Abstract
Background/Aim: We evaluated the efficacy of neoadjuvant chemotherapy (NACT) in reducing locally-advanced breast cancer (LABC) size, thus improving breast-conserving surgery (BCS) rates, as well as its long-term outcome. Patients and Methods: We analyzed 59 patients treated between 1999-2017 with NACT and subsequent surgery for LABC. Results: We observed a tumor size reduction in 95% of cases, resulting in downstaging in 62.7%. The average tumor shrinkage was 49%. Women with a reduction in tumor size >50% after NACT had better 10-year OS rates than women with a reduction ≤50% (p=0.025). NACT allowed to perform BCS in 44% cases, whereas the remaining 56% cases underwent mastectomy. Overall, we observed recurrences in 37.2% patients. Recurrence rates after BCS and mastectomy were 30.7% (6 loco-regional and 2 distant cases) and 42.4% (5 loco-regional and 9 distant cases), respectively (p=0.07). Conclusion: NACT confirmed its effectiveness in reducing mastectomy rates by approximately 50%, without increasing the risk of local or distant recurrences.
- Neoadjuvant chemotherapy
- long-term outcome
- down-staging
- tumor shrinkage
- breast conserving surgery
- recurrence
Neoadjuvant chemotherapy (NACT) is the standard of treatment for locally advanced breast cancer (LABC) and inflammatory breast carcinoma. It is also currently used to decrease mastectomy rates in women with large but operable tumors, as reported in previous studies (1, 2). The use of chemotherapy prior to surgery instead of post-operatively has not shown improvements in disease-free survival (DFS) or overall survival (OS) (3, 4). However, the tumor downstaging achieved after NACT allows to convert inoperable cases into operable ones and perform breast-conserving surgery (BCS) in a large proportion of patients for whom mastectomy would have been the only option for the control of loco-regional disease. The extent of clinical and pathologic response to NACT is obviously crucial to allow the use of BCS. Furthermore, a good clinical response to NACT is associated with a significant increase in DFS compared to patients with a minor response to therapy (1-5). Therefore, the introduction of NACT resulted in an increased use of BCS in patients with LABC. However, this may raise an issue in achieving adequate loco-regional control because of the difficulty of assessing tumor margins after the administration of NACT. Several studies investigated the oncologic safety of BCS after NACT, however data concerning loco-regional recurrence rates are discordant (6, 7). In particular, there are a few studies evaluating long-term results after NACT with controversial findings (8-10).
Patients and Methods
Study population. In a retrospective series, we selected all patients with LABC or inflammatory breast cancer who underwent NACT between September 1999 and September 2017 at our Breast Unit of the Gynecology section, Department of Health Science, Careggi Hospital, University of Florence. Overall, we identified 59 patients. We divided our population into two groups: patients who underwent BCS after NACT (n=26) and patients who underwent mastectomy after NACT (n=33). Clinical T (cT) and N (cN) stages were evaluated by physical examination, breast ultrasound, mammography and MRI before the start of NACT, according to the staging system set by the American Joint Committee on Cancer (AJCC) (11). The same examinations were repeated after NACT, to assess the response in terms of tumor shrinkage. All patients had biopsy-proven invasive breast cancer. Hormone receptor status was assessed by immunohistochemistry (IHC) and HER2 status was firstly evaluated by IHC and, if necessary, by fluorescence in situ hybridization (FISH).
Treatment. NACT consisted of intravenously administered anthracycline (9%) or combined anthracycline-taxane (91%) (median 6 cycles) (12). Trastuzumab was administered to patients with c-erbB2 (HER2)-overexpressing tumors (10.2%) (13). BCS or mastectomy (with or without immediate breast reconstruction) were performed according to clinical tumor response and tumor biological features. All patients treated with BCS underwent subsequent breast irradiation. Patients with extensive nodal involvement (cN2-3) or both clinically and biopsy-proven N1 disease prior to NACT, who remained clinically positive after the treatment (ycN1), underwent axillary dissection after NACT. Whereas, those with clinically positive, but not biopsy-proven involved lymph nodes or who down-staged from cN1 to ycN0 after NACT, underwent sentinel node biopsy after NACT, provided that axillary ultrasound scan performed after NACT did not reveal suspicious lymph nodes (14). In this latter case, patients were selected to undergo axillary dissection.
Response to NACT. We considered as complete pathological response (pCR) to NACT those cases with no evidence of residual invasive tumor in the surgical specimen of the breast and axillary lymph nodes. We considered as partial pathological response (pPR) when tumor was still present in the surgical specimen, although reduced in size. We defined as stable disease when no change in tumor size was observed and progressive disease when an increase in tumor size occurred during NACT.
