Machine Learning Frameworks to Predict Neoadjuvant Chemotherapy Response in Breast Cancer Using Clinical and Pathological Features

2021 ◽  
pp. 66-80
Author(s):  
Nicholas Meti ◽  
Khadijeh Saednia ◽  
Andrew Lagree ◽  
Sami Tabbarah ◽  
Majid Mohebpour ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Neoadjuvant Chemotherapy ◽  
Tumor Size ◽  
Locally Advanced ◽  
Her2 Status ◽  
Support Vector ◽  
Chemotherapy Response ◽  
Nottingham Grade ◽  
Er Status

PURPOSE Neoadjuvant chemotherapy (NAC) is used to treat locally advanced breast cancer (LABC) and high-risk early breast cancer (BC). Pathological complete response (pCR) has prognostic value depending on BC subtype. Rates of pCR, however, can be variable. Predictive modeling is desirable to help identify patients early who may have suboptimal NAC response. Here, we test and compare the predictive performances of machine learning (ML) prediction models to a standard statistical model, using clinical and pathological data. METHODS Clinical and pathological variables were collected in 431 patients, including tumor size, patient demographics, histological characteristics, molecular status, and staging information. A standard multivariable logistic regression (MLR) was developed and compared with five ML models: k-nearest neighbor classifier, random forest (RF) classifier, naive Bayes algorithm, support vector machine, and multilayer perceptron model. Model performances were measured using a receiver operating characteristic (ROC) analysis and statistically compared. RESULTS MLR predictors of NAC response included: estrogen receptor (ER) status, human epidermal growth factor-2 (HER2) status, tumor size, and Nottingham grade. The strongest MLR predictors of pCR included HER2+ versus HER2− BC (odds ratio [OR], 0.13; 95% CI, 0.07 to 0.23; P < .001) and Nottingham grade G3 versus G1-2 (G1-2: OR, 0.36; 95% CI, 0.20 to 0.65; P < .001). The area under the curve (AUC) for the MLR was AUC = 0.64. Among the various ML models, an RF classifier performed best, with an AUC = 0.88, sensitivity of 70.7%, and specificity of 84.6%, and included the following variables: menopausal status, ER status, HER2 status, Nottingham grade, tumor size, nodal status, and presence of inflammatory BC. CONCLUSION Modeling performances varied between standard versus ML classification methods. RF ML classifiers demonstrated the best predictive performance among all models.

Single nucleotide polymorphisms to predict for neoadjuvant chemotherapy in breast cancer according to estrogen receptor (ER) status.

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. 544-544
Author(s):  
Diane Pannier ◽  
Aurelie Dumont ◽  
Emmanuelle Tresch ◽  
Jinying Chen ◽  
Agnes Ducoulombier ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Estrogen Receptor ◽  
Neoadjuvant Chemotherapy ◽  
Locally Advanced Breast Cancer ◽  
Locally Advanced ◽  
Pathologic Complete Response ◽  
Breast Tumors ◽  
Breast Adenocarcinoma ◽  
Chemotherapy Response ◽  
Er Status

