St Gallen Consensus Research Articles

Abstract P2-11-15: Immunohistochemical assessment of Ki67 with the antibodies SP6 and MIB1 - A comparison of prognostic information and reproducibility

Abstract Background Proliferation is a key feature in breast cancer and also a clinical important factor for prognosis and treatment prediction. In the St Gallen Consensus of 2011, immunohistochemically analysed Ki67 was adopted as a surrogate marker to distinguish the “luminal A” from the “luminal B” subtype, in order to select patients with estrogen receptor positive disease expected to benefit from adjuvant chemotherapy. The mouse monoclonal antibody MIB1 is the generally accepted antibody for assessment of Ki67. However, other antibodies have been developed, e.g. the rabbit monoclonal antibody SP6. The assessment of Ki67 with MIB1 is unfortunately associated with a considerable lack of reproducibility. Rabbit monoclonal antibodies generally tend to have higher specificity without loss of sensitivity, compared to corresponding mouse monoclonal antibodies. SP6 has also been found to have reduced background staining compared to MIB1. According to these advantages, SP6 may be an alternative to MIB1 for routine staining of Ki67. Any methodological modification should, according to international recommendations, be compared against a clinically validated assay and demonstrate acceptable concordance before being introduced in clinical routine. The analysis of Ki67 with SP6 therefore needs to be evaluated and compared to MIB1 in a cohort of breast cancer patients with clinical follow-up. Aims To compare the antibodies SP6 and MIB1 for immunohistochemical assessment of Ki67 in primary breast cancer regarding prognostic strength and reproducibility of the evaluation. Methods Tissue microarray from a cohort of 237 premenopausal women with node-negative breast cancer was used for assessment of Ki67, with both SP6 and MIB1, by three different investigators. The 7th decile was applied for defining cut-off. Distant disease-free survival (DDFS) was used as endpoint and the follow-up was restricted to 5 years. Results Ninety per cent of the samples were classified into the same group, either high or low Ki67, irrespective of antibody used. Ki67 (high vs. low), analysed with both antibodies was associated to DDFS (34 events) in the univariable analyses (SP6: HR 2.6, 95% CI 1.3-5.2, p = 0.01 and MIB1: HR 2.8, 95% CI 1.4-5.7 p = 0.004) and showed borderline significance for DDFS in the multivariable analyses, also including HER2, age, and tumour size (SP6: HR 2.0, 95% CI 0.93-4.5, p = 0.074 and MIB1: HR 2.2, 95% CI 0.97-4.8, p = 0.058). The agreement between different assessors was somewhat higher for MIB1 than for SP6 (kappa-values 0.83-0.88 vs. 0.72-0.77). Conclusions SP6 was not superior to MIB1 and the two antibodies were comparable in the assessment of Ki67 for prognostic considerations in primary breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-11-15.

Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up

Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up

Association of Immunohistochemically Defined Molecular Subtypes with Clinical Response to Presurgical Chemotherapy in Patients with Advanced Breast Cancer

Gene expression profiling (GEP) has identified several molecular subtypes of breast cancer, with different clinico-pathologic features and exhibiting different responses to chemotherapy. However, GEP is expensive and not available in the developing countries where the majority of patients present at advanced stage. The St Gallen Consensus in 2011 proposed use of a simplified, four immunohistochemical (IHC) biomarker panel (ER, PR, HER2, Ki67/Tumor Grade) for molecular classification. The present study was conducted in 75 newly diagnosed patients of breast cancer with large (>5cm) tumors to evaluate the association of IHC surrogate molecular subtype with the clinical response to presurgical chemotherapy, evaluated by the WHO criteria, 3 weeks after the third cycle of 5 flourouracil, adriamycin, cyclophosphamide (FAC regimen). The subtypes of luminal, basal-like and HER2 enriched were found to account for 36.0 % (27/75), 34.7 % (26/75) and 29.3% (22/75) of patients respectively. Ten were luminal A and 14 luminal B (8 HER2 negative and 6HER2 positive). The triple negative breast cancer (TNBC) was most sensitive to chemotherapy with 19% achieving clinical-complete-response (cCR) followed by HER2 enriched (2/22 (9%) cCR), luminal B (1/6 (7%) cCR) and luminal A (0/10 (0%) cCR). Heterogeneity was observed within each subgroup, being most marked in the TNBC although the most responding tumors, 8% developing clinical-progressive-disease. The study supports association of molecular subtypes with response to chemotherapy in patients with advanced breast cancer and the existence of further heterogeneity within subtypes.

