You are here

Article Text

Article menu
Original article
Microvascular proliferation in luminal A and basal-like breast cancer subtypes
  1. Maria Ryssdal Kraby1,
  2. Kristi Krüger2,
  3. Signe Opdahl3,4,
  4. Lars J Vatten4,
  5. Lars A Akslen2,5,
  6. Anna M Bofin1
  1. 1Department of Laboratory Medicine, Children's and Women's Health, Norwegian University of Science and Technology, Trondheim, Norway
  2. 2Department of Clinical Medicine, Centre for Cancer Biomarkers, University of Bergen, Bergen, Norway
  3. 3Central Norway Regional Health Authority, Trondheim, Norway
  4. 4Department of Public Health and General Practice, Norwegian University of Science and Technology, Trondheim, Norway
  5. 5Department of Pathology, Haukeland University Hospital, Bergen, Norway
  1. Correspondence to Maria Ryssdal Kraby, Department of Laboratory Medicine, Children's and Women's Health, Norwegian University of Science and Technology, Erling Skjalgssons gate, Trondheim 7030, Norway; kraby{at}stud.ntnu.no

Abstract

Aims The aims of this study were to examine microvessel density (MVD), proliferating MVD (pMVD) and Vascular Proliferation Index (VPI) in basal-like phenotype (BP) and luminal A subtypes of breast cancer and to study their prognostic value.

Methods Dual-colour immunohistochemistry for von Willebrand factor and Ki67 was done on sections from 62 luminal A and 62 BP tumours matched for grade and selected from 909 breast cancers previously reclassified into molecular subtypes. Associations between MVD, pMVD and VPI, molecular subtypes and breast cancer prognosis were estimated using linear regression and survival analyses.

Results Both pMVD (difference 1.9 microvessels/mm2 (p=0.002)) and VPI (difference 1.7 percentage points (p=0.014)) were higher in BP tumours compared with luminal A. No clear difference between subtypes was found for MVD. However, only MVD was associated with prognosis. HR for breast cancer death for all cases was 1.10 (95% CI 1.02 to 1.18) per 10 vessels increase. Among luminal A tumours, HR was 1.22 per 10 vessels increase (p<0.001) and in BP it was 1.04 (p=0.37).

Conclusions High MVD was associated with poor prognosis in luminal A, but not in BP cancers. Vascular proliferation was higher in BP, indicating a more active angiogenesis than in luminal A tumours. The luminal A subgroup comprised mostly histopathological grade 3 cancers in this selected series, and further studies are needed to clarify whether MVD provides additional prognostic information for luminal A tumours irrespective of grade. This may contribute to stratification of this large group of patients and may aid in identifying tumours with a particularly good prognosis.

  • ANGIOGENESIS
  • BREAST CANCER
  • IMMUNOHISTOCHEMISTRY
  • BREAST PATHOLOGY
  • CANCER RESEARCH

http://dx.doi.org/10.1136/jclinpath-2015-203037

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Angiogenesis is a necessity for tumour growth.1 Without it, tumour size does not exceed 1–2 mm3.1–4 After vascularisation, tumour growth rate quickly increases and becomes exponential.2 ,4 A quantitative measure of angiogenesis within a tumour could therefore provide important prognostic information in patients with cancer.

    Microvessel density (MVD) is an established method for quantifying the number of vessels within a tumour. However, the method does not reflect ongoing angiogenesis.5 Proliferating MVD (pMVD) and Vascular Proliferation Index (VPI) describe endothelial cell proliferation within the tumour and may provide better estimates of ongoing angiogenesis.6–10 Recent studies have indicated that VPI is an independent prognostic factor in endometrial cancer,6 prostate cancer10 and breast cancer.8 ,9 Furthermore, high VPI has been associated with aggressive breast cancer features and with the basal-like phenotype (BP).7–9

    Gene expression studies have led to the identification of five biologically distinct breast cancer subtypes: luminal A, luminal B, BP, human epidermal growth factor receptor 2 (HER2) overexpressing and normal breast-like.11 ,12 These subtypes are characterised by different clinical outcomes. While the luminal A subtype has the best prognosis, the poorest outcomes are found in the BP and HER2 overexpressing subtypes.13

    Immunohistochemical (IHC) and in situ hybridisation (ISH) biomarkers have been used as surrogates for gene expression analysis in studies of breast cancer subtypes.7 ,9 ,14–21 Although not capable of identifying the intrinsic subtypes with complete consistency,22 ,23 IHC and ISH markers provide prognostic information similar to that of gene expression analyses.17 ,19 ,24 This is now an established approach in studies of large patient cohorts.17 ,19 ,21 ,24

    The aims of this study were to compare MVD, pMVD and VPI in the luminal A and BP breast cancer subtypes and to study the effect of these factors on prognosis in patients with breast cancer with long-term follow-up.

