Predictors of Neoadjuvant Chemotherapy Response in Breast Cancer: A Review

The Impact of Neoadjuvant Chemotherapy on Patients With T1N0M0 Triple-Negative and HER-2 Positive Breast Cancer: A Retrospective Analysis Based on the SEER Database

BACKGROUND

The utilization of neoadjuvant chemotherapy (NAC) originated in the treatment of locally advanced breast cancer (BC). Our study is designed to elucidate the effects of NAC on patients with T1N0M0 triple-negative and HER-2 positive BC.

METHODS

This study involved the selection of 10,614 patients diagnosed with T1N0M0 triple-negative and HER-2 positive breast cancer (BC) from the surveillance, epidemiology, and end results (SEER) database. To ascertain the impact of neoadjuvant chemotherapy (NAC) on T1a, T1b, and T1c N0M0 BC, we conducted multivariate Cox regression analyses. Similarly, we performed multivariate Cox regression analyses to compare the effects of neoadjuvant chemotherapy against adjuvant chemotherapy on T1N0M0 BC. The Kaplan-Meier method was employed to delineate survival curves for different molecular subtypes and clinical stages.

RESULTS

The data results from the SEER database reveal a significant enhancement of overall survival (OS) in T1c BC patients as a result of NAC. For T1b BC patients, NAC does not present any significant effect. Contrarily, NAC seems to adversely impact the OS of T1a triple-negative BC patients. However, the prognosis comparison between neoadjuvant and adjuvant chemotherapy for T1N0M0 breast cancer did not show any significant difference, with the exception of T1a triple-negative BC.

CONCLUSIONS

Patients with T1cN0M0 triple-negative and HER-2 positive BC may derive OS benefits from NAC. Additionally, NAC could be detrimental to T1a triple-negative BC.

High linoleic acid levels in red blood cells predict a poor response to neoadjuvant chemotherapy in human epidermal growth factor receptor type 2-positive breast cancer patients

OBJECTIVE

Polyunsaturated fatty acids are categorized as ω-3 or ⍵-6. Previous studies demonstrate that breast cancers display a high expression of fatty acid synthase and high fatty acid levels. Our study sought to determine if changes in plasma or red blood cell membrane fatty acid levels were associated with the response to preoperative (neoadjuvant) chemotherapy in non-metastatic breast cancer patients.

METHODS

Our prospective study assessed fatty acid levels in plasma and red blood cell membrane. Response to neoadjuvant chemotherapy was evaluated by the presence or absence of pathologic complete response and/or residual cancer burden.

RESULTS

A total of 28 patients were included. First, patients who achieved pathologic complete response had significantly higher neutrophil-to-lymphocyte ratio versus no pathologic complete response (P = 0.003). Second, total red blood cell membrane polyunsaturated fatty acids were higher in the absence of pathologic complete response (P = 0.0028). Third, total red blood cell membrane ⍵-6 polyunsaturated fatty acids were also higher in no pathologic complete response (P < 0.01). Among ⍵-6 polyunsaturated fatty acids, red blood cell membrane linoleic acid was higher in the absence of pathologic complete response (P < 0.01). Notably, plasma polyunsaturated fatty acid, ⍵-6, and linoleic acid levels did not have significant differences. A multivariate analysis confirmed red blood cell membrane linoleic acid was associated with no pathologic complete response; this was further confirmed by receiver operating characteristic analysis (specificity = 92.3%, sensitivity = 76.9%, and area under the curve = 0.855).

CONCLUSIONS

Pending further validation, red blood cell membrane linoleic acid might serve as a predictor biomarker of poorer response to neoadjuvant chemotherapy in non-metastatic human epidermal growth factor receptor type 2-positive breast cancer. Measuring fatty acids in red blood cell membrane could offer a convenient, minimally invasive strategy to identifying patients more likely to respond or those with chemoresistance.