544 Background: Neoadjuvant chemotherapy (NCT) using anthracyclines and taxanes is a standard treatment for locally advanced breast cancer and pathologic complete response (pCR) is a major prognostic factor for survival. Gene polymorphisms have been identified as modulators of chemotherapy response. Our study investigated constitutional variants of genes associated with a change in the response to neoadjuvant chemotherapy using taxanes and/or anthracyclines in patients with breast adenocarcinoma. Methods: From November 2007 to January 2012, 118 women with breast adenocarcinoma histologically proven, with no Her2 surexpression, receiving or having received a neoadjuvant chemotherapy with taxanes and/or anthracyclines were included in the study. NCT associated 3 FEC100 then 3 Docetaxel every 21 days. Genotyping of 46 SNPs was performed on germline DNA using real time PCR. pCR was correlated to clinical characteristics and genotypes using univariate logistic regression. Results: 21.2% had a pCR according to Sataloff classification. pCR is increased in SBRIII (p=0.009), estrogen receptor negative (p=0.005) and triple negative (p=0.006) tumors. 7 SNP are significantly associated with pCR in ER+ breast tumors (pCR=13.5%). Among these SNP, pCR is increased for patients carrying almost one G allele for SLCO1B3-rs11045585 (pCR=28.6%; p=0.032), for homozygotes GG for SHTM1-rs1979277 (pCR=24.3%, p=0.006) and for homozygotes CC for CYP1B1-rs1056836 (pCR=25.7%; p=0.003). Moreover, 4 SNPs are significantly associated with pCR in ER- breast tumors: ERCC1-rs11615 (carriers of almost one C allele: pCR=50%; p=0.030), CD24-rs52812045 (Homozygotes CC pCR=56.3%, p=0.033), CYP2B6-rs2279343(carriers of one or two G allele: pCR=52.6%; p=0.046) and GSTP1-rs1695 (carriers of one or two G allele: pCR=48%; p=0.050). Conclusions: Besides ER status, polymorphisms could be useful markers to predict response to anthracyclines/taxanes NCT in breast cancer. Furthermore, this work is the first describing ERCC1-rs11615, SLCO1B3-rs11045585 and SHTM1-rs1979277 as new potential genetic markers for NCT in breast cancer. (The first 3 authors contributed equally to this work.)

scholarly journals Concordance Between ER, PR, HER2 neu Receptors Before and After Neoadjuvant Chemotherapy in Locally Advanced Breast Cancer

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Heba F. Taha ◽  
Ola M. Elfarargy ◽  
Reham A. Salem ◽  
Doaa Mandour ◽  
Amira A. Salem ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Hormone Receptor ◽  
Growth Hormone Receptor ◽  
Locally Advanced Breast Cancer ◽  
Locally Advanced ◽  
Her2 Status ◽  
Breast Cancer Patients ◽  
Before And After ◽  
Tumor Phenotype

Abstract Background Introducing neoadjuvant chemotherapy (NCT) in a breast cancer patient may be associated with changes in estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth hormone receptor 2 (HER2) status. Method In our prospective cohort study, we evaluated the impact of change in estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth hormone receptor 2 (HER2) on the prognosis of breast cancer patients treated with neoadjuvant chemotherapy (NCT). We investigated 110 patients with locally advanced breast cancer for ER, PR and HER2 status of their lesions before and after NCT. Result For hormone receptor status (HR) (which include ER, PR) of the residual tumor of the patients after receiving NCT, 12 (10.9%) of them changed from HR (+) to HR (−) and 15 (13.6%) changed from HR (−) to HR (+). For HER2 status after NCT, 8 (7.3%) patients changed from HER2 (+) to HER2 (−) and 9 (8.2%) patients changed from HER2 (−) to HER2 (+). Triple negative (TN) tumor phenotype changes occurred in 17 (15.5%) patients. Patients for whom the HR status changed from positive to negative had poor prognosis for both disease-free survival (DFS) and overall survival (OS) in univariate survival analysis. Conclusion Changes in ER, PR, HER2 status and tumor phenotype in breast cancer patients after NCT had a negative prognostic impact and were associated with a poor prognosis.

scholarly journals Locally Advanced Breast Cancer (LABC): Real-World Outcome of Patients From Cancer Institute, Chennai

2021 ◽  
pp. 767-781
Author(s):  
Manikandan Dhanushkodi ◽  
Velusamy Sridevi ◽  
Viswanathan Shanta ◽  
Ranganathan Rama ◽  
Rajaraman Swaminathan ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Overall Survival ◽  
Neoadjuvant Chemotherapy ◽  
Triple Negative Breast Cancer ◽  
Tumor Size ◽  
Triple Negative ◽  
Locally Advanced Breast Cancer ◽  
Locally Advanced