    Materials and methods

    Study population

    Between 1956 and 1959, 25 897 women born between 1886 and 1928 in Nord-Trøndelag county, Norway, were invited to participate in a breast cancer survey organised by the Norwegian Cancer Society. Through information linkage with the Cancer Registry of Norway, these women were followed up for breast cancer occurrence from 1961 to 2008. Information on date and cause of death was available from the Cause of Death Registry.25–27 A total of 1393 primary breast cancers were diagnosed during the follow-up period. Of these, 909 tumours were retrieved from the archives of the Department of Pathology and Medical Genetics, St. Olav's Hospital, Trondheim, and reclassified into molecular breast cancer subtypes.21 Sixty-three tumours were classified as BP and 433 tumours were classified as luminal A. Of the BP tumours, 62 were suitable for inclusion in the present study. A luminal A case was selected from the same patient cohort for each BP case, matched for tumour grade if possible. Among the BP tumours, 52 were grade 3, 7 were grade 2 and 3 were grade 1. There were 42 grade 3, 17 grade 2 and 3 grade 1 luminal A tumours. The average grade for luminal A tumours was 2.6, compared with 2.8 for BP cancers.

    Specimen characteristics

    All cases were classified according to histopathological type and grade, and then reclassified into molecular subtypes using IHC and ISH as surrogates for gene expression analyses. Full-face sections were cut at 4 µm from formalin-fixed, paraffin-embedded tissue blocks, stained with haematoxylin-erythrosine-saffron (HES). Classification into histopathological type and grade was done by two experienced pathologists, according to WHO Classification of Tumours of the Breast28 and the Nottingham Grading System.29 ,30 In cases of disagreement, consensus was reached after discussion. Next, three 1 mm in diameter tissue cores were extracted from peripheral areas of each tumour and assembled in tissue microarrays (TMA). IHC staining was done on sections from the TMA blocks for the following markers: oestrogen receptor (ER), progesterone receptor (PR), Ki67, HER2, cytokeratin 5 and epithelial growth factor receptor 1. In addition, HER2 gene copy number status was estimated using chromogenic ISH. Tumours were reclassified into molecular subtypes according to the algorithm in figure 1.21

    Figure 1
    Figure 1

    Classification algorithm for molecular subtyping (Engstrom et al21). CK5, cytokeratin 5; ER, oestrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2.

    IHC staining

    For the present study, 4 µm full-face sections were retrieved from the freezer at −20°C. Pretreatment was carried out in a Pre-Treatment Link (DAKO). The sections were immersed in Dako Target Retrieval Solution buffer pH 6 (DAKO). The temperature of the solution was raised to 97°C and maintained for 20 min before cooling. Immunostaining was performed using Dako Autostainer Plus (DAKO). The process commenced with 8 min of enzyme blocking using Dual Endogenous Enzyme Block (DAKO). Primary antibodies, rabbit von Willebrand factor antibody (3.8 µg/L) (Polyclonal rabbit, A0082, Dako) and mouse Ki67 antibody (160 µg/L) (Clone MIB1, M7240, Dako), were mixed with Dako antibody diluent (DAKO) and incubated for 60 min at room temperature. Secondary antibodies, Southern Biotech alkaline phosphatase/goat anti-mouse and Dako EnVision Detection System-Peroxidase/rabbit, diluted 1:100, were incubated for 30 min (room temperature). Ki67 was visualised by applying Ferangi Blue (BIOCARE medical) for 15 min. After rinsing with dH2O, amino-ethyl-carbasol substrate chromogen (Dako) was applied for 15 min to visualise von Willebrand factor. Coverslipping with Dako Faramount aqueous medium (DAKO), von Willebrand factor positive cells showed reddish-brown cytoplasmic staining and Ki67 positive nuclei stained blue as seen in figure 2.

