Nomogram prediction model for pathological complete response of metastatic axillary lymph nodes of breast cancer after neoadjuvant chemotherapy Authors: Yinting Chen, Ge Song, Xiaohong Xu, Yuping Yang Published: 2026 ABSTRACT AbstractObjectiveTo evaluate the predictive value of a nomogram model combining ultrasound features and clinical indicators for pathological complete response (pCR) of metastatic axillary lymph nodes in breast cancer after neoadjuvant chemotherapy (NACT).MethodsA total of 109 women with histologically confirmed breast cancer and axillary lymph node metastasis were retrospectively analyzed. Patients were divided into pCR and non-pCR groups after NACT. A nomogram model was established using univariate and multivariate logistic regression analyses. The predictive performance of the clinical model, ultrasound model, and combined model was compared using the area under the curve (AUC).ResultsMolecular subtype, clinical stage, pre-NACT metastatic axillary lymph node subtype, post-NACT reduction rate of lymph node short diameter, and post-NACT reduction rate of primary tumor maximum diameter were independent predictors of axillary lymph node pCR. The combined model (AUC 0.844, 95%CI: 0.773–0.915, P < 0.05) showed significantly improved predictive ability compared with the clinical model (AUC 0.725, 95%CI: 0.628–0.823, P < 0.05) and ultrasound model (AUC 0.754, 95%CI: 0.665–0.844, P < 0.05).ConclusionThe nomogram combining ultrasound and clinical indicators effectively predicts pCR of metastatic axillary lymph nodes after NACT and provides a reference for surgical selection. However, the single‑center design, small sample size, and lack of external and prospective validation limit its generalizability. Multi‑center prospective validation is warranted.Keywords: Ultrasound, Neoadjuvant chemotherapy, Breast cancer, Axillary lymph nodes metastasis, Pathological complete response, Nomogram, Prediction model Introduction Axillary lymph node status is a key determinant of clinical staging and treatment planning in breast cancer [1]. In recent years, up to 74% of patients with positive axillary lymph nodes have achieved pathological complete response (pCR) after neoadjuvant chemotherapy (NACT) [2]. Axillary lymph node dissection (ALND) is the standard treatment for breast cancer patients with axillary lymph node metastasis, but it often leads to postoperative complications such as lymphedema and upper limb weakness, which reduce long-term quality of life. For patients whose axillary lymph nodes convert from positive to negative after NACT, sentinel lymph node biopsy (SLNB) has been increasingly used instead of ALND in multiple clinical trials. SLNB can avoid complications caused by ALND while achieving comparable efficacy, thus improving patients’ quality of life [3]. Both the European Society for Medical Oncology and the National Comprehensive Cancer Network recommend SLNB to replace ALND with suitable patients. However, multi-institutional clinical trials have shown that the false negative rate (FNR) of SLNB after NACT remains 13%–14% [4]. Marking positive lymph nodes before NACT can reduce FNR, but this method is costly and limited by marking materials, making it difficult to popularize [5]. Given the controversies in treatment selection, developing a prognostic prediction model to guide treatment decisions has important clinical value and application prospects. Some studies have established models to predict axillary lymph node pCR using clinical features [6], but none of these studies have incorporated ultrasound imaging features. Ultrasound is the preferred method for axillary imaging [7]. After NACT, the size, morphology, internal structure, and blood flow of axillary lymph nodes change significantly, and tumor regression patterns are diverse, leading to low accuracy of classic ultrasound indicators in diagnosing axillary pCR after NACT [8]. Therefore, this study constructed a nomogram prediction model based on clinical features and ultrasound imaging features to evaluate axillary lymph node pCR after NACT, aiming to provide a reference for the preferred use of SLNB after NACT. Materials and methods A retrospective analysis was performed on 109 female breast cancer patients who received NACT at the Affiliated Hospital of Guangdong Medical University from January 2019 to June 2022. All patients underwent core needle biopsy of breast masses and axillary lymph nodes before NACT, with pathological confirmation of invasive breast cancer and axillary lymph node metastasis. All patients underwent ALND after NACT. Patients were excluded if they had a history of breast or axillary surgery, previous