Abstract PD06-04: St. Gallen 2011 clinicopathological subtyping of breast cancer: impact of different proliferation assessment methods

Abstract Rationale: The St Gallen Consensus 2011 strongly recommends subtyping of breast cancer (BC). However, because there is no data from phase III trials to substanciate adjuvant therapy benefit gene expression arrays were not generally accepted to guide systemic therapy decisions. Instead, the consensus introduced clinicopathological BC subtypes (cpBCST) defined as sets of risk and predictive biomarkers which are established worlwide (Table 1). To discriminate Luminal A vs. Her2 negative Luminal B BCs the St Gallen Consensus recommends Ki67 for defining proliferation. Although, because no standardized methodology or cut-off definition for Ki67 is available so far ‘some alternative measure of proliferation’ was suggested. We analyzed the influence of different proliferation assessment methods on BC subtype distribution. Methods: cpBCST assessment including proliferation measurement by different variants of Ki67-Labeling-Index (KI67LI) was performed on 225 BCs. For comparison we used the Mitotic Activity Index (MAI) as the only prospectively evaluated and best reproducible proliferation measurement (Baak et al 2008). The Ki67-LI was based on counting any nuclear staining of distinctly defined target areas: (1) 100 tumor cells within the hot spot (Ki67-100), (2) 1020 tumor cells in 3 HPF in the tumor periphery including the hot spot (Ki67-1020periphery), and (3) 1020 tumor cells in 3 HPF including hot spot, cold spot, and an intermediate area (Ki67-1020spectrum). Staining quality and counting accuracy was tested as excellent by the 2011 circulation of the German Quality Initiative in Pathology. According to St Gallen, a cut-off <14% was used. The MAI was measured in 10 HPFs with a total area of 1.59 mm2 using a cut-off ≥10 as evaluated by Baak et al. 2008. Results: Conclusions: The delimination of Luminal A and Luminal B Her2 negative BCs according to the St. Gallen Consensus 2011 is highly influenced by the method of proliferation assessment. Using the recommended cut-off <14% all estimations of Ki67LI came out with Luminal A/Luminal B ratios (10–34% vs. 45–69%) which are much lower than those published (44–53% vs. 9–29%). As a consequence, this could lead to more cytotoxic treatment. Therefore, the biological relevance of KI67LI seems to be not unequivocal. Unlike this, proliferation measurement by MAI resulting in a Luminal A/Luminal B ratio of 52% vs. 29% seems more reliable. The differences may be due to the fact that mitotic figures reflect only the M phase, whereas Ki67 stains cells in nearly all phases of the cell cycle. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr PD06-04.

Abstract P2-10-12: How well predict the 2011 St Gallen early breast cancer surrogate phenotypes metastatic survival?

Abstract BACKGROUND: The five prognostic surrogate phenotypes, defined by the 2011 St Gallen consensus, predict metastatic relapse following early breast cancer treatment (Brouckaert et al Ann Oncol 2012). It is unknown to what extent these surrogate phenotypes defined on the primary tumor predict metastatic survival (MS) in a cohort of consecutive women with metachronous metastases. PATIENTS & METHODS: All 4318 patients with primary operable breast cancer, diagnosed between 01–01-2000 and 31–12-2009 and treated in our center were prospectively entered in our institutional database. We included 345 patients with a (distant) metastatic relapse during their follow-up. We observed metastatic survival as the time between the diagnosis of the relapse and death. Tumor subtype was defined according to the 2011 St Gallen recommendations in 5 groups: luminal A (ER+/HER2−, grade 1–2), luminal B1 (ER+/HER2−, grade 3), luminal B2 or luminal-HER2 (ER+/HER2+), HER2-like (ER−/HER2−) and triple-negative. We focused on the value of the primary tumor subtype as a major prognostic determinant, but also assessed age at diagnosis, tumor size, lymph node involvement, tumor subtype, PR-status, primary detection (screening vs. palpation), histology, adjuvant therapy, distant-metastasis free interval (DMFI), first-line metastatic hormonal therapy, recurrence sites (isolate or multiple) and metastasis detection (clinical vs. biochemical). RESULTS: In our cohort, we recorded a median metastatic survival (MS) of 22.3 months (95% CI: 19.7–25.6) and a 5-year survival probability of 15%. Significant differences in median MS were noted when considering the primary tumor subtype (cf. table). We observed an independent prognostic effect for multiple recurrence sites (HR = 1.78, 95% CI 1.37–2.32), first-line metastatic hormonal therapy (HR = 0.30, 95% CI 0.18–0.49), DMFI (HR = 0.92 CI 0.85–0.98) and, finally, primary tumor subtype. CONCLUSION: Our results showed that primary tumor subtype, DMFI, solitary recurrence and first-line metastatic hormonal therapy act as independent prognostic factors in metastatic breast cancer. In order to overcome the biological complexity of metastatic breast cancer, we must appreciate the primary tumor subtype in our therapeutic approach. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-10-12.