    Figure 2
    Figure 2

    Breast cancer section at ×400 stained with immunohistochemistry. von Willebrand factor positive cells display reddish-brown cytoplasm and Ki67 positive cells display blue nuclei.

    Scoring and reporting

    The most vascularised areas of the tumours were identified in the IHC-stained sections by two of the authors (MRK and AMB) prior to vessel counting. The number of microvessels was then counted in 10 high-power fields (×400) within the preselected areas. Each field comprised at least 50% tumour tissue. Sclerotic areas, fibrotic scars and areas close to normal breast tissue or necrosis were avoided. A countable microvessel was defined as a von Willebrand factor positive endothelial cell or cell cluster. A visible lumen was not necessary to fulfil the requirements of a vessel. However, in areas with long, twisted branches of endothelium or glomeruloid microvascular proliferations (vascular nests), each lumen was counted as a separate vascular unit.

    Proliferating microvessels were counted in the same 10 fields. A proliferating vessel was defined as a microvessel containing at least one endothelial cell nucleus with positive Ki67 staining. MVD and pMVD values were given as number of vessels per square millimeter, estimated as the average number of vessels per visual field divided by the visual field area of the microscope. VPI was defined as the ratio between pMVD and MVD, given in per cent.

    All cases were assessed by one observer (MRK) who had undergone a period of training prior to the study. Using a test series of 24 colon cancer sections, intraobserver and interobserver (MRK and KK) agreement were estimated using Kappa (κ) and Spearman's rho (ρ). Training was ended when the desired level of agreement, consistent over time, was achieved (κ>0.6 and ρ>0.8). This study complies with the REMARK reporting recommendations for tumour marker studies.31

    Statistical analyses

    Linear regression was used to estimate the association between breast cancer subtype and MVD, pMVD and VPI. Age (<65, 65–79, ≥80 years), grade (1, 2 and 3) and stage (I, II and III/IV) at diagnosis were adjusted for. Breast cancer prognosis was assessed using survival analyses, where patients were followed up until death from breast cancer, death from other causes or until 31 December 2010, whichever came first.21 Using Kaplan–Meier plots with log-rank tests, breast cancer specific survival (BCSS) was compared for patients with MVD, pMVD and VPI below and above the median value. Cox proportional hazards models were used to estimate risk of death from breast cancer per unit increase in MVD, pMVD and VPI, as well as for patients with values below and above the median value. HRs were calculated with 95% CIs adjusting for age, stage, grade and subtype (if applicable). Survival analyses were performed for all cases combined and for the two subtypes separately.

    To evaluate the robustness of the results, the analyses were repeated with adjustment for time period of diagnosis (10-year categories), narrower age categories, restriction to grade 3 tumours and with the cut-off value set at the 75th percentile for MVD, pMVD and VPI instead of the median. Proportionality between hazards was checked by comparing log minus log plots of survival and by performing tests based on Schoenfeld residuals, and assumptions were met. All analyses were performed using Stata V.13.1 (StataCorp LP, College Station, Texas, USA).

    Results

    Characteristics of the women and their tumours are presented in table 1. A total of 38.7% of the patients with luminal A tumours and 45.2% of the patients with BP tumours died from breast cancer during follow-up. The median age at diagnosis was 74 years for luminal A cases and 72 years for BP cases.

    View this table:
    Table 1

    Descriptive characteristics for the 124 breast cancer cases

    For all tumours combined, the median MVD was 66.9 microvessels/mm2 (IQR 49.6–94.6). The median pMVD was 1.7 microvessels/mm2 (IQR 0.6–3.5). The median VPI was 2.3% (IQR 1.1–5.9). The scatterplot in figure 3 shows the distribution of MVD and pMVD in luminal A and BP tumours.

    Figure 3
    Figure 3

    Distribution of luminal A and basal-like phenotype subtypes according to microvessel density (MVD) and proliferating MVD (pMVD). Predicted values were estimated by quadratic functions.

    Table 2 shows unadjusted and adjusted differences in mean MVD, pMVD and VPI according to tumour subtype. BP tumours had higher pMVD (Δ 1.9 proliferating microvessels/mm2, 95% CI 0.7 to 3.1, p=0.002) and VPI (Δ 1.7 percentage points, 95% CI 0.3 to 3.0, p=0.014) than luminal A tumours. No clear associations were found between subtypes and MVD score. There was little variation between unadjusted and adjusted estimates.

