TNM stage at diagnosis is more predictive of prognosis than pathological complete response in young breast cancer treated with neoadjuvant chemotherapy.

Abstract

Background Breast cancer is relatively rare in women younger than 35 years of age, accounting for 2–4% of the total number of breast cancer cases diagnosed each year in western countries and the USA 1–3. However, the proportion of young age-onset breast cancer was much higher in the Asian population 3–6. Breast cancer at a young age has been reported to have a more aggressive biological behavior and to be associated with worse prognosis compared with the disease in older patients 5,7–17. Higher incidence of recurrence and risk of death were detected in younger patients, even when more aggressive therapies were administered 10–14. Neoadjuvant chemotherapy (NCT) has been considered the standard treatment for locally advanced breast cancer patients. In human epidermal growth factor receptor 2 (HER2)-positive and triple-negative tumors, achievement of a pathological complete response (pCR) after NCT was associated with better survival outcomes 18. In recent studies, young breast cancer patients were reported to obtain higher pCR rates compared with older counterparts 19,20. However, it remains unclear whether pCR is a predictor for better prognosis in young breast cancer patients. In this study, we aimed to investigate chemotherapy response and its relation with prognosis in young breast cancer patients (<35 years old) who were treated with NCT. Patients and methods Patients From January 2003 to December 2013, patients with operable or locally advanced breast cancer who were treated with NCT at the Cancer Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College were reviewed systemically. The inclusion criteria for the study were as follows: (i) 35 years or younger or 60 years or older; (ii) diagnosis of invasive carcinoma confirmed by core needle biopsy before NCT and the lymph node status was evaluated by fine needle aspiration of palpable lymph node if applicable; (iii) immunohistochemical examination for estrogen receptor (ER), progesterone receptor (PgR), and HER2 status using tumor specimens from core needle biopsy; (iv) newly diagnosed breast cancer patients; (v) combination of paclitaxel and carboplatin (PC) in patients with triple-negative breast cancer as NCT; for other patients, a combination of cyclophosphamide and epirubincin (CE) as NCT with postoperative paclitaxel, or epirubincin in combination with paclitaxel (ET) as NCT 21; (vi) adequate hematologic, hepatic, and renal functions. The exclusion criteria for the study were as follows: (a) stage IV disease, bilateral breast cancer, male breast cancer, and patients complicated with other malignancies; (b) patients who did not have complete clinical information or immunohistochemistry; (c) patients who were lost to follow-up immediately after treatment. This study was a retrospective observational research and patients’ information was collected in the hospital database. There was no direct intervention in patients’ treatment or care. Therefore, a patient’s consent was not required. This study was approved by the Ethics Committee of Cancer Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College. Treatment Before the initiation of NCT, bilateral breast MRI or ultrasound, chest radiography, abdominal ultrasound, or computed tomography scans were performed to determine clinical stage. All patients were staged according to the 7th TNM (tumor–node–metastasis) staging system maintained by the American Joint Committee on Cancer (AJCC). All patients received NCT with CE, or ET, or PC regimens. CE-T regimen: cyclophosphamide 600 mg/m2 intravenously, day 1, and epirubicin 80 mg/m2 intravenously, day 1, q14d×4 cycles in NCT, followed by postoperative sequential paclitaxel 175 mg/m2 intravenously, day 1, q14d×4 cycles; ET regimen: epirubincin 75 mg/m2 intravenously, day 1, and paclitaxel 175 mg/m2 intravenously, day 2, q21d×4 cycles as NCT; and two cycles of ET regimen were repeated after surgery; PC regimen: paclitaxel 175 mg/m2 intravenously, day 1, and carboplatin area under the curve=5 intravenously, day 1, q21d×6 cycles as NCT. Mastectomy or breast-conserving surgery was performed within 1 month after the completion of NCT. Adjuvant radiotherapy was prescribed at the discretion of physicians mainly on the basis of the TNM stage before NCT, followed by endocrine therapy in cases with ER-positive or PgR-positive tumors. Trastuzumab was recommended in HER2-positive patients, but not compulsory. Efficacy evaluation Clinical efficacy was estimated every two cycles by clinical, mammographic, B-ultrasound