Prognosis of post-neoadjuvant therapy patients who underwent immediate breast reconstruction: a SEER-based, propensity-matched study

Objective The application of immediate breast reconstruction (IBR) for post-neoadjuvant therapy (NAT) patients was controversial. The aim of this study was to investigate the long-term survival outcomes of IBR for these patients. Methods Data between January 2010 and November 2017 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Propensity score matching (PSM) was performed to reduce the influence of confounding factors between the mastectomy alone group (MA) and the mastectomy with IBR group (IBR). The rates of 5 year breast cancer-specific survival (BCSS) were compared by Kaplan–Meier curves with log-rank test. Results The IBR was associated with improved 5-year BCSS in the IBR group before PSM (88.5 vs. 79.1%, P < 0.001). The proportion of IBR increased from 21.5% in 2010 to 28.2% in 2017. After PSM, a total of 9,610 patients were enrolled for survival analysis (4,805 in each group). In the complete response (CR) group, the 5-year BCSS rates did not differ (93.4 vs. 95.6%, P = 0.16). In the non-CR group, the 5-year BCSS rate was higher in patients who received IBR (82.5% 79.4%, P = 0.034). Conclusion In general, the application of IBR among post-NAT patients has steadily increased from 2010 to 2017. In the CR group, survival outcomes of post-NAT patients who received IBR were similar to those who received mastectomy alone. In the non-CR group, IBR was associated with potential survival benefits. More studies are expected to validate our findings.


Introduction
Neoadjuvant therapy (NAT) plays a vital role in the treatment of breast cancer, which includes neoadjuvant chemotherapy (NACT), neoadjuvant radiotherapy (NART), endocrine therapy and targeted therapy.NACT was originally recommended to treat locally advanced breast cancer, because of the effectiveness of downstaging tumors to facilitate surgery, as well as facilitating breast conservation in such cases [1,2].To date, the indications of NACT have gradually expanded to triple-negative, human epidermal growth factor receptor 2 positive (HER2 +), or early-stage breast cancer [3,4].Meanwhile, during the treatment, NART was typically used in conjunction with NACT to treat initial inoperable breast cancer [5].Furthermore, drug sensitivity of breast cancer patients could be studied during the treatment to guide subsequent therapy to improve outcomes.
Mastectomy followed by immediate breast reconstruction (IBR), involving autologous flap reconstruction and implantbased reconstruction, has gained popularity among patients.In comparison to delayed breast reconstruction, IBR offers better cosmetic outcomes, more positive psychological impact, and only one operation with less financial burden [6][7][8].However, the issue of IBR for post-NAT patients was controversial, with different conclusions being reached by different studies.Gouy et al. confirmed that IBR after NACT did not significantly affect the local relapse-free or distant disease-free survival [9].Wu et al. also demonstrated the comparable long-term oncologic outcomes of IBR with nipple-sparing or skin-sparing mastectomy to conventional mastectomy alone [10].However, according to a previous study which was based on data from the Korean Breast Cancer Society, IBR following mastectomy was associated with worse prognoses than mastectomy alone in non-pCR patients with advanced clinical stages of cancer [11].Meanwhile, a number of studies were concerned that NACT may increase local recurrence risk after breast-conserving surgery (BCS) [12,13].In addition, discrepancies in complication rates across studies hindered decision-making about whether to receive IBR in the setting of NAT [14][15][16].
In view of the personal preference of patients to breast reconstruction, it is difficult to conduct a randomized controlled trial to investigate IBR's outcomes.Meanwhile, prior studies, mainly retrospective single-center studies, provided insufficient evidence.Consequently, we used the SEER database to perform this large population-based research that investigate the long-term prognosis of post-NAT patients who underwent IBR.

Variables and definitions
An unmarried status was defined as divorced, separated, single, domestic partner, or widowed at the time of diagnosis.

