Impact of Pathologic Complete Response following Neoadjuvant Chemotherapy ± Trastuzumab in Locally Advanced Breast Cancer

Purpose This study was designed to examine the relationship between breast cancer molecular subtypes and pathological response to neoadjuvant chemotherapy (NAC) ± trastuzumab, in locally advanced breast cancer (LABC). Methods Female patients with LABC (T2–T4, N0–N2, and M0) who received neoadjuvant chemotherapy + trastuzumab if HER2+ subtype, followed by surgery and radiotherapy ± hormonal therapy, were identified. The primary endpoint was pathologic complete response (pCR) in the breast and axilla (ypT0/ypN0), with final analysis on disease-free survival (DFS) and overall survival (OS). Results Six hundred eighty-one patients with a median age of 44 years, premenopausal: 70%, median tumour size: 7.0 cm (range 4–11 cm), stage II B: 27% and III A/III B: 73%, ER+/HER2−: 40.8%, ER−/HER2−: 23%, ER+/HER2+: 17.7%, and ER−/HER2+: 18.5%. Overall pCR (ypT0/ypN0) was 23%. The pCR rates based on molecular subtypes were ER+/HER2−: 9%; ER+/HER2+: 29%; ER−/HER2−: 31%; and ER−/HER2+: 37%. At median follow-up of 61 months, ER+/HER2+ and ER+/HER2− subtypes had the best 5-year DFS and OS; meanwhile, ER−/HER2+ and ER−/HER2− subtypes had the worst. Conclusion Women with ER+/HER2− disease are the least likely to achieve pCR, with the highest rates in HER2+ and triple-negative subgroups. Degree of response is associated with OS; despite the comparatively higher likelihood of achieving pCR in ER−/HER2+ and triple-negative, these subgroups experience a survival detriment. We are consistent with the published data that patients who attain the pathological complete response defined as ypT0/ypN0 have improved outcomes.


Introduction
It has been established that locally advanced breast cancer (LABC) is an extremely heterogeneous disease that involves an extensive variety of biological phenomena. It is mostly diagnosed at advanced stages and has poor prognosis [1,2].
LABC has continued to serve as a serious problem with adverse outcomes in spite of all the revolutionary advancements made in context of cancer biology and introduction of targeted therapy for treatment of cancer [1]. In general, treatment of LABC is performed by considering it as a whole cohort. Clinical outcome is not predicted accurately with the TNM stage. However, the prognostic information can be refined through addition of biological characteristics which also prove to be very helpful in choosing suitable systemic treatments.
According to the findings of the CONCORD, a research conducted in Europe and the data published by the National Cancer Database, 4% of European and 8.5% of American patients suffering from breast cancer have LABC [3]. e situation is even worse in developing countries since the rate of incidence is between 33-77% [4][5][6][7]. In our healthcare setting, around 25% Saudi women are diagnosed with LABC. As per the Surveillance, Epidemiology and End Result (SEER) data, individuals suffering from stages IIIA and IIIB of breast cancer demonstrate five-year survival rates of 52% and 48%, respectively [1]. A promising approach for the treatment of LABC patients is the neoadjuvant chemotherapy (NAC). pCR has turned into a well-known surrogate marker indicating good long-term outcomes. Highly reduced risk of relapse and death has been reported for patients who have achieved a pCR regardless of their molecular subtype and initial stage [5,6,8].
Accurate pathological analysis following NAC is of great importance since it is helpful for the prediction of prognosis and for determining the effectiveness of the treatment. Classification of breast cancer using immunohistochemical marker expression or gene expression array data has recently been recognized. During the 13th St. Gallen International Breast Cancer Conference 2013, a novel surrogate intrinsic subtype of breast cancer was suggested to distinguish luminal A, luminal B (HER2−/HER2+), HER2 enriched, and triple-negative disease. With insufficient gene expression data for these subtypes, determination of subtype of cancer for an individual using immunohistochemical staining has proven to be a good approach in context of NAC [7,9,10]. Breast cancer subtypes have been shown to be associated with odds of achieving pCR in a meta-analysis on patient data [11]. Reduced rate of pCR together with favourable prognosis has been reported for luminal A subtype which is described as ER+, HER2−, and low KI 67 proliferative index (≤20%). Conversely, HER2+ and triple-negative patients had increased rate of pCR. An association has been found between luminal B subtype and intermediate rate of response [12]. HER2-positive/hormone receptor negative and triple negative subtype patients have demonstrated variable survival with or without a pCR [13]. Around 70% patients having triple negative (TN) disease and 40% HER2+ patients develop residual invasive carcinoma following surgery despite the progress made in context of NAC.
Anthracycline-based regimens represent an important treatment component in the management of patients with early-stage breast cancer, with reduced breast cancer mortality by 20-30% based on recent (EBCTCG) meta-analysis. It suggested that anthracycline should be used for patients with TNBC regardless of nodal status or HER2−/ER+ with significant nodal involvement [14]. On the other hand, taxanes were found to have a major clinical activity in breast cancer and proposed even in patients with low grade, ER+ tumor which are thought to be chromosomally stable [15]. is research is aimed at investigating the association between pathological complete response and outcomes in different molecular subtypes of breast cancer following neoadjuvant chemotherapy ± trastuzumab.