chemotherapy, interrupted NACT, or distant organ metastasis (Fig. 1). The study was approved by the Ethics Committee of the Affiliated Hospital of Guangdong Medical University (PJ2021-024) and conducted in accordance with the Declaration of Helsinki (2013 revision). Written informed consent was obtained from all participants. Flowchart of the study population. NACT: neoadjuvant chemotherapy treatment; ALND: Axillary lymph node dissection Ultrasound examination Breast and axillary ultrasound examinations were performed by ultrasound physicians with at least 3 years of clinical experience. Color Doppler ultrasound instruments including Siemens ACUSON X700, Toshiba Aplio400, Toshiba Aplio500, and Philips IU22 were used with high‑frequency linear array probes (5–15 MHz). Patients were placed in the supine or lateral position with hands behind the head to fully expose the breasts and axillae. Bilateral breasts and axillary lymph nodes were scanned in multiple planes. The maximum diameter of the primary tumor, as well as the long diameter, short diameter, subtype, calcification, and blood flow grade of metastatic axillary lymph nodes, were recorded before and after NACT. The reduction rates of the primary tumor maximum diameter and lymph node diameters were calculated. Image analysis Two ultrasound physicians with 3 and 5 years of experience respectively evaluated images in a double‑blind manner. The largest metastatic lymph node confirmed by biopsy was used as the observation target. Referring to previous literature [9], objective quantitative cutoffs cortical thickness were applied: cortical thickness was measured at the thickest part, and a cut‑off value of 0.3 cm was used to define abnormal thickening, which has been widely validated for evaluating metastatic axillary lymph nodes in breast cancer. Metastatic lymph node morphology was classified as follows: Type I: regular morphology, cortical thickness < 0.3 cm, intact central hilar structure. Type II: cortical thickness ≥ 0.3 cm, with eccentric or preserved central hilar structure. Type III: cortical thickness > 0.3 cm, with complete disappearance of hilar structure (Fig. 2). Morphological classification of metastatic lymph nodes. a Type I lymph nodes; b Type II lymph nodes (local cortical thickening); c Type II lymph nodes (uniform cortical thickening); d Type III lymph nodes Blood flow was graded according to Adler criteria: Grade 0, no blood flow signal; Grade I, a small amount of blood flow; Grade II, a moderate amount of blood flow; Grade III, a large amount of blood flow (Fig. 3). To ensure consistent observation before and after NACT, the same lymph nodes were identified according to their depth from the skin, surrounding adipose and fascial distribution, as well as the course and adjacent relationship of neighboring blood vessels (Fig. 4). Discrepant judgments were resolved by a senior ultrasound physician with 15 years of experience to reach a consensus. Adler grade of blood flow in metastatic lymph nodes. a Grade 0 blood flow; b Grade I blood flow; c Grade II blood flow; d Grade III blood flow A typical case of pCR of axillary lymph nodes after neoadjuvant chemotherapy. A patient with invasive breast cancer in the right breast, clinical stage IIB (T2N1M0), HER 2+; a Ultrasound image of metastatic axillary lymph nodes before NACT treatment; b Ultrasound image of metastatic axillary lymph nodes after 2 cycles of NACT treatment; c Ultrasound image of metastatic axillary lymph nodes after 4 cycles of NACT treatment; d Ultrasound image of metastatic axillary lymph nodes after 6 cycles of NACT treatment Pathological evaluation Resected lymph nodes after ALND were classified according to the AJCC staging system. Benign lymph nodes or isolated tumor cells (tumor diameter ≤ 0.2 mm and < 200 tumor cells on a single section) were defined as pCR. Micro‑metastasis (maximum diameter > 0.2 mm but ≤ 2.0 mm) and macro‑metastasis (maximum diameter > 2.0 mm) were defined as non‑pCR. Statistical processing SPSS 25.0 and R (4.2.0) software were used for statistical analysis. Measurement data conforming to normal distribution were expressed as mean ± standard deviation and compared using independent samples t-test. Enumeration data were described as case numbers and percentages and compared using χ² test or Fisher exact test. Multicollinearity among candidate predictive variables was assessed using the variance inflation factor (VIF). All variables included in the model had VIF < 2.0, indicating no significant multicollinearity. Multivariate logistic regression analysis was used to identify independent predictors of axillary lymph node pCR after NACT. A nomogram prediction model was