PURPOSE There are sparse data on the outcome of patients with locally advanced breast cancer (LABC). This report is on the prognostic factors and long-term outcome from Cancer Institute, Chennai. METHODS This is an analysis of untreated patients with LABC (stages IIIA-C) who were treated from January 2006 to December 2013. RESULTS Of the 4,577 patients with breast cancer who were treated, 2,137 patients (47%) with LABC were included for analysis. The median follow-up was 75 months (range, 1-170 months), and 2.3% (n = 49) were lost to follow-up at 5 years. The initial treatment was neoadjuvant concurrent chemoradiation (NACR) (77%), neoadjuvant chemotherapy (15%), or others (8%). Patients with triple-negative breast cancer had a pathologic complete response (PCR) of 41%. The 10-year overall survival was for stage IIIA (65.1%), stage IIIB (41.2%), and stage IIIC (26.7%). Recurrence of cancer was observed in 27% of patients (local 13% and distant 87%). Multivariate analysis showed that patients with a tumor size > 10 cm (hazard ratio [HR], 2.19; 95% CI, 1.62 to 2.98; P = .001), hormone receptor negativity (HR, 1.45; 95% CI, 1.22 to 1.72; P = .001), treatment modality (neoadjuvant chemotherapy, HR, 0.56; 95% CI, 0.43 to 0.73; P = .001), lack of PCR (HR, 2.36; 95% CI, 1.85 to 3.02; P = .001), and the presence of lymphovascular invasion (HR, 1.97; 95% CI, 1.60 to 2.44; P = .001) had decreased overall survival. CONCLUSION NACR was feasible in inoperable LABC and gave satisfactory long-term survival. PCR was significantly higher in patients with triple-negative breast cancer. The tumor size > 10 cm was significantly associated with inferior survival. However, this report acknowledges the limitations inherent in experience of management of LABC from a single center.

ROLE OF SONOMAMMOGRAPHY IN MONITORING THE RESPONSE OF NEOADJUVANT CHEMOTHERAPY IN MALIGNANT BREAST MASSES

QJM ◽  
2021 ◽  
Vol 114 (Supplement_1) ◽  
Author(s):  
Sherine George Moftah ◽  
Essam Mohamed Hafez ◽  
Walid Dahy Zidan
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Tumor Size ◽  
Locally Advanced ◽  
Residual Tumor ◽  
High Accuracy ◽  
Breast Ultrasound ◽  
Radiology Department ◽  
Duration Of Symptoms ◽  
Progressive Course

Abstract Background Breast cancer is a leading cause of death and disability between women, particularly young women, in low- and average -income countries (Porter, 2018). Neoadjuvant chemotherapy (NAC) is increasingly used in management patients with locally advanced breast cancer (LABC) (Mougalian et al., 2015). Mammography and breast ultrasound are the most widest diagnostic techniques used to evaluate primary tumor size and character at the time of diagnosis and monitoring the response to NAC (Kald et al., 2015). Aim of the work The purpose of this study is to assess and evaluate the role and the accuracy of Mammography and breast ultrasound modalities in monitoring the response of the breast cancer to the neoadjuvant chemotherapy. Patients and methods Twenty three patients presenting with pathologically proven malignant mass and conducted for neoadjuvant chemotherapy with age above 40 years-old. They were referred to the radiology department (mammography unit, Ultrasound unit and MRI unit) at El Salam oncology center, during the period between December 2018 to June 2019. Once a patient satisfied the inclusion and exclusion criteria for this study, an elaborate history was taken from all the patients which was followed by a thorough clinical evaluation, in which duration of symptoms, affected side, affected breast and family history. Patients were subjected to evaluation before starting neoadjunant chemotherapy by mammography, ultrasound and MRI as well as after finishing the neoadjuvant cycles which usually was ranging 3-4 cycles. Results Total of 23 patients with breast cancer were enrolled in this study. The patients were in the age group of 41-70 years with mean age of 54 years. Majority of patients had infiltrative ductal carcinoma IDC (16 cases -69.56% and the rest of cases were infiltrative lobular carcinoma (9 cases -39.13 %). After monitoring the response of the cases to neoadjuvant chemotherapy the results showed significant regressive course to the majority of cases ( 17 cases ) (73.91%) with the rest non respondent cases divided to cases with stationary ( 3 cases ) (13.04% ) and cases with progressive course ( 3 cases ) (13.04% ). The comparison between the results of Sonomammographic assessment of the breast lesions before and after the neoadjuvant chemotherapy cycles and also to MRI show close results. There were 5 cases that showed non-respondent (including stationary course and progressive course) in both Sonomammography and MRI and only one case that was respondent according to Sonomammography and non-respondent according to MRI breast (false +e). On the other hand there were 16 cases that had showed respondent fashion in both Sonomammography and MRI. Also there was one case that had showed non-respondent course according to Sonomammography and was respondent according to MRI results (false –ve). Sonomammography sensitivity was 94.12 % and specificity was 83.33% compared to MRI results, with high accuracy value reaching to 91.30 %. Conclusion The ability of mammography and breast ultrasound to accurately measure residual tumor size following neoadjuvant chemotherapy compared to Breast MRI measurement of the residual tumor demonstrated sensitivity of 94.12 % and specificity of 83.33% with high accuracy value reaching to 91.30 %. Even that the MRI is proved to be of high accuracy in detecting residual tumor, sonomammographic imaging isa cost-effective imaging modality than the MRI with sensitivity and specificity near to that of MRI.

