Prognostic value of inflammatory nutritional scores in locally advanced esophageal squamous cell carcinoma patients undergoing neoadjuvant chemoimmunotherapy: a multicenter study in China

Objective This study investigates the prognostic significance of inflammatory nutritional scores in patients with locally advanced esophageal squamous cell carcinoma (LA-ESCC) undergoing neoadjuvant chemoimmunotherapy. Methods A total of 190 LA-ESCC patients were recruited from three medical centers across China. Pre-treatment laboratory tests were utilized to calculate inflammatory nutritional scores. LASSO regression and multivariate logistic regression analyses were conducted to pinpoint predictors of pathological response. Kaplan-Meier and Cox regression analyses were employed to assess disease-free survival (DFS) prognostic factors. Results The cohort comprised 154 males (81.05%) and 36 females (18.95%), with a median age of 61.4 years. Pathological complete response (pCR) was achieved in 17.38% of patients, while 44.78% attained major pathological response (MPR). LASSO and multivariate logistic regression analyses identified that hemoglobin, albumin, lymphocyte, and platelet (HALP) (P=0.02) as an independent predictors of MPR in LA-ESCC patients receiving neoadjuvant chemoimmunotherapy. Kaplan-Meier and log-rank tests indicated that patients with low HALP, MPR, ypT1-2, ypN0 and, ypTNM I stages had prolonged DFS (P < 0.05). Furthermore, univariate and multivariate Cox regression analyses underscored HALP (P = 0.019) and ypT (P = 0.029) as independent predictive factors for DFS in ESCC. Conclusion Our study suggests that LA-ESCC patients with lower pre-treatment HALP scores exhibit improved pathological response and reduced recurrence rate. As a comprehensive index of inflammatory nutritional status, pre-treatment HALP may be a reliable prognostic marker in ESCC patients undergoing neoadjuvant chemoimmunotherapy.


Introduction
Esophageal cancer, a prominent cause of mortality among digestive tract malignancies, was the sixth leading cause of cancer-associated deaths worldwide in 2020, accounting for 544,000 fatalities (3.1% of all cancer deaths) (1).Esophageal cancer includes adenocarcinoma (AC) and squamous cell carcinoma (SCC) (2).Notably, esophageal squamous cell carcinoma (ESCC) comprises over 90% of cases in China (3,4).Frequently diagnosed at a locally advanced stage, esophageal cancer poses significant treatment challenges (5).While surgical resection is the primary treatment for locally advanced cases, the 5-year survival rates remain suboptima (6,7).Consequently, neoadjuvant therapy is recommended by the Chinese Society of Clinical Oncology (CSCO) and the National Comprehensive Cancer Network (NCCN) to enhance survival in patients with locally advanced ESCC (LA-ESCC) (8).
Immune checkpoint blockers (ICB) therapy has significantly progressed in treating metastatic or advanced ESCC (9,10).The NCCN guidelines now endorse immunotherapy for treating advanced esophageal cancer (EC) (11).Immune checkpoint inhibitors (ICIS), which can induce systemic anti-tumor immunity by releasing new antigens from dead tumor cells and stimulating the initiation and expansion of new antigen-specific T cells in tumors before surgical resection, have advantages over adjuvants in preclinical studies (12).Recent studies indicate that neoadjuvant chemoimmunotherapy is a viable and safe option for LA-ESCC patients, enhancing major pathological response (MPR) and pathological complete response (pCR) with manageable adverse effects (13)(14)(15).However, research on predictive factors for pathological response in LA-ESCC patients undergoing neoadjuvant chemoimmunotherapy remains scant, and survival prognosis in this demographic is underexplored.Presently, there is an absence of reliable pre-treatment indicator to predict the pathological response and survival in these patients.Therefore, identifying cost-effective and practical pre-treatment indicators is imperative for prognostication in LA-ESCC patients before neoadjuvant chemoimmunotherapy.
Hence, this study aims to investigate the role of inflammation and nutritional status in predicting pathological response and disease-free survival (DFS) in LA-ESCC patients undergoing radical surgery following neoadjuvant chemoimmunotherapy.The objective is to explore more cost-effective and efficient pre-treatment indicators for prognosis prediction in LA-ESCC patients, ultimately enhancing the postoperative prognosis management of esophageal cancer.