    View this table:
    Table 2

    Mean differences in MVD, pMVD and VPI according to tumour subtype based on linear regression

    Patients with MVD above the median showed a trend towards poorer BCSS than those with MVD below the median (p=0.066) (figure 4). HR for death from breast cancer was 1.10 (95% CI 1.02 to 1.18, p=0.008) per 10 vessels increase in MVD (table 3), adjusted for age, grade, stage and subtype. Neither pMVD nor VPI showed any clear prognostic association (figure 5A, B). HR was 1.04 (95% CI 0.96 to 1.13, p=0.28) for each single vessel increase in pMVD and 0.99 (95% CI 0.91 to 1.08, p=0.90) for every percentage point increase in VPI. Adjusting for age, grade, stage and subtype had little influence on the estimates.

    View this table:
    Table 3

    Risk of death from breast cancer according to MVD, pMVD and VPI

    Figure 4
    Figure 4

    Breast cancer specific survival according to microvessel density (MVD). Log rank test: p=0.066.

    Figure 5
    Figure 5

    (A) Breast cancer specific survival according to proliferating MVD (pMVD). Log rank test: p=0.73. (B) Breast cancer specific survival according to Vascular Proliferation Index (VPI). Log rank test: p=0.24.

    In separate analysis of each subtype, high MVD was associated with poorer survival for patients with the luminal A subtype (p=0.038) (figure 6), but not for patients with the BP subtype (p=0.7) (figure 6). After adjusting for age, stage and grade, HR was 1.22 (95% CI 1.09 to 1.37, p<0.001) per 10 vessels increase in the luminal A and 1.04 (95% CI 0.95 to 1.15, p=0.37) per 10 vessels increase in the BP subtype (table 4). No association was found between pMVD or VPI and prognosis in either subtype.

    View this table:
    Table 4

    Risk of death from breast cancer according to MVD, pMVD and VPI by tumour subtype

    Figure 6
    Figure 6

    Breast cancer specific survival according to microvessel density (MVD) and subtype. Log rank test for luminal A cases: p=0.038. Log-rank test for basal-like phenotype cases: p=0.7.

    The results were similar when the analyses were carried out with cut-off at the 75th percentile, with adjustments for time period at diagnosis, using different methods to adjust for age and in analyses restricted to grade 3 cases.

    Discussion

    In this study, there was no clear difference in MVD between BP and luminal A tumours. However, there was a clear difference in BCSS between luminal A tumours with high and low MVD, suggesting that MVD may be of prognostic significance in luminal A tumours. A higher number of proliferating vessels was found among BP tumours compared with luminal A tumours, although neither pMVD nor VPI was associated with BCSS.

    Tumour capillaries are different from normal vessels of the body. Tumour vessel diameter is typically three times larger, mural cell coverage is sporadic, and the vessel can suddenly end blindly or anastomose with itself.32 ,33 The capillary walls of tumour vasculature are about 10 times leakier than normal capillaries,34 and they are difficult to identify in HES-stained sections. In 1991, MVD was introduced by Weidner et al, who showed that risk of metastasis increased with increasing microvessel count. However, without an estimation of the proliferating fraction, it is not a measure of ongoing angiogenesis within a tumour.6 Some studies have found MVD to be a prognostic factor in cancer,35–37 while others have not.6 ,8 ,9 It has been suggested that measuring vascular proliferation may be a better estimate of actively ongoing angiogenesis within a tumour.6–10 ,32 ,38 In the present study, significantly more proliferating vessels were found in the BP compared with luminal A tumours. This finding is in accordance with those of other studies7 ,9 and is also in line with studies suggesting that the BP has more aggressive features than the luminal A subtype.12 ,20 ,39

    However, while pMVD and VPI have been associated with poor prognosis in previous studies,6–10 no such associations were found in the present study. It should be recognised that pMVD is generally low in breast cancer,8 and therefore prone to variations in estimation leading to a similar variation in VPI. Also, the two tumour categories in our study were matched for histological grade, and this could influence the results.

