The Breast 29 (2016) 147e152
Contents lists available at ScienceDirect
The Breast journal homepage: www.elsevier.com/brst
Original article
IHC4 score plus clinical treatment score predicts locoregional recurrence in early breast cancer Roopa Lakhanpal a, *, Ivana Sestak b, Bruce Shadbolt c, Genevieve M. Bennett d, Michael Brown d, Tessa Phillips d, Yanping Zhang e, Amanda Bullman f, Angela Rezo a a
FRANZCR, Department of Radiation Oncology, The Canberra Hospital, Canberra, Australia Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine, Queen Mary University, London, UK Centre for Advances in Epidemiology and IT, The Canberra Hospital, Canberra, Australia d RCPA, Department of ACT Pathology, The Canberra Hospital, Canberra, Australia e ACT and SE NSW Breast Cancer Treatment Group, ACT Health, GPO Box 825, ACT 2601, Australia f Department of ACT Pathology, The Canberra Hospital, Canberra, Australia b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 19 February 2016 Received in revised form 18 May 2016 Accepted 22 June 2016
Purpose: Immunohistochemical 4 (IHC4) score plus Clinical Treatment Score (CTS) is an inexpensive tool predicting risk of distant recurrence in women with early breast cancer (EBC). IHC4 score is based on ER, PR, HER2 and Ki67 index. This study explores the role of the combined score (IHC4 þ CTS) in predicting risk of locoregional recurrence (LRR) in women with EBC who had breast conservation surgery (BCS) without adjuvant radiotherapy (study group). The secondary objective was to evaluate the clinicopathological differences between our study group and women who had adjuvant radiation following BCS (control group). Methods and materials: Patients were selected from the local database over a 13-year period. IHC testing was done where results were missing. Combined scores were calculated using the appropriate formulae. Results: Patients in the study group (81 patients) had favorable clinicopathological features compared to the control group (1406 patients). The Cox regression indicated a statistically significant association between the combined score and the risk of LRR (p ¼ 0.03). The incidence of LRR was zero, 20% and 33.3% in the low, intermediate and high risk groups respectively (p ¼ 0.007). Margin status was the only variable not included in the combined score. The Cox regression analysis demonstrated that the combined score (p ¼ 0.02) and the ordinal measure of margins (p ¼ 0.03) were significant independent predictors of LRR. Conclusion: This is the first study of its kind. The IHC4 score þ CTS can be used to identify low risk women who can potentially avoid adjuvant radiotherapy. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Early breast cancer Adjuvant radiotherapy Immunohistochemical 4 score Locoregional recurrence
Introduction Breast cancer represents a heterogeneous group of diseases. Molecular profiling has helped stratify breast cancer into risk categories [1,2] to optimize systemic options and improve clinical outcomes. More refined assays such as the 21-gene expression profile (Oncotype Dx Recurrence Score) [3,4], the 70-gene
* Corresponding author. Pain Management Unit, The Canberra Hospital, Yamba Drive, Woden, ACT 2605, Australia. Tel.: þ61 401383934. E-mail address:
[email protected] (R. Lakhanpal). http://dx.doi.org/10.1016/j.breast.2016.06.019 0960-9776/© 2016 Elsevier Ltd. All rights reserved.
expression profile (Mammaprint) [5], and Mammostrat [6] have shown promise for predicting prognostic categories for personalized systemic treatment. An inexpensive gene expression-profiling tool is the Immunohistochemical 4 (IHC4) score [7] that measures the levels of four key proteins (estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor-2 (HER2), and human Ki67). Cuzick et al. [7] developed an overall prognostic score that combines the IHC4 score and the Clinical Treatment Score (CTS), providing prognostic information on the risk of distant recurrence similar to that provided by the 21-gene recurrence score. The CTS is based on clinical parameters such as tumor size, grade, nodal
148
R. Lakhanpal et al. / The Breast 29 (2016) 147e152
status, age, and endocrine treatment. This prognostic model was developed using data from the Arimidex, Tamoxifen, Alone or in Combination (ATAC) study [8] and validated on another cohort (the “Nottingham group”) [7]. The IHC4 score þ CTS is therefore applicable to post-menopausal women with ER positive breast cancer receiving adjuvant endocrine therapy without chemotherapy. The IHC4 score þ CTS is more cost effective than the above-mentioned genomic tools, as it can be calculated from parameters available in routine clinical practice. Unlike IHC4 score þ CTS, the 21-gene recurrence score does not take into account classical risk factors such as nodal status, tumor size, and grade. The molecular profiling of breast cancer to assess the risk of local recurrence (LR) risk after breast conservation surgery (BCS) is comparatively underdeveloped. A tool such as IHC4 score þ CTS may be useful in identifying a group of women with minimal risk in whom radiotherapy can be omitted. Given that LRR is a significant predictor of and has a temporal and proportional relationship to the risk of distant recurrence [9e11] we hypothesized that the IHC4 score þ CTS also predicts the risk of LRR which is the primary objective of our study. This hypothesis was tested in women who had BCS for early invasive breast cancer but no adjuvant radiotherapy (study group), and within that group, women who were post-menopausal and did not have adjuvant chemotherapy (the validated subgroup). The secondary objective was to evaluate the clinicopathological and survival differences between our study group and the cohort of women who had adjuvant radiation following BCS during the same period (control group). The purpose of this analysis is to determine if the score can be widely applied to all breast cancers patients or to a predefined subset.