established, and receiver operating characteristic (ROC) curves were plotted to calculate AUC. The Bootstrap method was used to test model calibration and plot calibration curves. Decision curve analysis (DCA) was performed to evaluate the clinical net benefit of the model. Results Factors associated with axillary lymph node pCR A total of 109 female patients aged 27–69 years were enrolled in this study. After NACT, 44 patients achieved axillary lymph node pCR (44/109, 40.36%), and 40 patients achieved primary tumor pCR (40/109, 36.69%). Among them, 33 patients achieved pCR in both the primary tumor and axillary lymph nodes (33/40, 82.5%), while only 11 patients achieved axillary pCR without primary tumor pCR (11/69, 15.9%). There were significant differences in molecular subtypes and clinical stages between the pCR group and non-pCR group (Table 1). Table 1Clinical characteristics of breast cancer patients with pCR and non-pCR in axillary lymph nodespCR(n = 44)non-pCR(n = 65)t/χ2 P Age49.08 ± 10.5549.08 ± 9.63t = 1.1320.337BMI0.5200.471≤ 2538(86.4)59(90.8)>256(13.6)6(9.2)Menopause0.8730.432no27(61.4)34(52.3)yes17(38.6)31(47.7)Obstetric history0.8780.564no2(4.5)1(1.5)yes42(95.5)64(98.5)Histological typing2.9820.161Invasive ductal carcinomas42(95.5)65(100.0)Others2(4.5)0Molecular subtypes8.6150.028Luminal A1(2.3)3(4.6)Luminal B21(47.7)42(64.6)HER 2+16(36.4)8(12.3)TN6(13.6)12(18.5)T stage7.0510.063T11(2.3)5(7.7)T228(63.7)26(40.0)T313(29.5)23(35.4)T42(4.5)11(16.9)N stage5.8620.054N127(61.4)26(40.0)N211(25.0)19(29.2)N36(13.6)20(30.8)Clinical staging of primary tumor7.8780.005II19(43.2)12(18.5)III25(56.8)53(81.5) Clinical characteristics of breast cancer patients with pCR and non-pCR in axillary lymph nodes For ultrasound features, significant differences were observed in pre-NACT metastatic axillary lymph node subtypes and blood flow grades between the two groups (P < 0.05). After NACT, significant differences were found in lymph node subtypes, short-diameter reduction rate of metastatic lymph nodes, and maximum diameter reduction rate of the primary tumor (Tables 2 and 3). Table 2Ultrasonographic imaging features pre-NACT in breast cancer patients with pCR and non-pCR of axillary lymph nodespCR(n = 44)Non-pCR(n = 65)t/χ2 P Long diameter of LN24.77 ± 12.1526.46 ± 13.19-0.6670.842Short diameter of LN12.68 ± 6.6714.31 ± 8.87-1.0030.650LN subtypes8.8110.008I2(4.5)1(1.5)II30(68.2)28(43.1)III12(27.3)36(55.4)calcification2.0690.271no44(100.0)62(95.4)yes03(4.6)Blood flow grade10.0900.0060/I10(22.7)28(43.1)II22(50.0)14(21.5)III12(27.3)23(35.4) Ultrasonographic imaging features pre-NACT in breast cancer patients with pCR and non-pCR of axillary lymph nodes Table 3Ultrasonographic imaging features post-NACT in breast cancer patients with pCR and non-pCR in axillary lymph nodespCR(n = 44)Non-pCR(n = 65)t/χ2 P Long diameter of LN13.43 ± 5.27616.00 ± 8.5181.7810.075Short diameter of LN6.20 ± 2.7168.40 ± 5.9892.2760.063LN subtypes14.8920.001I17 (38.6)8 (12.3)II18 (40.9)23 (35.4)III9(20.5)34 (52.3)calcification2.0690.271no44(100.0)62(95.4)yes0(0.0)3(4.6)Blood flow grade0.9820.6120/I32(72.7)42(64.6)II8(18.2)17(26.2)III4(9.1)6(9.2)Reduction rate of the maximum diameter of the primary tumor8.3420.004≥ 50%30(68.2)26(40)<50%14(31.8)39(60.0)Reduction rate of the long diameter of lymph nodes1.4930.236≥ 50%20(45.5)22(33.8)<50%24(54.5)43(66.2)Reduction rate of the short diameter of lymph nodes8.3420.004≥ 50%30(68.2)26(40.0)<50%14(31.8)39(60.0) Ultrasonographic imaging features post-NACT in breast cancer patients with pCR and non-pCR in axillary lymph nodes Logistic regression analysis Multivariate logistic regression analysis demonstrated that molecular subtypes, clinical stage, pre-NACT metastatic axillary lymph node subtypes, post-NACT reduction rate of primary tumor maximum diameter, and post-NACT reduction rate of lymph node short diameter were independent predictors of axillary lymph node pCR after NACT (Table 4). The reduction rate of the lymph node short diameter as a continuous variable was an independent predictor of axillary pCR. For clinical interpretability, a cutoff of 50% was used for illustration. Table 4Multivariate regression analysis of breast cancer patients with pCR and non - pCR in axillary lymph nodesβSEWaldOR(95%CI) P Molecular subtypesA--8.2051.0000.042B-1.5611.7220.8220.210(0.007-6.139)0.365HER 2+-1.2980.8292.4500.273(0.054-1.387)0.117TN0.6320.8720.5261.882(0.341-10.394)0.468Clinical stagesII/III2.1880.7098.9178.318(2.071-33.408)0.003LN subtypes pre-NACTI--8.5811.0000.014II2.7821.6172.96216.156(0.679-38.125)0.085III1.7790.6288.0295.926(1.731-20.287)0.005Blood flow grade0/I--4.5161.0000.105II-0.5010.6790.5450.606(0.160-2.291)0.460III0.8550.7071.4603.311(0.588-9.400)0.227LN