    When analysing all cases in this study, MVD was found to be a borderline prognostic factor. This is in accordance with previous studies.5 ,40 However, the results suggest that MVD is a prognostic factor in the luminal A subtype, but not in BP tumours. Cases were matched according to grade whenever possible, and therefore the luminal A cases in this study represent a high-grade subpopulation of the luminal A subtype comprising mostly grade 3 tumours. Despite this, a clear difference in survival was found between luminal A patients with high MVD and low MVD. This finding has, to the best of our knowledge, not been presented in any previous studies. To further explore and validate this finding, MVD should be investigated in a larger population of luminal A breast cancers including all grades.

    The majority of women in this cohort lived in an era when birth control pills, hormone replacement therapy and mammography screening were largely unavailable.25 ,26 Furthermore, they were diagnosed with breast cancer in a time period or at an age when adjuvant therapy was not an option, thus providing an opportunity to study the near natural course of the disease after surgery. However, it must be recognised that there may be differences between the tumours arising in this cohort and tumours occurring among women who have had access to mammography screening and menopausal hormonal treatment.

    Today, breast cancer treatment is based on histological type, histopathological grade, stage, proliferation rate and expression of ER, PR and HER2.41 In addition, breast cancer can be classified according to molecular subtypes.11 While the main focus has been to identify patients with breast cancer with poor prognosis, it is equally important to identify women with a particularly good prognosis to avoid unnecessary treatment. The results of this study underline the need to further study the prognostic and predictive value of MVD, pMVD and VPI. Vascular proliferation was higher in the BP, indicating that this subtype might be more susceptible to antiangiogenic treatment. Studies are needed to further explore whether pMVD and VPI could serve as predictive markers in the BP category.

    The luminal A subtype has the best prognosis of all subtypes.12 ,13 ,19 ,21 High MVD was significantly associated with poorer prognosis among high-grade luminal A tumours. If MVD provides additional prognostic information for these tumours, this might be helpful in further stratifying this large group of patients according to prognosis and may thus contribute to identifying patients with breast cancer with a particularly good prognosis that may receive unnecessary treatment today.

    Take home messages

    • Microvessel density is a prognostic factor in high-grade luminal A breast cancer.

    • Vascular proliferation is higher in the basal-like subtype of breast cancer.

    • Vascular proliferation is not associated with breast cancer specific survival in this series.

    Acknowledgments

    The authors thank the Department of Pathology and Medical Genetics, St. Olav's Hospital, Trondheim, for making the archives available for the study, the Cancer Registry of Norway for providing the patient data, and senior biomedical scientists Borgny Ytterhus, NTNU, and Lillian Hallseth, University of Bergen, for laboratory aspects of the study.

