Verification And Exploration Of Prognostic Factors For Triple-Negative Breast Cancer

Part Ⅰ:Retrospective Analysis of 967 Cases of Triple Negative Breast Cancer: Clinicopathologic and Recurrence FeaturesObjective:To explore the clinicopathological feature and prognosis for women with triple-negative breast cancer in Chinese people.Methods:Clinical data of 967 patients with histopathologically confirmed invasive ductal carcinoma and can be surgically removed triple-negative breast cancer, treated at Shanghai Cancer Center from Jan.2002 to Dec 2011, were enrolled. Clinical characteristics, recurrence and metastasis features, and 5-year breast cancer-specific survival rate were collected and analyzed.Results:Of the 967 patients, median age was 52 years (range 23-87). Sixty-six patients were confirmed to have family history of breast cancer, while 199 patients (20.6%) had family history of other malignant tumors. We identified more postmenopausal women(52.0%), more patients with larger tumor size (54.8%) and higher tumor grade (51.9%) in the patients. There were 118 patients accepted neoadjuvant chemotherapy, among which 25 patients received pathologic complete responce with a pCR rate of 21.2%. In the non-pCR patients,5-year recurrence rate was 39%,while 5-year BCSS was 75%, much lower than 80% of all the patients. Survival analysis showed that disease-free survival rates of triple negative breast cancer in five years were 90%(1-year),88%(2-year),84%(3-year) and 79%(5-year) the overall survival were 95%,89%,86% and 80%. By follow-up date, of the 967 TNBC patients,86 had recurrence and 210 (21.7%) had metastasis. The specific performance is that the risk of lung metastasis was the most (40.5%), and the others were, in order, brain metastasis (27.1%), liver metastasis (20.5%), bone metastasis (18.6%) and metastatic lymphadenopathy. The risk of recurrence and metastasis were high in the TNBC, mainly performed that the risk of distant metastasis is much higher which was the evidence of organ orientation. There were 141 patients died of breast cancer, and average survival time since recurrence and metastatic was 17.2 months.Conclusions:Triple-negative breast cancer has an aggressive clinical course, with larger tumor size and higher tumor grade. Risk of metastasis especially the risk of visceral metastasis is much higher, which suggests the organ orientation of metastasis.Part Ⅱ:Prognostic Factors of Triple-negative Breast CancerObjective:To investigate the prognostic factors related to triple-negative breast cancer by analyzing clinicopathologic characteristics, treatment and prognosis. Further evaluate the risk of survival associated with lymph node involvement and locoregional treatment.Methods:Clinical data of 849 patients histopathologically confirmed triple-negative breast cancer, without neoadjuvantchemotherapy, treated at Shanghai Cancer Center from Jan 2002 to Dec 2011. All the clinicopathological features and prognosis were available. Prognostic factors were retrospectively reviewed with Kaplan-Meier and Cox proportional hazards model. The risk of survival associated with lymph node involvement and locoregional treatment were further analyzed. The therapy of locoregional treatment includes breast-conserving therapy (BCT), modified radical mastectomy (MRM) and MRM with radiotherapy (RT).Results:Of the 849TNBC patients, median follow-up time was 43.5 months.Single factor analysis showed that the eld diagnosed age, postmenopausal, larger tumor size, axillary node involved, intravascular involved, and local treatment for MRM+RT were the prognostic risk factors of TNBC patients (P<0.05). Multivariate analysis indicated that menstrual status, tumor size and axillary node metastasis were independent prognostic factors of TNBC (P<0.001, P=0.009, P=0.001). For further discussing the prognosis associated with the number of lymph node involved, Kaplan-Meier estimates revealed that the more lymph node involved, the poorer prognosis patients have. Multivariate estimates revealed that lymph node classification is an independent prognostic factor in TNBC patients (P<0.001). Pairwise comparison by nodal status showed that when comparing NO with Nl