scholarly journals Machine learning predicts rapid relapse of triple negative breast cancer

2019 ◽  
Author(s):  
Yiqing Zhang ◽  
William Nock ◽  
Meghan Wyse ◽  
Zachary Weber ◽  
Elizabeth Adams ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Support Vector ◽  
Chi Square ◽  
Genomic Features ◽  
Immune Signatures ◽  
Metastatic Relapse ◽  
Significant Difference

ABSTRACTPurposeMetastatic relapse of triple-negative breast cancer (TNBC) within 2 years of diagnosis is associated with particularly aggressive disease and a distinct clinical course relative to TNBCs that relapse beyond 2 years. We hypothesized that rapid relapse TNBCs (rrTNBC; metastatic relapse or death <2 years) reflect unique genomic features relative to late relapse (lrTNBC; >2 years).Patients and MethodsWe identified 453 primary TNBCs from three publicly-available datasets and characterized each as rrTNBc, lrTNBC, or ‘no relapse’ (nrTNBC: no relapse/death with at least 5 years follow-up). We compiled primary tumor clinical and multi-omic data, including transcriptome (n=453), copy number alterations (CNAs; n=317), and mutations in 171 cancer-related genes (n=317), then calculated published gene expression and immune signatures.ResultsPatients with rrTNBC were higher stage at diagnosis (Chi-square p<0.0001) while lrTNBC were more likely to be non-basal PAM50 subtype (Chi-square p=0.03). Among 125 expression signatures, five immune signatures were significantly higher in nrTNBCs while lrTNBC were enriched for eight estrogen/luminal signatures (all FDR p<0.05). There was no significant difference in tumor mutation burden or percent genome altered across the groups. Among mutations, onlyTP53mutations were significantly more frequent in rrTNBC compared to lrTNBC (Fisher exact FDR p=0.009). To develop an optimal classifier, we used 77 significant clinical and ‘omic features to evaluate six modeling approaches encompassing simple, machine learning, and artificial neural network (ANN). Support vector machine outperformed other models with average receiver-operator characteristic area under curve >0.75.ConclusionsWe provide a new approach to define TNBCs based on timing of relapse. We identify distinct clinical and genomic features that can be incorporated into machine learning models to predict rapid relapse of TNBC.

scholarly journals Breast Cancer Type Classification Using Machine Learning

2021 ◽  
Vol 11 (2) ◽  
pp. 61
Author(s):  
Jiande Wu ◽  
Chindo Hicks
Keyword(s):  
Breast Cancer ◽  
Gene Expression ◽  
Machine Learning ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Genomic Research ◽  
Support Vector ◽  
Cancer Type ◽  
Classification Models

Background: Breast cancer is a heterogeneous disease defined by molecular types and subtypes. Advances in genomic research have enabled use of precision medicine in clinical management of breast cancer. A critical unmet medical need is distinguishing triple negative breast cancer, the most aggressive and lethal form of breast cancer, from non-triple negative breast cancer. Here we propose use of a machine learning (ML) approach for classification of triple negative breast cancer and non-triple negative breast cancer patients using gene expression data. Methods: We performed analysis of RNA-Sequence data from 110 triple negative and 992 non-triple negative breast cancer tumor samples from The Cancer Genome Atlas to select the features (genes) used in the development and validation of the classification models. We evaluated four different classification models including Support Vector Machines, K-nearest neighbor, Naïve Bayes and Decision tree using features selected at different threshold levels to train the models for classifying the two types of breast cancer. For performance evaluation and validation, the proposed methods were applied to independent gene expression datasets. Results: Among the four ML algorithms evaluated, the Support Vector Machine algorithm was able to classify breast cancer more accurately into triple negative and non-triple negative breast cancer and had less misclassification errors than the other three algorithms evaluated. Conclusions: The prediction results show that ML algorithms are efficient and can be used for classification of breast cancer into triple negative and non-triple negative breast cancer types.

scholarly journals Genomic Features of Rapid Versus Late Relapse in Triple Negative Breast Cancer 

2021 ◽  
Author(s):  
Yiqing Zhang ◽  
Sarah Asad ◽  
Zachary Weber ◽  
David Tallman ◽  
William Nock ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Logistic Regression ◽  
T Cell ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Late Relapse ◽  
Primary Tumors ◽  
Genomic Features ◽  
Immune Signatures ◽  
M1 Macrophage