    References

    1. ↵
      1. Folkman J
      . Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182–6. doi:10.1056/NEJM197108122850711
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Gimbrone MA Jr.,
      2. Leapman SB,
      3. Cotran RS, et al
      . Tumor dormancy in vivo by prevention of neovascularization. J Exp Med 1972;136:261–76. doi:10.1084/jem.136.2.261
      OpenUrlAbstract
    3. ↵
      1. Folkman J
      . Proceedings: tumor angiogenesis factor. Cancer Res 1974;34:2109–13.
      OpenUrlAbstract/FREE Full Text
    4. ↵
      1. Folkman J
      . What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 1990;82:4–6. doi:10.1093/jnci/82.1.4
      OpenUrlFREE Full Text
    5. ↵
      1. Hlatky L,
      2. Hahnfeldt P,
      3. Folkman J
      . Clinical application of antiangiogenic therapy: microvessel density, what it does and doesn't tell us. J Natl Cancer Inst 2002;94:883–93. doi:10.1093/jnci/94.12.883
      OpenUrlFREE Full Text
    6. ↵
      1. Stefansson IM,
      2. Salvesen HB,
      3. Akslen LA
      . Vascular proliferation is important for clinical progress of endometrial cancer. Cancer Res 2006;66: 3303–9. doi:10.1158/0008-5472.CAN-05-1163
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Nalwoga H,
      2. Arnes JB,
      3. Stefansson IM, et al
      . Vascular proliferation is increased in basal-like breast cancer. Breast Cancer Res Treat 2011;130: 1063–71. doi:10.1007/s10549-011-1740-7
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Arnes JB,
      2. Stefansson IM,
      3. Straume O, et al
      . Vascular proliferation is a prognostic factor in breast cancer. Breast Cancer Res Treat 2012;133:501–10. doi:10.1007/s10549-011-1785-7
      OpenUrlCrossRefPubMed
    9. ↵
      1. Kruger K,
      2. Stefansson IM,
      3. Collett K, et al
      . Microvessel proliferation by co-expression of endothelial nestin and Ki-67 is associated with a basal-like phenotype and aggressive features in breast cancer. Breast 2013;22:282–8. doi:10.1016/j.breast.2012.07.008
      OpenUrlCrossRefPubMed
    10. ↵
      1. Gravdal K,
      2. Halvorsen OJ,
      3. Haukaas SA, et al
      . Proliferation of immature tumor vessels is a novel marker of clinical progression in prostate cancer. Cancer Res 2009;69:4708–15. doi:10.1158/0008-5472.CAN-08-4417
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Perou CM,
      2. Sorlie T,
      3. Eisen MB, et al
      . Molecular portraits of human breast tumours. Nature 2000;406:747–52. doi:10.1038/35021093
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Sorlie T,
      2. Perou CM,
      3. Tibshirani R, et al
      . Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001;98:10869–74. doi:10.1073/pnas.191367098
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Sorlie T,
      2. Tibshirani R,
      3. Parker J, et al
      . Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA 2003;100:8418–23. doi:10.1073/pnas.0932692100
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Carey LA
      . Directed therapy of subtypes of triple-negative breast cancer. Oncologist 2010;15(Suppl 5):49–56. doi:10.1634/theoncologist.2010-S5-49
      OpenUrlFREE Full Text
    15. ↵
      1. Cheang MC,
      2. Voduc D,
      3. Bajdik C, et al
      . Basal-like breast cancer defined by five biomarkers has superior prognostic value than triple-negative phenotype. Clin Cancer Res 2008;14:1368–76. doi:10.1158/1078-0432.CCR-07-1658
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Dent R,
      2. Trudeau M,
      3. Pritchard KI, et al
      . Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res Research 2007;13(15 Pt 1):4429–34. doi:10.1158/1078-0432.CCR-06-3045
      OpenUrlCrossRef
    17. ↵
      1. Nielsen TO,
      2. Hsu FD,
      3. Jensen K, et al
      . Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 2004;10:5367–74. doi:10.1158/1078-0432.CCR-04-0220
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Arnes JB,
      2. Collett K,
      3. Akslen LA
      . Independent prognostic value of the basal-like phenotype of breast cancer and associations with EGFR and candidate stem cell marker BMI-1. Histopathology 2008;52:370–80. doi:10.1111/j.1365-2559.2007.02957.x
      OpenUrlCrossRefPubMed
    19. ↵
      1. Blows FM,
      2. Driver KE,
      3. Schmidt MK, et al
      . Subtyping of breast cancer by immunohistochemistry to investigate a relationship between subtype and short and long term survival: a collaborative analysis of data for 10,159 cases from 12 studies. PLoS Med 2010;7:e1000279. doi:10.1371/journal.pmed.1000279
      OpenUrlCrossRefPubMed
    20. ↵
      1. Carey LA,
      2. Perou CM,
      3. Livasy CA, et al
      . Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 2006;295:2492–502. doi:10.1001/jama.295.21.2492
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Engstrom MJ,
      2. Opdahl S,
      3. Hagen AI, et al
      . Molecular subtypes, histopathological grade and survival in a historic cohort of breast cancer patients. Breast Cancer Res Treat 2013;140:463–73. doi:10.1007/s10549-013-2647-2
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Parker JS,
      2. Mullins M,
      3. Cheang MC, et al
      . Supervised risk predictor of breast cancer based on intrinsic subtypes. JClin Oncol 2009;27:1160–7. doi:10.1200/JCO.2008.18.1370
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Prat A,