Value of ultrasound in assessing response to neoadjuvant chemotherapy in breast cancer

AIM

To analyse the utility of ultrasound in assessing response to neoadjuvant chemotherapy (NAC) and predicting residual cancer burden (RCB) index and pathological complete response (pCR) MATERIALS AND METHODS: This was a retrospective study with 417 patients over 7 years. The difference in longest diameter (LD) of the index lesion from baseline to end, baseline to mid, and mid to end was evaluated with respect to RCB class using logistic regression and ordered logistic regression.

RESULTS

Change in LD measurements from baseline to end, baseline to mid, and mid to end of chemotherapy as a predictor of RCB class show a negative relationship with a statistically significant association. This would suggest that a smaller change in LD measurements would be associated with an eventual higher RCB class. Change in LD measurements from baseline to end and baseline to mid chemotherapy as a predictor of pCR class show a negative relationship with a statistically significant association (p<0.05). This similarly indicates an inversely proportional relationship between changes in LD measurements and RCB class 0 for baseline to end and baseline to mid.

CONCLUSION

This study has shown significance in reducing LD measurements on ultrasound as a predictor of PCR and RCB class. This adds weight to the current practice of using ultrasound at the start, mid and end of chemotherapy cycles to monitor NACT responses.

Prognostic significance of platelet-to-lymphocyte ratio (PLR) in patients with breast cancer treated with neoadjuvant chemotherapy: a meta-analysis

OBJECTIVE

Platelet-to-lymphocyte ratio (PLR), known as a key systemic inflammatory parameter, has been proved to be associated with response to neoadjuvant therapy in breast cancer (BC); however, the results remain controversial. This meta-analysis was carried out to evaluate the prognostic values of PLR in patients with BC treated with neoadjuvant chemotherapy (NACT).

DESIGN

Meta-analysis.

DATA SOURCES

Relevant literature published on the following databases: PubMed, Embase, Web of Science databases and the Cochrane Library.

ELIGIBILITY CRITERIA

All studies involving patients with BC treated with NACT and peripheral blood pretreatment PLR recorded were included.

DATA EXTRACTION AND SYNTHESIS

Two researchers independently extracted and evaluated HR/OR and its 95% CI of survival outcomes, pathological complete response (pCR) rate and clinicopathological parameters.

RESULTS

The last search was updated to 31 December 2022. A total of 22 studies with 5533 patients with BC treated with NACT were enrolled in the final meta-analysis. Our results demonstrate that elevated PLR value appears to correlate with low pCR rate (HR 0.77, 95% CI 0.67 to 0.88, p<0.001, I2=75.80%, Ph<0.001) and poor prognosis, including overall survival (OS) (HR 1.90, 95% CI 1.39 to 2.59, p<0.001; I2=7.40%, Ph=0.365) and disease-free survival (HR 1.97, 95% CI 1.56 to 2.50, p<0.001; I2=0.0%, Ph=0.460). Furthermore, PLR level was associated with age (OR 0.86, 95% CI 0.79 to 0.93, p<0.001, I2=40.60%, Ph=0.096), menopausal status (OR 0.83, 95% CI 0.76 to 0.90, p<0.001, I2=50.80%, Ph=0.087) and T stage (OR 1.05, 95% CI 1.00 to 1.11, p=0.035; I2=70.30%, Ph=0.005) of patients with BC.

CONCLUSIONS

This meta-analysis demonstrated that high PLR was significantly related to the low pCR rate, poor OS and disease-free survival (DFS) of patients with BC treated with NACT. Therefore, PLR can be used as a potential predictor biomarker for the efficacy of NACT in BC.

Neoadjuvant chemotherapy for breast cancer: an evaluation of its efficacy and research progress

Neoadjuvant chemotherapy (NAC) for breast cancer is widely used in the clinical setting to improve the chance of surgery, breast conservation and quality of life for patients with advanced breast cancer. A more accurate efficacy evaluation system is important for the decision of surgery timing and chemotherapy regimen implementation. However, current methods, encompassing imaging techniques such as ultrasound and MRI, along with non-imaging approaches like pathological evaluations, often fall short in accurately depicting the therapeutic effects of NAC. Imaging techniques are subjective and only reflect macroscopic morphological changes, while pathological evaluation is the gold standard for efficacy assessment but has the disadvantage of delayed results. In an effort to identify assessment methods that align more closely with real-world clinical demands, this paper provides an in-depth exploration of the principles and clinical applications of various assessment approaches in the neoadjuvant chemotherapy process.