Patient selection and data collection
The study was conducted following the Declaration of Helsinki and was approved by the Ethics Committee of Zhongshan Hospital Xiamen University.We conducted a systematic review of the clinical data of patients with locally advanced ESCC who underwent neoadjuvant chemoimmunotherapy.Before treatment, comprehensive diagnostic evaluations were performed, encompassing electronic gastroscopy, chest and abdominal CT scans, upper gastrointestinal angiography, neck color ultrasound, and PET-CT scans as needed.Inclusion criteria of this research were: patients receiving treatment at Zhongshan Hospital Xiamen University, Fujian Medical University Union Hospital, Affiliated Hospital of Putian University, Union Hospital as well as Thoracic Surgery from January 2019 to April 2023; age between 18 and 75 years old; resectable stage II-IVa ESCC as determined by the American Joint Committee on Cancer (AJCC) 8th edition TNM staging system; underwent radical esophagectomy; completeness of medical records and follow-up.Exclusion criteria encompassed a pathological diagnosis of adenocarcinoma or other types of esophageal carcinoma types; tumor location at the esophagogastric junction or gastric proximal; cancer complicated with malignant diseases of other organs or previous therapy (including radiotherapy, chemotherapy, and submucosal resection); autoimmune disease or hematologic disease; heart, lung, or liver dysfunction, or acute infection.

Treatment and follow-up
Similar to our earlier research, the preoperative neoadjuvant chemoimmunotherapy treatment protocols were used (24).Patients received a minimum of two cycles of therapy, combining paclitaxel-based chemotherapy and PD-1 inhibitors.The PD-1 monoclonal antibodies used in this study included camrelizumab, pembrolizumab, sintilimab, tislelizuma, and toripalimab, administered at a dose of 200mg.Surgical resection followed 4-8 weeks post-treatment completion.Our department utilized the McKeown minimally invasive esophagectomy technique, including two-field lymphadenectomy and stomach-based digestive tract reconstruction.In cases of cervical lymph node involvement, a threefield lymphadenectomy was performed.Until now, the optimal adjuvant therapy following neoadjuvant chemoimmunotherapy and surgical intervention for patients with EC has yet to be clearly defined.The adjuvant treatment regimen varied among different institutions and involved the administration of adjuvant immunotherapy, radiotherapy, or chemotherapy.All patients were subject to regular follow-up assessments following the completion of their respective treatment modalities.Post-operation follow-up was performed every three months for the first three years, every six months for the subsequent three years, and the year after the fifth year.Follow-up procedures included routine blood examination, liver and kidney function, tumor markers and other laboratory tests, chest and abdominal CT scans, neck color ultrasound, upper gastrointestinal angiography, whole-body PET-CT, and bone scans.Magnetic resonance imaging was conducted and provided the needful.The AJCC/UICC 8th edition staging system and the modified Ryan protocol tumor regression grade (TRG) were used for evaluation (25).Major pathological response (MPR) was defined as the presence of less than 10% viable tumor cells in the resected tumor specimen, whereas pathological complete response (pCR) was defined as the absence of residual tumor cells in both the primary tumor and lymph nodes.Any recurrence, whether local, regional, or distant, was considered as such.Disease-free survival (DFS) was defined as the period from surgery until the first confirmed recurrence or death from any cause.The final follow-up period for this study concluded in June 2023.

Statistical analysis
Data analysis utilized SPSS 22.0 and R version 4.2.3.Normally distributed variables were presented as absolute numbers and percentages, mean, and standard deviation, while nonparametric variables were reported as median and interquartile range.Categorical variables were assessed using chi-square or Fisher's exact test, while continuous variables were compared using t-test or Mann-Whitney U test.The survminer package in R determined optimal cut-off values for PLR, LMR, NLR, SII, PNI, HALP, and SIRI.To identify potential predictors of MPR, we employed the least absolute shrinkage and selection operator (LASSO) logistic regression model, and these variables were included in a multivariate logistic analysis.To prevent multicollinearity, a stepwise regression analysis was employed.DFS was estimated using the Kaplan-Meier method, with survival differences assessed by a stratified log-rank test.Cox proportional-hazards model was employed for univariate and multivariate analyses on DFS, considering P<0.05 as statistically significant.

