ONCOPOOL – A European database for 16,944 cases of breast cancer

doi:10.1016/j.ejca.2009.09.009

European Journal of Cancer

Volume 46, Issue 1, January 2010, Pages 56-71

https://doi.org/10.1016/j.ejca.2009.09.009 Get rights and content

Abstract

ONCOPOOL is a retrospectively compiled database of primary operable invasive breast cancers treated in the 1990s in 10 European breast cancer Units. Sixteen thousand and nine hundred and forty four cases were entered, with tumours less than 5 cm diameter in women aged 70 or less (mean age 55).

Data

Data were date of birth, mode of diagnosis, pathology (size, lymph node status, grade, type, lympho-vascular invasion and hormone receptor) and therapies and outcome measures: first local, regional or distant recurrences, contralateral primary, date and cause of death.

Tumour characteristics

Mean diameter 1.8 cm, 66% lymph node negative, 24% 1–3 lymph nodes involved and 10% had 4 or more involved. Grade 1, 29%; Grade 2, 41%; and Grade 3, 30%. Polynomial relationships were established between grade, stage and size.

Seventy-five percent were oestrogen receptor (ER) positive. ER closely related to grade.

Outcomes

Overall Survival was 89% at 5 years from diagnosis, 80% 10 years and 73% 15 years; Breast Cancer-Specific survivals were 91%, 84% and 79%.

Survival strongly related to the Nottingham Prognostic Index (NPI).

Cases detected at screening had 84% 10-year survival, those presenting symptomatically 76%.

ER positive cases treated with adjuvant hormone therapy had a reduction in risk of death of 13% over those not receiving adjuvant therapy (p = 0.000). ER negative cases treated with chemotherapy showed a risk reduction of 23% over those not receiving chemotherapy (p = 0.000).

Introduction

ONCOPOOL (Pooling of European Data to Harmonise Translational Research in Breast Cancer) is a dataset compiled in a European Commission Framework 5 Project in the Quality of Life and Living Resources Programme. Cases of operable breast cancer from 12 European breast cancer Units in 10 European states, diagnosed in the 1990s were entered retrospectively, with the collection of data on the primary tumours being carried out between 2000 and 2003 and longer term follow-up data being collected since that date.

The SEER (surveillance, epidemiology and end results programme) database1, 2 with 851 citations has long been regarded as providing the best overall picture of the features of breast cancer at primary presentation and of the survival of primary breast cancer. Since SEER was published survival has improved enormously in breast cancer³; SEER is also entirely North American based. It is appropriate to have an up-to-date, standard setting, large European database available in the breast cancer literature.

In ONCOPOOL cases of operable primary breast cancer in women aged 70 or less and diagnosed consecutively in each Unit in periods defined by the Unit within 1990–1999 were entered retrospectively. Units mainly treated the population local to them so that few cases came from tertiary referral. All cases received their whole treatment for primary breast cancer under the care of their Unit. Data entry and follow-up were carried out by the Units, resulting in very good data capture at entry; these factors ensured that the dataset reflected the tumour features and behaviour in the whole population. Quality assurance has been applied to clinical, pathology and endocrine data.

A large range of tumour variables and outcomes has been inspected, including data on the method of detection, the treatment given, pathology and extra tumour factors to those standard for prognosis, such as HER2-neu, lympho-vascular invasion and outcome data on recurrences and the types of recurrence. Prognosis has been inspected by use of an index integrating prognostic factors, the well-recognised and validated Nottingham Prognostic Index (NPI).4, 5, 6, 7

Section snippets

Materials and methods

ONCOPOOL is a European Database for Primary Breast Cancers built as a European Community Framework 5 Project.

Patient ages

The mean ages at patient entry are shown for each Unit and overall in Table 1. Mean age for all cases was 55, with the range of means for the Units being 53–56 years.

Fig. 1 shows the percentage frequency of all cases entered into ONCOPOOL by patient age at diagnosis. With very few cases in women in their early twenties (only two were below 20), numbers of cancers diagnosed rose sharply (approaching an exponential), to 50 years of age after which numbers fell at an increasing rate; when correction

Discussion

This ONCOPOOL analysis provides information from a large dataset on Invasive Breast Cancers diagnosed in 10 European Specialist Breast Units in the 1990s. Details of the pathological make-up of tumours, of prognostic factors and of the inter-relation of these factors are recorded. Five- and 10-year survivals are given and the effects of interventions examined.