examinations, and MRI according to Response Evaluation Criteria in Solid Tumors 1.1 criteria. Pathological efficacy was assessed by pathologic report after surgery and imaging data. Efficacy evaluation indicators included pCR, clinical complete response (CR), clinical partial response (PR), overall objective response rate (ORR=CR+PR), clinical progressive disease (PD), and clinical stable disease (SD). pCR was defined as no histological evidence of malignancies or only in-situ residuals in breast tissue after surgery, and complete disappearance of lymph node metastasis. CR was defined as disappearance of all known lesions. PR was defined as at least a 30% decrease in the sum of the largest diameters of target lesions. PD was defined as at least a 20% increase in the sum of the largest diameters of target lesions or new lesions detected. SD was defined as a reduction in the largest sum diameters of tumors by no more than 30% or an increase of no more than 20%. Patients with overall objective response rate received planned treatment and sequential surgery. Patients who fulfilled the criteria for SD or PD at the initial efficacy evaluation received surgery as soon as possible and were then treated with alternative postoperative regimens. Statistical analysis All data were analyzed using SPSS medical statistical software (version 15.0; SPSS Inc., Chicago, Illinois, USA). Disease-free survival (DFS) was defined as the period from the date of receiving NCT to the date of recurrence or metastasis or last follow-up; overall survival (OS) was defined as the period from the date of receiving NCT to the date of death for any cause or last follow-up. Both OS and DFS were analyzed using the Kaplan–Meier method. Comparisons of OS or DFS between groups were performed using the log-rank test. A two-tailed P value less than 0.05 was considered statistically significant. The χ2 or Fisher’s exact test was performed to compare the clinicopathological variables and pCR rates between the subgroup of younger than 35 years of age and the subgroup of older than 60 years of age. Results Patient characteristics and treatment From January 2003 to December 2013, a total of 141 breast cancer patients were enrolled in this study and 74 (52.5%) of these patients were younger than 35 years old. As shown in Table 1, younger patients presented a tendency to have a tumor size of more than 5 cm (P=0.085), to have axillary positive nodes (P=0.118), and to have lymphovascular invasion (P=0.007) compared with their older counterparts (≥60 years old). As for the treatment, younger patients were more likely to choose breast-conserving surgeries (P=0.024) and to receive adjuvant radiotherapy after surgeries (P<0.001). For the entire cohort, the median number of NCT cycles was 4 (range: 4–6 cycles). The chemotherapy regimens were changed for a total of 31 patients who had a poor response to preoperative chemotherapies (including 26 SD and five PD). Overall, 52% (16/31) of these patients were in the young group, and the other 15 patients were older than 60 years. All 96 patients with preoperative or postoperative pathological ER-positive or PgR-positive breast cancer received different postoperative adjuvant endocrine therapies. In all, 18.9% (14/74) of patients in the young group and 16.4% (11/67) of patients in the old cohort received 1 year of trastuzumab after surgery or adjuvant chemotherapy.Table 1: Patients’ baseline characteristicsTreatment response All patients were evaluable for response to NCT. The rates of pCR, clinical PR (except clinical PR, but confirmed as pCR finally), clinical SD, and clinical PD were 12.8% (18/141), 65.2% (92/141), 18.4% (26/141), and 3.5% (5/141), respectively. pCR rates were similar between the young group and the old cohort (12.2 vs. 13.4%, P=0.821). The PC regimen tended to yield higher pCR rates in the young cohort, but this was not statistically significant. HER2-positive or inflammatory breast cancer patients in young group were also more likely to achieve pCR compared with their older counterparts. As shown in Table 2, there were no significant differences between the pCR rates of young and old patients across various subgroup analyses, including hormone receptor status, axillary lymph node metastasis from primary tumors, primary tumor size, HER2 expression status, and NCT regimens.Table 2: Clinical and pathological factors affecting pathological complete response ratesSurvival analysis By the last follow-up on 1 November 2015, with a median follow-up of 32 months, 38 relapse events and 11 deaths have occurred. Patients in the young group showed significantly lower 5-year DFS compared with their old counterparts (Fig. 1a, 62.2 