Predictive significance of HIF-1α, Snail, and PD-L1 expression in breast cancer

Currently, the prediction of breast cancer (BC) effectiveness to drug treatment is based on determining the expression level of steroid hormone receptors and human epidermal growth factor receptor type 2 (HER2). However, significant differences in individual response to drug treatment require the search for new predictive markers. Here, by comprehensively examining HIF-1α, Snail, and PD-L1 expression in BC tumor tissue, we demonstrate that high levels of these markers correlate with unfavorable factors of BC prognosis: the presence of regional and distant metastases and lymphovascular and perineural invasion. Analyzing the predictive significance of markers, we show that the most significant predictors of chemoresistant HER2-negative BC are a high PD-L1 level and a low Snail level, while in HER2-positive BC, only a high PD-L1 level is an independent predictor of chemoresistant BC. Our results suggest that using immune checkpoint inhibitors in these groups of patients may improve drug therapy effectiveness.

The Role of Stromal Tumour Infiltrating Lymphocytes (sTIL) Intensity and Programmed Death Ligand 1 () Expression in Breast Cancer Response to Neoadjuvant Therapy in Cipto Mangunkusumo Hospital (CMH), Indonesia

BACKGROUND

Pathological responses to neoadjuvant therapy were still relatively poor, especially in CMH. Studies had been done to search for predictors of response such as sTIL intensity and  expression, which is known to block sTIL action in killing cancer cells. This research assessed sTIL intensity and  expression as predictors of response to neoadjuvant therapy in breast cancer. The preliminary data might be used to better tailored breast cancer patient therapy, considering the availability of anti-PD-1/ PD- L1 immunotherapy nowadays.

OBJECTIVE

To assess TIL intensity,  expressions, and their roles as pathological predictors of breast cancer response to neoadjuvant therapy in Cipto Mangunkusumo Hospital (CMH).

METHOD

This was an observational analytic retrospective cohort study on breast cancer patients undergoing biopsy/review of biopsy specimens, receiving neoadjuvant therapy and mastectomy in CMH from January 2014 to December 2021. Sixty cases fulfilled the inclusion and exclusion criteria. Total sampling was done.   expression (immunohistochemistry, clone 22C3) and sTIL intensity (histopathology) was examined in the biopsy specimen. Linear regression analysis was done to determine the independent predictors of neoadjuvant therapy response (evaluated in the mastectomy specimen with residual cancer burden/ RCB score).

RESULTS

There were 60 female patients, median age 46 years old. 91,7% had invasive carcinoma of no special type. Median sTIL intensity was 10% (1%-70%). 58,3% patients had low sTIL intensity (≤10%). 28,3% patients had positive  expression (CPS ≥1). Only 8,3% patients had pCR, while 90% patients had RCB class II-III. Every 1% increase in sTIL intensity, no lymphovascular invasion, and taxane chemotherapy were predicted to lower RCB score by 0,058, 0,781, dan 0,594, respectively.  expression associated with pCR-RCB class I (p=0,048), but CPS score was not a predictor of RCB score in linear regression analysis.

CONSLUSION

sTIL intensity was an independent predictor of breast cancer response to neoadjuvant therapy in RSCM.  expression associated with pCR-RCB class I, but CPS score was not a predictor of RCB score.

Prediction of Tumor Progression During Neoadjuvant Chemotherapy and Survival Outcome in Patients With Triple-Negative Breast Cancer

OBJECTIVE

To investigate the association of clinical, pathologic, and magnetic resonance imaging (MRI) variables with progressive disease (PD) during neoadjuvant chemotherapy (NAC) and distant metastasis-free survival (DMFS) in patients with triple-negative breast cancer (TNBC).

