after the discovery of the lump in her breast.
Page  first reported that the average age of classical ICC is 53 years, younger than the age in mixed ICC (63 years). Venable reported that the mean age of patients with ICC is 58 years, including individuals whose tumors contain a cribriform pattern in less than 50% of cells . Although the age at diagnosis of the patient in the present case was 59 years, her actual age during discovery of the lump was 46 years, younger than in the Page and Venable reports. Our recent retrospective investigation of 51 ICC cases found that the average age of patients with pure ICC was 51 years (unpublished data). We thus speculate that the different ages at diagnosis may contribute to this difference because breast carcinomas may now be detected earlier with the use of more advanced detection techniques and increased awareness of self-examination. Second, ethnic differences might result in this disparity, which was confirmed by recent publications on IDC-NOS [10–12]. The ICC tumor may present as a mass but is frequently clinically occult, and may escape radiological detection. Thus, lesions are usually larger at presentation, although they grow slowly over time .
The patient in this case noted a lump with a diameter of 2.0 cm 13 years ago but did not consult a doctor. The tumor finally grew to a large mass with a diameter of 10 cm. The nipple, skin, and muscle were also involved. The patient’s pathology revealed well-differentiated invasive tumor cells because they were arranged in irregular islands with extensive gland lumen formation. Cytologically, the nuclear grade is usually low or moderate, and the karyokinesis is rare. The remaining components of this tumor (< 10%) comprised TC, which has been confirmed to be occasionally associated with ICC (23%) . Taylor and Norris  speculated that TC may, in some instances, serve as a differentiation bridge between intraductal carcinoma and ICC. Previous studies have indicated that in ICC there is expression of ER, but mostly there is expression of PR (69%), which could contribute to the favorable prognosis of this breast carcinoma [1, 5, 6]. ER and PR in this pure ICC case were found in 15% of the tumor cells even after 13 years of disease development. We thus speculate that the persistent expression of hormone receptors in pure ICC is a factor that influences the excellent prognosis of the carcinoma, given that the tumor at initial mass was positive for ER, since no publications of negative ER for pure ICC have been reported to date [1, 5, 6]. Previous publications [1, 5–7] indicated that only metastatic lymph nodes in pure ICC are arranged in a cribriform pattern, similar to primary ICC tumors; such findings were confirmed by our results. Tumor cells in one micro-infiltrating lymph node and in other paranodal tissues also displayed a similar morphology (Figure 4b and 4c).
To date, the prognostic significance of the clinicopathological characteristics of ICC patients has not yet to be well established because of the low incidence rate and lack of a standard definition for the carcinoma. Thus, treatment guidelines for ICC are mostly extrapolated from data based on IDC without clear validation. The case we reported was a late-stage (T4) breast carcinoma for which comprehensive therapy was carried out. Neoadjuvant chemotherapy was performed because of the large size of the mass and the presence of two enlarged axillary lymph nodes. Radical corrective surgery for cytoreduction and endocrine therapy followed., Thoracic radiotherapy was also performed because of the possibility of bone metastasis. Several recent publications have questioned whether adjuvant chemotherapy is suitable in ICC cases, because reports on the efficacy of adjuvant or neoadjuvant chemotherapy in patients with ER-positive disease are limited [14–16]. Colleoni  recommended no therapy or endocrine therapy alone for favorable histotypes (such as, tubular, cribriform, mucinous, papillary) with luminal tumors. Caution must be exercised when making such a decision. Neoadjuvant chemotherapy was partially effective for this patient because of the shrink in both the tumor and axillary lymph node. However, we cannot exclude the roles that postoperative local and systemic therapies played in the patient’s excellent prognosis. We believe that definition of specific niches for tailored research through international cooperation and prospective investigations of this topic are necessary to make progress and achieve a consensus on the treatment of individual patients with ICC.