Long-term outcomes of rare pure histological subtypes of breast cancer in a tertiary single-center retrospective cohort Authors: Mehmet Cem Fidan, Murad Guliyev, Murat Günaltılı, Zeliha Birsin, Ebru Çiçek, Hamza Abbasov, Emir Çerme, Selin Cebeci, Vali Aliyev, Berrin Papila, İpek Sertbudak, Tülin Öztürk, Gökmen Umut Erdem, Özkan Alan, Nebi Serkan Demirci, Çiğdem Papila Published: 2026 ABSTRACT BackgroundRare pure histological breast cancer subtypes represent a clinically heterogeneous group of tumors with diverse biological behaviours and prognostic profiles. Due to their rarity, data, particularly regarding long-term survival outcomes, are limited and controversial. This study aimed to evaluate the clinicopathological features and long-term survival outcomes of patients diagnosed with rare pure histological breast cancer subtypes in a single tertiary center.MethodsThis retrospective single-center study included 117 patients diagnosed with pure rare breast cancer histologies between November 2000 and February 2024. These included mucinous, metaplastic, micropapillary, and tubular carcinomas, as well as malignant phyllodes tumors. Clinicopathological features, treatment methods, recurrence risks, and survival outcomes were analyzed. 5-year and 10-year overall survival (OS) rates were calculated.ResultsThe most common subtype was mucinous carcinoma (41.9%), followed by metaplastic (25.6%), micropapillary (14.5%), tubular carcinoma (10.2%), and malignant phyllodes tumor (7.7%). Favourable biological features such as low grade, low Ki-67 proliferation index, and high hormone receptor expression were dominant in mucinous and tubular carcinomas, while metaplastic and micropapillary carcinomas frequently exhibited aggressive biological features such as high grade, high Ki-67, and more frequent lymph node involvement. In long-term follow-up, recurrence was most common in malignant phyllodes tumors, micropapillary carcinomas, and metaplastic carcinomas. Five- and ten-year overall survival rates differed significantly among subtypes, with the most favourable outcomes observed in tubular and mucinous carcinomas and relatively lower long-term survival observed in metaplastic and malignant phyllodes tumors.ConclusionsPure rare histological breast cancer subtypes appear to show heterogeneity in clinicopathological features and long-term survival outcomes. Due to the small number of patients in individual subgroups, these findings should be interpreted with caution. Our descriptive data suggest a potential prognostic role for histological subtype and highlight the importance of accurate pathological classification. Long-term follow-up data from tertiary care centers may provide useful descriptive information about the natural history and survival patterns of these rare breast cancer subtypes. Introduction According to GLOBOCAN 2022 data, breast cancer accounts for 11.6% of newly diagnosed cancer cases and is the second most common cancer type worldwide after lung cancer [1, 2]. Breast tumors are generally classified into two main classes: epithelial and non-epithelial. The World Health Organization (WHO) classification of breast tumors identifies 21 different invasive epithelial breast tumor subtypes [3] The most common of these are invasive ductal carcinoma (70–80%) and invasive lobular carcinoma (5–10%) [4, 5]. Rare epithelial tumors include tubular, mucinous, medullary, papillary, micropapillary, and apocrine carcinomas, while rare non-epithelial tumors include sarcomas, phyllodes tumors, and lymphomas. Individually, these rare histological subtypes account for a small proportion of all invasive breast cancers. Mucinous carcinoma represents approximately 1–7% of cases [6], tubular carcinoma 0.5–2% [4, 7] invasive micropapillary carcinoma less than 2% [4, 8, 9] and metaplastic carcinoma less than 1% [9] Malignant phyllodes tumors are even rarer, constituting less than 1% [10]. Due to the diagnostic challenges and the marked clinical and prognostic heterogeneity observed across breast cancer subtypes, careful pathological evaluation is essential, and obtaining a second pathological opinion from experienced centers is often recommended [11]. Therefore, data on the clinical behaviour and prognosis of these rare histological subtypes are limited, and the results are controversial. Current guidelines do not include a specific treatment for these rare subtypes due to their uncertain prognosis and lack of prospective studies; they are treated similarly to invasive ductal carcinomas. Mucinous, tubular, and medullary carcinomas generally have a good prognosis [12–14], while micropapillary, metaplastic, and neuroendocrine carcinomas are associated with more aggressive biological behaviour and poorer survival [6, 8]. Due to their rarity, it is difficult to collect significant data on these histological subtypes of breast cancer. A database containing large patient numbers is needed to obtain more precise information and improve a comprehensive understanding and appropriate management. The aim of this study was to evaluate the clinicopathological features, tumor size, lymph node status, stage, molecular subtype, treatment selection, and long-term survival outcomes of rare breast cancer subtypes