MATERIALS AND METHODS

This single-center retrospective study included 252 women with TNBC who underwent NAC between 2010 and 2019. Clinical, pathologic, and treatment data were collected. Two radiologists analyzed the pre-NAC MRI. After random allocation to the development and validation sets in a 2:1 ratio, we developed models to predict PD and DMFS using logistic regression and Cox proportional hazard regression, respectively, and validated them.

RESULTS

Among the 252 patients (age, 48.3 ± 10.7 years; 168 in the development set; 84 in the validation set), PD was occurred in 17 patients and 9 patients in the development and validation sets, respectively. In the clinical-pathologic-MRI model, the metaplastic histology (odds ratio [OR], 8.0; P = 0.032), Ki-67 index (OR, 1.02; P = 0.044), and subcutaneous edema (OR, 30.6; P = 0.004) were independently associated with PD in the development set. The clinical-pathologic-MRI model showed a higher area under the receiver-operating characteristic curve (AUC) than the clinical-pathologic model (AUC: 0.69 vs. 0.54; P = 0.017) for predicting PD in the validation set. Distant metastases occurred in 49 patients and 18 patients in the development and validation sets, respectively. Residual disease in both the breast and lymph nodes (hazard ratio [HR], 6.0; P = 0.005) and the presence of lymphovascular invasion (HR, 3.3; P < 0.001) were independently associated with DMFS. The model consisting of these pathologic variables showed a Harrell's C-index of 0.86 in the validation set.

CONCLUSION

The clinical-pathologic-MRI model, which considered subcutaneous edema observed using MRI, performed better than the clinical-pathologic model for predicting PD. However, MRI did not independently contribute to the prediction of DMFS.

DNA Methylation and Prospects for Predicting the Therapeutic Effect of Neoadjuvant Chemotherapy for Triple-Negative and Luminal B Breast Cancer

Despite advances in the diagnosis and treatment of breast cancer (BC), the main cause of deaths is resistance to existing therapies. An approach to improve the effectiveness of therapy in patients with aggressive BC subtypes is neoadjuvant chemotherapy (NACT). Yet, the response to NACT for aggressive subtypes is less than 65% according to large clinical trials. An obvious fact is the lack of biomarkers predicting the therapeutic effect of NACT. In a search for epigenetic markers, we performed genome-wide differential methylation screening by XmaI-RRBS in cohorts of NACT responders and nonresponders, for triple-negative (TN) and luminal B tumors. The predictive potential of the most discriminative loci was further assessed in independent cohorts by methylation-sensitive restriction enzyme quantitative PCR (MSRE-qPCR), a promising method for the implementation of DNA methylation markers in diagnostic laboratories. The selected most informative individual markers were combined into panels demonstrating cvAUC = 0.83 (TMEM132D and MYO15B markers panel) for TN tumors and cvAUC = 0.76 (TTC34, LTBR and CLEC14A) for luminal B tumors. The combination of methylation markers with clinical features that correlate with NACT effect (clinical stage for TN and lymph node status for luminal B tumors) produces better classifiers, with cvAUC = 0.87 for TN tumors and cvAUC = 0.83 for luminal B tumors. Thus, clinical characteristics predictive of NACT response are independently additive to the epigenetic classifier and in combination improve prediction.

Novel Radiomic Measurements of Tumor-Associated Vasculature Morphology on Clinical Imaging as a Biomarker of Treatment Response in Multiple Cancers

PURPOSE

The tumor-associated vasculature (TAV) differs from healthy blood vessels by its convolutedness, leakiness, and chaotic architecture, and these attributes facilitate the creation of a treatment-resistant tumor microenvironment. Measurable differences in these attributes might also help stratify patients by likely benefit of systemic therapy (e.g., chemotherapy). In this work, we present a new category of computational image-based biomarkers called quantitative tumor-associated vasculature (QuanTAV) features, and demonstrate their ability to predict response and survival across multiple cancer types, imaging modalities, and treatment regimens involving chemotherapy.

EXPERIMENTAL DESIGN

We isolated tumor vasculature and extracted mathematical measurements of twistedness and organization from routine pretreatment radiology (CT or contrast-enhanced MRI) of a total of 558 patients, who received one of four first-line chemotherapy-based therapeutic intervention strategies for breast (n = 371) or non-small cell lung cancer (NSCLC, n = 187).