Follow-up and clinical outcome. For the first 5 years after NACT and surgical treatment, patients underwent physical examination and tumor markers evaluation (CEA, Ca-15.3) every 6 months and mammography, breast ultrasound, abdominal ultrasound and bone scintigraphy every year. Every local or distant recurrence was registered. The median follow-up period was 60.5 months (range=12-180 months).The end-points analyzed were DFS and OS. DFS was measured from the date of surgical treatment to the date of the first loco-regional or systemic recurrence. OS was calculated from the date of surgical treatment to the date of the last follow-up or death from any cause. Loco-regional recurrence (LRR) was defined as recurrent disease in the ipsilateral breast or chest wall, in the ipsilateral axillary, supraclavicular, subclavian or internal mammary lymph nodes. Recurrence at any other site was considered as distant metastasis (DM).
Compliance with ethical standards. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Statistical analysis. The DFS and OS rates were calculated using the Kaplan-Meier method and compared by the log-rank test. The associations between patients' features were assessed by Fisher's Exact test. A p-value≤0.05 was considered statistically significant. All analyses were performed using the R Project for Statistical Computing version 3.5.2, R Foundation for Statistical Computing, Vienna, Austria.
Results
The median age at diagnosis was 47 years (range=21-74 years). Patients' clinical and pathological features are reported in Table I. All patients were treated with NACT. In 6 cases (10.1%) neoadjuvant trastuzumab was also given. We observed a reduction in tumor size in 56 out of 59 tumors (95%), with an average tumor shrinkage of 49% and a down-staging in 37 out of 59 tumors (62.7%). A detailed description of tumor stage before and after NACT is shown in Table II. Excluding the 8 patients presenting with inflammatory cancer, we observed a down-staging in 35 out 51 tumors (68.6%). We observed 2 pCR (3.4%), 54 pPR (91.6%), 3 patients (5%) with stable disease, while no patient had disease progression during NACT.
We divided the patients into two groups, according to the extent of tumor shrinkage after NACT: >50% (n=24; 40.7%) vs ≤50% (n=35; 59.3%). We observed 5 recurrences (20.8%) and 1 death (4.1%) in the >50% shrinkage group, whereas we had 17 recurrences (48.5%) and 11 deaths (31.4%) in the ≤50% shrinkage group (p=0.01). We then analyzed the 10-year DFS and OS according to response to NACT. The difference between the two groups was statistically significant in terms DFS (p=0.05) and of OS (p=0.025) (Figure 1).
Overall, 26 patients were treated with BCS (44%) and 33 patients underwent mastectomy (56%) after NACT. Not including the 8 patients with inflammatory cancer, for whom mastectomy was required in any case, NACT allowed to spare a mastectomy in 51% of patients. We also assessed the relationship between tumor histological types and tumor shrinkage after NACT. Patients with ductal carcinoma were more likely to have a >50% tumor shrinkage compared to those with lobular carcinoma (53.1% vs. 17.7% respectively, p<0.01, Fisher's exact test). Therefore, patients with ductal carcinoma were more likely to undergo BCS (59.2%) compared to patients with lobular carcinoma (41.2%).
DFS and OS were analyzed according to tumor pathological response after NACT (ypT) (Figure 2). Patients with less advanced tumor stage after NACT (ypT0, ypT1 or ypT2) had better DFS (p<0.001) and OS (p<0.001) than patients with more advanced tumor stage after NACT (ypT3-ypT4). DFS and OS were also computed comparing the axillary lymph nodes response after NACT (ypN) (Figure 3). Patients with less advanced stages after NACT (ypN0 or ypN1) had better DFS (p<0.001) and OS (p<0.001) than those with more advanced stages (ypN2-ypN3).
Overall, we observed 6 loco-regional recurrences (LRR), 2 distant metastasis (DM), 2 deaths in the BCS group and 5 LRR, 9 DM, and 10 deaths in the mastectomy group. Recurrences after BCS occurred after an average of 46 months. Considering only the LRR, they occurred after an average of 58.8 months among patients treated with BCS, whereas they occurred after an average of 31 months among those treated with mastectomy. None of the patients who relapsed locally after BCS died of the tumor, although in 4 cases a subsequent mastectomy was needed (after an average of 66 months). Regarding the two patients who died after BCS, they died for distant metastasis (bone, pleura or liver). Regarding the 10 patients who died after mastectomy, 6 of them had inflammatory cancer. Overall, the incidence of relapse after mastectomy was not significantly different from that observed after BCS (p=0.07, Fisher's exact test), although there was a trend toward prevalence of LRR after BCS and DM after mastectomy. In fact, the incidence of DM was 27.2% after mastectomy versus 7.7% after BCS (p=0.055).
Discussion
The main purposes of this study were to evaluate the efficacy of NACT in 59 patients with LABC in reducing the tumor size, allowing a conservative surgery and analyze the long-term results related to the risk of local and distant recurrences. One should consider that all patients in our study should have undergone mastectomy if they had not received NACT. Thanks to NACT administration, a conservative surgery was feasible in 26 out of 59 cases (44%) whereas in the remaining 33 cases (56%) we performed a mastectomy. Excluding the 8 patients with inflammatory cancer, who must be treated with mastectomy anyway, mastectomy rate decreased to 51%, according to previous studies' results (1, 2).