Patient characteristics
The study enrolled 190 patients diagnosed with LA-ESCC, comprising 154 males (81.05%) and 36 females (18.95%), with a median age of 61.4 years (IQR: 57.0-67.0years).Tobacco use was reported by 56.32% of the patients, while 30.53% had a history of alcohol consumption.Additionally, hypertension was present in 19.47% of the patients, and diabetes in 5.26%.The majority of primary tumors were located in the middle (53.16%) or lower portion (38.94%)

Predictors for MPR using LASSO regression and logistic analysis
The objective of this study was to identify prognostic indicators influencing MPR in patients with locally advanced ESCC undergoing neoadjuvant chemoimmunotherapy.We conducted a comprehensive analysis of the medical records of 190 patients, focusing on their general condition, coexisting diseases, specifics of the treatment received, and laboratory results.Employing LASSO regression, we identified several key factors, including HALP, SIRI, ASA status, and ALB(albumin), as significant predictors of MPR (Figures 1A, B).These factors were further examined through multivariate logistic regression analyses (Table 2).Our analysis highlighted HALP (OR = 0.335, 95% CI = 0.125,0.807,P = 0.02) as an independent predictor of MPR in this patient cohort.Additional statistical evaluations, including the Receiver Operating Characteristic curve (ROC), calibration curve, and Decision Curve Analysis (DCA), are detailed in the Supplementary File.The findings of our analyses are succinctly summarized in Table 2.

Patient characteristics grouped by HALP
This study focused on further validating the prognostic significance of the HALP in predicting DFS in ESCC patients treated with neoadjuvant chemoimmunotherapy.We utilized the Kaplan-Meier method and Cox regression analyses to assess the correlation between HALP and clinical outcomes.For this analysis, we included only those patients who had a follow-up period exceeding three months, resulting in a cohort of 181 patients.Within this group, 30 (16.57%) patients were classified under H-HALP, and 151 (83.43%) patients under L-HALP.Clinical characteristics grouped by HALP are presented in Table 3.Our findings indicated a significant association of L-HALP with MPR (P=0.007) and an association of H-HALP with high recurrence rates (P<0.001) and possibly with advanced ypTNM stages (P = 0.054).No significant differences were observed in other clinical information (P > 0.05).These results support the potential of HALP as a reliable prognostic marker for DFS in LA-ESCC patients undergoing neoadjuvant chemoimmunotherapy.

Categorization of recurrence patterns in ESCC patients post-treatment
In this study, patients were stratified based on their initial posttreatment recurrence patterns into three distinct categories: local recurrence, distant metastasis, and concurrent local and distant metastasis.Among the 29 patients who experienced recurrence, 17 (58,62%) presented with local recurrence, which encompassed locoregional lymph node metastasis and anastomotic site recurrence.Additionally, 3 (10.34%)developed metastasis, while 9 (31.03%)patients exhibited both local recurrence and distant metastasis.Distant metastasis consisted of peritoneal metastasis and non-regional lymph node metastasis.Recurrence in two patients was confirmed through biopsy, whereas the remaining cases were identified via imaging examinations.