Procedure and Data Collection.
is study was approved by the Institutional Review Board (IRB). e study involved patients who were histologically diagnosed with unilateral locally advanced breast cancer (T2 ≥ 4 cm, T3 or T4, N0-N2, and M0) of noninflammatory nature and of stages II B, III A, and III B and who were given treatment using NAC approach during 2005-2014 at this setting through an electronic database which was prospectively maintained. Subjects included in this review were females who were diagnosed with LABC and were treated with NAC plus trastuzumab when HER2 positive and definitive surgery and locoregional radiotherapy ± hormonal therapy as per indication. Clinical and pathological data were collected including age, tumour size, histopathological subtype, hormone receptors (HR) and HER2 status, type of chemotherapy, number of given chemotherapy cycles, kind of surgery executed, and condition on last visit. Diagnosis of invasive breast cancer was performed through true-cut needle biopsy. Immunohistochemistry of pretreatment biopsy was used to determine estrogen (ER), progesterone receptors (PR), and HER2. No subject having residual disease demonstrated repeated ER, PR, or HER2 postoperatively. When 10% or more, tumour cells were stained for ER and/or PR, and hormonal receptor (HR) was considered positive. IHC was conducted to determine HER2 status which was verified through fluorescent in situ hybridization (FISH). e status was taken to be positive when it was equivocal (2+) on IHC and when FISH ratio was more than 2. Clinical examination was conducted with caliper prior to every treatment cycle in order to evaluate the clinical response. As per the Response Evaluation Criteria in Solid Tumour (RECIST) version 1.1, the response was categorized as progressive disease (PD), partial response (PR), stable disease (SD), and complete response (CR). e opinion of the breast surgeon as well as the patient served to be the basis for the decision to execute breast conservative surgery (BCS) or MRM. Adjuvant endocrine therapy for at least five years is suggested for individuals having HRpositive tumours. In case of premenopausal females, tamoxifen or aromatase inhibitors + LHRH agonists were given, and in case of postmenopausal females, aromatase inhibitors are given. Adjuvant radiotherapy was administered to individuals treated with BCS. Patients demonstrating pathological positivity of four or more axillary lymph nodes or those demonstrating at least clinical stage III disease or tumour size ≥5 cm at the time of diagnosis were treated with postmastectomy radiotherapy. pCR was defined as the complete absence of viable invasive tumour cells on pathologic examination in the breast and axillary lymph nodes, including surgical margins (ypT0 and ypN0). Patients with residual carcinoma in situ only were considered to have no pCR. Individuals demonstrating residual carcinoma only in situ were said to exhibit no pCR. As per our institutional guidelines, a written informed consent was taken from all the patients prior to initiation of chemotherapy ± targeted therapy.
Four main chemotherapy approaches were used for patients who were grouped accordingly. First group was given with anthracycline alone (A no T) comprising of doxorubicin or epirubicin together with cyclophosphamide. Anthracycline and taxanes (T ± trastuzumab) were given to the 2 nd group, and this system comprised of epirubicin/cyclophosphamide after which docetaxel or weekly paclitaxel ± trastuzumab was given doxorubicin/cyclophosphamide after which docetaxel or weekly paclitaxel ± trastuzumab was given. Anthracycline followed by platinum-based chemotherapy (A-P) was given to the 3 rd group.
is approach made use of epirubicin or doxorubicin after cisplatin/docetaxel ± trastuzumab. Taxanes alone (TC with no A) were given to the 4 th group, and this comprised of docetaxel/cyclophosphamide ± trastuzumab. Seventeen cycles of trastuzumab were administered to HER2positive patients.