ONCOPOOL resembles SEER in the accrual of tumour pathology, inter-relation of pathological factors, recording of treatments and relation

Contributors

The submission was from Nottingham City Hospital and was supported by EUSOMA (European Society of Breast Cancer Specialists). The project was designed and the protocol and proposal were written by R.W.B. (Project coordinator and Grant Holder) and by B.H.-S, and they and the local coordinators of all the partner Units (including the Tenovus Centre) were co-applicants for grant support.

We thank all the contributors listed below for their work on this project. In addition we thank Dr. Claire

Conflict of interest statement

None declared.

Acknowledgements

ONCOPOOL (Pooling of European Data to Harmonise Translational Research in Breast Cancer) was supported by a European Commission Framework 5 Project Grant; we thank the EC.

References (20)

V. Beral et al.
Sudden fall in breast cancer death rates in England and Wales
Lancet
(1995)
R.W. Blamey et al.
Survival of invasive breast cancer according to the Nottingham Prognostic Index in cases diagnosed in 1990–1999
Eur J Cancer
(2007)
R.W. Blamey et al.
Reading the prognosis of the individual with breast cancer
Eur J Cancer
(2007)
R.D. Macmillan et al.
Sentinel node biopsy for breast cancer may have little to offer four-node samplers: results of comparison study
EJC
(2001)
G. D’Eredita et al.
Prognostic factors in breast cancer: the predictive value of the Nottingham Prognostic Index in patients with a long term follow-up and treated in a single institution
Eur J Cancer
(2001)
J. Lundin
The Nottingham Prognostic Index – from relative to absolute risk prediction
Eur J Cancer
(2007)
D.E. Henson et al.
Relationship among outcome, stage of disease, and histologic grade for 22,616 cases of breast cancer. The basis for a prognostic index
Cancer
(1991)
L.A.G.R. Ries et al.
Survival from breast cancer according to tumor size and nodal status
Surg Oncol Clin North Am
(1994)
J.L. Haybittle et al.
A prognostic index in primary breast cancer
Br J Cancer
(1982)
M.H. Galea et al.
The Nottingham Prognostic Index in primary breast cancer
Breast Cancer Res Treat
(1992)

There are more references available in the full text version of this article.

Cited by (87)