An interesting finding of our study was the impact of tumor shrinkage after NACT on DFS and OS. Indeed, patients who achieved a >50% tumor size reduction after NACT had a significantly better DFS and OS than those who did not achieve this target (Figure 1). Therefore, a greater tumor response to NACT not only increased the use of BCS, but it also resulted in an improved outcome. Our findings are consistent with previous studies showing that a pCR to NACT was associated with better outcome (15-18). However, the impact of response to NACT on outcome has not yet been established by universally accepted criteria (19, 20), thus further studies are needed to assess the prognostic role of pCR and tumor size reduction.
Furthermore, we found that a greater lymph node response to NACT (ypN0-1) was significantly associated with longer DFS and OS, allowing patients with negative axillary lymph node after NACT (excluding N2-N3 patients before NACT) to undergo only sentinel lymph node biopsy, instead of axillary dissection and, therefore, preventing dissection side effects. This approach is supported also by the recent ACOSOG Z1071 trial (21), which confirmed the accuracy of axillary ultrasound after NACT in addressing patients towards sentinel lymph node biopsy or axillary dissection, as appropriate. Previous studies from our group also reported the efficacy of axillary ultrasound in selecting patients suitable for sentinel lymph node biopsy (22, 23).
We also noticed that response to NACT was different depending on tumor histological type. Patients with ductal carcinoma responded better to NACT, compared to those with lobular carcinoma, who more often required a mastectomy (Table I). This result was partly due to the reduced efficacy of NACT in this specific histological type, as well as to the intrinsic characteristics of lobular carcinoma, with a higher incidence of multifocal and multicentric disease (24, 25). We could also suggest that the different response to NACT depends on specific biomolecular pathways in different histological/biomolecular cancer types, as previous studies revealed (26). Further studies are needed to better establish these findings.
Concerning recurrences after surgery, approximately one third of patients relapsed (22 cases out of 59; 37.2%). Analyzing the site and the time of relapse onset, we found that the pattern of relapse after mastectomy was not significantly different from that observed after BCS (p=0.07, Fisher's exact test), although there was a trend toward prevalence of local recurrences after BCS and DM after mastectomy, as shown in several previous studies (4, 10, 27). We found a greater incidence of local recurrences after BCS compared to mastectomy, but the difference was not significant (23.0% vs. 15.1% respectively, p=0.48), in accordance with previous studies (10). It is interesting to note that the local recurrence event did not affect patients' outcome. In fact, all of the patients with LRR after BCS were still alive and disease-free after an average follow-up of 8 years. On the other hand, DM were more frequent in patients treated with mastectomy (9 cases out of 11; 81.8%) revealing a higher aggressiveness of the primary tumor. DM obviously resulted in a reduced OS for such patients: 9 patients out of 11 (81.8%) affected by DM died after an average distance of 38 months. These findings suggest that tumors treated with mastectomy following NACT are possibly more aggressive than tumors suitable for BCS after NACT. Focusing the attention on the cases with LRR after BCS, one could question that if such patients had undergone mastectomy in the first instance, they would have probably avoided relapsing and undergoing a second surgery. However, it is interesting to note that among the 6 patients who had a LRR after BCS, 2 were treated with a second BCS, as they presented small isolated cutaneous recurrences around the scar; the remaining 4 patients required mastectomy. Therefore, considering all the patients treated with BCS after NACT, after a long-term follow-up, only 4 cases out of 26 (15.3%) needed a salvage mastectomy. Whereas, in the remaining 22 patients (84.7%) the use of NACT allowed to avoid such a mutilating surgery without compromising patients' survival.
Advantage and disadvantages of NACT. Our findings confirmed the effectiveness of NACT to reduce tumor size and to increase BCS rates in LABC, without increasing the risk for local or distant recurrences. Mastectomy rate decreased by 51% among patients with non-inflammatory LABC. A reduction of tumor size >50% was significantly associated with improved long-term outcome. The trend towards prevalence of LRR after BCS and DM after mastectomy observed in the current study represents an interesting finding. The relatively higher incidence of local recurrence after BCS seems not to compromise cure, because these patients can successfully undergo either a second surgical excision or a salvage mastectomy. The occurrence of DM, more frequent after mastectomy, is obviously associated with a poor outcome. Further studies, with larger series of patients, are required to better establish the evidence of these associations.
Footnotes
* These Authors contributed equally to this study.
Authors' Contributions
TS: Conceptualization, data curation, formal analysis, investigation, founding acquisition, writing original draft and editing; MG: Conceptualization, data curation, formal analysis, investigation, writing original draft and editing; IR, AVa, KT and AVi: Data curation, critical review, editing; JN and SB: Methodology, formal analysis, investigation.
Conflicts of Interest
All Authors declare that there are no conflicts of interest regarding this study.
- Received December 19, 2019.
- Revision received January 14, 2020.
- Accepted January 15, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved