Introduction
The prevalence of breast cancer has continuously increased worldwide over the past few decades, with a particularly drastic increase in developing countries [
1,
2].Breast cancer has emerged as the most prevalent female malignant neoplasia, with morbidity predicted to continue to increase in the future [
3]. GLOBOCAN 2020 latest cancer burden data show breast Cancer has replaced lung cancer as the most common cancer in the world [
4]. All the number of new cases of global breast cancer reached 2.26 million, accounting for the fifth most common cause of cancer death overall. There were 416 000 new breast cancer cases and deaths in China 117,000 cases, ranking the first highest incidence of female cancer [
5]. Although advancements have been made in early detection and aggressive treatment in recent decades, the health quality and prognosis of breast cancer patients remain poor.
Currently, Neoadjuvant chemotherapy (NAC) is recognized as an indispensable treatment option for locally advanced breast cancer [
6,
7].The principal role of NAC is to downstage cancers, so that inoperable patients can undergo surgery, or patients who are not suitable for breast-conserving surgery (BCS) can obtain breast-conserving opportunities to satisfy their aesthetic needs. In addition, the response to NAC can be used to evaluate drug sensitivity and guide follow-up adjuvant therapy [
8‐
10]. Patients who achieve pathologic complete response (pCR) after NAC may have a long-term survival benefit, although differences have been observed among breast cancer subtypes [
11]. Many breast cancer patients who cannot reach pCR after NAC, and the prognosis of non-pCR patients greatly varies [
12,
13].Therefore, to maximize the effect of neoadjuvant therapy regimens in distinct breast cancer subtypes, a more precise pathological evaluation system is urgently required to guide clinicians to develop personalized treatment protocols and improve the prognosis of patients.
The residual cancer burden (RCB) index is a scoring system for assessing residual lesions after NAC based on breast tumors and regional lymph nodes proposed in 2007 [
14]. Subsequently, much clinical evidence from America and Europe has proven that the RCB system is effective, repeatable and useful for the pathological evaluation of different subtypes of breast cancer after treatment [
15‐
18]. The RCB system is considered as a long-term prognostic indicator for NAC treatment and has been demonstrated to be a better predictor of overall survival than most evaluation systems [
19]. RCB index and classification could help determine the most appropriate treatment plans for patients with all breast cancer subtypes. Residual cancer burden (RCB) continuous index and classification were independently and strongly prognostic for all breast cancer phenotypes. RCB index also was tightly associated with prognosis over long-term follow-up [
20].In recent years, the RCB system has been gradually recognized in Asia, and the 2021 version of the Chinese Society of Clinical Oncology (CSCO) guidelines added the RCB index as a post-NAC evaluation system. Our research mainly aimed to validate the contact between the RCB score and prognosis in the Chinese population by analyzing real-world data.
Methods
Patients and data collection
In this retrospective study, we included breast cancer patients who received NAC at Shandong Cancer Hospital and Institute and Liaocheng Peoples Hospital between 2016 and 2020. "Neoadjuvant chemotherapy" was used as the appropriate keyword to search for breast cancer patients via the medical record system. The patients who underwent NAC and were diagnosed with breast cancer for using the patient interface of the hospital electronic medical record system, utilizing the keyword "nejuvant chemotherapy".We excluded patients who did not receive surgery after NAC. Among these patients,223 (87.8%) underwent radical mastectomy, and 31 (12.2%) underwent BCS. All of the patients after BCS had completed radiotherapy. We also gathered clinicopathological data, including the onset age, menopausal status, clinical stage, pretreatment estrogen receptor (ER) and progesterone receptor (PR) levels and human epidermal growth factor receptor 2 (HER2) and Ki-67 statuses, type of operation, posttreatment T stage and N stage, presence of lymphatic vessel invasion (LVI), chemotherapy regimen, targeted therapeutic options, and Miller-Payne grade.
The enrolled patients were classified according to ER,PR and HER2 status as follows: ER-positive or PR-positive and HER2-negative was defined as HR + /HER2-; ER-positive or PR-positive and HER2-positive was defined as HR + /HER2 + ; ER- negative, PR-negative, and HER2-positive was defined as HR-/HER2 + ; ER-negative, PR-negative, and HER2-negative was defined as triple-negative breast cancer (TNBC). ER and PR were positively stained in at least 1% of nuclei. HER2 positivity was defined as an immunohistochemistry score of 3 + or 2 + with HER2 gene amplification by fluorescence in situ hybridization.
Miller-Payne grading system
The Miller-Payne grading system is routinely used by the two hospitals to assess the pathologic response after NAC. The criteria of classification were as follows.
Grade 1: No change or some alteration to individual malignant cells but no reduction in overall cellularity.
Grade 2: A minor loss of tumor cells(up to 30% loss),but high overall cellularity.
Grade 3: Estimated 30–90% reduction in tumor cells.