subtypes post-NACTI--3.6061.0000.165II1.1010.7402.2143.008(0.705-12.829)0.137III1.1120.6303.1113.041(0.884-10.463)0.078Reduction rate of the maximum diameter of the primary tumor<50% OR ≥50%-1.2730.6354.0130.280(0.081-0.973)0.045Reduction rate of the short diameter of lymph nodes<50% OR ≥50%-1.2450.6343.8630.288(0.083-0.996)0.049 Multivariate regression analysis of breast cancer patients with pCR and non - pCR in axillary lymph nodes Establishment of the nomogram prediction model A nomogram for predicting axillary lymph node pCR after NACT was established based on the above five independent predictors (Fig. 5). Each variable was assigned a corresponding score, and the total score corresponded to the probability of axillary pCR. Nomogram for predicting the probability of achieving axillary lymph node pCR post-NACT. MT: Molecular subtypes; LN: Lymph nodes subtypes pre-NACT; CS: Clinical stages; LNR: Reduction rate of the short diameter of lymph nodes; PTR: Reduction rate of the maximum diameter of the primary tumor ROC curve analysis showed that the AUC of the combined model was 0.844 (95% CI: 0.773–0.915), which was significantly higher than that of the clinical model (AUC = 0.725, 95% CI: 0.628–0.823) and ultrasound model (AUC = 0.754, 95% CI: 0.665–0.844) (Fig. 6). Although the 95% confidence intervals overlapped among the three models, the combined model showed a trend toward improved predictive performance, supporting the incremental value of integrating ultrasound and clinical features. The calibration curve revealed excellent consistency between the predicted probability and actual observation of pCR (Fig. 7). Decision curve analysis (DCA) demonstrated that the nomogram provided positive net benefit within a threshold probability range of approximately 15% to 90%, indicating favorable clinical utility for predicting axillary pCR in this range (Fig. 8). ROC curve of the predictive performance of the prediction model Calibration curve of the predictive nomogram model. The calibration curve shows the observed and predicted axillary lymph node pCR rates in breast cancer patients post-NACT DCA curve of the predictive nomogram model. In the DCA curve, the horizontal coordinate represents the threshold probability, and the value on the vertical coordinate represents the net benefit. The blue curve shows the predicted benefits of the model. Take the intersection of the blue curve and the gray line as the starting point and the intersection with the black line as the end point. Patients corresponding to this range can benefit Discussion The primary goal of NACT is to downstage tumors for surgical treatment, maximize breast preservation, and improve long-term prognosis. Individualized treatment strategies can also be formulated according to tumor response [10]. Previous studies reported that the axillary pCR rate after NACT was approximately 10% higher than that of the primary breast tumor [11], which was consistent with the results of the present study. The high axillary pCR rate indicates that ALND is not necessary for all node-positive breast cancer patients, thereby reducing over-treatment and surgery-related complications. In this study, 82.5% of patients with primary tumor pCR also achieved axillary pCR, whereas only 15.9% of patients without primary tumor pCR achieved axillary pCR. The axillary pCR rate in HER2 + patients reached 66.7%, which was higher than other subtypes, suggesting that HER2 + status is a favorable predictor of axillary pCR, in line with previous reports [12]. This may be attributed to the high sensitivity of HER2-positive lesions to trastuzumab-based chemotherapy [13]. Clinical stage was also an independent predictor. The pCR rate was 61.3% of stage II patients and 32.1% in stage III patients. Small-volume tumors are more chemosensitive, leading to more prominent volume reduction under the same chemotherapy dose, which is consistent with previous findings [14–16]. Previous studies have investigated the predictive value of ultrasound in evaluating metastatic lymph nodes. Reported sensitivity of preoperative axillary ultrasound ranges from 37% to 100%, with specificity ranging from 69% to 92%. Ultrasound exhibits higher sensitivity than MRI and PET, as well as higher accuracy than clinical evaluation alone [17, 18]. Cortical thickening and hilar disappearance are recognized indicators of metastatic lymph nodes [19, 20]. In this study, lymph nodes were classified into three types based on cortical thickness (0.3 cm) and hilar structure. The pCR rates of Type I, II, and III lymph nodes before NACT were 66.7% (2/3), 51.7% (30/58), and 