Propensity score matching
Propensity score matching (PSM) was carried out in a 1:1 ratio in the comparison between the MA group and the IBR group using the R package "MatchIt", with the parameters of the "nearest" method and a caliper of 0.02.This was done to lessen the impact of confounding factors on prognosis.Age, year of diagnosis, race, marital status at diagnosis, grade, histology, T stage, N stage, CPM, RNT, radiation, systemic therapy, estrogen receptor (ER) status, progesterone receptor (PR) status, HER2 status, and median household income were the baseline characteristics used for matching.

Statistical analysis
All analyses were performed using R software (version 4.2.1;http:// www.r-proje ct.org/).To compare normally distributed data that were expressed as mean ± standard deviation (x ± s), Student's t-test was utilized.Non-normally distributed data that were reported as median (interquartile range, IQR) were analyzed using the Mann-Whitney U test.Categorical data was presented as n (%) and was then examined using the Chi-square test or the Fisher's exact test.Kaplan-Meier method with log-rank test was used to assess 5 year BCSS.The statistical significance levels were both two-sided and set at P < 0.05.

Results
We obtained information on 31,054 female patients aged 20 to 80 who had M0 stage breast cancer and had undergone NAT between 2010 and 2017.In Fig. 1, the screening flow is displayed.A total of 24,537 eligible patients were eventually enrolled, of whom 9,017 underwent breast-conserving surgery, 9,064 had a mastectomy alone and 6,456 underwent mastectomy with IBR.The MA group and the IBR group were further compared.

Demographic features
Trends in the application of the three treatments are shown in Fig. 2. Generally, the proportion of post-NAT patients who underwent either breast-conserving surgery (28.8 to 41.5%) or mastectomy with IBR (21.5 to 28.2%) was increasing from 2010 to 2017, and the proportion of mastectomy alone was declining (49.7 to 30.3%).Table 1 reports the comparison of the MA group and the IBR group.Mastectomy with IBR was associated with younger, white or married patients (P < 0.001, respectively).In addition, patients in both groups mostly earned between $50,000 and $70,000, while the patients with income higher than $70,000 were more in the IBR group (P < 0.001).

Clinical and pathological characteristics
With the exception of laterality, every clinical and pathological characteristic between groups was statistically significant (Table 1 T3-4 and N1-3 stage breast cancer.In the IBR group, CR was more prevalent (32.0 vs. 25.2%),whereas in the MA group, NR was more prevalent (10.4 vs. 7.2%, P < 0.001).
The most prevalent breast subtype was HR + /HER2-(40.8% in the MA group and 37.8% in the IBR group), whereas HR-/HER2 + was the least common (12.5% in the MA group and 11.1% in the IBR group).

Therapeutic characteristics
In the IBR group compared to the MA group, CPM was more prevalent (55.2% vs. 31.6%,P < 0.001).In the IBR group, implant-based reconstruction (34.2%) was the most common.In the both groups, systemic therapy was given to the vast majority of the patients.However, radiation was given to more patients in the MA group (60.7 vs. 49.7%,P < 0.001).
According to subgroup analysis, the 5 year BCSS rate was higher in the IBR group of the non-CR group (82.5 vs. 79.4%,P = 0.034) (Fig. 5), but did not differ in the CR or the unknown-response groups.There were no statistically significant between-group differences across the various stages after further stratification by the AJCC stage.