Abstract Background: Triple-negative breast cancer (TNBC) is a heterogeneous disease and we have previously shown that rapid relapse of TNBC is associated with distinct sociodemographic features. We hypothesized that rapid versus late relapse in TNBC is also defined by distinct clinical and genomic features of primary tumors.Methods: Using three publicly-available datasets, we identified 453 patients diagnosed with primary TNBC with adequate follow-up to be characterized as ‘rapid relapse’ (rrTNBC; relapse/death ≤2 years of diagnosis), ‘late relapse’ (lrTNBC; >2 years) or ‘no relapse’ (nrTNBC: >5 years no relapse/death). We explored basic clinical and primary tumor multi-omic data, including whole transcriptome (n=453), and whole genome copy number and mutation data for 171 cancer-related genes (n=317). Association of rapid relapse with clinical and genomic features were assessed using Pearson chi-squared tests, t-tests, ANOVA, and Fisher exact tests. We evaluated logistic regression models of clinical features with subtype versus two models that integrated significant genomic features.Results: Relative to nrTNBC, both rrTNBC and lrTNBC had significantly lower immune signatures and immune signatures were highly correlated to anti-tumor CD8 T-cell, M1 macrophage, and gamma-delta T-cell CIBERSORT inferred immune subsets. Intriguingly, lrTNBCs were enriched for luminal signatures. There was no difference in tumor mutation burden or percent genome altered across groups. Logistic regression mModels that incorporate genomic features significantly outperformed standard clinical/subtype models in training (n=63 patients), testing (n=63) and independent validation (n=34) cohorts, although performance of all models were overall modest. Conclusions: We identify clinical and genomic features associated with rapid relapse TNBC for further study of this aggressive TNBC subset.

scholarly journals Genomic features of rapid versus late relapse in triple negative breast cancer

BMC Cancer ◽  
2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yiqing Zhang ◽  
Sarah Asad ◽  
Zachary Weber ◽  
David Tallman ◽  
William Nock ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Logistic Regression ◽  
T Cell ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Late Relapse ◽  
Primary Tumors ◽  
Genomic Features ◽  
Immune Signatures ◽  
M1 Macrophage

Abstract Background Triple-negative breast cancer (TNBC) is a heterogeneous disease and we have previously shown that rapid relapse of TNBC is associated with distinct sociodemographic features. We hypothesized that rapid versus late relapse in TNBC is also defined by distinct clinical and genomic features of primary tumors. Methods Using three publicly-available datasets, we identified 453 patients diagnosed with primary TNBC with adequate follow-up to be characterized as ‘rapid relapse’ (rrTNBC; distant relapse or death ≤2 years of diagnosis), ‘late relapse’ (lrTNBC; > 2 years) or ‘no relapse’ (nrTNBC: > 5 years no relapse/death). We explored basic clinical and primary tumor multi-omic data, including whole transcriptome (n = 453), and whole genome copy number and mutation data for 171 cancer-related genes (n = 317). Association of rapid relapse with clinical and genomic features were assessed using Pearson chi-squared tests, t-tests, ANOVA, and Fisher exact tests. We evaluated logistic regression models of clinical features with subtype versus two models that integrated significant genomic features. Results Relative to nrTNBC, both rrTNBC and lrTNBC had significantly lower immune signatures and immune signatures were highly correlated to anti-tumor CD8 T-cell, M1 macrophage, and gamma-delta T-cell CIBERSORT inferred immune subsets. Intriguingly, lrTNBCs were enriched for luminal signatures. There was no difference in tumor mutation burden or percent genome altered across groups. Logistic regression mModels that incorporate genomic features significantly outperformed standard clinical/subtype models in training (n = 63 patients), testing (n = 63) and independent validation (n = 34) cohorts, although performance of all models were overall modest. Conclusions We identify clinical and genomic features associated with rapid relapse TNBC for further study of this aggressive TNBC subset.

scholarly journals Machine Learning Helps Identify New Drug Mechanisms in Triple-Negative Breast Cancer

2018 ◽  
Vol 17 (3) ◽  
pp. 251-259 ◽  
Author(s):  
Arjun P. Athreya ◽  
Alan J. Gaglio ◽  
Junmei Cairns ◽  
Krishna R. Kalari ◽  
Richard M. Weinshilboum ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
New Drug