Methods and materials Patient population This is a prospective cohort study of women with early, unilateral, invasive breast cancer amenable to and treated with breast conservation surgery over a 13-year period (July 1997 to June 2010) in the Australian Capital Territory and the South East New South Wales (ACT&SENSW) region. The ACT&SENSW Breast Cancer Treatment Group Quality Assurance Project began accruing patients in 1997. Ninety-five percent of all breast cancer patients from ACT and SENSW region have given informed consent for their demographic indicators, disease pathology and treatment details to be recorded in this database and for their disease status to be followed up annually. The cohort was divided into those who had breast conservation surgery and adjuvant radiotherapy called the “control group”, and those that did not have radiotherapy after breast conservation surgery called the “study group”. A subgroup of patients from the study group, labeled ‘the validated subgroup’, was created that had the same inclusion and exclusion criteria as the original study by Cuzick et al. [7]. This group includes women who were postmenopausal and ER-positive and/or PR-positive and excluded patients who were both ER-negative and PR-negative and had received adjuvant chemotherapy. The data were analyzed for: 1) the entire study group and 2) the validated subgroup. Patients were excluded due to insufficient tumor (3 þ 0.559 Gr2 þ 0.970 Gr3 þ 0.130 Age65 0.149 Ana)} where N, T, Gr denote categories of nodal status, tumor size, grade respectively and Ana denotes treatment with anastrozole versus tamoxifen. The IHC4 score plus CTS were added for each patient to get a combine score. Patients were categorized as having low risk (1.909) [7]. These cut-offs correlate with a 10-year distant recurrence risk less than 10%, 10e20%, and greater than 20%, respectively in the transATAC group [12]. As margin status is an important predictor of the risk of LRR and it was not included in the IHC4 þ CTS, we decided to explore the relationship between margin status and IHC4 þ CTS in predicting risk of LRR. We extracted data on margin status for the validated subgroup and categorized it into two groups (margin less than 5 mm and margins greater than 5 mm). The association between margin data relative to the IHC4 þ CTS in predicting the risk of LRR was determined. The pre-specified primary endpoint for this analysis is the time to ipsilateral LRR. LRR was defined as recurrence of the same histological type of invasive breast cancer as the original invasive breast cancer in the ipsilateral treated breast and/or ipsilateral lymph nodes. Time to ipsilateral LRR is the instantaneous probability of the event at 5 years as described in the Kaplan Meier Curve. Patients on this were censored if they died without local recurrence. The Local Regional Recurrence Free Survival includes invasive ipsilateral tumor and/or ipsilateral lymph nodes. The rates of distant recurrence, overall survival, and disease-free survival, the overall median survival time, and the median time to death from breast cancer were also recorded for the study group and the control group. The follow up data was available till 31-7-2012.
Laboratory methods Formalin-fixed, paraffin-embedded (FFPE) tumor blocks were used for IHC analysis performed on tissue sections cut at 4 microns. All antibodies were pre-diluted by Ventana Medical Systems. Staining was performed on the Ventana Benchmark Ultra using SP1 and 1E2 antibodies for ER and PR staining, respectively. All of the ER-stained slides were reviewed in order to classify the proportion of cells that stained strongly, moderately, and weakly for ER in order to record the “H score” [7] in the IHC4 formula. Ventana antiHER2/neu (4B5) rabbit monoclonal primary antibody was used for IHC determination of HER2 status. Tumors were considered positive if the IHC staining score was 3þ or 2þ. Ki67 analysis was performed using 30-9 stain and manual counting. To avoid interobserver bias, the “H score” and the Ki67 score were each reported by a single pathologist.
R. Lakhanpal et al. / The Breast 29 (2016) 147e152 Table 1 Patient demographics and tumor characteristics. Study
Control
p Value
2.5% 13.4% 84%
9.6% 24.5% 65.6%