Abstract P3-06-19: Ki-67 mRNA as a predictor for response to neoadjuvant chemotherapy in primary breast cancer

Abstract Background: Stratification of molecular subtypes is of outmost importance for clinical decision making according to the 12th St Gallen Consensus conference (1). The molecular subtypes were originally identified by genome-wide mRNA analysis of fresh tissue specimen (2). Conventional immunehistochemistry assessment using ER, PR, HER2 and Ki67 approximate these subtypes with ∼ 80% concordance (3). Here we have tested a predefined Single-Sample-Predictor for molecular subtyping based on mRNA assessment of ESR1, PGR, HER2, Ki67 and RACGAP1 in a prospective-retrospective study design. Materials and Methods: Pretreatment core cut biopsies from n=100 patients with PBC treated within a randomized phase II trial (1) of anthracylin/taxane based NAC were examined. Immunohistochemistry was performed for the Ki67 antigen on an automated IHC platform (Dako Techmate 500). Ki-67 were assessed either by visual scoring (vIHC) or by quantitative image analysis (qIHC). RNA from formalin-fixed, paraffin embedded (FFPE) routine biopsies was extracted using a bead-based extraction method (STRATIFYER XTRAKT kits). Fresh tissue samokes were prepared by using Qiagen kits. ESR1, PGR, HER2, Ki-67 and RACGAP1 as well as CALM2 as a house keeping gene were measured via a multiplex quantitative RT-PCR (qPCR). Stratification into molecular subtypes was done by stepwise analysis of singular genes using predefined cut-off values. Kaplan-Meier survival estimates, patitioning test and correlation analyses were performed using the SAS JMP® 9.0.0 software. Results: Concordance between RNA based and IHC-based subtyping into Luminal A, Luminal B, HER2 positive and Triple Negative breast cancer by using ESR1, PGR, HER2, Ki-67 and RACGAP1 was at 85%. mRNA based analysis of fixed tumor specimen performed equally well as fresh tissue analysis. The rate of pathological complete Remission for the predefined molecular subtypes has been for Luminal A 0%, Luminal B 19%, HER2 positive 21% (without Trastuzumab) und Triple Negative Tumors 22%. qPCR from fresh tissue specimen resulted in almost identical classifications. The 5-year overall survival rate was significantly different between subtypes as expected: Luminal A 100%, Luminal B 85%, HER2 positiv 80% und triple negativ 60%. Luminal tumors predominantly metastasized into the bones. Cancer related deaths within the first three years were restricted tot he Triple negative Tumortype. Conclusion: The Single-Sample-Predictor based on assessment of only five mRNA Markers (ESR1, PGR, HER2, Ki-67 and RACGAP1) reliably stratified into molecular subtypes by using either fixed or fresh pretreatment core needle biopsies. The mRNA based assessment enables objective and highly reproducible subtyping for clinical routine diagnostics. 1) Goldhirsch et al., Annals of Oncology 2011 2) Perou et al., Nature 2000 3) Sotiriou&Pusztai, NEJM 2009 4) Schneeweiss et al, Ann Oncol 2011 Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-06-19.