HALP and disease-free survival of LA-ESCC
We aimed to explore prognostic factors influencing DFS in ESCC patients and further substantiate the predictive capacity of the HALP score.The median follow-up period was 16(range:3-38) months.Kaplan-Meier analysis revealed a significantly higher DFS in the L-HALP group compared to the H-HALP group (P=0.00056, Figure 2A).Additionally, we examined DFS variations across different postoperative pathological states.Our findings indicated that patients with MPR (P=0.013, Figure 2E) or ypTNM stage I (P=0.026, Figure 2D) exhibited extended DFS.Conversely, patients presenting with ypT3-4 (P=0.0003, Figure 2B) or ypN+ (P=0.0069, Figure 2C) had a poorer prognosis.Although not statistically significant, patients who achieved pCR showed a trend towards longer DFS (P=0.29, Figure 2F).Notably, the median DFS was reached only in the ypT3-4 and H-HALP subgroups, with durations of 25.0 and 21.0 months, respectively.In summary, our analysis suggests that HALP, ypT, ypN, ypTNM stage, and MPR are collectively indicative of survival outcomes in ESCC patients.

Kaplan-Meier survival curves of L-HALP subgroup and H-HALP subgroup stratified by ypT stage
The Kaplan-Meier survival curves were used to assess the DFS of the high and low HALP groups stratified by the ypT stage.Our findings demonstrated that patients within the L-HALP subgroup who were at ypTNM stage I stage exhibited significantly longer DFS (P=0.0089, Figure 3A).In contrast, while the H-HALP subgroup at ypTNM stage I also showed a trend towards extended DFS, this difference did not reach statistical significance (P=0.064, Figure 3B).Notably, only patients in the H-HALP subgroup with ypTNM II/ III/IV stages achieved the median DFS, recorded at 18.0 months.Further investigations are warranted to confirm these observations.

