Abstract
Background/Aim: It is possible that the degree of enhancement on contrast-enhanced spectral mammography (CESM), a new diagnostic method, might provide prognostic information for breast cancer patients. Therefore, in a group of 82 breast cancer patients, we analyzed the prognostic significance of degree and pattern of enhancement on CESM as well as its relation to: (a) breast cancer immunophenotype (based on ER/PR/HER2 status) (b) podoplanin expression in cancer stroma (lymphatic vessel density plus podoplanin-positivity of cancer-associated fibroblasts), and (c) other histological parameters. Materials and Methods: For each tumor the intensity of enhancement on CESM was qualitatively assessed as strong or weak/medium, while the pattern – as homogenous and heterogenous. Results: Herein we report, for the first time, that strong and heterogenous enhancement on CESM was related to unfavorable disease-free survival of breast cancer patients (p=0.005). Moreover, the strong enhancement was more frequent in large and node-positive tumors (pT>1, pN>0) (p=0.002), as well as in carcinomas with podoplanin-sparse stroma (p=0.008). Conclusion: Intensity and pattern of enhancement on CESM might provide (together with the results of other diagnostic imaging methods) not only the confirmation of presence or absence of tumor, but also prognostic information.
- Contrast-enhanced spectral mammography (CESM)
- prognostic significance
- podoplanin-positive cancer stroma
- breast cancer immunophenotype
Contrast-enhanced spectral mammography (CESM), also known as dual-energy mammography is a relatively new breast imaging technique, which combines full-field digital mammography with the intravenous injection of an iodinated contrast medium (1, 2). This technique allows for imaging and at the same time detection of tumor neoangiogenesis as a classic sign of malignant tumors as well as for the detection of other breast lesions (1, 3). It is possible to achieve more accurate assessment of tumor vascularity using CESM than Doppler ultrasound examination (4). CESM is used in establishing a diagnosis in patients with suspicious breast tumors where basic mammography and ultrasound examinations are not sufficient and it is capable of detecting cancers that are not visible in standard mammography setting (2, 5). CESM sensitivity is significantly higher in comparison to mammography (MG) (2, 6, 7). It is particularly useful for women with dense breasts when tumor detection based on MG is more difficult. However, the number of patients examined with CESM is still relatively small. There are several reasons of that situation. Firstly, CESM radiation dose per patient is about 20% higher than in conventional MG. The second reason is relatively small number of installations and lack of application experience.
It is worth mentioning, that CESM was approved by FDA in 2011 and since then it has been included into diagnostics in the Centre of Oncology Cracow Branch. It was the first installation in Poland and in Eastern Europe.
The differences in the intensity and pattern of enhancement between breast carcinomas might result from the differences in the amount of contrast that leaked out from the blood vessels (capillary filtration) and timely arrested in the interstitium. The above-mentioned retention of contrast might be a consequence of impaired structure of tumor blood vessels (the mechanism of permeability edema) and/or dysfunction of lymphatic vessels (the mechanism of lymphedema). The latter, on the other hand, might result from the following factors: (1) physical obstruction of lymphatics exerted by growing tumors or lymphatic vessel invasion, (2) destruction of the existing lymphatics or lack of tumor lymphangiogenesis, (3) collapse of lymphatic vessels, which prevents their filling with tissue fluid, which is caused by disruption (proteolytic degradation during inflammation or tumor growth) of filaments anchoring lymphatic endothelium to extracellular matrix proteins. In case of lymphatic vessel obstruction transcapillary filtration continues until interstitial pressure rises to equal net filtration pressure and interstitial colloid osmotic pressure equals plasma colloid osmotic pressure (8).