25.0% (12/48), respectively. Type III lymph nodes present severe structural destruction, which may block drug penetration and lead to poor chemotherapy response. Multivariate regression analysis showed that there was a statistically significant difference between the groups with a reduction of < 50% and ≥ 50% in the maximum diameter of the primary tumor post-NACT in the axillary lymph node pCR group. Previous studies have proven that the change in the size of the primary tumor after NACT is correlated with lymph node metastasis, indicating that the lymph node status is an indicator of tumor spread ability. However, there are not many studies on the response of lymph nodes themselves before and post-NACT [20, 21]. In this study, by recording the long and short diameters of lymph nodes before and post-NACT, it was found that a reduction of ≥ 50% in the short diameter between the pCR group and the non-pCR group was an independent influencing factor for predicting axillary lymph node pCR which is also confirmed by Li et al. [22]. As for imaging, the measurement of the short diameter of lymph nodes has higher repeatability than that of the long axis [23]. According to the RECIST 1.1 criteria, it is divided into complete response (lesion disappearance), partial response (lesion diameter reduction of more than 30%), stable disease (between partial response and disease progression), and disease progression (lesion diameter increase of at least 20%). Clinically, when the diameter of tumor is reduced by 30%, NACT is considered effective for the tumor. Referring to previous studies on the efficacy of NACT in breast cancer [24, 25], chemotherapy-sensitive breast cancer patients could get 30% reduction in the diameter of the primary tumor after two cycles of NACT. In this study, however, the enrolled patients had completed NACT (4 to 6 cycles or more). To achieve pCR, the criterion for post-NACT effectiveness should be based on an additional 30% reduction beyond the 30% reduction. That is, the threshold established in this study requires that the diameter of the lymph nodes should be reduced by (1–0.7 × 0.7 = 0.51) post- NACT. Although MRI and PET/CT were not used in this study due to low clinical utilization, their roles in axillary pCR prediction deserve mention. Multiparametric MRI is increasingly used for treatment response assessment and provides reliable information for predicting axillary metastasis after NACT. PET/CT shows high specificity but limited sensitivity in evaluating axillary status. Compared with these modalities, ultrasound is more widely available, cost-effective, and radiation-free, making it the first-line tool for repeated axillary evaluation during NACT. Quantitative ultrasound features derived from routine examinations also enable stable and feasible prediction of axillary pCR. In this study, the ultrasound features were refined and added to the clinical model, and a nomogram was drawn. Based on the developed nomogram, breast cancer patients who can achieve axillary lymph node pCR post-NACT can be better risk stratified. The predictive model of axillary pCR post-NACT for breast cancer with positive axillary lymph nodes combined with ultrasound features has good discrimination ability, which improves the predictive ability of the model using only clinical features. The predictive nomogram model will provide information for the downgrading of axillary treatment in patients post-NACT and can help clinicians more appropriately assess the risks and benefits of SLNB post-NACT. The combined model showed a numerically higher AUC than the clinical and ultrasound-alone models, although overlapping CIs suggest that larger - sample validation is still needed. In our study, the DCA curve of the model shows that the net benefit predicted by the model is consistently higher than that of the full-treatment strategy. After rigorous statistical analysis, we found that when the threshold probability is in the range of 0.15–0.9, the net benefit predicted by the model is always higher than that of the full- treatment strategy and the non-treatment strategy. When the probability of a patient having lymph node pCR is lower than 0.15, the net benefit predicted by the model may be similar to that of the non-treatment strategy. In this case, ALND should be considered. When the probability of a patient having lymph node pCR in the range of 0.15–0.9, the doctors can first perform SLNB. If the SLNB result is positive, ALND can then be considered; if the SLNB result is negative, ALND can be avoided, thereby reducing surgical trauma and complications for the patient. When the probability of a patient having lymph node pCR is higher than 