Abstract PD06-05: St. Gallen 2011 clinico-pathological subtypes of breast cancer: concordance of core biopsies and related surgical specimens

Abstract Rationale: Whereas the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011 considered molecular typing as the adequate model representing the heterogeneity of breast cancers, it did not generally accept gene expression array criteria for clinical decision making. When the classification of tumours by gene expression array criteria and immunohistochemical analysis of ER, PR and HER2 are discrepant, patients should be managed according to the current immunohistochemical markers. Therefore, for systemic therapy decision the St Gallen consensus Conference 2011 did recommend clinicopathologically defined breast cancer subtypes (cpBCST) to guide therapeutic choices. There are many reasons to provide patients and doctors these information already preoperatively. To evaluate relevance of St. Gallen 2011 cpBCST core biopsy (CB) results we estimated their concordance with postoperative cpBCSTs. Methods: cpBCSTs of CB and related surgical specimen (SSP) of 379breast cancers (BC) (62 Grade 1, 157Grade 2, 160 grade 3) were assessed according to the St Gallen definition using the prospectively estimated results for Estrogen, Progesteron, and Her2-status. To differentiate between Luminal A and the Her2 negative subgroup of Luminal B we retrospectively measured the mitotic activity index (MAI) with a cut-off of ≥10 for high versus low as evaluated by Baak et al. 2008. Concordance between cpBCST results of CB and SSP and PPV of cpBCST of CB were calculated. Results: None of the Luminal A BCs was overestimated in CB, whereas 49% (53/108) of all Luminal B or 66% (55/83) of the Luminal B (Her2−) BCs were underestimated as Luminal A, for all other subtypes, the agreement was 100%. In general, there was no overestimation as a more unfavourable CPBCST in CB. The overall underestimation rate came out with 14%. Whereas the PPV in CB is 100% for Luminal B, Her2 positive, and Triple negative CPBCSTs, it is only 78% for Luminal A (198/254). The reason is underestimation of proliferation in CB. Conclusions: BC subtypes classified according to the St. Gallen clinicopathological definitions on CB show a high agreement with corresponding SSP results. The respective PPVs are also generally high. This means that clinicopathological subtyping on CB can be used for preoperative decision making in the vast majority of cases. However, for the Her2 negative subgroup of Luminal B subtype in CB the agreement with SSP is lower due to underestimation of proliferation in CB. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr PD06-05.

Applying the 2011 St Gallen panel of prognostic markers on a large single hospital cohort of consecutively treated primary operable breast cancers

Many easily measurable and readily available factors are now established as being prognostic in primary operable breast cancer. We here applied the 2011 St Gallen surrogate definition for breast cancer subclassification using tumor grade instead of Ki67. Four thousand three hundred and eighteen consecutive patients who had surgery for primary operable breast cancer (1 January 2000 and 31 December 2009) in UZ Leuven excluding primary metastastic male breast cancers and those receiving neoadjuvant therapy. Five different surrogate phenotypes were created using the combined expression of estrogen receptor, progesterone receptor, human epidermal growth factor receptor-2 together with tumor grade. Disease-free interval (DFI), distant metastastis-free interval (DMFI), locoregional relapse-free interval (LRRFI), breast cancer-specific survival (BCSS) and overall survival (OS) were calculated. Surrogate phenotypes present with significant differences in DFI, DMFI, LRRFI, BCSS and OS. 'Luminal A' tumors presented with the best outcome parameters but the effect weakened at longer follow-up. The four surrogate markers, agreed upon by the 2011 St Gallen consensus, defined five prognostic surrogate phenotypes in a large series of consecutively treated breast cancer patients. Their prognostic value changed with longer follow-up. The added value of gene expression profile over classical pathological assessment remains to be defined.

Complementary role of Ki67 index and 70-gene signature (MammaPrint) high-risk patients in the St Gallen risk group with uncertain chemotherapy suggestion.