RESULTS

Across four chemotherapy-based treatment strategies, classifiers of QuanTAV measurements significantly (P < 0.05) predicted response in held out testing cohorts alone (AUC = 0.63-0.71) and increased AUC by 0.06-0.12 when added to models of significant clinical variables alone. Similarly, we derived QuanTAV risk scores that were prognostic of recurrence-free survival in treatment cohorts who received surgery following chemotherapy for breast cancer [P = 0.0022; HR = 1.25; 95% confidence interval (CI), 1.08-1.44; concordance index (C-index) = 0.66] and chemoradiation for NSCLC (P = 0.039; HR = 1.28; 95% CI, 1.01-1.62; C-index = 0.66). From vessel-based risk scores, we further derived categorical QuanTAV high/low risk groups that were independently prognostic among all treatment groups, including patients with NSCLC who received chemotherapy only (P = 0.034; HR = 2.29; 95% CI, 1.07-4.94; C-index = 0.62). QuanTAV response and risk scores were independent of clinicopathologic risk factors and matched or exceeded models of clinical variables including posttreatment response.

CONCLUSIONS

Across these domains, we observed an association of vascular morphology on CT and MRI-as captured by metrics of vessel curvature, torsion, and organizational heterogeneity-and treatment outcome. Our findings suggest the potential of shape and structure of the TAV in developing prognostic and predictive biomarkers for multiple cancers and different treatment strategies.

Correlation Between Immune-Related Genes and Tumor-Infiltrating Immune Cells With the Efficacy of Neoadjuvant Chemotherapy for Breast Cancer

Background: In the absence of targeted therapy or clear clinically relevant biomarkers, neoadjuvant chemotherapy (NAC) is still the standard neoadjuvant systemic therapy for breast cancer. Among the many biomarkers predicting the efficacy of NAC, immune-related biomarkers, such as immune-related genes and tumor-infiltrating lymphocytes (TILs), play a key role.

Methods: We analyzed gene expression from several datasets in the Gene Expression Omnibus (GEO) database and evaluated the relative proportion of immune cells using the CIBERSORT method. In addition, mIHC/IF detection was performed on clinical surgical specimens of triple-negative breast cancer patients after NAC.

Results: We obtained seven immune-related genes, namely, CXCL1, CXCL9, CXCL10, CXCL11, IDO1, IFNG, and ORM1 with higher expression in the pathological complete response (pCR) group than in the non-pCR group. In the pCR group, the levels of M1 and γδT macrophages were higher, while those of the M2 macrophages and mast cells were lower. After NAC, the proportions of M1, γδT cells, and resting CD4 memory T cells were increased, while the proportions of natural killer cells and dendritic cells were decreased with downregulated immune-related genes. The results of mIHC/IF detection and the prognostic information of corresponding clinical surgical specimens showed the correlation of proportions of natural killer cells, CD8-positive T cells, and macrophages with different disease-free survival outcomes.

Conclusion: The immune-related genes and immune cells of different subtypes in the tumor microenvironment are correlated with the response to NAC in breast cancer, and the interaction between TILs and NAC highlights the significance of combining NAC with immunotherapy to achieve better clinical benefits.

Predicting dynamic response to neoadjuvant chemotherapy in breast cancer: a novel metabolomics approach