The above-mentioned description is confirmed by our previous report (9). In this study, in the group of breast lesions (benign lesions and carcinomas), strong enhancement on CESM was related to low density of lymphatic vessels, assessed on formalin-fixed paraffin embedded tissue sections using podoplanin as a marker. In breast tissue and breast carcinomas expression of the aforementioned marker, has been observed not only in lymphatic vessels but also in myoepithelial cells surrounding normal breast ducts/in situ carcinomas and in cancer associated fibroblasts (10-14). Density of podoplanin-positive lymphatic vessels and frequency of cancer-associated fibroblasts (CAFs) with podoplanin expression was found to be correlated both to molecular breast cancer subtype (15) and to survival of breast cancer patients (10-14). Therefore, we decided to study the expression of podoplanin in cancer stroma (lymphatic vessels and CAFs) and its relation to both breast cancer immunophenotype and survival of breast cancer patients. The present study is the first one addressing this issue.
To the best of our knowledge, prognostication based on CESM results has not been reported so far. Moreover, little is known about the relationship between CESM results and histological/immunophenotypic characteristics of breast cancer. Our team is actually the first one who undertook the above mentioned challenge (9).
Nevertheless, the idea of prognostication based on radiological features is not a new one. Published papers showed the influence of selected parameters of Contrast-Enhanced Magnetic Resonance on patients' survival rate (16-18). Additionally, some studies reported the relationship between selected features on MG, ultrasonography or MRI and breast cancer immunophenotype or density of blood vessels (19-25).
We believe that in the future sensitivity of imaging methods will be equal to 100% (sensitivity of CESM is close to this value ranging from 72-100% (9, 26-28) and decoding tumor phenotype and prognostication/prediction based on their results will be possible (such approach is called radiomics). As a consequence, it might result in excluding the necessity of performing core needle biopsy for diagnostics and prognostication.
Therefore, in a group of 82 breast cancer patients, we decided to analyze the relationship between the intensity and the pattern of enhancement on CESM and: (i) disease-free survival of breast cancer patients (ii) breast cancer subtype (based on ER/PR/HER2 status), (iii) podoplanin positivity of cancer stroma and (iv) other histological parameters. To the best of our knowledge this type of analysis is being performed for the first time.
Materials and Methods
Patients. CESM was applied to patients with glandular breast structure reported in the previously performed MG or with a lesion requiring further diagnostic procedures. Patients with lesions enhancing on CESM (Figure 1 b, d, f, h) were subjected to core biopsy (in case of lesion >1 cm) or vacuum-assisted core biopsy (in case of lesion ≤1 cm) guided by ultrasonography or MG. Material obtained during biopsies was histopathologically verified. In the present study we retrospectively investigated 84 invasive breast carcinomas diagnosed in 82 patients (in 2 patents two separate tumors were found) between 2011 and 2012. Mean patients' age was 57.2±11.5 (SD). The clinical, histological, immunophenotypic and treatment characteristics of patients and their tumors are depicted in Table I.
This study was performed in compliance with the Declaration of Helsinki and it received the approval of Ethical Committee at the Regional Medical Chamber in Krakow (decision from 15.10.2013).
All patients received standard therapy according to individual indications based on clinical, histological and immunophenotypic features(15). The surgery was performed as first fundamental therapy (Table I). Adjuvant therapy was applied according to presence of histological and immunophenotypic predictive factors. Radiotherapy was applied in 70 patients: after breast-conserving surgery (56 cases) as a standard of breast-conserving therapy, and in 14 patients after radical mastectomy (lymph-node positive patients and individuals with high grade tumors). The indications for chemotherapy included: positive lymph nodes, high proliferation index, poor histological tumor differentiation and human epidermal growth factor receptor 2 (HER2) overexpression or triple negative immunophenotype (15). Hormonotherapy (53 patients) and targeted anti-HER (21 patients) were applied in patients with immunopositivity of estrogen/progesterone receptors (ER, PR) and overexpression of HER2, respectively.
Methodology
Contrast-Enhanced Spectral Mammography (CESM). Digital mammography device (GE Healthcare, SenoBright, Chalfont St-Giles, UK) was used for all CESM examinations, which were performed with dual-energy CESM acquisitions. An intravenous injection of non-ionic contrast agent (Ultravist 370, 1.5 ml/kg of body weight) was given using a power injector (Covidien, Optistar™ Elite Injector Cincinnati, US) to each patient before the exposure, at a rate of 3 ml/s with a bolus chaser of saline. After the initiation of contrast agent administration (2 and 4 min, for mediolateral oblique view (MLO) and craniocaudal view (CC), respectively), a pair of exposures (low- and high-energy) in each view (MLO and CC) were performed automatically. Combination of low-energy and high-energy images was done using a proper image processing, allowing to generate subtracted images with contrast agent uptake information in each view.