      2. Perou CM
      . Deconstructing the molecular portraits of breast cancer. Mol Oncol 2011;5:5–23. doi:10.1016/j.molonc.2010.11.003
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Cheang MC,
      2. Chia SK,
      3. Voduc D, et al
      . Ki67 index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst 2009;101:736–50. doi:10.1093/jnci/djp082
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Kvale G,
      2. Heuch I
      . A prospective study of reproductive factors and breast cancer. II. Age at first and last birth. Am J Epidemiol 1987;126:842–50.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Kvale G,
      2. Heuch I,
      3. Eide GE
      . A prospective study of reproductive factors and breast cancer. I. Parity. Am J Epidemiol 1987;126:831–41.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Opdahl S,
      2. Alsaker MD,
      3. Janszky I, et al
      . Joint effects of nulliparity and other breast cancer risk factors. Br J Cancer 2011;105:731–6. doi:10.1038/bjc.2011.286
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Lakhani SR,
      2. Ellis IO,
      3. Schnitt SJ, et al
      . World Health Organization (eds) WHO Classification of Tumours of the Breast. International Agency for Research on Cancer (IARC), Lyon, 2012.
    29. ↵
      1. Elston CW,
      2. Ellis IO
      . Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 1991;19:403–10. Histopathology 2002;41(3A):151–2, discussion 52–3. doi:10.1111/j.1365-2559.1991.tb00229.x
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Robbins P,
      2. Pinder S,
      3. de Klerk N, et al
      . Histological grading of breast carcinomas: a study of interobserver agreement. Hum Pathol 1995;26:873–9. doi:10.1016/0046-8177(95)90010-1
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. McShane LM,
      2. Altman DG,
      3. Sauerbrei W, et al
      . REporting recommendations for tumor MARKer prognostic studies (REMARK). Breast Cancer Res Treat 2006;100:229–35. doi:10.1007/s10549-006-9242-8
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Eberhard A,
      2. Kahlert S,
      3. Goede V, et al
      . Heterogeneity of angiogenesis and blood vessel maturation in human tumors: implications for antiangiogenic tumor therapies. Cancer Res 2000;60:1388–93.
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Weinberg RA
      . The biology of cancer. New York: Garland Science, Taylor & Francis Group, LLC, 2007:560–1.
    34. ↵
      1. Weinberg RA
      . The biology of cancer. New York: Garland Science, Taylor & Francis Group, LLC, 2007:544, 560–1.
    35. ↵
      1. Ogawa Y,
      2. Chung YS,
      3. Nakata B, et al
      . Microvessel quantitation in invasive breast cancer by staining for factor VIII-related antigen. Br J Cancer 1995;71:1297–301. doi:10.1038/bjc.1995.251
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. Hollingsworth HC,
      2. Kohn EC,
      3. Steinberg SM, et al
      . Tumor angiogenesis in advanced stage ovarian carcinoma. Am J Pathol 1995;147:33–41.
      OpenUrlPubMedWeb of Science
    37. ↵
      1. Bosari S,
      2. Lee AK,
      3. DeLellis RA, et al
      . Microvessel quantitation and prognosis in invasive breast carcinoma. Hum Pathol 1992;23:755–61. doi:10.1016/0046-8177(92)90344-3
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Fox SB,
      2. Gatter KC,
      3. Bicknell R, et al
      . Relationship of endothelial cell proliferation to tumor vascularity in human breast cancer. Cancer Res 1993;53:4161–3.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Perou CM
      . Molecular stratification of triple-negative breast cancers. Oncologist 2011;16(Suppl 1):61–70. doi:10.1634/theoncologist.2011-S1-61
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. Weidner N,
      2. Semple JP,
      3. Welch WR, et al
      . Tumor angiogenesis and metastasis--correlation in invasive breast carcinoma. N Engl J Med 1991;324:1–8. doi:10.1056/NEJM199101033240101
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Goldhirsch A,
      2. Wood WC,
      3. Coates AS, et al
      . Strategies for subtypes--dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 2011;22:1736–47. doi:10.1093/annonc/mdr304
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Abstract in Norwegian

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Handling editor Cheok Soon Lee

    • Contributors MRK: vessel counting; statistical analyses; main author of the manuscript. KK: training in vessel counting; contributed to the manuscript. SO: statistical analyses; contributed to the manuscript. LJV: contributed to the manuscript. LAA: design of the study; contributed to the manuscript. AMB: design of the study; vessel counting; contributed to the manuscript. Contributors listed in the Acknowledgements section—Borgny Ytterhus: laboratory methods, optimalisation; Lillan Hallseth: laboratory methods, advisory.

    • Funding Liaison Committee between the Central Norway Regional Health Authority (RHA) and the Norwegian University of Science and Technology (NTNU).

    • Competing interests None declared.

    • Ethics approval The study was granted approval including dispensation from the general requirement of patient consent by the Regional Committee for Medical and Health Sciences Research Ethics (REK, Midt-Norge, ref. nr: 836/2009).

    • Provenance and peer review Not commissioned; externally peer reviewed.