Study Endpoints and Statistical Analysis.
e primary endpoint of this study was to determine the rate of pCR according to molecular subtypes. e secondary endpoints were to determine DFS (defined as the interval between the date of surgery and the date of disease recurrence or death before recurrence). Diseasefree patients were censored at the last follow-up date. OS is defined as the interval between the date of diagnosis and the date of death from any cause. Surviving patients were censored at the last follow-up date. e pCR rate was calculated for each molecular subtype, and the Fisher exact test was used to evaluate the relationship between the subtype and the pCR rate. Descriptive statistics were calculated using the median and the range for continuous variables and compared among different studying groups using the Wilcoxon test. Categorical variables, frequencies, and percentages were calculated and compared among different groups using the chi-squared test. e probabilities of OS and DFS were calculated using the Kaplan-Meier estimator. Survival curves were compared using the log-rank test. p value <0.05 was considered significant. Statistical analysis was applied using SPSS version 20.

Enrollment and Demographics.
ree thousand and two hundred forty patients were seen at our oncology centre during the period between January 2005 and December 2014. 770 (25%) patients had LABC, and 625 (19%) had stage IV disease. 89 patients from the 770 LABC patients were excluded. Figure 1 presents a flow diagram of the patient cohort. As per the criterion of this study, 681 cases were reviewed. Table 1 shows the baseline clinical features of the population of this study. e median age of 44 year, and the majority of the patients were premenopausal (70%), stages III A and B (73%), ER+ HER2− (40.8%), ER− HER2− (23%), ER+ HER2+ (17.7%), and ER− HER2+ (18.5%). BCS and MRM were performed in 12% and 88%, respectively.

Efficacy.
All patients demonstrated a clinical response during a follow-up of 38-92 months (median, 61 months). Out of 274 HER2+ individuals, pCR was achieved in 82 (33%). Conversely, 74 (17%) out of 434 HER2− patients achieved pCR (ypT0/ypN0). Overall pCR was found to be 23%. ere was a significant difference in the achievement of pCR in different subtypes. 37% ER−/HER2+, 31% ER−/ HER2−, 29% ER+/HER+, and 9% ER+/HER2− achieved pCR. Table 2 summarizes pathologic responses of patients receiving NAC. On the basis of chemotherapy regimens ± trastuzumab, subjects were categorized into four groups. As shown in Figure 1, A + T, A no T, A + platinum, and T no A were given to 380 (56%), 142 (21%), 91 (13%), and 68 (10%) patients. Pertuzumab was not recommended for any patient. Likewise, genetic testing was not conducted for triple-negative subjects. Data relating to suitability or preference of the patients for breast-conserving surgery was not collected owing to small rate (12%) of this surgery in this population.