diagnosed at our center. Evaluating these features, particularly because of their long-term outcomes, could contribute to understanding the nature of these rare tumors, contributing to clinical approaches and improving patient management. Materials and methods Research methodology This retrospective study, conducted at a single center, involved 117 patients diagnosed with pure rare breast cancer histology from November 2000 to February 2024. Because tumor heterogeneity can lead to variability in clinical behavior and potentially confound survival analyses, we aimed to provide a more accurate characterization of the natural course of each distinct subtype by limiting the cohort to pure histological forms. The inclusion criteria consisted of histopathological confirmation of the rare breast cancer subtype, along with immunohistochemical assessment of estrogen receptor (ER), progesterone receptor (PgR), and human epidermal growth factor receptor 2 (HER2) status in patients. Patients lacking adequate clinicopathological documentation and with uncertain clinical outcomes were excluded. Patient population Clinicopathological data were obtained from databases and medical records. Data were collected regarding patient characteristics, tumor histopathological features, administered treatments and their classifications, as well as the clinical course of the patients. Assessments of ER, PgR, and HER2 status in patients were conducted following the guidelines established by the American Society of Clinical Oncology and the American College of Pathologists [15, 16]. Following the St. Gallen consensus recommendations [17], we established a threshold of 20% for Ki-67 and PgR expression levels, classifying patients into Ki-67-low (< 20%) and Ki-67-high (≥ 20%); PgR-low (< 20%) and PgR-high (≥ 20%). Patients exhibiting HER2 immunohistochemistry (IHC) scores of 0, + 1, +2, and lacking gene amplification via in situ hybridization (ISH) were designated as HER2 negative. Conversely, patients with HER2 IHC scores of + 2 accompanied by gene amplification through ISH, as well as those with IHC + 3, were categorized as HER2 positive [15]. Additional rare histological subtypes, such as adenoid cystic carcinoma, secretory carcinoma, and apocrine carcinoma, were identified during the study period; however, these were excluded from the current analysis due to single-case representation (n = 1 each), which prevented meaningful statistical and survival assessment. Outcomes of the study Overall survival (OS) was defined as the time from diagnosis to death from any cause or to the date of the last follow-up. Survival outcomes were estimated using the Kaplan-Meier method. Five- and ten-year survival rates were derived from the Kaplan-Meier survival function. Survival was also assessed using a life table at monthly intervals in subtypes where a median overall survival time could not be obtained, and the cumulative survival probabilities at the end of the respective intervals were used to estimate long-term survival rates. Ethical considerations The study was conducted in accordance with the principles of the Declaration of Helsinki and received approval from the local clinical research ethics committee (date: September 18, 2024; number: E 83045809-604.01.01.01–1094495). The retrospective design of this study eliminated the need for informed consent. Statistical analysis Survival analyses were performed separately for each histological subtype. Pooled survival estimates combining epithelial carcinomas and malignant phyllodes tumors were not performed. Due to the small number of patients within individual subgroups, formal statistical comparisons between subtypes (e.g., log-rank tests) and multivariable analyses were not conducted; accordingly, all results are descriptive. Statistical analysis was performed using SPSS version 27. First, normal distribution was tested for quantitative variables. Descriptive data were presented as percentages, means, and medians. Overall survival (OS) was defined as the time from diagnosis to mortality or the last follow-up in any given case. Results Baseline characteristics of patients A total of 117 patients with pure rare histological subtypes of breast cancer were included in the study. The distribution of histological subtypes was as follows: mucinous carcinoma (n = 49; 41.9%), metaplastic carcinoma (n = 30; 25.6%), micropapillary carcinoma (n = 17; 14.5%), tubular carcinoma (n = 12; 10.2%), and malignant phyllodes tumor (n = 9; 7.7%). Baseline demographic and clinicopathological characteristics according to histological subtype are summarized in Table 1. The median age at diagnosis differed among subtypes, with tubular carcinoma representing the oldest patient group (median age: 69 years) and malignant phyllodes tumors the youngest (median age: 41 years). All patients in the mucinous and tubular carcinoma subgroups presented with grade 1 or 2 tumors. On the other hand, 90% (n = 27) of patients with the metaplastic subtype and 52.9% (n = 9) of patients with micropapillary cancer showed grade 3 tumors. Luminal A was the dominant molecular subtype in mucinous (95.9%) and tubular (100%) carcinomas, while Luminal B was most common in micropapillary carcinoma (58.8%). Metaplastic