Search

BackgroundPrognostic factors in breast cancer are often measured on a continuous scale, but categorized for clinical decision-making. The primary aim of this study was to evaluate if accounting for continuous non-linear effects of the three factors age at diagnosis, tumor size, and number of positive lymph nodes improves prognostication. These factors will most likely be included in the management of breast cancer patients also in the future, after an expected implementation of gene expression profiling for adjuvant treatment decision-making.MethodsFour thousand four hundred forty seven and 1132 women with primary breast cancer constituted the derivation and validation set, respectively. Potential non-linear effects on the log hazard of distant recurrences of the three factors were evaluated during 10years of follow-up. Cox-models of successively increasing complexity: dichotomized predictors, predictors categorized into three or four groups, and predictors transformed using fractional polynomials (FPs) or restricted cubic splines (RCS), were used. Predictive performance was evaluated by Harrells C-index.ResultsUsing FP-transformations, non-linear effects were detected for tumor size and number of positive lymph nodes in univariable analyses. For age, non-linear transformations did, however, not improve the model fit significantly compared to the linear identity transformation. As expected, the C-index increased with increasing model complexity for multivariable models including the three factors. By allowing more than one cut-point per factor, the C-index increased from 0.628 to 0.674. The additional gain, as measured by the C-index, when using FP- or RCS-transformations was modest (0.695 and 0.696, respectively). The corresponding C-indices for these four models in the validation set, based on the same transformations and parameter estimates from the derivation set, were 0.675, 0.700, 0.706, and 0.701.ConclusionsCategorization of each factor into three to four gro, Funding Agencies|Swedish Cancer Society; Swedish Research Council; Gunnar Nilsson Cancer Foundation; Swedish Breast Cancer Association; Swedish Cancer and Allergy Foundation; Mrs. Berta Kamprad Foundation; Anna and Edwin Bergers Foundation; Skane County Councils Research and Development Foundation; Governmental Funding of Clinical Research within the National Health Service

Background: The St Gallen surrogate definition of the intrinsic subtypes of breast cancer consist of five subgroups based on estrogen receptor (ER), progesterone receptor (PgR), human epidermal growth factor receptor type 2 (HER2), and Ki-67. PgR and Ki-67 are used for discriminating between the Luminal A-like and Luminal B-like (HER2-negative) subtypes. Histological grade (G) has prognostic value in breast cancer; however, its relationship to the St Gallen subtypes is not clear. Based on a previous pilot study, we hypothesized that G could be a primary discriminator for ER-positive/HER2-negative breast cancers that were G1 or G3, whereas Ki-67 and PgR could provide additional prognostic information specifically for patients with G2 tumors. To test this hypothesis, a larger patient cohort was examined. Patients and methods: Six hundred seventy-one patients (amp;gt;= 35 years of age, pT1-2, pN0-1) with ER-positive/HER2-negative breast cancer and complete data for PgR, Ki-67, G, lymph node status, tumor size, age, and distant disease-free survival (DDFS; median follow-up 9.2 years) were included. Results: Luminal A-like tumors were mostly G1 or G2 (90%) whereas Luminal B-like tumors were mostly G2 or G3 (87%) and corresponded with good and poor DDFS, respectively. In Luminal B-like tumors that were G1 (n = 23), no metastasis occurred, whereas 14 of 40 Luminal A-like tumors that were G3 metastasized. In the G2 subgroup, low PgR and high Ki-67 were associated with an increased risk of distant metastases, hazard ratio (HR) and 95% confidence interval (CI) 1.8 (0.95-3.4) and 1.5 (0.80-2.8), respectively. Conclusions: Patients with ER-positive/HER2-negative/G1 breast cancer have a good prognosis, similar to that of Luminal A-like, while those with ER-positive/HER2-negative/G3 breast cancer have a worse prognosis, similar to that of Luminal B-like, when assessed independently of PgR and Ki-67. Therapy decisions based on Ki-67 and PgR might thus be restricted to the subgroup, Funding Agencies|Swedish Cancer Society; Gunnar Nilsson Cancer Foundation; Mrs. Berta Kamprad Foundation; Anna and Edwin Bergers Foundation; Swedish Breast Cancer Association (BRO); Swedish Cancer and Allergy Foundation; Scientific Committee of Blekinge County Council; Skane County Councils Research and Development Foundation; Governmental Funding of Clinical Research within the National Health Service