Example images of mammography (MG): a, e – of the left breast in mediolateral oblique (MLO) projection; c, g – of the left breast in craniocaudal (CC) projection; and contrast-enhanced spectral mammography (CESM): b, d (MLO and CC respectively)-weak/medium/heterogenous enhancement, f, h (MLO and CC respectively)– strong/homogenous enhancement.
All images obtained on CESM were evaluated by three radiologists, who were blind to patients' history. For each tumor enhancing on CESM, the intensity and the pattern of enhancement were evaluated. The architecture distortion was visualized (Figure 1a, c, g, e). The enhancement of contrast agent uptake was qualitatively assessed as weak/medium (Figure 1b, d) or strong (Figure 1f, h), while the pattern as heterogeneous (Figure 1b, d) or homogenous (Figure 1f, h).
Immunohistochemistry
Immunohistochemistry (IHC) staining. Details of immunohistochemistry are depicted in Table II. In double-staining procedure, after application of the visualization system (Table II) VIP (violet color) was used as peroxidase substrate for visualization of CD34 (Figure 2a-c: white arrow), while DAB (brown color) for podoplanin (Figure 2d-f: black arrow, 2g-i: asterisk). Eventually, slides were counterstained with Mayer's hematoxylin.
Immunohistochemistry (IHC) evaluation. Lymphatic vessels were defined as strongly podoplanin-stained structures with lymphatic vessel characteristics (with or without visible lumen), clearly distinguishable from other tissue structures and cells.
For the assessment of the distribution of podoplanin-positive lymphatic vessels (DPV) the whole specimen was examined (over 4 microscopic fields located in intratumoral, peripheral and peritumoral area). Based on the percentage or absence of lymphatics, each field was classified as positive (with at least one podoplanin-positive vessel) or negative (without lymphatics). Finally, the percentage of fields with at least one lymphatic vessel was calculated. DPV ≤18% (median value of DPV) was considered as low, while DPV >18% as high.
Clinical characteristic for 82 breast cancer patients.
Podoplanin-rich stromal cancer-associated fibroblasts (CAFs) were defined as spindle-shaped stromal cells with indistinct cell borders, presenting strong podoplanin expression (slightly weaker or equal to podoplanin expression in lymphatic endothelium), found in more than 50% of overall stromal area (Figure 2g, h, asterisk). Lack of podoplanin expression in tumor stroma or podoplanin CAFs positivity in ≤50% of overall stromal area was specified as podoplanin-sparse CAFs (Figure 2a-f).
Based on the above-mentioned parameters we classified tumor stroma as: (1) podoplanin-sparse (with low DPV and podoplanin-sparse CAFs) and (2) podoplanin-rich (with high DPV or podoplanin-rich CAFs).
Ki-67 labelling index (Ki-67LI) was assessed in over 1000 cancer cells and expressed as percentage Ki-67 nuclear immunopositivity.
Evaluation of estrogen/progesterone receptor and human epidermal growth factor receptor 2 expression. For the purposes of the present study, statuses of ER/PR receptor and HER2 expression, which have been established during routine diagnostic procedures, were retrieved from patients' files. ER or PR receptors were considered as immunopositive if nuclear staining was present in >1% of tumor cells (14, 15). Overexpression of HER2 was tested using HercepTest (Dako Denmark A/S, Glostrup Denmark). In case of an unclear result (expression assessed as 2+), amplification of HER2 gene was verified using fluorescence in situ hybridization (FISH) – PathVysion HER2 DNA Probe (Abbot Molecular) and overexpression of HER2 protein/amplification of HER2 gene was evaluated according to the standards recommended at that time (15).