carcinoma was predominantly triple-negative (66.7%). Consistent with these distributions, mucinous and tubular carcinomas were uniformly hormone receptor–positive and HER2-negative, whereas HER2 positivity was more frequently observed in metaplastic (20%) and micropapillary (23.5%) subtypes. Low proliferative activity (Ki-67 < 20%) predominated in mucinous (95%) and tubular (100%) carcinomas, whereas higher Ki-67 levels (≥ 20%) were commonly observed in metaplastic (83.3%) and micropapillary (58.8%) carcinomas. Radiologically, the largest tumors were observed in malignant phyllodes tumors (median: 44 mm), while tubular carcinomas had the smallest median tumor size (10 mm). Pathologically, the subtype with the largest tumor size was malignant phyllodes tumor (median: 65 mm; min-max: 35–140 mm). The majority of patients were diagnosed at early stages (T1–T2; 87.1%). De novo metastatic disease was observed in three patients (mucinous, n = 1; metaplastic, n = 2). Clinically, node-positive disease was most frequently observed in the micropapillary carcinoma subtype (76.5%). Clinically, node-negative disease predominated in the other subtypes. Clinically and pathologically, more advanced node positivity was observed in the micropapillary carcinoma subtype (cN2 or cN3 disease rate 23.5%). Pathological node positivity and more advanced node positivity (pN2 or pN3) were more frequent in the micropapillary carcinoma subtype. In total, 3 patients had de novo metastatic disease (mucinous carcinoma, n = 1; metaplastic carcinoma, n = 2). A total of four patients received systemic treatment (1 patient with metaplastic carcinoma experienced progression under neoadjuvant therapy). The subtypes with the most frequent recurrence after definitive treatment were, respectively, malignant phyllodes tumor (n:3; 33%), micropapillary carcinoma (n:5; 29.4%), and metaplastic carcinoma (n:8; 26.7%). Table 1Baseline clinicopathological characteristics of patientsClinicopathologicalvariablesHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign Filloidesn=49n=30n=17n=12n=9AgeMean ± Std67 ± 1351 ± 1256 ± 753 ± 1043 ± 12Median (Min. - Max.)67 (35 - 89)53 (29 - 80)56 (44 - 70)52 (42 - 74)41 (30 - 62)Menopausal StatusPremenopausal, n (%)48.21343.3529.4433.3666.7Postmenopausal, n (%)4591.81756.71270.6866.7333.3PR %20<%20, n (%)4387.8310.01164.712100.000.0≥%20, n (%)612.22790.0635.300.09100.0HER-2 StatusNegative, n (%)49100.02480.01376.512100.000.0Positive, n (%)00.0620.0423.500.000.0Ki67<%20, n (%)4795.9516.7741.212100.09100.0≥%20, n (%)24.12583.31058.800.000.0GradeGrade 1–2, n (%)49100.0310.0847.112100.000.0Grade 3, n (%)00.02790.0952.900.000.0Reseptor StatusLuminal A, n (%)4795.900.0211.812100.000.0Luminal B, n (%)24.1413.31058.800.000.0HR positive HER-2 positive, n (%)00.013.3317.600.000.0HR negative HER-2 positive, n (%)00.0516.715.900.000.0Triple-negative, n (%)00.02066.715.900.000.0clinical T (Radiological tumor size, mm)Mean ± Std28 ± 1637 ± 1842 ±4910 ± 456 ± 32Median (Min. - Max.)25 (4 - 67)30 (10 - 87)22 (7- 184)10 (7 - 18)44 (16 - 110)clinical N Stage0, n (%)3979.61860.0741.212100.09100.01, n (%)1020.41240.0635.300.000.02, n (%)00.000.0211.800.000.03, n (%)00.000.0211.800.000.0(y)pT (Pathological tumor size, mm)Mean ± Std25 ± 1642 ± 2629 ± 469 ± 367 ± 36Median (Min. - Max.)23 (2 - 65)34 (15 - 130)20 (0 - 180)10 (4 - 15)65 (35 - 140)(y)pT StageT00100.000.00T1a24.113.3317.618.300.0T1b612.226.7211.8975.000.0T1c1428.613.3317.6216.700.0T22142.92273.3635.300.0444.4T3510.2310.015.900.0555.6T4a00.000.000.000.000.0T4b00.000.000.000.000.0T4c00.000.000.000.000.0T4d00.000.000.000.000.0ypN StageN03673.51756.7635.312100.09100.0N1i36.113.315.900.000.0N1a714.3620.0423.500.000.0N1b12.0310.000.000.000.0N1c00.013.315.900.000.0N2a00.013.3211.800.000.0N2b12.000.015.900.000.0N3a00.013.315.900.000.0N3b00.000.015.900.000.0N3c00.000.000.000.000.0MetastasisNo, n (%)4898.030100.01588.212100.09100.0Yes, n (%)12.000.0211.800.000.0RecurrenceNo, n (%)4591.82273.31270.612100.0666.7Yes, n (%)36.1826.7529.400.0333.3HR: Hormone receptor Table 1Baseline clinicopathological characteristics of patientsClinicopathologicalvariablesHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign Filloidesn=49n=30n=17n=12n=9AgeMean ± Std67 ± 1351 ± 1256 ± 753 ± 1043 ± 12Median (Min. - Max.)67 (35 - 89)53 (29 - 80)56 (44 - 70)52 (42 - 74)41 (30 - 62)Menopausal StatusPremenopausal, n (%)48.21343.3529.4433.3666.7Postmenopausal, n (%)4591.81756.71270.6866.7333.3PR %20<%20, n (%)4387.8310.01164.712100.000.0≥%20, n (%)612.22790.0635.300.09100.0HER-2 StatusNegative, n (%)49100.02480.01376.512100.000.0Positive, n (%)00.0620.0423.500.000.0Ki67<%20, n (%)4795.9516.7741.212100.09100.0≥%20, n (%)24.12583.31058.800.000.0GradeGrade 1–2, n (%)49100.0310.0847.112100.000.0Grade 3, n (%)00.02790.0952.900.000.0Reseptor StatusLuminal A, n (%)4795.900.0211.812100.000.0Luminal B, n (%)24.1413.31058.800.000.0HR positive HER-2 positive, n (%)00.013.3317.600.000.0HR negative HER-2 positive, n (%)00.0516.715.900.000.0Triple-negative, n (%)00.02066.715.900.000.0clinical T (Radiological tumor size, mm)Mean ± Std28 ± 1637 ± 1842 ±4910 ± 456 ± 32Median (Min. - Max.)25 (4 - 67)30 (10 - 87)22 (7- 184)10 (7 - 18)44 (16 - 110)clinical N Stage0, n (%)3979.61860.0741.212100.09100.01, n (%)1020.41240.0635.300.000.02, n (%)00.000.0211.800.000.03, n (%)00.000.0211.800.000.0(y)pT (Pathological tumor size, mm)Mean ± Std25 ± 1642 ± 2629 ± 469 ± 367 ± 36Median (Min. - Max.)23 (2 - 65)34 (15 - 130)20 (0 - 180)10 (4 - 15)65 (35 - 140)(y)pT StageT00100.000.00T1a24.113.3317.618.300.0T1b612.226.7211.8975.000.0T1c1428.613.3317.6216.700.0T22142.92273.3635.300.0444.4T3510.2310.015.900.0555.6T4a00.000.000.000.000.0T4b00.000.000.000.000.0T4c00.000.000.000.000.0T4d00.000.000.000.000.0ypN StageN03673.51756.7635.312100.09100.0N1i36.113.315.900.000.0N1a714.3620.0423.500.000.0N1b12.0310.000.000.000.0N1c00.013.315.900.000.0N2a00.013.3211.800.000.0N2b12.000.015.900.000.0N3a00.013.315.900.000.0N3b00.000.015.900.000.0N3c00.000.000.000.000.0MetastasisNo, n (%)4898.030100.01588.212100.09100.0Yes, n (%)12.000.0211.800.000.0RecurrenceNo, n (%)4591.82273.31270.612100.0666.7Yes, n (%)36.1826.7529.400.0333.3HR: Hormone receptor Baseline clinicopathological characteristics of patients HR: Hormone receptor Treatment approaches are detailed in Table 2. In all subtypes, most patients received adjuvant therapy after an upfront surgical approach. No patient with mucinous, tubular, or malignant phyllodes subtypes received neoadjuvant therapy. Neoadjuvant treatment was applied only in metaplastic (n:7; 23.3%) and micropapillary (n:4; 23.5%) subtypes. The breast-conserving surgical approach was most frequently preferred in tubular carcinoma (n:11, 91.7%), malignant phyllodes tumor (n:7, 77.8%), and mucinous carcinoma (n:34, 69.4%) subtypes, while mastectomy was more frequently performed in metaplastic carcinoma (n:17, 56.7%) and micropapillary carcinoma (n:10, 58.9%) subtypes. In 8.3% of patients who underwent surgery (upfront or after neoadjuvant treatment) (total n: 10 patients; mucinous n:3, micropapillary n:2, metaplastic n:2, malignant phyllodes tumors n:2, tubular carcinoma n:1), re-excision was performed due to positive surgical margins. In 8 of these patients, the initial surgical approach was breast-conserving surgery, while in 2 patients (micropapillary carcinoma), the initial surgical approach was mastectomy. Table 2Treatment characteristicsClinicopathological variablesHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign Filloidesn=49n=30n=17n=12n=9TherapyOnly Surgery, n (%)24.100.000.0325.0888.9Surgery plus AdjTx, n (%)4693.92273.31164.7975.0111.1NeoTx + Surg ± AdjTx, n (%)00.0723.3423.500.000.0Systemic Treatment, n (%)12.013.3211.800.000.0Surgery TypeBCS plus SLNB, n (%)2449.0620.000.0975.0777.8BCS-ALND, n (%)1020.4516.7423.5216.700.0Mastectomy plus SLNB, n (%)918.4516.7211.818.3222.2Mastectomy plus ALND, n (%)510.21240.0847.100.000.0BCS plus SLNB -> ALND, n (%)00.013.315.900.000.0Neoadjuvant TherapyNo, n (%)49100.02376.71376.512100.09100.0Yes, n (%)00.0723.3423.500.000.0Adjuvant TherapyNo, n (%)36.1310.0211.8216.7777.8Yes, n (%)4693.92790.01588.21083.3222.2Adjuvant Radiation TherapyNo, n (%)1020.41033.3635.318.3777.8Yes, n (%)3979.62066.71164.71191.7222.2Re-excisionNo, n (%)4693.92893.31588.21191.7777.8Yes, n (%)36.126.7211.818.3222.2AdjTx: Adjuvant therapy, ALND: Axillary lymph node dissection, BCS: Breast-conserving surgery, NeoTx: Neoadjuvan therapy, SLNB: sentinel lymph node biopsy, Table 2Treatment characteristicsClinicopathological variablesHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign Filloidesn=49n=30n=17n=12n=9TherapyOnly Surgery, n (%)24.100.000.0325.0888.9Surgery plus AdjTx, n (%)4693.92273.31164.7975.0111.1NeoTx + Surg ± AdjTx, n (%)00.0723.3423.500.000.0Systemic Treatment, n (%)12.013.3211.800.000.0Surgery TypeBCS plus SLNB, n (%)2449.0620.000.0975.0777.8BCS-ALND, n (%)1020.4516.7423.5216.700.0Mastectomy plus SLNB, n (%)918.4516.7211.818.3222.2Mastectomy plus ALND, n (%)510.21240.0847.100.000.0BCS plus SLNB -> ALND, n (%)00.013.315.900.000.0Neoadjuvant TherapyNo, n (%)49100.02376.71376.512100.09100.0Yes, n (%)00.0723.3423.500.000.0Adjuvant TherapyNo, n (%)36.1310.0211.8216.7777.8Yes, n (%)4693.92790.01588.21083.3222.2Adjuvant Radiation TherapyNo, n (%)1020.41033.3635.318.3777.8Yes, n (%)3979.62066.71164.71191.7222.2Re-excisionNo, n (%)4693.92893.31588.21191.7777.8Yes, n (%)36.126.7211.818.3222.2AdjTx: Adjuvant therapy, ALND: Axillary lymph node dissection, BCS: Breast-conserving surgery, NeoTx: Neoadjuvan therapy, SLNB: sentinel lymph node biopsy, Treatment characteristics AdjTx: Adjuvant therapy, ALND: Axillary lymph node dissection, BCS: Breast-conserving surgery, NeoTx: Neoadjuvan therapy, SLNB: sentinel lymph node biopsy, Adjuvant treatment approaches varied according to histological subtype (Table 3). Adjuvant endocrine therapy was predominantly administered in mucinous and tubular carcinomas, reflecting their hormone receptor–positive profiles. In contrast, adjuvant chemotherapy was more frequently used in metaplastic and micropapillary carcinomas. Adjuvant radiotherapy was commonly applied following breast-conserving surgery across all subtypes. Malignant phyllodes tumors were mainly treated with surgical resection, with adjuvant radiotherapy administered in selected cases, while adjuvant chemotherapy was used infrequently. Table 3Summary of adjuvant therapy regimensAdjuvant therapyHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign PhyllodesOnly letrozole261220Only anastrozole110350Only Tamoxifen10110Tamoxifen → AI ± OFS11010TC → AI12000FEC → AI40000FEC → Tamoxifen + OFS20000AC-T → AI01100AC-T05000FEC06000TC01000Capecitabine02000AC-T → Tamoxifen00100AC-T → Tamoxifen → AI00200Trastuzumab02100Trastuzumab + AI00200T-DM1 + AI01000TH01000TH → AI01000AC-TH03100AC-THP + AI00100TCH01000IMA00001AI: aromatase inhibitor, AC-T: doxorubicin and cyclophosphamide followed by taxol, AC-TH: doxorubicin and cyclophosphamide followed by taxol and trastuzumab, AC-THP: doxorubicin and cyclophosphamide followed by taxol, trastuzumab and pertuzumab, FEC: 5-fluorouracil, epirubicin and cyclophosphamide, IMA: ifosfamide, mesna and doxorubicin, OFS: ovarian function suppression, TC: docetaxel and cyclophosphamide, T-DM1: ado-trastuzumab emtansine, TH: paklitaxel and trastuzumab, TCH: taxol, carboplatin and trastuzumab Table 3Summary of adjuvant therapy regimensAdjuvant therapyHistological SubtypeMucinousMetaplasticMicropapillaryTubularMalign PhyllodesOnly letrozole261220Only anastrozole110350Only Tamoxifen10110Tamoxifen → AI ± OFS11010TC → AI12000FEC → AI40000FEC → Tamoxifen + OFS20000AC-T → AI01100AC-T05000FEC06000TC01000Capecitabine02000AC-T → Tamoxifen00100AC-T → Tamoxifen → AI00200Trastuzumab02100Trastuzumab + AI00200T-DM1 + AI01000TH01000TH → AI01000AC-TH03100AC-THP + AI00100TCH01000IMA00001AI: aromatase inhibitor, AC-T: doxorubicin and cyclophosphamide followed by taxol, AC-TH: doxorubicin and cyclophosphamide followed by taxol and trastuzumab, AC-THP: doxorubicin and cyclophosphamide followed by taxol, trastuzumab and pertuzumab, FEC: 5-fluorouracil, epirubicin and cyclophosphamide, IMA: ifosfamide, mesna and doxorubicin, OFS: ovarian function suppression, TC: docetaxel and cyclophosphamide, T-DM1: ado-trastuzumab emtansine, TH: paklitaxel and trastuzumab, TCH: taxol, carboplatin and trastuzumab Summary of adjuvant therapy regimens AI: aromatase inhibitor, AC-T: doxorubicin and cyclophosphamide followed by taxol, AC-TH: doxorubicin and cyclophosphamide followed by taxol and trastuzumab, AC-THP: doxorubicin and cyclophosphamide followed by taxol, trastuzumab and pertuzumab, FEC: 5-fluorouracil, epirubicin and cyclophosphamide, IMA: ifosfamide, mesna and doxorubicin, OFS: ovarian function suppression, TC: docetaxel and cyclophosphamide, T-DM1: ado-trastuzumab emtansine, TH: paklitaxel and trastuzumab, TCH: taxol, carboplatin and trastuzumab Survival status Survival outcomes are presented separately for each histological subtype. In our study, median OS was obtained only in the mucinous breast cancer subtype; median OS could not be achieved in other subtypes. The median follow-up duration for the mucinous subtype was 124.6 months, ranging from 5.4 to 267 months. The median OS was 208.4 months (Fig. 1). The 5-year survival rate was 93.7%, while the 10-year survival rate was 78.4%. The median follow-up duration for the metaplastic subtype was 111.6 months, with a range of 9 to 175.2 months. The 5-year survival rate was 89.6%, while the 10-year survival rate was 77.6%. The median follow-up duration for the tubular subtype was 154.6 months, with a range of 110.2 to 298.3 months. The 5-year survival rate was 100%, while the 10-year survival rate was 91.7%. The median follow-up duration for the micropapillary subtype was 161 months, with a range of 2.3 to 238.8 months. The 5-year survival rate was 88.2%, and the 10-year survival rate was also 88.2%. The median follow-up duration for malignant phyllodes tumors was 96 months, with a range of 9.7 to 264.8 months. The survival rate at 5 years is 77.8%, and the survival rate at 10 years is also 77.8%. Fig. 1Kaplan-Meier curve in mucinous breast carcinoma subtype Kaplan-Meier curve in mucinous breast carcinoma subtype Discussion Rare breast cancer subtypes with pure histopathology account for approximately 10% of all invasive breast cancer subtypes [18, 19] Although epithelial carcinomas and malignant phyllodes tumors differ in pathogenesis and treatment approaches, biological and clinical heterogeneity also exists among epithelial carcinoma subtypes. Therefore, each histological entity was evaluated separately, and all survival analyses were performed independently. Mucinous breast carcinoma is a more prevalent subtype compared to other rare breast cancer subtypes [18, 19]and was the most frequently identified in our analysis, comprising 49 patients (41.9%). Mucinous breast cancer is typically more commonly diagnosed in individuals over 65 years of age [13, 20, 21]. The median age at diagnosis in our study was 67 years. Mucinous breast carcinomas are often low grade (grade 1–2) and have hormone receptor positivity and HER2-negative tumors [6, 13, 20, 21]. In our study, all patients with this subtype were in grade 1–2 and in the luminal A or B groups. It is a subtype that is usually detected at an early stage and has a low risk of axillary metastasis [6, 13, 19–22]. The probability of diagnosis in the de novo metastatic stage is quite low [19, 21]. In our study, one patient (2%) was diagnosed at the de novo metastatic stage. The median tumor size was 25 mm clinically and 23 mm pathologically. 