Neoadjuvant chemotherapy (NACT) outcomes vary according to breast cancer (BC) subtype. Since pathologic complete response is one of the most important target endpoints of NACT, further investigation of NACT outcomes in BC is crucial. Thus, identifying sensitive and specific predictors of treatment response for each phenotype would enable early detection of chemoresistance and residual disease, decreasing exposures to ineffective therapies and enhancing overall survival rates. We used liquid chromatography−high‐resolution mass spectrometry (LC‐HRMS)‐based untargeted metabolomics to detect molecular changes in plasma of three different BC subtypes following the same NACT regimen, with the aim of searching for potential predictors of response. The metabolomics data set was analyzed by combining univariate and multivariate statistical strategies. By using ANOVA–simultaneous component analysis (ASCA), we were able to determine the prognostic value of potential biomarker candidates of response to NACT in the triple‐negative (TN) subtype. Higher concentrations of docosahexaenoic acid and secondary bile acids were found at basal and presurgery samples, respectively, in the responders group. In addition, the glycohyocholic and glycodeoxycholic acids were able to classify TN patients according to response to treatment and overall survival with an area under the curve model > 0.77. In relation to luminal B (LB) and HER2+ subjects, it should be noted that significant differences were related to time and individual factors. Specifically, tryptophan was identified to be decreased over time in HER2+ patients, whereas LysoPE (22:6) appeared to be increased, but could not be associated with response to NACT. Therefore, the combination of untargeted‐based metabolomics along with longitudinal statistical approaches may represent a very useful tool for the improvement of treatment and in administering a more personalized BC follow‐up in the clinical practice.

Development of Gene Expression-Based Random Forest Model for Predicting Neoadjuvant Chemotherapy Response in Triple-Negative Breast Cancer

Simple Summary

Only 20–50% of patients with triple negative breast cancer achieve a pathological complete response from neoadjuvant chemotherapy, a strong indicator of patient survival. Therefore, there is an urgent need for a reliable predictive model of the patient’s pathological complete response prior to actual treatment. The purpose of this study was to develop such a model based on random forest recursive feature elimination and to benchmark the performance of the proposed model against existing predictive models. Our study suggests that an 86-gene-based random forest model associated to DNA repair and cell cycle mechanisms can provide reliable predictions of neoadjuvant chemotherapy response in patients with triple negative breast cancer.

Abstract

Neoadjuvant chemotherapy (NAC) response is an important indicator of patient survival in triple negative breast cancer (TNBC), but predicting chemosensitivity remains a challenge in clinical practice. We developed an 86-gene-based random forest (RF) classifier capable of predicting neoadjuvant chemotherapy response (pathological Complete Response (pCR) or Residual Disease (RD)) in TNBC patients. The performance of pCR classification of the proposed model was evaluated by Receiver Operating Characteristic (ROC) curve and Precision Recall (PR) curve. The AUROC and AUPRC of the proposed model on the test set were 0.891 and 0.829, respectively. At a predefined specificity (>90%), the proposed model shows a superior sensitivity compared to the best performing reported NAC response prediction model (69.2% vs. 36.9%). Moreover, the predicted pCR status by the model well explains the distance recurrence free survival (DRFS) of TNBC patients. In addition, the pCR probabilities of the proposed model using the expression profiles of the CCLE TNBC cell lines show a high Spearman rank correlation with cyclophosphamide sensitivity in the TNBC cell lines (SRCC =0.697, p-value =0.031). Associations between the 86 genes and DNA repair/cell cycle mechanisms were provided through function enrichment analysis. Our study suggests that the random forest-based prediction model provides a reliable prediction of the clinical response to neoadjuvant chemotherapy and may explain chemosensitivity in TNBC.

Role of neutrophil-to-lymphocyte ratio as a prognostic biomarker in patients with breast cancer receiving neoadjuvant chemotherapy: a meta-analysis

OBJECTIVE

The neutrophil-to-lymphocyte ratio (NLR) is recognised as a suitable prognostic biomarker in patients with breast cancer. Nevertheless, the efficacy of this biomarker in predicting the pathological complete response (pCR) and survival in patients with breast cancer receiving neoadjuvant chemotherapy (NACT) is still controversial. This meta-analysis aimed to identify the association between baseline NLR and the prognosis of patients with breast cancer treated with NACT.

DESIGN

Meta-analysis.

DATA SOURCES

Relevant literature published before 1 May 2021 was searched using the Cochrane Library, Embase, PubMed and the Web of Science databases.

ELIGIBILITY CRITERIA

All studies involving patients with breast cancer treated with NACT and peripheral blood pretreatment NLR recorded as a dichotomous variable were included.

DATA EXTRACTION AND SYNTHESIS

Two researchers independently extracted and evaluated OR/HR and its 95% CIs of survival outcomes and clinicopathological parameters.

RESULTS

A total of 19 studies were identified. From each study, the impact of NLR on the pCR, OR and HR, with their 95% CIs were extracted and combined using either a random or fixed-effects model. The results indicate that a higher pCR in patients with a low NLR (OR 1.620, 95% CI 1.209 to 2.169, p＜0.001). In addition, an elevated NLR predicted lower disease-free survival (HR 2.269, 95% CI 1.557 to 3.307, p<0.001) and overall survival (HR 1.691, 95% CI 1.365 to 2.096, p<0.001) in patients with breast cancer treated with NACT.

CONCLUSIONS

NLR is a suitable biomarker for predicting pCR and survival in patients with breast cancer receiving NACT.

Factors Affecting Pathologic Complete Response Following Neoadjuvant Chemotherapy in Breast Cancer: Development and Validation of a Predictive Nomogram

Background There is an increasing need to develop a more accurate prediction model for pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) in breast cancer. Purpose To develop a nomogram based on MRI and clinical-pathologic variables to predict pCR. Materials and Methods In this single-center retrospective study, consecutive women with stage II-III breast cancer who underwent NAC followed by surgery between January 2011 and December 2017 were considered for inclusion. The women were divided into a development cohort between January 2011 and September 2015 and a validation cohort between October 2015 and December 2017. Clinical-pathologic data were collected, and mammograms and MRI scans obtained before and after NAC were analyzed. Logistic regression analyses were performed to identify independent variables associated with pCR in the development cohort from which the nomogram was created. Nomogram performance was assessed with the area under the receiver operating characteristic curve (AUC) and calibration slope. Results A total of 359 women (mean age, 49 years ± 10 [standard deviation]) were in the development cohort and 351 (49 years ± 10) in the validation cohort. Hormone receptor negativity (odds ratio [OR], 3.1; 95% CI: 1.4, 7.1; P = .006), high Ki-67 index (OR, 1.05; 95% CI: 1.03, 1.07; P < .001), and post-NAC MRI variables, including small tumor size (OR, 0.6; 95% CI: 0.4, 0.9; P = .03), low lesion-to-background parenchymal signal enhancement ratio (OR, 0.2; 95% CI: 0.1, 0.6; P = .004), and absence of enhancement in the tumor bed (OR, 3.8; 95% CI: 1.4, 10.5; P = .009) were independently associated with pCR. The nomogram incorporating these variables showed good discrimination (AUC, 0.90; 95% CI: 0.86, 0.94) and calibration abilities (calibration slope, 0.91; 95% CI: 0.69, 1.13) in the independent validation cohort. Conclusion A nomogram incorporating hormone receptor status, Ki-67 index, and MRI variables showed good discrimination and calibration abilities in predicting pathologic complete response. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Imbriaco and Ponsiglione in this issue.

A Decision-Making Supporting Prediction Method for Breast Cancer Neoadjuvant Chemotherapy

Neoadjuvant chemotherapy (NAC) may increase the resection rate of breast cancer and shows promising effects on patient prognosis. It has become a necessary treatment choice and is widely used in the clinical setting. Benefitting from the clinical information obtained during NAC treatment, computational methods can improve decision-making by evaluating and predicting treatment responses using a multidisciplinary approach, as there are no uniformly accepted protocols for all institutions for adopting different treatment regiments. In this study, 166 Chinese breast cancer cases were collected from patients who received NAC treatment at the First Bethune Hospital of Jilin University. The Miller–Payne grading system was used to evaluate the treatment response. Four machine learning multiple classifiers were constructed to predict the treatment response against the 26 features extracted from the patients’ clinical data, including Random Forest (RF) model, Convolution Neural Network (CNN) model, Support Vector Machine (SVM) model, and Logistic Regression (LR) model, where the RF model achieved the best performance using our data. To allow a more general application, the models were reconstructed using only six selected features, and the RF model achieved the highest performance with 54.26% accuracy. This work can efficiently guide optimal treatment planning for breast cancer patients.