PD-L1 expression and TOP3A mutation as prognostic factors for adjuvant chemotherapy in triple negative breast cancer.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 541-541
Author(s):  
Xue Wang ◽  
Peng Yuan ◽  
Feng Du ◽  
Lina Cui ◽  
Fangchao Zheng ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Adjuvant Chemotherapy ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Disease Free Survival ◽  
Genetic Alterations ◽  
Functional Enrichment ◽  
Pathway Enrichment Analysis ◽  
Genetic Characteristics ◽  
Significant Difference

541 Background: Triple negative breast cancer (TNBC) is a highly aggressive subtype of breast cancer that is markedly heterogeneous and lacks specific targets. The aim of this study is to explore potential predictors and therapeutic targets based on clinical and genetic characteristics. Methods: 138 patients with triple-negative breast cancer after surgical treatment were 1:1 randomly assigned to the paclitaxel combined with carboplatin (TCb) group or the epirubicin combined with cyclophosphamide sequential paclitaxel (EC-T) adjuvant chemotherapy group. PD-L1 was retrospectively analyzed by surgically resected specimens, and 733 cancer-related genes were detected by NGS. Pathway enrichment analysis was performed using DAVID for functional enrichment genetic alterations. Cox regression models and Kaplan-Meier were used to evaluate disease-free survival (DFS). Results: In this study, there was no significant difference in DFS between the TCb and EC-T groups. 31 (22.5%) of 138 TNBC patients were positive for PD-L1 expression, including 15 (10.9%) patients positive for PD-L1 in tumor cells (TCs) and 29 (21.0%) patients positive for PD-L1 in tumor-infiltrating immune cells (TICs). Patients with positive PD-L1 expression, either in TCs or TICs, achieved better DFS [HR=0.13 (95% CI: 0.02-0.93), p=0.016], the difference was also shown in the EC-T group [HR=0 (95% CI: 0- inf), p=0.037], but not in the TCb group [HR=0 (95% CI: 0.04-2.1), p=0.189]. In addition, we identified 7 patients with mutations in DNA topoisomerase IIIα(TOP3A), a homologous recombination (HR)-related gene, and patients with mutations in this gene had worse DFS than those without mutations [HR=4]. However, there was no statistically significant association between BRCA mutation and response to either therapeutic regimens. Conclusions: In this TNBC patient population, immunohistochemistry (IHC) and NGS analyses identified potential prognostic markers. PD-L1 positive and TOP3A mutation were significantly associated with early triple-negative breast cancer prognosis.

Expression of Transgelin in Triple Negative and Non Triple Negative Breast Cancer: A Differential Study

2020 ◽  
Vol 106 (1_suppl) ◽  
pp. 20-20
Author(s):  
NS Tolba ◽  
AS Alsedfy ◽  
SW Skandar ◽  
YM El-Kerm
Keyword(s):  
Breast Cancer ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Molecular Subtypes ◽  
High Rate ◽  
Targeted Treatment ◽  
Her2 Status ◽  
Altered Expression ◽  
Wide Range ◽  
Significant Difference

Introduction: Triple negative breast cancer (TNBC) is defined by the absence of ER expression, PR expression and HER2 amplification. No targeted treatment is available for TNBC and chemotherapy remains the best therapeutic option. However, in the case of recurrence or chemo-resistance, therapeutic options are very limited. TNBC presents a high rate of proliferation and is highly aggressive having low survival rate. As the complexity of this disease is being simplified over time, new targets are also being discovered for the treatment of this disease. Therefore, there is still need for new biomarkers, which would serve for targeted treatment. Transgelin was proposed as a new potential cancer biomarker. Altered expression of Transgelin has been described in a wide range of cancers, often with contradictory results. The aim of the study was to compare Transgelin expression across molecular subtypes of breast cancer, to identify if it can be used as a future molecular targeted protein for TNBC. Material and Methods: Transgelin immunohistochemistry was applied on 60 retrospectively collected paraffin blocks of patients presenting with invasive breast carcinoma (NST) having different molecular subtypes. Blocks were collected between 2015 and 2016 from Pathology department, Medical Research Institute, Egypt. Her2 equivocal cases were excluded from the study. Results: Transgelin expression was positive in 23 cases and negative in 37 cases. There was a statistically significant difference between (Transgelin +) and (Transgelin -) cases being highly expressed in TNBC in comparison to other molecular subtypes. It was also highly expressed in tumors with large size, high grade, positive lymph-vascular invasion status & lymph node metastasis. There was no statistically significant difference between (Transgelin+) and (Transgelin-) as regards age and Her2 status. Conclusions: Transgelin is an aggressive biomarker differentially expressed among the molecular breast cancer subtypes with high expression in TNBC. Transgelin may provide a potential target for future treatment of TNBC.