Based on ER/PR/HER2 expression we defined four immunophenotypes: Luminal A subtype (expression of ER or PR and HER2 immunonegativity); luminal B (ER/PR immunopositivity and HER2 overexpression), HER2-overexpressing (ER/PR immuno-negativity and HER2 overexpression), triple-negative phenotype (ER/PR/HER2 immunonegativity).
Statistical analysis. The STATISTICA v.12 software (StatSoft, Inc. Tulsa, OK, USA) was used for all calculations. p-Value <0.05 was considered significant. Fisher test (two rows and two columns) or Pearson χ2 test (more than two columns) was applied for the assessment of independence between two categorical variables.
Kaplan-Meier method was used for estimation of cumulative survival probabilities while log-rank test for evaluation of differences between survival rates. In the present study we analyzed disease-free survival (DFS), which was defined as the time (number of months) from surgery to the occurrence of breast cancer failure (distant metastases and/or local recurrence). The joint effect of covariates, which were significant in the univariate analysis were analyzed using Cox proportional hazard model with stepwise regression procedure.
Results
Results of CESM. Weak/medium enhancement on CESM was observed in 36 tumors (Figure 1b, d), while strong in 48 (Figure 1f, h). Independently, 20 homogenous (Figure 1f, h) and 64 heterogeneous patterns (Figure 1b, d) of enhancement on CESM were noted.
Immunohistochemical visualization of podoplanin-positive lymphatic vessels (brown color, microphotograph d-f) and CD34-positive blood vessels (violet color, microphotograph a-i), and podoplanin immunopositivity of cancer-associated fibroblasts: CAFs (brown color, microphotograph g-i and e - weak immunostaining). a-c – CD34-expressing blood vessels (white arrow) without visible podoplanin-positive lymphatic vessels; d-f – CD34-expressing blood vessels and podoplanin-positive lymphatic vessels (black arrow), e – weak CAFs immunostaining; g-i - podoplanin-immunopositive CAFs (asterisk): g-h – strong immunopositivity, i - medium immunopositivity.
Results of immunohistochemistry: distribution of podoplanin-positive lymphatic vessels (DPV), podoplanin expression in stromal cancer-associated fibroblasts (CAFs) and Ki-67 labelling index. The expression of podoplanin was assessed in 69 breast carcinomas, in intratumoral, peripheral and peritumoral areas. Podoplanin-positive lymphatic vessels were present in 45 cases (Figure 2d-f), while they were not found in the remaining 24 (46,6%) cases (Figure 2a-c; g-i). Mean value of DPV was 28.7%±33.3 (SD), while median value – 18%. There were 36 carcinomas with low DPV (DPV≤18%) and 33 with high DPV (DPV>18%).
Details of immunohistochemistry.
Podoplanin-rich CAFs were observed in 8 out of 69 (11.6%) analyzed carcinomas (Figure 2g-i), while in 61 cases podoplanin-sparse CAFs were found (Figure 2a-f).
Finally, we found podoplanin-rich cancer stroma (DPV+ podoplanin-rich CAFs) in 36 tumors (47.8%), while podoplanin-sparse stroma – in 33 carcinomas (52.2%).
Ki-67LI was assessed in 66 cases. Mean value of Ki-67LI was 28.8%±19.0 (SD), while median was 26.6%. In the following analyses the cut-off value for Ki-67LI was set at the mean value 28%.
Relationship between podoplanin expression in stromal cells and histological parameters. Podoplanin-sparse stroma was observed significantly more frequently in luminal A breast cancer subtype (p=0.004, Figure 3e). Moreover, podoplanin-rich stroma was insignificantly more frequent in: pT1 tumor stage (p=0.166, Figure 3a) and in carcinomas with high proliferative potential (p=0.090, Figure 3f, high Ki-67LI). No relationship (p>0.05) was found between expression of podoplanin in cancer stroma and pN (Figure 3b), tumor grade (Figure 3c) and histological type of breast cancer, (Figure 3d).
Relationship between the results of CESM and histological/biological parameters. Strong enhancement on CESM was found more frequently in: (i) large tumors (pT>1) (p=0.002, Figure 4a), (ii) node-positive carcinomas (p=0.002, Figure 4b), (iii) in tumors with podoplanin-sparse stroma vs. tumors with podoplanin-rich stroma (p=0.008, Figure 4f).