Neoadjuvant chemotherapy in locally advanced breast cancer.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. e11034-e11034
Author(s):  
Sami sahnoun Soraya
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Advanced Breast Cancer ◽  
Locally Advanced Breast Cancer ◽  
Locally Advanced ◽  
Molecular Classification ◽  
Advanced Breast ◽  
Her2 Positive ◽  
Significant Difference ◽  
Er Status

e11034 Background: Neoadjuvant chemotherapy(NAC)is one of the treatment options for locally advanced breast cancer. In this study, we evaluated the efficacy and safety of 4 cycles of NAC doxorubicine,docetaxel and cyclophosphamide(TAC),correlation between the response to NAC and molecular classification sub-groups and between the pCR and the time to progression(TTP). Methods: This is a prospective study from January 2005 to December 2008.110 pts with locally advanced breast cancer stage III.All pts have received 4 cycles of NAC based on docetaxel 75 mg/m², doxorubicine 50 mg/m² and cyclophosphamide 500 mg/m² every 3 weeks, followed by surgery.101 pts were assessed, since 9 of them have progressed on treatment and came out of the study. Pts were stratified according to age, menopausal status, histopathological analysis (luminal tumors(ER-positive and HER2-negative), triple-negative tumors (TN)and HER2-positive tumors), response to the treatment and survival. The median follow up of patients was 39 months. The statistical study was done using SPSS 17. Results: The median age was 41(23–65).30% of pts were younger than 35 and 80% were premenopausal. 55% luminal tumors(56 pts), 33% HER2 positive(33 pts) and 12 % TN(12 pts).CRR was estimated at 89%(37% of CR and 63% of PR).There were 23, 7% of pCR according to Chevallier’s classification. In luminal, TN and HER2-positive pCR rates were 16%(9 of 56), 66,6%(8 of 12), and 21,2%(7 of 33) respectively. Multivariate analysis showed that the ER status was the only significant predictor of pCR(P = 0.025).HER2 status was not significantly associated with pCR(P= 0,423).TTP was 50 months. In luminal tumors, TN and HER2-positive tumors the TTP was respectively 59, 52 and 49 months. There was not a significant difference in TTP between the pCR(51 months) and the non-pCR group(44 months)(CI 95, p= 0.109).Grade III/IV toxicity included neutropenia(22%), febrile neutropenia(6,5%), mucositis(13%), and diarrhea(4%). Conclusions: Breast cancer occurs in young women in Algeria. In this study, neoadjuvant TAC was effective and well tolerated. The ER status was the only significant predictor of pCR. The molecular classification group with the highest percentage of pCR was the TN group. pCR was not associated with a better prognosis.