vs. 77.8%, P=0.037). However, no significant difference in 5-year OS was observed between the young group and the old cohorts (Fig. 1b, 84.0 vs. 94.8%, P=0.212).Fig. 1: (a) Kaplan–Meier curves of disease-free survival (DFS) in the young group (N=74) and the elderly group (N=67); Kaplan–Meier curves of overall survival (OS) in the young group (N=74) and the elderly group (N=67).In the group of young patients, pCR was not a significant predictor for DFS (Fig. 2a, P=0.408), whereas significant differences were observed with respect to the ascending TNM stage at diagnosis (Fig. 2b, P=0.001). In the subset of old patients, neither pCR nor TNM stage was a prognostic factor against DFS (Fig. 2c, P=0.129 and Fig. 2d, P=0.174).Fig. 2: Disease-free survival (DFS) curves by chemotherapy response (a) and TNM stage at diagnosis (b) in young patients; DFS curves by chemotherapy response (c) and TNM stage at diagnosis (d) in old patients. pCR, pathological complete response.To further explore the difference in survival between young and old groups, we carried out a stratified analysis. As shown in Fig. 3, the 5-year DFS was lower in the young group than in the old group within the subgroup of patients who presented with inflammatory breast cancer [56.5 vs. 75.1%, hazard ratio (HR)=2.47, P=0.044], with a large primary tumor (46.7 vs. 81.4%, HR=2.93, P=0.053), with lymph node involvement (57.7 vs. 74.8%, HR=2.40, P=0.025), and with higher TNM stage at diagnosis (46.7 vs. 67.1%, HR=2.52, P=0.027). Subgroup analyses on OS were carried out across predefined subsets and no significant difference was observed (Fig. 4).Fig. 3: Forest plot of disease-free survival. Hazard ratio more than 1 indicated that old patients had better survival outcome. CI, confidence interval; ER, estrogen receptor; PgR, progesterone receptor; TNBC, triple-negative breast cancer.Fig. 4: Forest plot of overall survival. Hazard ratio more than 1 indicated that old patients had better survival outcome. CI, confidence interval; ER, estrogen receptor; NA, not available; PgR, progesterone receptor; TNBC, triple-negative breast cancer.Discussion Although there was no consensus definition for young breast cancer, it had been widely believed that breast cancer at a young age had a more aggressive biological behavior compared with the disease arising from older patients. Tumors in younger women present with a higher grade, a higher T stage, a higher N stage, and more dedifferentiation, and have a higher proliferating fraction and more vascular invasion 5,7–17. In our study, the clinicopathological characteristics of young patients were consistent with previous researches. Keegan et al.22 and Colleoni et al. 23 found that higher proportions of HER2-positive tumors occurred in young patients. It was reported that about 25% of all breast cancers are ER or PgR negative, but a large proportion of hormone receptor-negative tumors occur in young women 13,15,17,24. Azim et al. 11 reported that there was a significantly higher proportion of basal-like tumors (34.3%) in young patients compared with those aged 41–52 (27.7%). No significant differences in the breast cancer subtype pattern were observed in this study and this could be attributed to the limited sample size of our study. It has been shown that younger age was associated with a less favorable prognosis 5,7–17. Tang et al. 15 found that with a follow-up of 54 months, inferior 5-year DFS (72 vs. 83%, P<0.01) and 5-year OS (87 vs. 93%, P <0.01) were observed in patients aged younger than 40 years compared with those aged 40–50 years. In our study, worse 5-year DFS was observed in young patients (62.2 vs. 77.8%, P=0.037), which was consistent with the results in previous studies. Stratified analysis showed that the predictive value of age on DFS was proven in the subgroup of patients with high-risk factors, including inflammatory breast cancer, large primary tumor, positive axillary lymph node, or higher TNM tumor stage at diagnosis, whereas no difference in the 5-year DFS was observed in patients without the above-mentioned risk factors. Similarly, in the study of Han et al. 5, there was no significant difference in DFS between young and old groups in lymph node-negative patients (P=0.223), whereas the younger group showed worse prognosis among lymph node-positive patients (P<0.001). In a Korean study, patients in the very young group with lymph node metastasis had poorer 5-year OS (70 vs. 83%, P<0.001) and DFS (58 vs. 74%, P <0.001) than their older counterparts; in patients without lymph node metastasis, the survival outcomes did not differ significantly between the two groups 17. Therefore, it is a key point to identify high-risk