87.7% of the patients presented with T1-T2 tumors. Lymph node involvement was identified in ten patients (20.4%) through radiological and clinical assessment, whereas the pathological assessment revealed a rate of 24.5% (n: 12). These data in our study were consistent with the data in the literature [6, 13, 18, 19, 21]. Mucinous breast tumors are amenable to breast-conserving surgery due to their early-stage detection. Research performed from 1990 to 2010 showed that the rates of breast-conserving procedures ranged from 20% to 55% [6, 19, 21, 22]. Our study revealed a rate of 69.4%. The difference may result from advances in surgical methods, improved access to radiotherapy, and improvements in radiotherapy techniques. In our study, 8 patients (16.3%) with mucinous breast carcinoma received adjuvant chemotherapy, and 39 patients (79.6%) received adjuvant radiotherapy. In a large China-based population study by Cao et al. and in a large retrospective series covering 11,400 cases from 1973 to 2002, radiotherapy rates were 25% to 30% [13, 22]. Recent data shows that this rate was approximately 60–90% [6, 12, 19] Vo et al. found adjuvant chemotherapy use rates of 13%, which align with our study; however, our chemotherapy rate is lower than the 40–60% reported in the previous research. In our cohort, chemotherapy administration was not determined by histology alone but was considered in the presence of established high-risk features, including nodal involvement, higher tumor grade, larger tumor size, or unfavorable receptor status. The relatively low chemotherapy rate in our study likely reflects the predominance of low-grade, hormone receptor–positive tumors and the advanced age of many patients in this subgroup.The differences in how chemotherapy and radiotherapy are used now compared to the past may be due to changes in treatment standards, a better understanding of how mucinous carcinoma works, better radiotherapy techniques, and the fact that patients from a more recent time period were included. The median follow-up duration for the mucinous subtype was 124.6 months (range 5.4–267 months), and the median OS was 208.4 months. In our study, 17 of the 49 patients with mucinous carcinoma died, whereas only three patients experienced documented recurrence during follow-up. The difference between mortality and recurrence rates is likely related to the older age and comorbidity profile of this cohort rather than aggressive tumor behavior. The median age at diagnosis was 67 years, and most patients were diagnosed at an older age. The discrepancy between the low recurrence rate and elevated overall mortality in the mucinous group highlights the limitation of overall survival (OS) as a surrogate for oncologic outcomes in elderly populations. Since OS includes both cancer-related and non-cancer-related deaths, it may not fully reflect the true cancer-specific prognosis in biologically favorable subtypes. The survival rate at five years was 93.7%, while at ten years it was 78.4%. The rates observed, while lower than those in larger sample sizes reported in the literature (91.6% and 89%), had a shorter median follow-up in these studies (66 months and 84 months) [13, 19]. On the other hand, another analysis with a median follow-up of 10.6 years reported a 10-year survival rate of 74.5% [12] Longer follow-up, particularly in low-risk tumor subtypes, may lead to higher rates of late-stage mortality. The longer follow-up period in our study may explain the lower survival rate. Metaplastic breast carcinoma, a rare histological subtype, accounts for less than 1% of invasive breast cancers [19]. Metaplastic carcinomas represent a subtype exhibiting aggressive characteristics [9, 23]. These tumors are typically high grade, exhibit a high Ki67 index, present larger sizes (T3-4), show reduced expression of hormonal receptors, and are frequently triple-negative [9, 21, 24]. At the molecular level, metaplastic carcinomas demonstrate marked transcriptomic heterogeneity and are most classified into basal-like or claudin-low intrinsic subtypes according to PAM50-based analyses [25, 26]. The claudin-low subtype, in particular, has been associated with enrichment of epithelial–mesenchymal transition signatures and relative chemoresistance [25, 27]. In histological subtypes exhibiting aggressive clinical behavior, such as metaplastic carcinomas, precision oncology approaches can be critical for identifying driver mutations (e.g., BRCA1, PIK3CA) and selecting targeted therapies specific to these mutations [28]. In our study, 83.4% of patients in the metaplastic carcinoma subgroup were hormone negative, and 66.7% of these were classified as triple negative. The majority exhibited aggressive tumor characteristics, as 83.3% had a Ki67 index greater than 20%, and 90% showed grade 3 disease. Approximately 10% of metaplastic breast cancers show up as de novo metastatic cases [29]. Diagnosis typically occurs at an early or locally advanced stage, and owing to their aggressive characteristics, neoadjuvant therapy is often preferred in treatment approaches [24]. Because they are tumors resistant to conventional chemotherapy and exhibit aggressive characteristics, mastectomy and axillary lymph node dissection rates are high even after neoadjuvant treatments [30–32]. In our study, two patients (6.7%) presented