Comparison of the Nottingham Prognostic Index and OncotypeDX© recurrence score in predicting outcome in estrogen receptor positive breast cancer
2022, Breast
Traditionally, Nottingham prognostic index (NPI) informed prognosis in patients with estrogen receptor positive, human epidermal growth factor receptor-2 negative, node negative (ER+/HER2-/LN-) breast cancer. At present, OncotypeDX© Recurrence Score (RS) predicts prognosis and response to adjuvant chemotherapy (AC).
To compare NPI and RS for estimating prognosis in ER + breast cancer.
Consecutive patients with ER+/HER2-/LN- disease were included. Disease-free (DFS) and overall survival (OS) were determined using Kaplan-Meier and Cox regression analyses.
1471 patients met inclusion criteria. The mean follow-up was 110.7months. NPI was calculable for 1382 patients: 19.8% had NPI≤2.4 (291/1471), 33.0% had NPI 2.41–3.4 (486/1471), 30.0% had NPI 3.41–4.4 (441/1471), 10.9% had NPI 4.41–5.4 (160/1471), and 0.3% had NPI>5.4 (4/1471). In total, 329 patients underwent RS (mean RS: 18.7) and 82.1% had RS < 25 (270/329) and 17.9% had RS ≥ 25 (59/329). Using multivariable Cox regression analyses (n = 1382), NPI independently predicted DFS (Hazard ratio (HR): 1.357, 95% confidence interval (CI): 1.140–1.616, P < 0.001) and OS (HR: 1.003, 95% CI: 1.001–1.006, P = 0.024). When performing a focused analysis of those who underwent both NPI and RS (n = 329), neither biomarker predicted DFS or OS. Using Kaplan Meier analyses, NPI category predicted DFS (P = 0.008) and (P = 0.026) OS. Conversely, 21-gene RS group failed to predict DFS (P = 0.187) and OS (P = 0.296).
In our focused analysis, neither NPI nor RS predicted survival outcomes. However, in the entire series, NPI independently predicted both DFS and OS. On the 40th anniversary since its derivation, NPI continues to provide accurate prognostication in breast cancer, outperforming RS in the current study.
Mitotic counts in one high power field in breast core biopsies is equivalent to counts in 10 high power fields
2022, Pathology
Mitotic rate is an important prognostic predictor in invasive breast carcinoma. Current guidelines recommend counting mitoses from 10 contiguous high power fields (HPFs) in the core biopsy. We propose a method to score mitotic activity in 1 HPF at the most mitotically active area of the tumour edge, or the interface between invasive tumour and benign breast tissue. We propose a score of 1, 2, or 3, corresponding to ≤1, 2, or ≥3 mitoses in 1 HPF, respectively.
A total of 141 breast core biopsies with corresponding surgical excisions were blindly examined. We counted the number of mitotic figures in 1 HPF and in 10 contiguous HPFs in the core biopsy and compared with the mitotic count from 10 contiguous HPFs in the excision which is considered the gold standard. Concordance rates and interobserver agreement rates were calculated.
The concordance rate was 82.3%, 78.7% and 82.3% between 1 HPF versus 10 HPFs in the core biopsy, 1 HPF in the core biopsy versus 10 HPFs in the excision and 10 HPFs in the core biopsy vs 10 HPFs in the excision, respectively. In the core biopsy, all three investigators agreed in 73.8% and 83.7% of the cases using the 1 HPF method and the 10 HPFs method, respectively; in the excision specimen, agreement was reached in 82.3% of the cases. The 1 HPF method showed similar concordance rate and interobserver agreement compared to the conventional method in the prediction of the mitotic score in the excision in all score groups. When stratified by mitotic score, the 1 HPF method predicted superior correlation with excision in the score 1 group than the 10 HPFs method, but not in the score 2 or 3 groups. From these findings we conclude that the proposed 1 HPF method can be used in clinical practice to grade invasive breast carcinomas in core biopsies, with the possibility of being utilised in small biopsies with less than 10 HPFs of invasive carcinoma.
PET imaging in breast cancer
2022, Nuclear Medicine and Molecular Imaging: Volume 1-4
This chapter describes the ways PET/CT imaging is utilized in metastatic breast cancer, from staging to treatment response of osseous and non-osseous metastatic disease. We focus on the three main PET radiotracers used for breast cancer at this time: [¹⁸F]fluorodeoxyglucose ([¹⁸F]FDG), [¹⁸F]Sodium Fluoride ([¹⁸F]NaF) and [¹⁸F]Fluoroestradiol ([¹⁸F]FES). We describe foundational and recent studies supporting PET/CT imaging of late stage and metastatic breast cancer, identifying the preferred radiotracer in a range of clinical scenarios. We conclude with future directions of the field.
Synthesis and preliminary evaluation of a PET-FI bimodal imaging agent targeting estrogen receptor
2021, Bioorganic and Medicinal Chemistry Letters
Estrogen receptor is an attractive target for the diagnosis and treatment of breast cancer. This article reports for the first time a dual-modality imaging agent targeting estrogen receptor that can use PET imaging to diagnose breast cancer and utilize fluorescence imaging to achieve intraoperative navigation. Fluorescence experiments show that [^natGa] 1 has typical aggregate induced emission characteristics. Above the critical concentration, [^natGa] 1 can form biocompatible nanomicelles. [^natGa] 1 can quickly light up estrogen receptor positive MCF-7 cells. Cell uptake experiments show that [⁶⁸Ga] 1 is mediated by estrogen receptor. Therefore, [^nat/68Ga] 1 shows the characteristics of highly sensitive diagnosis and visualization of breast cancer, and can be used as a lead compound for the development of a novel PET-FI bimodal imaging agent targeting the estrogen receptor.