young breast cancer patients, and implement a tailored and aggressive treatment to improve the outcomes of these patients. It had been reported that improved survival outcomes were observed in patients with pCR compared with those with residual tumor 18,25–29. Further analysis showed that pCR was associated with better DFS in ER or PgR-positive/HER2-negative with grades 1–2, HER2-positive, and triple-negative disease, but not for those with ER or PgR-positive/HER2-positive, and ER or PgR-positive/HER2-negative with grade 3 tumors 18. In addition, in a meta-regression analysis of 29 heterogeneous neoadjuvant trials, pCR was not suggested to be a surrogate end point for DFS and OS in patients with breast cancer 30. At the 2016 annual meeting of American Society of Clinical Oncology, Robidoux et al. 31 reported 5-year outcomes of the NSABP protocol B-41 and found that long-term outcomes correlated with pCR status. Additional analyses indicated that pCR was related to a significant improvement in survival rates in the ER-negative subset, but not in the ER-positive cohort. In our study, young patients achieving pCR showed similar 5-year DFS compared with those with non-pCR (P=0.408). Therefore, pCR may be not an appropriate surrogate for prognosis in young breast cancer patients treated with NCT. The TNM staging system maintained by the AJCC is considered the most clinically useful cancer staging system for cancers. Orucevic et al. 32 reported on 782 Caucasian women diagnosed with invasive ductal carcinoma who were grouped according to TNM stage and molecular phenotype. The results supported the traditional TNM staging as a continued relevant predictive tool for breast cancer outcomes even with the emerging prognostic impact of tumor biomarkers (ER/PgR/HER2). Carey et al.33 suggested that classification of residual tumor in the breast and axillary surgical specimens after NCT using the AJCC TNM staging system could predict distant relapse and survival. In our study, young patients with higher TNM stage at diagnosis showed worse survival outcomes. Therefore, TNM stage may be more predictive of prognosis than pCR in breast cancer patients treated with NCT. This study is limited by its retrospective nature, nonrandomized design, small sample size, and relatively short follow-up duration. Trastuzumab was not covered by medical insurance in China and was much more expensive than common chemotherapy drugs. Therefore, some of the HER2-positive patients could not afford the expense. The diversity of NCT regimens and difference in the proportions of patients receiving trastuzumab between young and old cohorts may also have influenced the results of this study. Further prospective studies are warranted to validate the results. In recent years, researches have confirmed that breast cancer arising in young women is a unique disease entity driven by complex biologic processes extending beyond hormone receptors and hereditary cancer syndromes. Anders et al. 34 found 367 significant gene sets among young women’s tumors that specifically distinguished them from tumors arising in older women, including those related to immune function, the mammalian target of rapamycin/rapamycin pathway, hypoxia, BRCA1, stem cells, apoptosis, histone deacetylase, and multiple oncogenic signaling pathways. In the study of Azim et al. 11, breast cancer in the young was enriched with processes related to immature mammary epithelial cells and growth factor signaling. There was also downregulation of apoptosis-related genes. The identification of genomic pathways specific to breast cancer arising in young patients provided a better understanding of the interplay of age and contributing biologic processes and a unique opportunity to explore therapeutic targets. Conclusion Young breast cancer patients treated with NCT present more aggressive clinicopathological features and worse prognosis compared with their elderly counterparts. TNM stage at diagnosis may be more predictive of prognosis than pCR in young breast cancer patients with NCT. The underlying biology of young breast cancer needs to be elucidated and the development of tailored treatment is crucial. Acknowledgements The authors thank all doctors and nurses of the Department of Medical Oncology for their help with the realization of this study. Binghe Xu and Jiayu Wang designed the research, analyzed the data, and wrote the paper; Jingjing Wang analyzed the data and wrote the paper; Jiayu Wang, Qing Li, Pin Zhang, Peng Yuan, Fei Ma, Yang Luo, Ruigang Cai, Ying Fan, Shanshan Chen, Qiao Li, Binghe Xu selected the cases and analyzed the clinical data. Conflicts of interest There are no conflicts of interest.