Pathological responses and survival outcomes in patients with locally advanced breast cancer after neoadjuvant chemotherapy: a single-institute experience

Background Pathological complete response (pCR) is a surrogate for the efficacy of neoadjuvant chemotherapy (NCT) in locally advanced breast cancer (LABC). We analyzed the predictive clinical factors for pathological responses and survival outcomes in a cohort of Egyptian patients. Methods We evaluated the medical records of patients with breast cancer who received NCT in our academic institute. Survival curves were estimated with the Kaplan-Meier method. Cox proportional models were used for multiple regression analysis. Results Our cohort included 368 patients with a median age of 48 years (range 21–70). The median follow-up time was 3 years. The clinical tumor stage (T3–4) represented 58%, with 80% having positive axillary nodes. The luminal subgroup prevailed by 68%. The objective response rate (ORR) reached 78%, and 16% of patients achieved pCR. The clinical node stage and optimal chemotherapy were associated with higher ORR (p = 0.035 and p = 0.001, respectively). Predictors of pCR were clinical T-stage (p = 0.026), high Ki-67 index > 20 (p = 0.05), and receiving optimal chemotherapy (p = 0.014). The estimated 3-year disease free-survival (DFS) was 53%. Receptor status, achieving ORR, and pCR were associated with better DFS with hazard ratios of 0.56, p = 0.008; 0.38, p = 0.04; and 0.28, p = 0.007, respectively. Conclusions Luminal tumors still draw benefit from neoadjuvant chemotherapy in terms of clinical response and breast conservative surgery. Treatment escalation to those who did not achieve pCR requires more investigation, given a higher recurrence rate in real-world experience.


Background
Breast cancer is the commonest female malignancy, with 276,480 new cases and 42,170 deaths expected in the USA in 2020 [1]. In Egypt, breast cancer represented 32% of newly diagnosed female cancers [2,3] and ranked as the first cause of female cancer deaths in the World Health Organization's report 2014 [2]. A major proportion of our patients present with locally advanced breast cancer (LABC). Neoadjuvant chemotherapy (NCT) is increasingly adopted by our breast cancer multidisciplinary teams (MDTs). Besides its usefulness to downstage inoperable LABC and increased rates of conservative breast surgery in the operable cases, NCT proved to be

Open Access
Journal of the Egyptian National Cancer Institute an excellent platform for studying different prognostic factors for long-term outcomes, especially pathological complete response (pCR) [4]. In this cohort of Egyptian breast patients treated at a single academic institute, we looked for clinicopathological criteria of our LABC patients to analyze rates of pCR across different subtypes, predictors of these rates, and association with long-term disease outcomes.

Treatment regimens and surgical interventions
All medically fit patients received anthracycline/taxanes containing regimen, consisting of doxorubicin 60 mg/ m 2 and cyclophosphamide 600 mg/m 2 (AC) once every 3 weeks for four cycles followed by paclitaxel (T) 80 mg/ m 2 weekly for 12 cycles. Most of the patients with HER2positive disease received trastuzumab concurrently with taxanes (8 mg/kg loading dose and then 6 mg/kg every 3 weeks). Following NCT, all patients were discussed within MDT for surgical interventions, either mastectomy or breast conservative surgery.

Clinical response
The objective response rate (ORR) to NCT (defined as more than 50% decrease in the largest tumor dimension by clinical examination) in 122 evaluable patients was 78% (95/122). We limited this analysis to patients with serial clinical tumor size reported numerically in the visits during neoadjuvant treatment; 25.5% (31/122) of patients achieved a clinical complete response (CR), 52.5% (64/122) had a partial response (PR), 20.5% (25/122) with stable disease (SD), and two patients developed disease progression (DP). ORR was significantly higher in patients who completed an optimal course of NCT (≥ 6 cycles) compared to receiving sub-optimal course, i.e., < 6 cycles (62% vs. 15%, p = 0.001). The clinical nodal stage was significantly associated with ORR, specifically node-positive patients achieving higher ORR compared to node-negative patients in the simple regression analysis (48% vs. 10%, p = 0.035). No significant difference in ORR was observed according to HR status (41% vs. 21%, p = 0.111). There was a numerically higher ORR in younger patients, i.e., < 50 years (81.5% vs. 73.2%, p = 0.2).

Pathological response
Surgical specimens of 280 patients were evaluable for pathologic response assessment. The median pathologic tumor size was 3 cm (range 0-13). The median number Residual tumor tissue of grade 2 was reported in 36% of specimens with adverse pathological features, including lymphovascular invasion (LVI) reported in 32% and extracapsular extension reported in 13% of the surgical specimens. A total of 44 patients (15.7%) achieved pCR in both breast and axilla (ypT0-is, ypN0), and 55 patients (20%) achieved pCR only in the breast (ypT0). Postoperative pathological outcomes are summarized in Table 2.
In the current cohort, 20% of patients had complete adjuvant endocrinal therapy data, for which 50% were on luteinizing hormone-releasing hormone (LHRH) agonist in addition to oral endocrinal therapy (tamoxifen or aromatase inhibitors). Due to the insufficient long-term data, this factor was not included in the regression analysis of survival outcomes.