patients, there was no significant difference in OS (P=0.071). However, NO and N1 shows a better prognosis respectively compared with N2 and N3 (P<0.05). When comparing prognosis of different locoregional treatment for TNBC with Kaplan-Meier, patients accepted breast conserving therapy had a better prognosis than those accepted modified radical mastectomy (with/without radiotherapy) (P<0.001). Cox multivariate shows no difference between patients with BCT or MRM without RT in OS and DFS (P=0.103) after adjustment for confounders. For patients with T1-2N0 TNBC treated with BCT or MRM without RT, there were no significant differences in OS or DFS.Conclusions:(a) Menstrual status, tumor size and lymph node status were independent prognostic factors of TNBC. (b) Patients with more than four lymph nodes involved may have a poor prognosis. (c) TNBC patients treated with BCT have no significant difference compared with those treated with MRM in T1-2N0. Early stage TNBCs were still the candidates for conservative surgery therapy. significant difference in OS (P=0.071). However, NO and N1 shows a better prognosis respectively compared with N2 and N3 (P<0.05). When comparing prognosis of different locoregional treatment for TNBC with Kaplan-Meier, patients accepted breast conserving therapy had a better prognosis than those accepted modified radical mastectomy (with/without radiotherapy) (P<0.001). Cox multivariate shows no difference between patients with BCT or MRM without RT in OS and DFS (P=0.103) after adjustment for confounders. For patients with T1-2N0 TNBC treated with BCT or MRM without RT, there were no significant differences in OS or DFS.Part Ⅲ:the Prognostic Value of Ki-67 Labeling Index in Patients with Triple-Negative Breast CancerObjective:The clinicopathological importance of Ki-67 labeling index (LI) in breast cancers has been intensely studied; however, the prognostic relevance in triple negative (TN) subtype has not been demonstrated. The aim of this analysis was to investigate the relevance of Ki-67 LI and age for the identification of different prognostic subgroups in TNBC patients.Methods:A total of 571 female TNBC patients were operated and diagnosed at our institution from January 2002 to June 2011. All the clinical-pathological features of 571 patients were available and categorized by Ki-67 LI and age of diagnosis. The cut-off values for Ki-67 LI and diagnosed age were selected using the medians. For survival analysis, the Kaplan-Meier method and the log-rank test were used. Cox proportional hazards models were fit to determine the association of Ki-67LI and diagnosed age with breast-cancer specific survival (BCSS) outcomes after adjustment for disease characteristics. A varying-coefficient Cox model was used to describe the effect of Ki-67 on the BCSS outcomes changing with diagnosed age after adjustment for disease characteristics.Results:Of the 571 patients, median age was 50 years, median Ki-67 LI 35%(range 0-97.5%), and median follow-up time of 44.5 months (range 2.2-139.1 months). The 5-year BCSS was 83%. Based on the analysis results, there was no prognostic significance in Ki-67 LI in all patients. When analyzing age of diagnosis as a continuous variable, the hazard ratio (HR) of Ki-67 LI>35% to Ki-67 LI≤35% of breast cancer-specific mortality (BCSM) increased in a S-shaped curve with increasing diagnosed age up to about 50 years old and remained higher risks for high Ki-67 LI. When focused on the patients being diagnosed at age ≤50 years, the BCSS was 84%, univariate survival analysis revealed that Ki-67 LI ≤35%(p= 0.018) were significantly associated with shorter BCSS, with an adjusted RRlow Ki-67 LI vs. high of 0.36 (95% CI 0.135-0.995, P=0.04). However, in the patients being diagnosed at age> 50 years, there was no significant differences between low Ki-67 LI and high Ki-67 LI (adjusted P=0.22,>0.05).Conclusions:Lower cellular proliferation measured by Ki-67 has poor prognosis relevance in TNBC of patients diagnosed ≤50 years old, which is different with Luminal A subtype. Further validation of the clinical significance of Ki-67 LI need to be demonstrated, which is essential for this group of breast tumors. proportional