Abstract
Introduction and objectives Up to 30% of early breast cancer (BC) patients treated with neoadjuvant chemotherapy will have systemic relapse during the follow-up. The integration of clinical, analytical and molecular parameters associated to tumor biology or host immune response could help to better stratify the prognosis of these patients. The aim of this study was to analyze the prognostic ability of immune-related and proliferation markers in combination with clinical parameters in patients with early BC treated with neoadjuvant chemotherapy. Methods and materials Retrospective and single-center cohort of BC patients treated with neoadjuvant chemotherapy between 2001 and 2010. We analyzed the following pre-treatment biological markers: neutrophil-to-lymphocyte ratio (NLR) in peripheral blood and CD3+-tumor-infiltrating lymphocytes (TIL), interferon-gamma and interleukin-10 in tumor samples by using a tissue microarray. Gene expression of AURKA, MYBL2, MKI67 and CTNNB1 in RNA from tumor samples was also evaluated by quantitative real-time polymerase chain reaction (qRT-PCR). Survival analysis was performed using Cox regression. The predictive capacity of the regression models was evaluated using AIC (Akaike Information Criterion) index, ROC curves and Harrell's C statistics. Results A total of 121 patients were included. Median age: 56 years. Cancer stage at diagnosis: 16% IIA, 28% IIB, 33% IIIA, 7% IIIB and 16% IIIC. Molecular subtype: 64% hormone receptor-positive (12% HER2-positive), 11% HER2-positive and 22% triple-negative. Pathological complete response (pCR): 16.5%. Median follow-up: 12 years. In the univariate analysis, NLR (HR 1.23, 95% CI 1.11-1.36; p<0.001), TIL (HR 0.89, CI95% 0.81-0.98; p=0, 02), AURKA (HR 1.02, 95% CI 1.01-1.04; p<0.001) and MYBL2 (HR 1.10, 95% CI 1.03-1.19; p=0.007) showed prognostic value for overall survival (OS). In the multivariate analysis, including staging after neoadjuvant therapy (HR 6.54, 95% CI 1.36-31.49; p=0.02), NLR (HR 1.33, 95% CI 1.08-1.64; p=0.008), TIL (HR 0.84, 95% CI 0.73-0.97; p=0.21), AURKA (HR 1.05, 95% CI 0.99-1.10; p=0.055) and MYBL2 (HR 1.14, 95% CI 1.00-1.31; p=0.04) remained as independent predictive variables in a regression analysis. Consecutive addition of these biomarkers to a regression model based on staging after neoadjuvant treatment progressively increased the discrimination accuracy of the models. These differences were more marked for the predictive model that included the four biological parameters identified in the multivariate analysis: NLR, TIL, AURKA and MYBL2 (Table 1). Conclusion In our cohort, the creation of a prognostic model integrated by clinical factors together with proliferation and both tissue and circulating immune biomarkers demonstrated high predictive capacity for OS. The validation of these findings in independent cohorts could impact in patient’s management.
. | AIC . | AUC ROC (95% CI) . | p . | Harrell's C-index (95% CI) . | p . |
---|---|---|---|---|---|
Model 1: ypTN | 274 | 0,76 (0,58-0,94) | Ref* | 0,74 (0,57-0,91) | Ref* |
Model 2: ypTN + NLR | 187 | 0,78 (0,61-0,95) | 0,83 | 0,80 (0,67-0,93) | 0,53 |
Model 3: ypTN + NLR + TIL | 85 | 0,85 (0,73-0,98) | 0,30 | 0,83 (0,73-0,94) | 0,30 |
Model 4: ypTN + NLR + TIL + + AURKA + MYBL2 | 57 | 0,91 (0,80-1) | 0,11 | 0,89 (0,81-0,97) | 0,13 |
. | AIC . | AUC ROC (95% CI) . | p . | Harrell's C-index (95% CI) . | p . |
---|---|---|---|---|---|
Model 1: ypTN | 274 | 0,76 (0,58-0,94) | Ref* | 0,74 (0,57-0,91) | Ref* |
Model 2: ypTN + NLR | 187 | 0,78 (0,61-0,95) | 0,83 | 0,80 (0,67-0,93) | 0,53 |
Model 3: ypTN + NLR + TIL | 85 | 0,85 (0,73-0,98) | 0,30 | 0,83 (0,73-0,94) | 0,30 |
Model 4: ypTN + NLR + TIL + + AURKA + MYBL2 | 57 | 0,91 (0,80-1) | 0,11 | 0,89 (0,81-0,97) | 0,13 |
*Ref: model 1 was the reference category that was used for comparison of AUC and C-index.
AIC: Akaike Information Criterion. AUC ROC: Area Under the ROC Curve. ROC: Receiver Operating Characteristic.
Citation Format: Esmeralda Garcia-Torralba, Beatriz Álvarez-Abril, Carlos Bravo-Pérez, Esther Navarro Manzano, Pilar de la Morena Barrio, Alejandra Ivars Rubio, Elisa García-Garre, Gema Marín Zafra, Francisco Ayala de la Peña, Elena García-Martínez. Development of prognostic models based on clinical, immune-related and proliferation factors in early breast cancer patients treated with neoadjuvant chemotherapy [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P4-07-03.