Discussion
is study was aimed at analysis of the predictive and prognostic accuracy of molecular subtypes in females with LABC who were subjected to neoadjuvant chemotherapy ± trastuzumab. Molecular subtype was found to be a helpful tool in prediction of pCR. High pCR rate together with distinct survival outcome were demonstrated by triplenegative and HER2+ patients in comparison to luminal subtype. Response rates of 58% and 91% after neoadjuvant chemotherapy have been reported by earlier studies with 3-62% pCR [13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29]. However, subjects of these studies were not LABC patients only. Since this research involved 681 LABC patients, it is amongst largest studies investigating the association between molecular subtypes and responses to NAC. Variable NAC sensitivities were exhibited by different subtypes. Highest pCR rates were observed in ER−/HER2− and ER−/HER2+ and the lowest pCR rate by ER+/HER2−. Hence, the molecular subtype was proved as a strong independent predictor of OS and pCR. Achievement of pCR correlated with improved 5-year survival in all subtypes (Figures 2(a) and 2(b), Table 3) In context of neoadjuvant therapy, achievement of pCR has been found to be strongly correlated with improved event-free and overall survival (EFS&OS) [13,30]. Cortazar et al. conducted a pooled analysis involving 12 NAC trials (11955 patients) and reported that pCR in axilla as well as breast was more strongly linked with improved OS and EFS as compared to pCR in breast only. Moreover, patients with pCR exhibited 56% reduction in recurrence risk as   compared to those with residual disease. pCR can therefore be regarded as a promising prognostic tool with 71%, 92%, and 84% reduction in death risk in ER+/HER−, ER−/HER2+ and ER−/HER2− subtypes. pCR rate of 7.5% has been reported by CTneoBC study involving analogous luminal population (ER+/HER−) which is comparable to population of this research. pCR rate of TN patients (33%) was quite comparable to our finding (31%). e pCR rate of HER2+ patients was higher for ER− instead of ER+ tumour (50% vs. 31%) in comparison with our data (37% vs. 29%), respectively. ese pCR rates were comparable and, however, lesser than CTneoBC research [31].
During phase II trial of neoadjuvant (FEC100) after which cisplatin/docetaxel together with trastuzumab was given to HER2+ LABC patients, we found pCR rates in ER−/ HER2+ and ER+/HER2+ to be 62% and 56%, respectively. Only 36% pCR rate was exhibited by TN patients. is treatment approach is included in the institution's guidelines with pertuzumab being added recently for treating HER2+ and TN subtypes [16].
Gentile et al. have also analysed responses of LABC patients towards NAC and found variable rates of pCR in different molecular subtypes. Overall pCR rate was found to be 25%. pCR rates for ER+/HER2−, HER2+, and TN patients Journal of Oncology 5 were found to be 7%, 48%, and 23%, respectively. Increased pCR demonstrated by HER2+ patients could be because of utilization of pertuzumab. However, patients in this study had disease of more advance stages, and the median tumour size was 6 cm. Moreover, the median age was 52 years. ese studies indicate that the response to chemotherapy ± targeted therapies is not dependent on the tumour size. In fact, tumour biology affects pCR [31]. is research has validated the finding that patients with pCR exhibit desirable outcome [32]. During this research, patients with pCR demonstrated better five-year DFS. In particular, five-year DFS in patients with pCR was found to be 92% in comparison with 67% in patients without pCR, thereby making the difference between the two as 25% (P < 0.001). Moreover, poor DFS was found in patients with residual disease irrespective of the molecular subtype. It implies that absence of response towards NAC indicates unfavourable outcomes in all subtypes. Reports indicate association between luminal A tumours and favourable survival outcomes in comparison with other subtypes. Conversely, TN and HER2+ patients exhibited most unfavourable outcomes [32,33].
6094 females with invasive breast cancer were analysed through the Surveillance Epidemiology and End Result cancer registry (SEER) data, and it was found that molecular subtypes affected the four-year breast cancer specific survival. OS pattern was best in ER+/HER2+ patients, followed 20 in ER+/HER2− and ER−/HER2+ patients. Triple-negative patients exhibited worst OS [34]. Spring et al. recently conducted a meta-analysis involving 52 studies with 27,895    [35]. Intriguingly, our results were in agreement with these data as shown in Figure 3(a) and (b) and Table 3. During this research, the relapse rate was highest (45%) in ER+/HER2− patients followed by ER−/HER− (23.6%) and ER−/HER2+ (20.7%). ER+/HER2+ demonstrated lowest relapse rate (10.6%). A majority of relapses were seen in individuals who exhibited residual disease postoperatively. Similarly, earlier studies have reported variable rates of relapses in different individuals having different molecular subtypes. Differences in timing and pattern of relapses have also been reported in different molecular subtypes. TN patients relapsed within initial 3-5 years of the disease and the risk reduced to marginal following this period. Relapse risk was persistent and was five years in ER+ patients [32,33,36]. During this research, 30% ER+/HER2− patients relapsed persistently  Journal of Oncology 7 following 5 years. However, others relapsed during initial 2-5 years. Recurrence risk in HER2+ patients was considerably lesser than other subtypes possibly because of the fact that trastuzumab was given to all subjects included in this study. e pCR rates determined in this research are comparable with previously reported rates and greater for TN and ER−/HER2+ in contrast to ER+/HER2− patients. However, over 65% of HER2+ and 70% of TN patients had residual disease when their surgery was performed. ese patients could be treated with adjuvant capecitabine (CTRATE-X trial) or trastuzumab emtansine (KATHERINE trail) [37,38].
Correlation between pCR and improved OS and DFS (Table 3) is in agreement with earlier reports. Researchers are examining numerous approaches in postneoadjuvant setting like targeted therapy and immunotherapy. Clinical trials investigating treatment of residual disease can benefit from detailed information relating to genomic and molecular mechanisms. Technological advancements made in the field of biology during past years such as DNA sequencing, fieldresolved infrared spectroscopy, and acquisition of gene expression profiles have improved our knowledge relating to tumour biology, development, and progression of tumour and drug resistance [39,40].
Database of our institution proved to be very helpful in investigating this crucial aspect of LABC. Besides the retrospective design, this research has other limitations as well. Results could have been affected by unknown biases. KI-67 was not available for more grouping. e researchers did not consider the nature of adjuvant hormonal therapy, duration, and agents which could have influenced the OS. ough molecular subtypes indicated OS independently, numerous prospective clinical trials have not seen such trend.