We found no relationship (p>0.05) between enhancement on CESM and: tumor grade (Figure 4c), histological type of cancer (Figure 4d), breast cancer immunophenotype (Figure 4e) and Ki-67LI (Figure 4g). However, in luminal A tumors strong enhancement on CESM was insignificantly more frequent as compared to neoplasms with non-luminal A subtype (p=0.241, Figure 4e).
There was no correlation between the pattern of enhancement on CESM and other studied parameters (p>0.05).
Survival analysis. In the analyzed group the follow-up time ranged between 1 and 60 months with median value 50 months. During this period four patients developed distant metastases, while 2 patients – metastases and local recurrences. The 5-year disease-free survival was estimated as 87.0%.
The univariate analysis revealed poorer survival rate in patients with carcinomas presenting: (a) strong and heterogenous enhancement on CESM (vs. weak/medium or homogenous/ring-like pattern enhancement on CESM; p=0.005, Figure 5a), (b) Ki-67LI>28% (vs. Ki-67LI≤28%; p=0.029, Figure 5b) and (c) pT2-4 tumors (vs. pT1; p=0.008, Figure 5c). Other analyzed parameters (pN, tumor grade, histological type of the tumor and expression of podoplanin in tumor stroma) did not influence patient's survival rate significantly (p>0.05). Cox multivariate analysis with stepwise regression procedure revealed a heterogenous pattern and strong enhancement on CESM and Ki-67LI>28% as significant independent negative prognostic factors (p=0.063, RR (relative risk)=4.9; 0.9≤95%CI (confidence interval)<26.6 and p=0.077, RR=8.7; 0.8≤95%CI<95.3), respectively; p-value at significance border).
Bar graphs presenting relation between podoplanin expression in cancer stroma and: a – pT; b – pN; c – tumor grade; d – histological type of the tumor; e – breast cancer subtype; f – Ki-67LI. p-values presented on the plots are from Fisher test or Pearson χ2 test. Luminal A subtype: Expression of estrogen receptor (ER) or progesterone receptor (PR) and immunonegativity of human epidermal growth factor receptor 2 (HER2); luminal B subtype: ER/PR immunopositivity and HER2 overexpression; HER2: ER/PR immunonegativity and HER2 overexpression and triple-negative phenotype: ER/PR/HER2 immunonegativity; podoplanin-sparse tumors: low (≤18%) percentage of fields with podoplanin-positive lymphatic vessels (DPV) and sparse podoplanin expression in cancer associated fibroblasts (CAFs); podoplanin-rich tumors: high DPV or podoplanin-rich CAFs; number in/over each bar represents number of cases.
Bar graphs presenting relation between intensity of enhancement on CESM and: a – pT; b – pN; c – tumor grade; d – histological type of the tumor; e – breast cancer subtype; f – podoplanin expression in breast cancer stroma; g - Ki-67LI. p-values presented on the plots are from Fisher test or Pearson χ2 test. All abbreviations are defined in the legend of Figure 3.
Disease-free survival of patients with invasive ductal breast cancer stratified by: a) enhancement on CESM, b) Ki-67 expression, c) pT stage.
Discussion
In our study prognostic significance of selected CESM features was found for the first time: strong and heterogenous enhancement on CESM was related to poor patients' outcome. In this study, the aforementioned correlation was additionally confirmed by the relationship between strong enhancement on CESM and nodal involvement or large tumor size.
As mentioned before prognostication based on CESM features has not been reported so far. However, previous reports have proved that selected features of pretreatment images obtained using Contrast-Enhanced Magnetic Resonance, provide significant independent prognostic information (together with well-known prognostic/predictive markers) (16, 18). Among prognostic parameters (influencing both overall and disease-free survival) assessed using Contrast-Enhanced Magnetic Resonance there were: tumor longest dimension, tumor volume, vascular kinetics (maximum enhancement index, rate of enhancement, AUC, signal enhancement ratio), texture, and shape-based metrics (16, 17). In previous reports, also, the rim enhancement showed significant correlation with shorter survival and poor prognosis in patients with triple-negative breast cancer (29). As reported previously, in breast cancer patients, some enhancement parameters (background parenchymal enhancement, its coefficient and perfusion parameters) significantly correlated with poor prognostic factors (30).