Grade 4: A marked disappearance of tumor cells (more than 90% loss of tumor cells) such that only small clusters or widely dispersed individual cells remain.
Grade 5: No malignant cells identifiable in sections from the site of the tumor; only vascular fibroelastotic stroma containing macrophages often remains. However, ductal carcinoma in situ (DCIS) may be present.
RCB score calculation
Neither of the two hospitals routinely evaluated the pathology by the RCB system. Thus, two pathologists from Shandong Cancer Hospital reevaluated the postoperative pathology of the 254 patients according to the requirements of the RCB evaluation system and input the data into the network calculator (
www.mdanderson.org/breastCancer_RCB) to calculate the RCB index. Then, according to the cutoff values of 1.36 and 3.28, the patients were further categorized into four different RCB classes: RCB 0 (equal to pCR), RCB I(minimal burden), RCB II (moderate burden) and RCB III(extensive burden).The specific classification methods were as follows: Pathological complete response (pCR), defined by the exclusion of any residual cancer.RCB score of 0 was defined as pCR, RCB score greater than 0, less than or equal to 1.36 was defined as RCB grade I, RCB score greater than 1.36 and less than or equal to 3.28 was defined as RCB grade II, and RCB score greater than 3.28 was defined as RCB grade III [
20].
Routine survival tracking
The patients were followed up for a long time via outpatient reexamination, telephone and e-mail. All cases received a standard postsurgical records, with scheduled clinical visits and imaging examinations every 3 months during the first year, every 6 months during the subsequent 2 years, and once yearly thereafter.
The primary follow-up endpoint was RFS, with the interval from the operation to the first occurrence of disease relapse and distant metastasis. The Kaplan–Meier survival curve is a commonly used statistical method to assess the probability of survival or occurrence of an event in patients within a specific time frame. In the case of breast cancer patients treated with NAC, the Kaplan–Meier survival curve can be used to evaluate the probability of relapse.
Statistical analysis
The data analysis was performed with SPSS V.25 and GraphPad Prism 8.0.2.The clinically significant pathological features were screened via a Cox regression model, and the log-rank test was performed. Logistic regression was used to identify the factors associated with pCR. GraphPad Prism 8.0.2 was used to draw the survival curves of RFS and pCR. The diagnostic efficiency was judged by the receiver operating characteristic (ROC) curve, including the area under curve (AUC), specificity and sensitivity. P ≤ 0.05 was defined as statistically significant.
Discussion
Breast cancer includes multiple molecular subtypes and is a highly heterogeneous solid tumor [
21]. NAC has been proven to increase the radical resection rate and breast preservation rate. The treatment efficacy varies from person to person, and the clinical response is a method of early evaluation [
22,
23].In this paper, we emphatically analyzed the viability of the RCB model and its influence on the prognosis after NAC. Our cohort was a collection of high-risk cases (Table
1: TNBC 18.9% and HER2-positive 20.9%), and the distribution of these subtypes was similar to that in the RCB validation groups performed by Symmans et al. [
14,
16]. Although the research has some shortcomings, we found that the survival prediction of RCB was similar to that in previous studies.
The Miller-Payne grading system is an accepted model that compares preoperative and postoperative tumor tissues, and is extensively used in neoadjuvant efficacy evaluation in domestic hospitals [
24]. According to the percentage of cell density reduction in primary tumor foci, the system categorizes NAC efficacy from class 1 to class 5 [
25]. Although it concisely and visually depicts the critical parameters associated with breast carcinomas and guides the selection of subsequent clinical treatment, it does not meticulously assess the postoperative pathology, particularly in patients with lymph node metastasis. And the MP system is not sufficiently comprehensive to measure the curative effect of tumor treatment due to the evaluation of only primary breast lesions. Moreover, following effective NAC for smaller tumors, the decrease in tumor cell density is more obvious than that in larger tumors, which indicates that the change in tumor cell density alone is not sufficiently comprehensive and objective to evaluate the therapeutic effect of tumor treatment [
26]. In contrast, the RCB system has more meticulous requirements for specimen collection and microscopic evaluation after NAC. The RCB score contains information on the tumor foci and positive lymph nodes. The long and short diameters of the tumor foci, number of positive lymph nodes, proportion of the primary tumor beds that contain infiltrating cells and maximum diameter of the axillary lymph node metastasis are used to calculate the score after the NAC [
27]. And this system has been gradually recognized in China over the years.