Discussion
This study frames an Egyptian experience with neoadjuvant systemic treatment for LABC represented by this cohort treated at a single-academic institute. It sheds light on our unique clinicopathological criteria of this advanced disease stage and explores their prognostic value. The identified predictive factors shall serve as informative tools for oncologists in a similar setting to tailor the management and follow-up plan.
We frequently encounter younger female patients with LABC. The median age of our cohort is 48 years, which is 14 years younger than the Western population median (62 years) [5]. This is a consistent observation with other national data [5,6]. For example, in a study comparing Egypt's Gharbia Cancer Registry (GCR) and the United States Surveillance, Epidemiology, and End Results (SEER) registries, Egyptian GCR cases were, on average, over 10 years younger than US SEER cases, with nearly 19% of GCR cases ≤ 40 years of age as compared to only 6% of US SEER cases [5].   High-risk biological profiles were evident in our cohort, with more than 50% having either HER2-positive or triple-negative breast cancer (TNBC) subtypes. Although some of these breast tumors were amenable for upfront BCS, our tumor board prefers to start neoadjuvant therapy for these higher-risk features. Concerning smaller breast sizes, BCS was sometimes very challenging in larger T2-3 cases. T4 patients represented 32% and were indicated for upfront mastectomy. All the above mandates special clinical practice implications like tendency to treatment escalation and increased need for fertility preservation counseling.
Our cohort comprised relatively more locally advanced cases compared to Western data [5]. Clinical stages IIIA and IIIB constituted 58% of our cohort, with 80% having node-positive disease compared to 37% in EBCTCG data [4] and 46% in Cortazar et al. [7]. Also, the incidence of luminal subtype was relatively higher than other reported cohorts (68%), with a relatively lower prevalence of TNBC (16.4%). This is in concordance with previous reports from Egypt [8,9], suggesting a profile of locally advanced breast cancer in Egyptian patients with younger age, more luminal disease, and more advanced clinical T and N stages. This, in part, explains the relatively higher rates of clinical response despite similar pCR rates compared to international studies [10][11][12].
The clear effect of optimal chemotherapy course on the ORR, pCR, and survival outcomes confirms the importance of patient counseling to comply with preplanned treatment courses. This is crucial in limitedresource settings when completing chemotherapy cycles could sometimes be overlooked if a remarkable response is achieved during NCT. We noticed that this practice was evident outside specialized breast cancer centers when patients were then referred to us for operative/postoperative management.
Achieving pCR in our cohort was significantly associated with survival outcomes. This comes in concordance with a recently published large meta-analysis by Laura et al., including more than 27,000 patients who were evaluated from 52 studies on NCT. Patients who   [13]. pCR remained associated with significantly improved event-free survival (EFS) even with the absence of adjuvant chemotherapy (HR 0.36, 95% PI 0.27-0.54, n = 18,462). Accumulating evidence has pointed to the poor outcomes observed in patients who failed to achieve pCR after NCT and the need for treatment intensification accordingly. For example, adjuvant capecitabine improved both DFS and OS when combined with standard adjuvant therapy for HER2-negative patients with residual invasive disease [14]. In addition, Trastuzumab emtansine (T-DM1) halved the risk of recurrence or death compared to adjuvant trastuzumab in HER2-positive patients who failed to attain pCR after standard NCT with anti-Her2 therapy [15].
However, while many oncologists would focus on achieving pCR as a landmark of the success of neoadjuvant therapy, more information could be discussed in this setting. For example, in a study by Symmans and colleagues on a pool of 5160 patients from the I-SPY platform, the subdivision of the pathological response using the MD Anderson residual cancer burden (RCB) into four levels of response could provide additional prognostic data to that provided by pCR rates only [16]. This might be much more valuable in our situation with more patients with the large initial disease and luminal subtype (less likely to reach pCR). Such information is crucial when evaluating patients for further adjuvant treatment in patients with residual disease after NCT.
In our low-resource setting, with 78% ORR, neoadjuvant therapy for our young locally advanced patient cohort could allow for significant tumor size reduction (enhancing more breast conservation). Additionally, NCT adds to the long-term survival outcomes by early addressing the expected micrometastatic disease and offers a platform to formulate highly effective and simple prognostication models [17,18]. This study is one of the few analyses that look into detailed outcomes of Egyptian LABC patients. However, our study is limited by its retrospective nature and depends on a single-center experience. Also, insufficient long-term adjuvant endocrine therapy data leads to its exclusion from the regression analysis. Limitations faced by our patients to complete the full course of therapy were sometimes attributed to delayed reimbursement by other sponsoring entities. Furthermore, larger multicenter studies are needed to validate our findings.