scholarly journals Clinicopathological, Treatment and Event-Free Survival Characteristics in a Moroccan Population of Triple-Negative Breast Cancer

2020 ◽  
Vol 14 ◽  
pp. 117822342090642
Author(s):  
Fatima Zahra Mouh ◽  
Meriem Slaoui ◽  
Rachid Razine ◽  
Mohammed EL Mzibri ◽  
Mariam Amrani
Keyword(s):  
Breast Cancer ◽  
Retrospective Study ◽  
Survival Rate ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Menopausal Status ◽  
Age At Menarche ◽  
Event Free Survival ◽  
Free Survival ◽  
Significant Difference

Introduction: Triple-negative breast cancer (TNBC) is a group of breast carcinoma characterized by the lack of expression of estrogen and progesterone hormone receptors (ER, PgR) and HER2. This form is also characterized by its aggressiveness, a low survival rate, and the absence of targeted therapies. This study was planned to evaluate the clinical features, treatment, and prognosis characteristics of TNBC in a population of Moroccan patients. Methods: In this retrospective study, a total of 905 patients diagnosed with breast cancer at the National Institute of Oncology in Rabat, Morocco, have been included. Based on molecular subtype, patients were divided into 2 categories: TNBC and non-TNBC patients. Data were recorded from patients’ medical files and analyzed using SPSS 13.0 software (IBM). Results: Overall, 17% of the patients had TNBC. At diagnosis, the median age of TNBC cases was 47 years, with extreme ages of 40 and 55 years. The median follow-up time was 30 months (10-53 months) and the 3-year survival rate was 76%. No significant difference was observed among the patients in terms of age at diagnosis, age at menarche, age at the time of first birth, nulliparity, oral contraception, and family history of breast cancer. Menopausal status and the number of pregnancy were significantly higher in the non-TNBC group. The percentage of grade 3 (G3) tumors was higher in the TNBC group ( P < .001). Using neoadjuvant, adjuvant chemotherapy and radiotherapy, a net benefit in the event-free survival was registered for the 2 groups. Conclusions: This retrospective study was very informative and showed that women with TNBC had a less favorable prognosis than non-TNBC cases. Clinical data demonstrated that risk factors including age, premenopausal status, parity, hormonal contraceptive use, advanced disease, and a high histologic grade were independently associated with TNBC. However, large tumors and high Scarff-Bloom and Richardson grade prevail in TNBC cases with a higher incidence of lymph node metastases.

Demography and treatment outcomes of triple-negative breast cancer: A single institution study.

2011 ◽  
Vol 29 (27_suppl) ◽  
pp. 179-179
Author(s):  
S. E. James ◽  
P. Vos ◽  
T. Biswas
Keyword(s):  
Breast Cancer ◽  
Treatment Outcomes ◽  
Median Survival ◽  
Triple Negative Breast Cancer ◽  
White Women ◽  
Triple Negative ◽  
Neoadjuvant Treatment ◽  
Stage I ◽  
Chinese Patients ◽  
Significant Difference

179 Background: Triple negative breast cancer (TNBC) constitutes about 11-20% of all cases of breast cancer. TNBC has particularly aggressive clinical course and worse prognosis especially in African American (AA) women compared to white women. This retrospective study aims to investigate TNBC in the eastern North Carolina (NC) population, report treatment outcomes and determine the rate of recurrence. Methods: 202 TNBC patients treated at our institution between 1/01 and 11/09 were included in this analysis. Eastern NC has an AA population higher than the national average (30.2% vs. 12.4%) which is reflected in our patient demographics with 111 (55%) AA, 84 (41.6%) White, 5 (2.5%) Hispanic and 2 (1%) Chinese patients. Stage distribution was 41 (20%) stage I, 97 (48%) stage II, 37 (18%) stage III, 16 (8%) stage IV disease. Six patients (3%) had a diagnosis of DCIS and 5 (3%) were unknown. Majority was (79%) poorly differentiated carcinoma followed by moderately differentiated (13%) carcinoma. Variables were compared between groups using two-sample t-tests. Kaplan Meier curves were generated for overall survival (OS). Group was compared by log rank test. Results: The median FU time was 4.73 years (1.2–10.2). Median age at diagnosis was 53 (22-88) years overall, 51 years for AA and 57 years for White (p=0.01) patients. The overall median survival was 32.18 months with no difference between AA (32.39 months) and white (29.7 months) patients (p=0.785). 160 patients with stage I-III disease received chemotherapy with 96 patients (60%) receiving adjuvant and 64 (40%) receiving neoadjuvant therapy. The median survival for adjuvant patients was 37.08 months (6.87–113) vs 26.48 months (1.51–87.49) for neoadjuvant patients (p=0.001). 27 (14%) patients with stage I-III diseases experienced a local recurrence, 31 patients (15%) experienced distal recurrence with a median time to failure of 25.6 months. Conclusions: This study demonstrated no significant difference in median survival between AA and white patients; however, AA women were diagnosed at a younger age compared to white women. Patients treated with adjuvant chemotherapy had a significantly greater median survival when compared to patients receiving neoadjuvant treatment.