with de novo metastatic disease. However, recurrence occurred in 26.8% (n: 8) of patients during follow-up. Although the clinical lymph node positivity rate in the early stages was 40% and 23.3% received neoadjuvant therapy, the pathological lymph node positivity rate was 43.7%. This outcome supports the literature’s observation that this subtype responds poorly to conventional chemotherapy. Furthermore, the fact that 56.7% of patients in our study underwent mastectomy and 60% underwent axillary lymph node dissection is consistent with the data reported in the literature. The 5-year overall survival for metaplastic breast cancer reported in the literature is 70% to 90% [24, 29, 30] The median follow-up for metaplastic breast carcinoma in our study was 111.6 months, with a range of 9 to 175.2 months. The survival rate after five years was 89,6%, and the survival rate after ten years was 77,6%. These findings are consistent with the literature and provide additional long-term outcome data for this rare entity. Micropapillary breast carcinomas account for less than 2% of all breast cancers [9, 21]. While they are highly ER and PR positive and have luminal A and B biological subtypes, they are generally characterized by larger tumor sizes than other subtypes, high grade, and frequent lymph node involvement [8, 9, 19, 21, 33] Large series report approximately 90% ER/PR positivity, approximately 60–90% luminal A-B subtype, 30–60% grade 3 tumor, and 60–65% lymph node metastasis [19, 21]. The data obtained in our study (88.2% ER/PR positive, 70.6% Luminal A-B biological subtype, radiological and pathological lymph node involvement rates of 59.8% and 64.7%, respectively, and 52.9% grade 3 tumors) were found to be consistent with the literature. The incidence of de novo metastatic disease has been documented in the literature to range from 0.7% to 5% [19, 21, 33, 34]. The finding of this rate of 11.8% (n = 2) in our study may be due to our relatively small sample size. The 5-year survival rates reported for micropapillary breast carcinoma in the literature range from 82% to 97% [8, 21, 34, 35] In our study, the median follow-up for the micropapillary subtype was 161 months (range, 2.3–238.8.3.8 months), a relatively long follow-up period, and both 5- and 10-year survival rates were 88%. Our data, despite the small number of patients in the micropapillary carcinoma subtype, contribute to and support the literature because of the relatively long follow-up period for survival. Tubular carcinomas are rare breast tumors that constitute approximately 0.5–1.5% of breast cancers [18, 21]. They are generally well differentiated, low grade, and favourable prognosis tumors [36–39]. They are smaller in size and less likely to metastasize to lymph nodes [37, 39]. They are often hormone-positive tumors [37, 38]. These are tumors with low rates of mastectomy and axillary lymph node dissection [7, 37]. Consistent with previous studies, all patients in our study with the tubular carcinoma subtype had grade 1–2, luminal A tumors. The median tumor size was 10 mm radiologically and pathologically. All patients were node-negative radiologically and pathologically. Only 8.3% (n: 1) of the patients underwent mastectomy, and 16.7% (n: 2) underwent axillary lymph node dissection. Although all patients in our study were clinically node-negative, the reason for two patients undergoing axillary lymph node dissection may be explained by the fact that sentinel sampling of lymph nodes was not routinely used in clinical practice during the years the patients underwent surgery (patients operated on in 2000 and 2003). Tubular breast carcinomas have long-term survival rates. The 5-year survival rate reported in the literature is between 94% and 100% [7, 21, 38]. In our study, the tubular subtype had a relatively long median follow-up of 154.6 months (range, 110.2–298.3.2.3 months), with a 5-year survival rate of 100% and a 10-year survival rate of 92%. Our findings are consistent with the literature, and the long follow-up period adds descriptive long-term survival data that may complement the existing evidence for this rare subtype. Phyllodes tumors of the breast account for less than 1% of all breast cancers [40] According to the WHO, they are divided into three groups: benign, borderline, and malignant [3]. The incidence of malignant phyllodes tumors has been reported to range from 10% to 44% in large series [40–44]. Phyllodes tumors are generally seen in middle-aged women [41, 43, 45, 46], and the median tumor size is between 3 and 6.5 cm [41, 42, 46, 47]. In our study, the median age at presentation was 43 years, the median tumor size was 44 mm (range, 16–110 mm) radiologically and 65 mm (range, 35–140 mm) pathologically, consistent with the literature. The standard treatment for phyllodes tumors is microscopically margin-negative resection (R0). Adequate surgical margins are considered ≥ 1 cm [48], and positive surgical margins are associated with high recurrence rates [42]. Axillary lymph node dissection is generally not necessary because the risk of axillary lymph node metastasis is low [44, 47, 49]. Breast-conserving surgery appears to be as effective as total mastectomy if adequate surgical margins