579 Background: St Gallen (SG) Consensus Panel on adjuvant treatment recommends the use of proliferation markers and multigene assays when choosing the appropriate treatment in addition to traditional parameters. Ki67 predictive/prognostic value is widely accepted; we tested its role in increasing the accuracy of risk prediction among Mammaprint (MP) /High(HR) or /Low Risk (LR) breast cancer patients with a SG risk uncertain for chemotherapy selection. Methods: MP was determined (courtesy of Agendia, Amsterdam) in 3 Italian Hospitals on 305 consecutive samples of breast cancer patients. We focused on SG “Intermediate Risk” Group (HER2 neg, no VI and: ER>10<50%, or G2 or Ki67 >15<30% or N1a or T>2<5cm), where the indication for adjuvant CHT still remains uncertain. In MP/HR cases, a “low” Ki67 value (<15%) was used to eliminate CHT suggestion and viceversa an “intermediate” Ki67 value (>15<30%) in MP/LR cases. Results: Overall, 72/305 pts (26.6%) were in SG intermediate risk group with a median age of 64 years (26-98). Ki67 was non available in 4.1% of cases, MP rejected in 16.6%, HR was detected in 39 (54.2%), whereas LR in 21 (29.2%). Ki67 resulted low in 23 MP/HR cases (59%), intermediate in 6 MP/LR cases (28.5%) and in 3 out 12 MP/Rejected cases (25%). The overall concordance between MP and Ki67 was 28/57 (49.1%), MP rejected excluded. Overall MP suggested 39/60 CHT (65%), Ki67 29/69 (42%). Conclusions: The risk of relapse is a continous variable and MP evaluates it in a dichotomy (low vs high) wich could decrease the accuracy of the test. In the selected SG Group, the average risk of 5 yr relapse is around 20% (EBCTCG, Lancet 2011) and MP HR cases in our experience account for 55%. A low Ki67 value could help avoid more than 20% of cytotoxic treatments suggested by MP. In the same way, according to an ER value <50% (2 cases), an intermediate Ki67 value could suggest a more aggressive treatment in a further 13% of cases MP LR (6/21) or Rejected (3/12). MP probably overestimates the risk in 20% of cases and underestimates it in further 20%. Ki67 could be usefull for a more personalized treatment in patients where the indication for adjuvant chemotherapy added to endocrine treatment is uncertain.

Role of Ki67 labeling index (LI) in subdividing homogeneous ER-positive grading groups of early breast cancer (EBC) in distinct biologic entities.

539 Background: Ki67LI is a relevant marker EBC. The 2009 St Gallen Consensus considered Ki67LI as an important parameter for the addition of chemotherapy to endocrine therapy in hormonereceptor-positive EBC, suggesting three cutoffs: low <15%, intermediate 15-30% and high >30%. Our study aimed to evaluate if these cutoffs can be used in different subgroups of EBC, or if to apply different cutoffs in distinct biological setting. Methods: Ki67LI 1 was identified by immunohistochemical staining in 3802 EBC pts treated from 1995 to 2008. Median age was 61. The relationship with clinic-pathological parameters and the prognostic significance of KI67LI was investigated in all EBCs and in subgroups of ER+ cases based on homogeneous grading (656 G1, 1535 G2, 1113 G3). Results: Median Ki67LI values were 22% in all cases and 10, 20 and 36% in ER+ G1, G2 and G3 respectively. High Ki67LI was significantly (p<0.001) associated with younger age, ductal type, greater size, positive N, poor G, absent or low ER /PR, positive HER-2, triple negative subtypes, larger use of chemo±hormonotherapy. At median f-up of 51 months DFS and OS were 84 and 85% respectively in high, 89 and 90% in intermediate, 92 and 94% respectively in low Ki67LI group (p <0.001). There were 456 relapses (12%): 207 (5.4%) in high, 136 (3.6%) in intermediate and 113 (3%) in low Ki67LI group. Using median ki67LI value for different homogenous ER+ grading groups (ER+GG) we stratified these populations in low and high risk. The results are shown in the table. Conclusions: Ki67 LI confirms to be a significant prognostic biomarker for DFS and OS in EBC, associated with other clinical-pathological characteristics. Cutoffs are different into ER+GG. They can cathegorize at least two biological entities in every grading group providing additional prognostic information in planning therapies and outcome prediction. Subgroups At median follow-up DFS (%) P value At median follow-up OS (%) P value G1 High Low 90 95 0.058 93 97 0.005 G2 High Low 93 87 <0.001 94 90 0.001 G3 High Low 87 88 NS 85 87 0.008

When to start an aromatase inhibitor: Now or later?

Among women who have undergone surgery for breast cancer, the risk of recurrence, especially distant metastases, peaks 1-2 years postsurgery. Recent clinical trial evidence suggests that initial adjuvant therapy with an aromatase inhibitor (AI) can reduce this early risk of recurrence. According to the recently updated St Gallen consensus statement, initial AI therapy is the preferred strategy for adjuvant endocrine therapy in postmenopausal women with hormone receptor-positive breast cancer.