Experimental models implicate protease activated receptors (PARs) as important sensors of the proteolytic tumor microenvironment during breast cancer development. However, the role of the major PARs, PAR-1 and PAR-2, in human breast tumors remains to be elucidated. Here, we have investigated how PAR-1 and PAR-2 protein expression correlate with established clinicopathological variables and patient outcome in a well-characterized cohort of 221 breast cancer patients. Univariable and multivariable hazard ratios (HR) were estimated by the Cox proportional hazards model, distant disease-free survival (DDFS) and overall survival by the Kaplan-Meier method, and survival in different strata was determined by the log-rank test. Associations between PARs and clinicopathological variables were analyzed using Pearson's χ2-test. We find that PAR-2 associates with DDFS (HR = 3.1, P = 0.003), whereas no such association was found with PAR-1 (HR = 1.2, P = 0.6). Interestingly, the effect of PAR-2 was confined to the ER-positive sub-group (HR = 5.5, P = 0.003 vs. HR = 1.2 in ER-negative; P = 0.045 for differential effect), and PAR-2 was an independent prognostic factor specifically in ER-positive tumors (HR = 3.9, P = 0.045). On the contrary, PAR-1 correlated with worse prognosis specifically in the ER-negative group (HR = 2.6, P = 0.069 vs. HR = 0.5, P = 0.19 in ER-positive; P = 0.026 for differential effect). This study provides novel insight into the respective roles of PAR-1 and PAR-2 in human breast cancer and suggests a hitherto unknown association between PARs and ER signaling that warrants further investigation.

Proliferation, either as the main common denominator in genetic profiles, or in the form of single factors such as Ki67, is recommended for clinical use especially in estrogen receptor-positive (ER) patients. However, due to high costs of genetic profiles and lack of reproducibility for Ki67, studies on other proliferation factors are warranted. The aim of the present study was to evaluate the prognostic value of the proliferation factors mitotic activity index (MAI), phosphohistone H3 (PPH3), cyclin B1, cyclin A and Ki67, alone and in combinations. In 222 consecutive premenopausal node-negative breast cancer patients (87% without adjuvant medical treatment), MAI was assessed on whole tissue sections (predefined cut-off ≥10 mitoses), and PPH3, cyclin B1, cyclin A, and Ki67 on tissue microarray (predefined cut-offs 7th decile). In univariable analysis (high versus low) the strongest prognostic proliferation factor for 10-year distant disease-free survival was MAI (Hazard Ratio (HR)=3.3, 95% Confidence Interval (CI): 1.8-6.1), followed by PPH3, cyclin A, Ki67, and cyclin B1. A combination variable, with patients with MAI and/or cyclin A high defined as high-risk, had even stronger prognostic value (HR=4.2, 95%CI: 2.2-7). When stratifying for ER-status, MAI was a significant prognostic factor in ER-positive patients only (HR=7.0, 95%CI: 3.1-16). Stratified for histological grade, MAI added prognostic value in grade 2 (HR=7.2, 95%CI: 3.1-38) and grade 1 patients. In multivariable analysis including HER2, age, adjuvant medical treatment, ER, and one proliferation factor at a time, only MAI (HR=2.7, 95%CI: 1.1-6.7), and cyclin A (HR=2.7, 95%CI: 1.2-6.0) remained independently prognostic. In conclusion this study confirms the strong prognostic value of all proliferation factors, especially MAI and cyclin A, in all patients, and more specifically in ER-positive patients, and patients with histological grade 2 and 1. Additionally, by combining two proliferation factors, an e

OBJECTIVE: The concordance index (c-index) is the standard way of evaluating the performance of prognostic models in the presence of censored data. Constructing prognostic models using artificial neural networks (ANNs) is commonly done by training on error functions which are modified versions of the c-index. Our objective was to demonstrate the capability of training directly on the c-index and to evaluate our approach compared to the Cox proportional hazards model. METHOD: We constructed a prognostic model using an ensemble of ANNs which were trained using a genetic algorithm. The individual networks were trained on a non-linear artificial data set divided into a training and test set both of size 2000, where 50% of the data was censored. The ANNs were also trained on a data set consisting of 4042 patients treated for breast cancer spread over five different medical studies, 2/3 used for training and 1/3 used as a test set. A Cox model was also constructed on the same data in both cases. The two models' c-indices on the test sets were then compared. The ranking performance of the models is additionally presented visually using modified scatter plots. RESULTS: Cross validation on the cancer training set did not indicate any non-linear effects between the covariates. An ensemble of 30 ANNs with one hidden neuron was therefore used. The ANN model had almost the same c-index score as the Cox model (c-index=0.70 and 0.71, respectively) on the cancer test set. Both models identified similarly sized low risk groups with at most 10% false positives, 49 for the ANN model and 60 for the Cox model, but repeated bootstrap runs indicate that the difference was not significant. A significant difference could however be seen when applied on the non-linear synthetic data set. In that case the ANN ensemble managed to achieve a c-index score of 0.90 whereas the Cox model failed to distinguish itself from the random case (c-index=0.49). CONCLUSIONS: We have found empirical evidence

The aim was to confirm a previously defined prognostic index, combining a proliferation marker, histological grade, and estrogen receptor (ER) in different subsets of primary N0/N1 chemo-naïve breast cancer patients.