Among multimodal imaging techniques, which can be trusted to assess radiologic N0 stage in patients with breast cancer?
2019, Imagerie de la Femme
L’extension tumorale aux ganglions lymphatiques chez les patientes atteintes d’un cancer du sein est un facteur pronostique majeur. Même s’il existe différentes méthodes de biopsie des ganglions lymphatiques, cette technique est maintenant largement utilisée et acceptée comme norme de soins dans le monde entier. Mais ces techniques ont un taux de faux négatifs de 10 à 15 %, y compris en cas de ganglions macrométastatiques. L’évaluation clinique étant limitée, la multimodalité est couramment utilisée pour prédire l’extension ganglionnaire. L’échographie est l’examen de première intention. Elle doit être bilatérale comparative et doit explorer les ganglions axillaires, ceux de la chaîne mammaires internes, rétropectoraux et sus-claviculaires. Selon leur aspect échographique, les ganglions lymphatiques axillaires peuvent être divisés en trois groupes : d’aspect normal, incertain ou suspect. L’aspiration à l’aiguille fine ou la microbiopsie sont les deux techniques utilisées pour prélever un échantillon de ganglions lymphatiques et doivent être réalisées chez des patientes présentant des ganglions lymphatiques indéterminés ou suspects. L’IRM, la TDM et la TEP-TDM peuvent être utilisées pour évaluer l’étendue de l’extension ganglionnaire, principalement chez les patientes atteintes des tumeurs malignes de plus de 3cm ou N+ échographqiue. Cet article propose une description des meilleures méthodes d’imagerie actuelle pour statuer sur le stade N0 radiologique (rN0) avant la prise en charge thérapeutique des cancers du sein.
Lymph node involvement in breast cancer is a major prognostic factor. Even if different methods for sentinel lymph node biopsy exist, this technique is now widely practiced and accepted as standard of care worldwide. But those techniques have a rate of 10 to 15% of false negative, including macrometastatic nodes. Clinical evaluation being limited, multimodality imaging is commonly used to predict lymph node involvement in patient with breast cancer. Ultrasound is the first-line exam and has to be bilateral and must explore axillary, internal mammary, retropectoral and over-clavicular nodes. According to the morphologic evaluation of lymph nodes using ultrasound, axillary lymph nodes can be divided in three groups: normal, uncertain and suspicious. Fine-needle aspiration or core-needle biopsy are the two techniques used to have a sample of lymph nodes, and must be performed in patients with uncertain or suspicious lymph nodes. MRI, CT and PET-CT can be used to assess the extent of node involvement, mostly in patient with proven malignant lymph nodes.
Consideration of breast cancer subtype in targeting the androgen receptor
2019, Pharmacology and Therapeutics
Citation Excerpt :
Among invasive breast cancers, 75% express the estrogen receptor (ER) and 20–30% overexpress the human epidermal growth factor receptor 2 (HER2). These patients can benefit from therapy that targets ER or HER2, resulting in superior overall survival (OS) in both the curative and non-curative setting (Blamey et al., 2010; Gibson, Dawson, Lawrence, & Bliss, 2007; Swain et al., 2015). However, there is still a need to improve disease outcome, leading to a constant search for new drug targets.
The androgen receptor (AR) is a drug target in breast cancer, and AR-targeted therapies have induced tumor responses in breast cancer patients. In this review, we summarized the role of AR in breast cancer based on preclinical and clinical data. Response to AR-targeted therapies in unselected breast cancer populations is relatively low. Preclinical and clinical data show that AR antagonists might have a role in estrogen receptor (ER)-negative/AR-positive tumors. The prognostic value of AR for patients remains uncertain due to the use of various antibodies and cut-off values for immunohistochemical assessment. To get more insight into the role of AR in breast cancer, we additionally performed a retrospective pooled analysis to determine the prognostic value of the AR using mRNA profiles of 7270 primary breast tumors. Our analysis shows that a higher AR mRNA level is associated with improved disease outcome in patients with ER-positive/human epidermal growth factor receptor 2 (HER2)-negative tumors, but with worse disease outcome in HER2-positive subgroups. In conclusion, next to AR expression, incorporation of additional tumor characteristics will potentially make AR targeting a more valuable therapeutic strategy in breast cancer.

View all citing articles on Scopus

^a: Co-applicant for grant support.

^b: Clinical QA.

^c: Data analyst.

^d: Endocrine QA.

^e: Pathology QA.

View full text

ONCOPOOL – A European database for 16,944 cases of breast cancer

Abstract

Data

Tumour characteristics

Outcomes

Introduction

Section snippets

Materials and methods

Patient ages

Discussion

Contributors

Conflict of interest statement

Acknowledgements

Lancet

Eur J Cancer

Eur J Cancer

EJC

Eur J Cancer

Eur J Cancer

Relationship among outcome, stage of disease, and histologic grade for 22,616 cases of breast cancer. The basis for a prognostic index

Cancer

Survival from breast cancer according to tumor size and nodal status

Surg Oncol Clin North Am

A prognostic index in primary breast cancer

Br J Cancer

The Nottingham Prognostic Index in primary breast cancer

Breast Cancer Res Treat