scholarly journals Ki-67 Expression by Immunohistochemistry and Quantitative Real-Time Polymerase Chain Reaction as Predictor of Clinical Response to Neoadjuvant Chemotherapy in Locally Advanced Breast Cancer

2017 ◽  
Vol 2017 ◽  
pp. 1-8 ◽  
Author(s):  
Prihantono Prihantono ◽  
Mochammad Hatta ◽  
Christian Binekada ◽  
Daniel Sampepajung ◽  
Haryasena Haryasena ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Clinical Response ◽  
Locally Advanced Breast Cancer ◽  
Locally Advanced ◽  
Chemotherapy Response ◽  
Ki 67 ◽  
Qrt Pcr ◽  
Polymerase Chain ◽  
Response To Neoadjuvant Chemotherapy

Background. Chemotherapy has become a standard of treatment in managing breast cancer. To achieve proper treatment for the right patients, the predictive marker is needed. Ki-67 is a biomarker of proliferation for solid tumor. Studies mentioned association of Ki-67 expression with chemotherapy response. The study aims are to evaluate whether Ki-67 expression detected by immunohistochemistry (IHC) and quantitative real-time polymerase chain reaction (qRT-PCR) may predict clinical response to neoadjuvant chemotherapy in breast cancer.Methods. This study utilized a longitudinal study. IHC and qRT-PCR methods were used for detection of Ki-67 expression. Chemotherapy response was calculated using RECIST. Data were analyzed with Chi-square and Wilcoxon’s test.Results. There were 48 subjects in this study. Analysis of Ki-67 expression with chemotherapy response has a significant correlation withp=0.025(<0.05), OR: 1.69, confidence interval (95% CI) 1.022–2.810. Analysis of Ki-67 mRNA expression with chemotherapy response has a significant correlationp=0.002(<0.05), OR: 6.85, confidence interval (95% CI) 1.064–44.193. Detection of Ki-67 expression using IHC and qRT-PCR has similar results,p=0.012(<0.05).Conclusion. These results suggest that Ki-67 expression detected by both IHC and qRT-PCR is considered to be a predictor of clinical response to neoadjuvant chemotherapy in locally advanced breast cancer.

scholarly journals Neoadjuvan Chemotheraphy Responses to Local Advanced Breast Cancer Patients at Dr Moh Hoesin Hospital Palembang

2020 ◽  
Vol 3 (3) ◽  
pp. 51-58
Author(s):  
Aldo Giovanno ◽  
Mgs. Irsan Saleh ◽  
Nur Qodir ◽  
Mulawan Umar
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Advanced Breast Cancer ◽  
Cancer Patients ◽  
Univariate Analysis ◽  
Locally Advanced ◽  
Radical Mastectomy ◽  
Advanced Breast ◽  
Chemotherapy Response ◽  
Breast Cancer Patients

Breast cancer is a malignancy which invaded breast tissue in the form of ductal or lobular. One most therapywhich is given is neoadjuvant chemotherapy. Neoadjuvant Chemotherapy can reduce tumor size so that surgerycan be performed with good breast removal with Modification of Radical Mastectomy (MRM) and BreastConservative Therapy (BCT). This purpose from this research is to find out neoadjuvant chemotherapy response inLocally Advanced Breast Cancer Patients which has received chemotherapy treatment in RSUP dr MohammadHoesin Palembang. This observational descriptive study was conducted at RSUP Mohammad Hoesin Palembang inthe period between October until November 2019. The sample of this study was locally advanced breast Cancerpatients who underwent chemotherapy that met the inclusion and exclusion criteria. The data were obtained byinterviews and observed medical records from the patients which were then analyzed by univariate analysis usingSPSS version 25. In this study there were 34 locally advanced breast cancer patients who fulfilled the inclusion andexclusion criteria. 24 of 34 patients (70,6%) received positive response and 10 of 34 patients (29,4%) receivednegative response.

Sign in / Sign up

Export Citation Format

Share Document