Comparing guidelines for adjuvant endocrine therapy in postmenopausal women with breast cancer: a coming of age

Following surgery for early breast cancer, the standard of care for postmenopausal women is adjuvant therapy with any combination of radiation therapy, endocrine therapy, chemotherapy and/or targeted therapy. Clinicians rely on many tools, including guidelines, to make these treatment decisions. Such guidelines include the St Gallen consensus statement, the American Society of Clinical Oncology guidelines and the National Comprehensive Cancer Network guidelines, as well as various regional and national guidelines. Recommendations may vary, because different methods and criteria were used to assess the strength of supporting data. This article provides an overview of global guidelines for the adjuvant treatment of breast cancer and points out the major differences. Ongoing changes are highlighted, particularly those regarding the adjuvant endocrine treatment of postmenopausal women with breast cancer. While previous guidelines recommended tamoxifen alone, all major guidelines now recommend using third-generation aromatase inhibitors either in sequence with tamoxifen or as upfront treatment.

Accumulation of p53 determined by immunohistochemistry as a prognostic marker in node negative breast cancer; analysis according to st gallen consensus and intrinsic subtypes

This study evaluated the prognostic impact of p53 accumulation by immunohistochemistry (IHC) in lymph node-negative breast cancer (LNN-BC), and in subgroups of St Gallen consensus and intrinsic subtypes. A total 845 with a pathologic diagnosis of LNN-BC patients that underwent surgery at the National Cancer Center, Korea between 2001 and 2005 were retrospectively reviewed. The median age was 48 years (range: 25-85) and median follow-up period was 66.0 months (range: 9-101). Univariate analysis determined that tumor size, estrogen receptor (ER), progesterone receptor (PgR), p53, and Ki-67 were significant for disease free survival (DFS). Of these factors, PgR negativity (HR 3.57; 95% CI 1.26-10.09; P = 0.01) and p53 positivity (HR 3.17; 95% CI 1.51-6.65; P = 0.002) were independent prognostic factors in multivariate analysis. In the subanalysis for 4 intrinsic subtypes (luminal A, luminal B, HER2-overexpression, and triple-negative subtypes) and risk groups by St Gallen consensus, there were significant differences of DFS rates by p53 (5-year DFS rate, luminal A; 97.2% for p53 (-) vs 93.8% for p53 (+); P = 0.03, triple-negative subgroups; 94.1% vs 78.7%; P = 0.002, intermediate-risk group; 96.5% vs 90.7%; P = 0.003). P53 has prognostic power in LNN-BC, and gives the additional prognostic information for intrinsic subtypes and St Gallen consensus.

Molecular profiling currently offers no more than tumour morphology and basic immunohistochemistry

The management of breast cancer patients is still guided based on a constellation of clinicopathological features, including prognostic markers derived from careful histo-pathological analysis of tumours, namely tumour size, histological grade, presence of lymph node metastasis and vascular invasion [1-3]. Despite the huge amount of resources allocated to translational research endeavours, only three predictive markers are utilised to define the therapy of breast cancer patients: oestrogen receptor (ER) and progesterone receptor (PR), the predictive markers of response to endocrine therapy, and human epidermal growth factor receptor 2 (HER2), the molecular target of trastuzumab and lapatinib. These parameters are then used in conjunction either in the form of guidelines (for example, St Gallen's consensus criteria) or included in multivariable algorithms (for example, Adjuvant!Online) for clinical decision making [1-3]. Albeit seemingly simplistic, this approach has been shown to be clinically relevant, given that predictions made with Adjuvant!Online do correlate with the actual outcome of breast cancer patients [4], and, most importantly, the use of this framework to define the systemic therapy of breast cancer patients has contributed to the steady decline in the mortality of breast cancer patients [5]. Although eective, this approach is not sucient for the potential of individualised therapy to be realised. The promise of high throughput technologies, and in particular of gene expression profiling with microarrays, has been of apocalyptic dimensions [6-9]. The objectivity of the methodology coupled with the elaborate, if not mind boggling [10], bioinformatic approaches to answer clinically relevant questions have led some of the proponents of this technology to compare histopathology with some rituals practiced by ancient tribes [7], and some experts in the field predicted back in 2000 that microarrays would make conventional diagnostic techniques obsolete [6]. Microarrays and their derivatives have undoubtedly contributed to our understanding of breast cancer (for reviews, see [1,2]). They have provided direct evidence to demonstrate that breast cancer is a heterogeneous disease at the molecular level [11], that ER-positive and -negative diseases are fundamentally different [11-14], that molecular subtypes of breast cancer do exist [11,15-18], and that some special histological types of breast cancer are distinct entities at the molecular level [19-22]. Furthermore, they have led to the development of a molecular taxonomy that is currently being tested in clinical trials [16], and of prognostic 'gene signatures', some of which have already been approved by the US Food and Drug Administration [1,2,13,23].