Discussion
To our knowledge, the current work was the first SEER database-based research on the application of IBR for post-NAT patients.Our study showed a steady increase in IBR for post-NAT patients over the past few years.Meanwhile, IBR was proper and worthy of consideration for post-NAT patients, which might improve the 5-year BCSS.Several reasons gave rise to the increase of IBR in the setting of NAT.On the one hand, the use of NAC has gradually increased since 2010, of which the aim gradually expanded from breast conservation to personalized treatment [18,19].The population base of our study kept growing for this reason.On the other hand, the Women's Health and Cancer Rights Act, which was passed in the US in 1998, mandated comprehensive coverage for breast reconstruction following mastectomy.An earlier study showed that the percentage of women getting breast reconstruction increased steadily over time [20].Furthermore, several studies investigated the application of breast reconstruction in patients who were traditionally considered at high risk, and had achieved acceptable outcomes, which further improved the increase of IBR [21][22][23].Before-PSM comparison showed IBR following NAT was associated with younger age, married status, earlier tumor stage and higher income, which were consistent with previous researches [11,24,25].Analysis of prognosis showed the IBR was associated with improved BCSS.However, this beneficial association should be cautious, although several small-sample, single-center studies had demonstrated the feasibility and safety of IBR following NAT [9,10,26].IBR lowered breast cancer-specific mortality in a prior sizable population-based trial, according to Bezuhly M et al.However, they also observed that these results were due to imbalances in socioeconomic factors and access to care rather than insufficient adjustment for tumor features and disease severity [27].When the confounding effect of family income was taken into account, another study showed that immediate postmastectomy reconstruction had little benefits for survival [28].In contrast, the previous Korean nationwide study confirmed that IBR following mastectomy was associated with worse prognoses than mastectomy alone among patients with advanced clinical stage tumor of the non-pCR group [11].
Although PSM was performed in this present work, some confounding factors were still unbalanced due to their unavailability in the SEER database such as baseline health status.Patients at a poor health status such as obesity, cardiovascular disease, and smoking history might not consider the IBR, but were associated with worse prognosis.We believe this potential imbalance caused the improved 5-year BCSS in the non-CR group.Additionally, the impact of IBR on personal life quality and psychology should also be considered.A previous metaanalysis demonstrated a better health-related quality of Another study revealed that reconstruction following a mastectomy significantly affects body image and sexual function [30].Furthermore, a recent study demonstrated that poor HRQOL might increase the mortality risk [31].Some patients with inoperable breast cancer in our study received NAT for breast conservation or reconstruction purpose, but might failed because of the severity of the disease (non-CR), which caused anxiety and depression.This potential psychological fluctuation during the treatment was a probable adverse factor, which led to a worse HRQOL with worse prognosis.However, in the absence of relevant data in the SEER database, we were unable to investigate the impact of personal life quality and psychology.Further researches to estimate HRQOL of the patients who received breast reconstruction after NAT are required.
There are several limitations in the use of the SEER database.Firstly, selection bias was inevitable in this retrospective study.Secondly, due to the lack of data on local recurrence, adequate evaluation of the oncology outcome is limited.Meanwhile, we are unable to access the perioperative complications among different treatments because of the unavailable data.Finally, this work was only based on population in the United States, and more realword studies in other countries or regions are expected.Nonetheless, our retrospective work is able to provide some valuable reference in the absence of randomized controlled trials.

Conclusion
The application of IBR in post-NAT patients has steadily increased.In the CR group, survival outcomes of post-NAT patients who received IBR were similar to those who received mastectomy alone.In the non-CR group, IBR was associated with potential survival benefits.More studies are expected to validate our findings.

Fig. 2
Fig. 2 Trends in applications of breast-conserving, mastectomy alone and mastectomy with immediate breast reconstruction from 2010 to 2017.IBR, immediate breast reconstruction

Fig. 3
Fig. 3 Breast cancer-specific survival rates of the MA versus IBR groups before propensity score matching.BCSS breast cancer-specific survival, MA mastectomy alone, IBR immediate breast reconstruction

Fig. 4
Fig. 4 Breast cancer-specific survival rates of the MA versus IBR groups after propensity score matching.A All the patients; B The CR group; C The non-CR group; D The unknown-response group.BCSS

Fig. 5
Fig. 5 Breast cancer-specific survival rates of the MA versus IBR groups of the non-CR subgroup after propensity score matching.A Stage-I; B Stage-II; C Stage-III; D Unknown-stage.BCSS breast can-

Table 1
Demographic and baseline characteristics of the mastectomy alone group and the immediate breast reconstruction group