are obtained and adjuvant radiotherapy is administered [40, 45] In our study, 77.8% of patients (n: 7) underwent breast-conserving surgery. No patient underwent axillary sampling or dissection during their initial surgery. Adjuvant radiotherapy may be beneficial in terms of local control in malignant phyllodes tumors [49–51]. The role of adjuvant chemotherapy is controversial [10, 50, 52]. In our study, two patients underwent re-excision due to positive surgical margins. One patient received adjuvant ifosfamide, mesna, doxorubicin (IMA), and seven patients (77.8%) received adjuvant radiotherapy. Malignant phyllodes tumors of the breast metastasize distantly in 24–34% of cases [40, 52, 53] In our study, two patients (22.2%) experienced recurrence (Patient 1: unresectable chest wall recurrence; Patient 2: lung, liver, and supraclavicular lymphadenopathy recurrence) and received first-line chemotherapy. Both patients died during follow-up. Patient 2 was diagnosed at 34 weeks of gestation. After her initial surgery, she underwent a second surgery due to positive surgical margins and received adjuvant radiotherapy and adjuvant IMA therapy. However, metastasis developed during adjuvant therapy and was treated with gemcitabine-docetaxel. Despite systemic treatment, the disease progressed rapidly, and the patient died within one year of diagnosis. Malignant phyllodes tumors of the breast during pregnancy are extremely rare, and data in the literature are limited to case reports. There are no specific treatment guidelines, and treatment is generally guided by case reports in the literature. In our case, the tumor can progress rapidly and be fatal [54], or it can be monitored for an extended period without progression [55, 56]. Generally, 5-year survival rates reported in the literature for malignant phyllodes tumors of the breast are between 60% and 80%. Long-term outcome data are important because the survival patterns of rare breast cancer subtypes may differ from those of more prevalent invasive ductal carcinoma… long-term follow-up is essential for better characterizing the natural history and survival patterns of these uncommon tumors. Some forms, such as mucinous and tubular carcinomas, usually have favorable early outcomes but may come back later. On the other hand, more aggressive varieties, such as metaplastic or micropapillary carcinomas, are likely to come back sooner. Because these clinical courses are so different and there isn’t a lot of large-scale prospective data, long-term follow-up is essential to have a better idea of how long people live and when they get sick again in these unusual cases. This study has limitations in several aspects. First, the retrospective and single-center design of the study may lead to selection bias and make the generalizability of the results difficult. Second, although the total cohort size was reasonable because the tumors we considered in our study were rare, the histological subgroups were still quite small when considered separately. Limited sample sizes prevented statistical comparisons between subtypes and multivariate survival analyses; therefore, all findings should be interpreted as descriptive and hypothesis-forming rather than definitive. Third, the inclusion of epithelial carcinoma and malignant phyllodes tumors may affect comparative interpretation due to their histological and biological differences; therefore, survival analyses were performed separately for each subtype. Fourth, over the long study period (2000–2024), treatment methods improved. For example, systemic therapies and radiotherapy techniques improved, which may have altered survival outcomes. Also, statistics on survival and cause of death for cancer patients were not always available, so overall survival was chosen as the primary target. This may help explain the differences in recurrence and mortality rates, especially in older groups with mucinous carcinoma. Finally, molecular profiling data were not available to all members of the group, which made it difficult to understand the biology in more detail. Conclusions This study reveals the clinical, pathological, and survival characteristics of rare histological subtypes of breast cancer obtained from a tertiary healthcare setting. While mucinous and tubular carcinomas generally follow a good prognosis with low grade and strong hormone receptor positivity, metaplastic and micropapillary carcinomas are characterized by more aggressive biological behaviour, a higher risk of recurrence, and the need for more intensive treatment. These findings are consistent with existing literature; however, the small number of patients in the subgroups limits the strength of the results, and statistical comparisons between subtypes were not performed. Our findings may contribute to the current literature and support the importance of accurate pathological classification in the clinical management of these rare tumors. Our findings may contribute to the existing literature and support the importance of accurate pathological classification in the clinical management of these rare tumors. Larger, multicenter, prospective studies are needed to confirm these results and improve treatment approaches for rare breast cancer subtypes.