Conclusions
Responses to NAC and outcomes of patients with different subtypes of LABC differed substantially. erefore, molecular subtype can be regarded as strong independent predictor of pCR and OS. Improved five-year survival was seen in patients with pCR irrespective of the subtype. In comparison to ER−/HER+ patients, ER+/HER2+ patients exhibited improved five-year OS and DFS and lesser rate of relapse. ER+/HER+ patients with pCR did not show relapse at all. pCR (ypT0/ypN0) can serve as promising surrogate endpoint for TN and ER−/HER2+ patients but not for ER+/ HER2− patients. Residual cancer burden (RCB) index may serve as a substitute surrogate endpoint for these patients after neoadjuvant setting serves as an opportunity for adjuvant clinical trials for testing novel drugs in populations with high proportions of high-risk individuals with residual tumour following neoadjuvant therapies.

NAC:
Neoadjuvant chemotherapy LABC: Locally advanced breast cancer pCR: Pathologic complete response BCS: Breast-conserving surgery MRM: Modified radical mastectomy SEER: Surveillance, epidemiology, and end result TN: Triple-negative IRB: Institutional Review Board HR: Hormone receptors ER: Estrogen receptors PR: Progesterone receptors IHC: Immunohistochemistry FISH: Fluorescent in situ hybridization CR: Complete response SD: Stable disease PR: Partial response PD: Progressive disease RECIST: Response evaluation criteria in solid tumours DFS: Disease-free survival OS: Overall survival REC: Research Ethics Committee.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
is study does not contain any studies with animals performed by any of the authors. All procedures performed in this study were following the ethical standards of institution and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. is work was approved by their institution' IRB (RAC # 2061020).

Consent
e Research Ethics Committee (REC) approved a consent procedure that did not include, or which altered, some or all the elements of informed consent (i.e., a waiver of informed consent).

Conflicts of Interest
Dr. Taher Al-Tweigeri has received speaking honoraria from Roche, Novartis, and Lilly; received travel support from Roche and Novartis; and has served on an advisory committee for Roche, Lilly, and Novartis. Other authors declare that they have no conflicts of interest.