hazards models were fit to determine the association of Ki-67LI and diagnosed age with breast-cancer specific survival (BCSS) outcomes after adjustment for disease characteristics. A varying-coefficient Cox model was used to describe the effect of Ki-67 on the BCSS outcomes changing with diagnosed age after adjustment for disease characteristics.Results:Of the 571 patients, median age was 50 years, median Ki-67 LI 35%(range 0-97.5%), and median follow-up time of 44.5 months (range 2.2-139.1 months). The 5-year BCSS was 83%. Based on the analysis results, there was no prognostic significance in Ki-67 LI in all patients. When analyzing age of diagnosis as a continuous variable, the hazard ratio (HR) of Ki-67 LI>35% to Ki-67 LI≤35% of breast cancer-specific mortality (BCSM) increased in a S-shaped curve with increasing diagnosed age up to about 50 years old and remained higher risks for high Ki-67 LI. When focused on the patients being diagnosed at age ≤50 years, the BCSS was 84%, univariate survival analysis revealed that Ki-67 LI ≤35%(p= 0.018) were significantly associated with shorter BCSS, with an adjusted RRlow Ki-67 LI vs. high of 0.36 (95% CI 0.135-0.995, P=0.04). However, in the patients being diagnosed at age> 50 years, there was no significant differences between low Ki-67 LI and high Ki-67 LI (adjusted P=0.22,>0.05).Conclusions:Lower cellular proliferation measured by Ki-67 has poor prognosis relevance in TNBC of patients diagnosed ≤50 years old, which is different with Luminal A subtype. Further validation of the clinical significance of Ki-67 LI need to be demonstrated, which is essential for this group of breast tumors.Part Ⅳ:the Prognostic Value of BMI in Patients with Triple-Negative Breast CancerPart IV:the prognostic Value of BMI in Patients with Triple-Negative Breast CancerPurpose:Triple-negative breast cancer is a subtype of breast tumor with unique characteristics in terms of clinicopathological presentation, prognosis, and response to therapy. Obesity/overweight is confirmed to be associated with poorer outcomes in patients with hormone receptor-positive breast cancers. This association is not well established for women with TNBC. In this study, the prognostic effects of body mass index (BMI) on clinical outcome were evaluated in patients with TNBC.Methods:A total of 1106 patients who met the inclusion criteria were identified from January 2002 to June 2012. Clinical and biological features were collected to evaluate the relation between body mass index (BMI:weight in kilograms divided by the square of height in meters) and breast cancer specific survival (BCSS) after controlling for other clinically significant variables. Kaplan-Meier method and log-rank test were applied in the univariate survival analysis. Cox proportional hazard model was used to investigate the relevance between BMI and BCSS outcomes after adjustment for clinical and pathologic chatacteristics.Results:Six hundredand fifty-six patients (59.3%) were normal/underweight (BMI ≤24 kg/m2),450 patients (40.7%) were overweight (BMI>24 kg/m2). The median follow-up time was44.8 months. The 5-year BCSS was87%. Compared with normal/underweight patients, the multivariate hazard ratio (HR) for BCSS was 1.18 (95% confidence interval,0.97-1.44) for overweight patients in the whole population. Similarly, BCSS was not associated with BMI category in postmenopausal women; overweight patients had an HR of death of 0.94 (95% CI,0.74-1.21) compared with normal/underweight patients. In contrast, the multivariate hazard ratio (HR) for BCSS was 1.46 (95% confidence interval,1.03-2.07; p=0.04) for overweight patients in premenopausal women. A Cox regression analysis identified overweight (HR,1.46; 95% CI,1.03-2.07), positive nodal involvement (HR,12.43; 95% CI,4.32-35.78), and tumor size larger than 5 cm (HR,2.53; 95% CI,1.71-3.73) as independent prognostic factors for BCSS in premenopausal women.Conclusions:According to this analysis, overweight women with TNBC are at a greater risk of poorer prognosis than normal/underweight women inpremenopausal population. If validated, these findings should be taken into consideration for the development of targeted preventive programs.