0.9, SLNB can be directly accepted. This study has some limitations. First, this is a single-center retrospective study with a relatively small sample size of only 109 patients. The single-center design may introduce selection bias and restrict the generalizability of the nomogram to other clinical settings with different patient demographics, ultrasound devices, and operator experience. The small sample size may also reduce the statistical power and stability of the logistic regression and model calibration, potentially affecting the precision of the predicted probabilities. Second, the nomogram established in this study lacks external validation, which is a major limitation for evaluating its reliability and clinical applicability in independent populations. Future studies are necessary to validate it in a larger multi - institutional cohort. In addition, in this study, the proportion of patients undergoing other imaging examinations before and post-NACT was relatively low. Therefore, only the relationship between ultrasound features and axillary lymph node pCR post-NACT was analyzed. Multiple imaging methods should be combined as much as possible to establish a multimodal predictive model to improve the accuracy of the model. Finally, biopsy-confirmed lymph nodes were not clipped or marked for long-term identification, which may have overestimated the predictive accuracy of model. Although we attempted to re-evaluate the same lymph nodes based on anatomical landmarks and adjacent structures, the lack of standardized lymph node marking represents a major limitation and may introduce bias in assessing lymph node response. To address these limitations, we propose a detailed multi-center prospective validation plan: (1) Multi-center design: Enroll consecutive patients from at least 3 independent medical centers with different geographic and clinical settings. (2) Sample size: Target a total sample size ≥ 300 to ensure statistical stability and external validity. (3)Prospective design: Use standardized ultrasound protocols, central image review, and predefined pathological endpoints to reduce bias. (4) Outcome measures: Primary outcome is axillary pCR; secondary outcomes include AUC, calibration, sensitivity, specificity, and FNR of SLNB. (5) Statistical analysis: Perform external validation to test AUC, calibration curves, and DCA in the independent validation cohort. Identifying patients more likely to achieve axillary lymph node pCR is an important factor for performing SLNB and avoiding unnecessary ALND post-NACT. This helps to accurately evaluate the axillary status, formulate surgical plans, and reduce surgical complications. In this case, our nomogram will help screen out patients more likely to achieve axillary lymph node pCR post-NACT, which will contribute to the design of future clinical trials, but only after rigorous multi‑center prospective validation can it be widely used in clinical practice. Acknowledgements This research is based on the data collected by Affiliated Hospital of Guangdong Medical University, and thanks to the doctors of this hospital for their help throughout the research process. Abbreviations Pathological complete response Neoadjuvant chemotherapy treatment Area under the curve Axillary lymph node dissection Sentinel lymph node biopsy False negative rate Body mass index Primary tumor grade Regional lymph node involvement grade Pulse repetition frequency Receiver operating characteristic curve Decision curve analysis Author contributions C wrote the paper, collected data, and made charts; S collected data; Y and X guided data collection, reviewed the paper, and made revisions. Funding This work was supported by the high-level talents scientific research start-up funds of the Affiliated Hospital of Guangdong Medical University [GCC2024029] and the Science and Technology Bureau Project of Zhanjiang [Grant No. 2025A502034]. Data availability The data underlying this article will be shared on reasonable request to the corresponding author. Declarations Ethics approval and consent to participate Ethics approval and consent to participate This study was conducted in accordance with the principles of the Declaration of Helsinki. Ethical approval was obtained from the Affiliated Hospital of Guangdong Medical University (PJ2021–024), and informed consent was secured from all participants. Consent for publication Applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Xiaohong Xu, Email: 13828297586@139.com. Yuping Yang, Email: 286270774@qq.com. References Associated Data Data Availability Statement The data underlying this article will be shared on reasonable request to the corresponding author.