Breast cancer assessment tools and optimizing adjuvant therapy

Recommendation of systemic adjuvant therapy and choice of optimal agents for early-stage breast cancer remains a challenge. Adjuvant therapy is indicated on the assumption of residual micrometastatic disease. Adjuvant assessment tools for prognosis and prediction of treatment benefit, including Adjuvant! Online, the St Gallen Consensus, Oncotype DX(®) and MammaPrint(®), aid clinical decision making. However, all of these tools have limitations that must be considered in their judicious application. Clinicopathological based tools are critically dependent on accurate, standardized measurement of parameters. Multigene tools are appealing for their objectivity and reproducibility, particularly regarding analysis of proliferation, but these approaches still overlook the biological heterogeneity within tumors evidenced by distinct cell subpopulations with different genomic patterns and function. The greatest treatment challenge remains for patients assessed as intermediate risk of relapse, a problem not overcome by multigene tools. Remarkable diversity in breast cancer dictates that adjuvant management must be biologically driven. Future identification of predictive biomarkers for specific chemotherapy sensitivity may allow targeted use of available agents, including anthracyclines, taxanes and DNA damaging agents. The presence of drug targets and targetable signaling pathways, rather than molecularly defined subgroups, may ultimately drive treatment decisions.

Use of ER/PR/HER2 subtypes in conjunction with the 2007 St Gallen Consensus Statement for early breast cancer

BackgroundThe 2007 St Gallen international expert consensus statement describes three risk categories and provides recommendations for treatment of early breast cancer. The set of recommendations on how to best treat primary breast cancer is recognized and used by clinicians worldwide. We now examine the variability of five-year survival of the 2007 St Gallen Risk Classifications utilizing the ER/PR/HER2 subtypes.MethodsUsing the population-based California Cancer Registry, 114,786 incident cases of Stages 1-3 invasive breast cancer diagnosed between 2000 and 2006 were identified. Cases were assigned to Low, Intermediate, or High Risk categories. Five-year-relative survival was computed for the three St Gallen risk categories and for the ER/PR/HER2 subtypes for further differentiation.Results and DiscussionThere were 9,124 (13%) cases classified as Low Risk, 44,234 (65%) cases as Intermediate Risk, and 14,340 (21%) as High Risk. Within the Intermediate Risk group, 33,735 (76%) were node-negative (Intermediate Risk 2) and 10,499 (24%) were node-positive (Intermediate Risk 3). For the High Risk group, 6,149 (43%) had 1 to 3 positive axillary lymph nodes (High Risk 4) and 8,191 (57%) had four or more positive lymph nodes (High Risk 5).Using five-year relative survival as the principal criterion, we found the following: a) There was very little difference between the Low Risk and Intermediate Risk categories; b) Use of the ER/PR/HER2 subtypes within the Intermediate and High Risk categories separated each into a group with better five-year survival (ER-positive) and a group with worse survival (ER-negative), irrespective of HER2-status; c) The heterogeneity of the High Risk category was most evident when one examined the ER/PR/HER2 subtypes with four or more positive axillary lymph nodes; (d) HER2-positivity did not always translate to worse survival, as noted when one compared the triple positive subtype (ER+/PR+/HER2+) to the triple negative subtype (ER-/PR-/HER2-); and (e) ER-negativity appeared to be a stronger predictor of poor survival than HER2-positivity.ConclusionThe use of ER/PR/HER2 subtype highlights the marked heterogeneity of the Intermediate and High Risk categories of the 2007 St Gallen statements. The use of ER/PR/HER2 subtypes and correlation with molecular classification of breast cancer is recommended.