As an effective postoperative pathological response evaluation system, RCB has been validated in many countries and regions. A classic clinical test (protocol MDACC-LAB98–240), with the longest cohort follow-up time of 13 years, revealed no difference in survival between low RCB grades; in contrast, poor prognosis was mainly associated with the higher RCB class, which was assessed by Symmans et al. [
16]. This conclusion was validated in another study by Müller, H. D et al.,who enrolled 184 cases [
15]. In our retrospective study, we found that patients classified as RCB I (HR of RCB I vs. RCB 0 = 0.712,
p = 0.680) could have a good prognosis and a low risk of recurrence. As expected, patients with a higher RCB class had worse survival outcomes, as confirmed by the Cox multivariate analysis, where RCB II (HR of RCB II vs. RCB 0 = 3.270,
p = 0.018) and RCB III (HR of RCB III vs. RCB 0 = 5.108,
p = 0.002) were significant factors. The results are consistent with those obtained by others. Ki-67 represents cell proliferation and is a recognized risk factor in breast cancer patients [
28,
29]. Our study show that patients had a shorter RFS with high pathologic T stage after NAC, which suggest that lesions with high proliferative capacity may have worse outcomes. The results are also consistent with findings in other studies, Li-Yun Xie et al. revealed that a higher pre-neoadjuvant clinical T stage and N stage were independent predictors for an increased risk of tumor recurrence. Similar research results have also been found by Mariko Asaoka, supporting the conclusion of our study [
30,
31].
Young age is a known risk factor for long-term survival in patients who undergo BCS and are not treated with NAC [
32,
33]. This view was verified by a meta-analysis of large-scale prospective tests of BCS, which suggested that younger female patients had a higher 10-year locoregional recurrence rate(LRR) [
34]. Nevertheless, once the patients were treated with NAC, we could not able to assess the role of age in predicting survival outcomes. A large and authoritative EORTC 10994/BIG 1–00 study showed that younger age was not a risk factor for local recurrence(LR) [
35], and another study by Müller, H. D et al. from Europe did not separately analyze the age [
15]. Another study included 263 cases, with a cutoff value of 50 years, and mainly analyzed the impact of younger age on LR after NAC. The results revealed that patients < 50 years could have higher pCR rates, and young age could have a better outcome after NAC [
36]. Our study divided the cases by age into two sets, with 102 patients (40.2%) > 50 years, and we concluded that older age (> 50 years) would have a higher rate of relapse; however, we have no evidence to verify that younger age was highly predictive of pCR.
Our binary logistic regression analysis reveal that the phenotypic subtype was the unique associated factor in models that included age, stage, and chemotherapy regimens, and we found that patients with HER2-positive breast cancer, particularly TNBC, had higher pCR rates than HR-positive/HER2-negative patients. Similar conclusions have been observed in other studies [
37,
38]. Increased RFS with pCR occurred regardless of the clinicopathological characteristics, including HR-positive/HER2-negative patients [
39]. Finally, ROC curves were used to evaluate the prognostic efficiency of the RCB and Miller-Payne scoring systems for RFS, including calculation of the AUC, which demonstrated the favorable diagnostic efficiencies of the RCB and Miller-Payne scoring systems, with AUCs of 0.691 and 0.342, respectively. Taken together, these data suggest that the two systems are promising predictors for breast cancer patients treated with NAC.
According to the "NCCN Guidelines Version 2023 Breast Cancer," for TNBC patients who do not achieve pCR, oral capecitabine for 1 year may be considered as a treatment option. Multiple studies on stage III disease have shown that postoperative radiation therapy can improve local control, even for patients who have achieved pCR with NAC [
40]. Additionally, the use of preoperative systemic therapy can provide important prognostic information based on treatment response. Extra attention should be given to patients with RCB grade 3. Other subtypes do not require further adjuvant treatment, but we can enhance their follow-up process and reduce the time needed for reviews.
However, there are some limitations in this study. The follow-up period is relatively short, while the survival time of breast cancer patients is relatively long. Therefore, we were unable to obtain the overall survival time of patients to include it in our research.In the future, we will continue to expand the number of case samples and increase the follow-up time to obtain more convincing survival data. Second, our study may have introduced selection bias because the data came from only two hospitals. Last, due to the diversity of NAC regimens, 86.6% of cases had received the same kind of chemotherapy regimen.
Although the RCB system has more detailed requirements for evaluation, it can only be used to evaluate postoperative pathology. The cell density of thick-needle aspiration specimens before NAC and surgical specimens after NAC can’t be compared as in the Miller-Payne system and can’t reflect the contrast gap before and after NAC, so it also has some limitations. Kim JY,et al.combined RCB score with the Ki67 to form a "residual proliferative tumor load" (residual proliferative cancer burden, RPCB) system, and the RPCB score provided richer prognostic information and had a higher predictive efficiency [
41]. Recent studies have combined tumor infiltrating lymphocytes (TILs) with the RCB score as an original evaluation system, particularly in TNBC, which also shows good prospects [
42,
43].
Our team analyzed the differentially expressed genes of the resected tissue following NAC, and it is believed that promising biomarkers with prognostic value will be found soon. In addition, Shandong Cancer Hospital took the lead in performing internal breast lymph node biopsy in China [
44‐
48]. We also envision combining the internal breast lymph node information with the RCB system to develop a new, more comprehensive and accurate postoperative pathological evaluation system.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.