The other factors, which influenced patient's survival in the present study - Ki-67LI and tumor size – are well known prognostic factors (15).
In the present study for the first time we found the relationship between strong enhancement on CESM and sparse stromal podoplanin expression, which was defined as low lymphatic vessel density and podoplanin-sparse CAFs (enhancement pattern was not related to expression of podoplanin in cancer stroma). On the other hand, in carcinomas with medium/weak enhancement, podoplanin-rich stroma was more frequent than podoplanin-sparse stroma. One of the logical explanations of this phenomenon is that low number of stromal lymphatics may result in impaired removal of contrast from interstitium, which is observed on CESM as strong enhancement (8, 9). However, the Authors are aware of the fact that not only the density of lymphatic and blood vessels are responsible for stromal contrast retention. There is a large portion of additional biological factors, related to cancer stroma remodeling, which are related to tumor progression and additionally might be associated with degree/pattern of enhancement on CESM. For example, destruction of extracellular matrix (ECM) induced by inflammation or cancer growth, might increase interstitial fluid pressure and prevent movement of fluid (and contrast) from interstitium to lymphatic or blood vessels. Another potential mechanism is overproduction of collagen, and mucopolysaccharides by cancer-associated fibroblasts (CAFs) (which might be observed in selected carcinomas), which might increase the density and volume of ECM and therefore may produce a highly negative interstitial fluid pressure (myxedema). The above-mentioned mechanisms might be interesting aims of future translational research.
Interestingly, in our study luminal A subtype and low Ki-67LI were more frequent in tumors with podoplanin sparse stroma, that is in accordance with previous reports, in which low number of lymphatics or low frequency of CAFs was observed more frequently in steroid receptor-positive carcinomas (9-13).
In the previous reports a relationship was found between MRI parameters and blood vessel density (22-25), confirming MRI as a method for angiogenesis assessment, however, to the best of our knowledge, lymphangiogenesis and stromal podoplanin expression were not compared to MRI results previously.
Additionally, we herein report an insignificant relationship between enhancement on CESM and molecular breast cancer subtype – in luminal A breast cancers strong enhancement on CESM was insignificantly more frequent. In some sense, this confirms other authors reports in which triple-negative immunophenotype (which is known as a biologically aggressive subtype) presents a few radiological features characteristic for benign lesions. It was described as a very hypoechogenic, lobulated mass with indistinct or microlobulated margins, with an abrupt interface, sometimes pseudo-benign, while on MRI presenting distinct demarcation, regular edges, hyperintensity on T2-weighted signals and, importantly, a crown-shaped enhancement (19-21).
We are aware that the number of patients included in the present study is relatively small. This is a consequence of the following facts: (1) CESM is not routine a screening method for breast cancer, (2) in CESM radiation dose per patient is about 20% higher than in conventional MG (3) it is a new diagnostic method (it was approved by FDA in 2011 and since 2011has been included in Centre of Oncology Cracow Branch) and therefore the number of patients with follow-up sufficient for survival analysis is still small. It is worth mentioning that about 3-4% of patients diagnosed with breast cancer were examined with CESM in 2011-2015 in our Institution.
Our results need confirmation in a larger patients group. Additionally, we will try to make a prognostic algorithm based on CESM and other imaging methods and try to answer in a more detailed way (based on biological characteristics of examined tumors), why studied carcinomas present different patterns and intensity of enhancement on CESM.
Conclusion
Our results may suggest that intensity and pattern of enhancement on CESM might bring (together with the results of diagnostic imaging methods) not only the confirmation of presence or absence of tumor, but also prognostic information.
Footnotes
Conflicts of Interest
The Authors have declared that no competing interest exists.
- Received November 17, 2017.
- Revision received December 16, 2017.
- Accepted December 18, 2017.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.