Efficacy of platinum-based chemotherapy in advanced triple-negative breast cancer in association with homologous recombination deficiency.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e13051-e13051
Author(s):  
Yimeng Chen ◽  
Lina Cui ◽  
Bei Zhang ◽  
Xiaochen Zhao ◽  
Binghe Xu
Keyword(s):  
Breast Cancer ◽  
Homologous Recombination ◽  
Free Group ◽  
Triple Negative Breast Cancer ◽  
Triple Negative ◽  
Cox Regression ◽  
Deleterious Mutation ◽  
Homologous Recombination Deficiency ◽  
Platinum Based Chemotherapy ◽  
Significant Difference

e13051 Background: Previous studies have identified that at least 50% of triple negative breast cancer (TNBC) harbor mutation characteristics of homologous recombination deficiency (HRD). Thus, more sophisticated research into comprehensive genomic profiling of HRD is urgently needed. Whereas BRCA1/2-deficient advanced TNBC patients are sensitive to treatment with platinum, it is not yet clear whether HRD status could predict platinum response. Methods: 3DMed-HRD algorithm was developed based on loss of heterozygosity score (LOH), telomeric allelic imbalance score (TAI) and large-scale state transition score (LST) as previously described. HRD status was defined as HRD positive (deleterious mutation in BRCA1/2 or HRD score ≥30) or HRD negative (no deleterious mutation in BRCA1/2 and HRD score < 30). Tumor samples from 207 TNBC patients were analyzed by next-generation sequencing. Deleterious or suspected deleterious mutations were included for analysis. Among the overall cohort, 34 patients with advanced TNBC treated with chemotherapy were analyzed. Cox regression model was applied to evaluate the relationship between HRD status and clinical outcomes. Results: Deleterious BRCA1/2 mutations were detected in 22.2% (46/207) of TNBC patients as well as 54.6% (113/207) were defined as HRD positive. The most frequent mutations in HRD-positive patients were TP53 (93.5%), MYC (29.0%), PIK3R1 (22.6%), PTEN (22.6%) and MCL1 (19.4%), while TP53 (77.8%), MYC (29.6%), PIK3CA (18.5%), KMT2C (14.8%) and RIT1 (14.8%) enriched in HRD-negative patients. Mutations in DNA Damage Response (DDR), P53, Checkpoint and Receptor Tyrosine Kinase (RTK) pathways were most involved in HRD positive patients. In advanced TNBC cohort, 19 patients received platinum-based and 15 received platinum-free chemotherapy in the first-line treatment. The progression-free survival (PFS) of the platinum-based group was longer than that of the platinum-free group (media PFS 9.1 vs 2.2 months, HR 0.44, 95%CI, 0.20-0.94, P = 0.009). In HRD-positive patients, median PFS was significantly longer in platinum-based group (N = 6) than platinum-free group (N = 8) (media PFS 13.6 vs 1.9 months, HR 0.30, 95%CI, 0.09-0.95, P = 0.008). No significant difference in PFS between platinum-based and platinum-free group (p = 0.332) in patients with HRD-negative tumors. Patients with mutations in homologous recombination (HR) pathway had a worse PFS compared to wild-type in platinum-free group (1.4 vs 3.0 months, p = 0.050). Conclusions: Our findings illustrate the potential of HRD status as a marker to guide chemotherapy in advanced TNBC. In HRD positive patients, platinum-based chemotherapy might be a preferable regimen. Patients with mutations in HR pathway had a shorter PFS in platinum-free group. Prospective study with a larger sample-size is needed for further validation of our findings.

Sign in / Sign up

Export Citation Format

Share Document