Global gene expression profiles, consisting mainly of genes associated with proliferation, have been shown to subdivide histological grade 2 breast cancers into groups with different prognosis. We raised the question whether this subdivision could be done using a single proliferation marker, cyclin A. Furthermore, we combined cyclin A (CA), histological grade (G), and estrogen receptor-ER (E) into a new variable, CAGE. Our aim was to investigate not only the prognostic importance of cyclin A alone but also the value of the combination variable CAGE. In 219 premenopausal node-negative patients, cyclin A was assessed using immunohistochemistry on tissue microarrays. High cyclin A was defined as above the seventh decile of positive cells. Only 13% of the patients received adjuvant systemic therapy. Cox proportional hazards regression was used to model the impact of the factors on distant disease-free survival (DDFS). Cyclin A divided histological grade 2 tumors into two groups with significantly different DDFS (hazard ratio [HR]: 15, P < 0.001). When stratifying for ER status, cyclin A was a prognostic factor only in the ER positive subgroup. We found that CAGE was an independent prognostic factor for DDFS in multivariate analysis (HR: 4.1, P = 0.002), together with HER2. CAGE and HER2 identified 53% as low-risk patients with a 5-year DDFS of 95%. A new prognostic variable was created by combining cyclin A, histological grade, and ER (CAGE). CAGE together with HER2 identified a large low-risk group for whom adjuvant chemotherapy will have limited efficacy and may be avoided.

Introduction: Breast cancer is a profoundly heterogeneous disease with respect to biologic and clinical behavior. Gene-expression profiling has been used to dissect this complexity and to stratify tumors into intrinsic gene-expression subtypes, associated with distinct biology, patient outcome, and genomic alterations. Additionally, breast tumors occurring in individuals with germline BRCA1 or BRCA2 mutations typically fall into distinct subtypes. Methods: We applied global DNA copy number and gene-expression profiling in 359 breast tumors. All tumors were classified according to intrinsic gene-expression subtypes and included cases from genetically predisposed women. The Genomic Identification of Significant Targets in Cancer (GISTIC) algorithm was used to identify significant DNA copy-number aberrations and genomic subgroups of breast cancer. Results: We identified 31 genomic regions that were highly amplified in > 1% of the 359 breast tumors. Several amplicons were found to co-occur, the 8p12 and 11q13.3 regions being the most frequent combination besides amplicons on the same chromosomal arm. Unsupervised hierarchical clustering with 133 significant GISTIC regions revealed six genomic subtypes, termed 17q12, basal-complex, luminal-simple, luminal-complex, amplifier, and mixed subtypes. Four of them had striking similarity to intrinsic gene-expression subtypes and showed associations to conventional tumor biomarkers and clinical outcome. However, luminal A-classified tumors were distributed in two main genomic subtypes, luminal-simple and luminal-complex, the former group having a better prognosis, whereas the latter group included also luminal B and the majority of BRCA2-mutated tumors. The basal-complex subtype displayed extensive genomic homogeneity and harbored the majority of BRCA1-mutated tumors. The 17q12 subtype comprised mostly HER2-amplified and HER2-enriched subtype tumors and had the worst prognosis. The amplifier and mixed subtypes contained tumors f

A large proportion of breast cancer patients are treated with adjuvant chemotherapy after the primary operation, but some will recur in spite of this treatment. In order to achieve an improved and more individualised therapy, our knowledge in mechanisms for drug resistance needs to be increased. We have investigated to what extent cDNA microarray measurements could distinguish the likelihood of recurrences after adjuvant CMF (cyclophosphamide, methotrexate and 5-fluorouracil) treatment of premenopausal, lymph node positive breast cancer patients, and have also compared this with the corresponding performance when using conventional clinical variables. We tried several gene selection strategies, and built classifiers using the resulting gene lists. The best performing classifier with odds ratio (OR) = 6.5 (95% confidence interval (CI) = 1.4-62) did not outperform corresponding classifiers based on clinical variables. For the clinical variables, calibrated on the samples, either using all the clinical parameters or the Nottingham Prognostic Index (NPI) parameters, the areas under the receiver operating characteristics (ROC) curve were 0.78 and 0.79, respectively. The ORs at 90% sensitivity were 15 (95% CI = 3.1-140) and 10 (95% CI = 2.1-97), respectively. Our data have provided evidence for a comparable prediction of clinical outcome in CMF-treated breast cancer patients using conventional clinical variables and gene expression based markers. (c) 2006 Elsevier Ltd. All rights reserved.