134 Accumulation of p53 determined by immunohistochemistry as a prognostic marker in node negative breast cancer; analysis according to St Gallen consensus and intrinsic subtypes

134 Accumulation of p53 determined by immunohistochemistry as a prognostic marker in node negative breast cancer; analysis according to St Gallen consensus and intrinsic subtypes

Small Beginnings: Do They Matter? The Importance of Lymphovascular Invasion in Early Breast Cancer

Small Beginnings: Do They Matter? The Importance of Lymphovascular Invasion in Early Breast Cancer

Triple-negative receptor status and the 2007 St. Gallen consensus guidelines for early breast cancer

547 Background: The 2007 St Gallen consensus statement describes three risk categories and provides guidelines for treatment. We previously described the distribution of ER/PR/HER2 in breast cancer and analyzed the triple-negative subtype. We now examine the 2007 St Gallen consensus-risk categories with respect to the ER/PR/HER2 subtypes. Methods: Using the population-based California Cancer Registry from 2000 through 2006, we identified 63,925 cases of stages I-III first primary invasive female breast cancer with known status of ER, PR, and HER2. We retrospectively assigned cases to risk categories using pathologic tumor size and grade, age, and status of ER, PR, HER2, and lymph nodes. Extensive peritumoral vascular invasion was not readily available and could not be utilized. We determined the ER/PR/HER2 subtypes for each risk category and performed five-year relative survival. Results: Five year relative survival was 95% or greater for the low risk (LR), node-negative intermediate risk (NNIR), and node positive (1–3) intermediate risk (NPIR) categories. The 5-year relative survival for the node positive (1–3) high-risk (NPHR) category was 85% and 74% for the node positive (&gt; 3) high-risk (NPHR4) category; both had significantly worse survival than the low- and intermediate-risk categories. Classification of the risk categories by ER/PR/HER2 subtype showed no differences between the subtypes in the low risk category and clear differences in all other risk categories. All ER-negative subtypes had the worst survival within the NNIR, NPHR, and NPHR4 categories. The triple negative subtype had the worst survival in all of these risk categories. The greatest differences in 5-year relative survival were noted in the NPHR4 category with survival ranging from 85% in the ER+/PR+/HER2- subtype to 50% in the triple negative subtype. Conclusions: The St Gallen risk model clearly separates early breast cancer into three risk categories and helps with treatment decisions. Use of the ER/PR/HER2 subtypes within these categories clearly illustrates the heterogeneity of the intermediate and high risk categories and may prove to be important in this era of tailored therapy. No significant financial relationships to disclose.

Guideline-conform systemic treatment improves survival of breast cancer patients.

Abstract Abstract #2115 Purpose: The influence of guidelines for systemic therapy of early-stage breast cancer on survival of patients is still not clear. We investigated to which extend adherence to the St Gallen expert consensus recommendations can improve outcome of the disease.&amp;#x2028; Methods: Clinical characteristics and survival data of 2454 patients with early breast cancer which received primary therapy between 1992 and 2005 at a cancer referral center were collected by patient chart review, queries to local death registries and physicians involved in follow-up care. Patients were assigned one of three risk categories and an algorithm based on the St Gallen Expert Consensus 2005 on primary therapy of early-stage breast cancer was developed to estimate risk reduction by compliance to consensus recommendations using different Cox proportional hazard models.&amp;#x2028; Results: According to the St Gallen consensus recommendations, indication for endocrine therapy was classified as mandatory in 78.3% (n=1922) of patients and as optional in 3.7% (n=92). A significant risk reduction by guideline-compliant endocrine therapy in patients with mandatory indication (HR = 0.62, CI 0.48 - 0.81) and also in patients with optional indication for endocrine therapy (HR = 0.20, CI 0.04 - 0.90) was estimated after adjustment for prognostic factors, co-existing morbidity and adjuvant treatment measures. Based on the consensus, indication for chemotherapy was classified as mandatory in 42.5% (n=1044) of the patients and as optional in 53.7% (n=1318). A significant risk reduction was observed for both subgroups of patients after adjustment for prognostic factors, co-existing morbidity and adjuvant treatment measures (HR = 0.80, CI 0.59 - 1.09 and HR = 0.53, CI 0.33 - 0.86). Model based estimates of overall survival for the subgroup of patients for whom endocrine therapy is mandatory and chemotherapy is optional is shown in figure 1, even more pronounced effects were found in patients for whom both endocrine therapy and chemotherapy are mandatory.&amp;#x2028; &amp;#x2028; &amp;#x2028; &amp;#x2028; Conclusion: Systemic adjuvant therapy in accordance with the St Gallen consensus recommendations 2005 is clearly associated with prolonged survival. Chemotherapy and endocrine therapy are beneficial when applied in accordance with the indications of St. Gallen consensus recommendations. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 2115.