Discussion
In recent years, the landscape of cancer treatment has increasingly pivoted towards immunotherapy, recognized as a breakthrough in therapeutic strategies.Compelling evidence from randomized phase III trials substantiates the efficacy of immunotherapy as a primary treatment modality either as a standalone approach or synergistically combined with chemotherapy, specifically for patients with advanced or metastatic esophageal squamous cell carcinoma (ESCC) (9, 10, 28).Building on this foundation, researchers have begun to explore the potential benefits of utilizing neoadjuvant chemotherapy combined with immunotherapy to treat locally advanced ESCC.Studies in this domain have revealed that adverse reactions associated with neoadjuvant immunotherapy are manageable, culminating in high incidences of R0 resection, commendable pCR rates, and limited postoperative complication (29,30).However, the predictive factors influencing therapeutic outcomes remain elusive.With this in mind, the present investigation aims to identify vital prognostic factors that may impact MPR and DFS in patients with locally advanced ESCC treated with neoadjuvant chemoimmunotherapy.
Our investigation yielded that 17.38% of ESCC patients attained a pCR, while 44.78% achieved a MPR following neoadjuvant chemoimmunotherapy.In the current study, we initially explored the role of pre-treatment inflammatory nutritional scores on pathological response prediction in ESCC with neoadjuvant chemoimmunotherapy.Among the cohort with locally advanced ESCC undergoing neoadjuvant chemoimmunotherapy, we employed Lasso regression for an initial screening of potential predictive factors associated with MPR.This analysis pinpointed four pivotal predictors: the HALP, SIRI, ASA status, and serum albumin (ALB).Further, multifactor logistic regression analysis ratified HALP as an independent predictor of MPR.
Subsequently, we delved into the correlation between HALP and prognosis.To ascertain the prognostic significance of HALP and other variables for DFS, Kaplan-Meier curves and Cox regression analyses were utilized.Our results demonstrate that both HALP and ypT stage stand as robust prognostic markers for DFS in patients with LA-ESCC post-neoadjuvant chemoimmunotherapy.These insights furnish   Neutrophils are widely recognized as the primary source of vascular endothelial growth factor (VEGF), which can promote both tumor proliferation and neovascularization (31).Concurrently, there is increasing evidence of cancer progression being intertwined with the activation of circulating lymphocytes (32).Furthermore, a growing body of research underscores the complex roles of platelets in the advancement of cancer (33).In addition, composite inflammation and nutrition scores, derived from various hematological parameters, have been documented to possess prognostic significance across diverse cancer types, including ESCC (20,21).The HALP score, which evaluates both immune and nutritional status, has emerged as a notable prognostic marker in various cancers, such as gastrointestinal and urogenital cancers (34,35).Moreover, the Systemic Immune-Inflammation Index (SIRI), calculated using absolute neutrophil, monocyte, and platelet counts, has been identified as a prognostic element in patients with hepatocellular carcinoma and lung cancer (36,37).Similarly, the lymphocyte-to-monocyte ratio (LMR) has been established as an independent predictor for achieving R0 resection in surgeries for hilar cholangiocarcinoma (38).Notably, while previous research has associated a low HALP score with poor survival, our study posits that a low HALP score may be a predictive factor for MPR and DFS in locally advanced ESCC patients undergoing neoadjuvant chemoimmunotherapy.This phenomenon warrants further exploration.Firstly, the precise role of platelets in tumor biology remains a subject of ongoing investigation.Some studies correlate an increase in pre-treatment platelet counts with enhanced neoadjuvant response (39).The team led by Tim F. Greten has found that platelets inhibit tumor growth and metastasis by releasing P2Y12-dependent CD40L (40).Others have found that an elevated baseline platelet count is an independent predictor of worse overall survival (41).Hence, the interaction between the bioactive factors released by platelets and the tumor immune environment in different situations remains to be studied.Platelet count is a critical component of HALP.Secondly, in this population of locally advanced ESCC patients receiving neoadjuvant chemoimmunotherapy, inflammatory factors might be activated during treatment, contributing to favorable responses (39).Such inflammatory states can precipitate the release of proinflammatory mediators from leukocytes, promoting lymphocyte apoptosis and platelet activation, while also impacting the metabolism of erythrocytes and albumin, potentially leading to reductions in the HALP score.Further studies are needed to confirm these findings and to elucidate the underlying mechanisms.
Additionally, our study suggests that in patients with ESCC undergoing neoadjuvant chemoimmunotherapy, ASA status may emerge as an independent predictor of MPR.The ASA score, a prevalent metric for evaluating preoperative physical status in surgical patients.not only reflects the patient's overall health but has also been acknowledged as a reliable forecaster of postoperative complications and survival in various cancer types (42,43).It is critical to highlight that despite employing stepwise regression analysis to mitigate multicollinearity, some degree of overfitting remain, attributed to the constrained sample size(OR: 0.356, 95%  (44,45).Nevertheless, even in instances where patients attain pCR or MPR during surgery, the prospect of a complete cure remains uncertain.Moreover, it is observed that some patients may manifest a favorable prognosis without achieving pathological remission; hence, a multifaceted evaluation encompassing various factors is crucial for prognostic assessment.Therefore, our research also delves into the influence of disparate postoperative pathological outcomes and staging on DFS in LA-ESCC patients treated with neoadjuvant chemoimmunotherapy.Utilizing the Kaplan-Meier method for survival curves construction   and the log-rank test for survival differences assessment, our findings reveal that patients with an MPR, ypT1-2, ypN0, or ypTNM I stage exhibit prolonged DFS (P<0.05).While patients with pCR appeared to have an extended DFS, the difference was insignificant (P>0.05).Moreover, multivariate Cox regression analysis identified ypT stage as an independent prognostic indicator for DFS in ESCC.Given the limitations of follow-up duration and sample size, subsequent research is warranted to explore the enduring impact of postoperative pathological outcomes and staging on survival.
Although our study identified HALP as a good prognostic factor for ESCC patients, several limitations exist.Primarily, the small sample size and the retrospective nature of this study constrain its broad applicability Consequently, the prognostic validity of the HALP score in ESCC patients undergoing neoadjuvant chemoimmunotherapy necessitates further corroboration in a more extensive cohort.Secondly, our investigation focused exclusively on the predictive value of inflammatory and nutritional scores, without delving into other potential prognostic indicators like tumor biomarkers.Thirdly, the optimal neoadjuvant treatment strategy for locally advanced ESCC remains a subject of ongoing debate.Therefore, our findings might not extend to other emerging treatment modalities, such as neoadjuvant radiochemotherapy combined with immunotherapy.Fourthly, the intricate mechanisms underpinning the relationship between inflammatory and nutritional scores and both pathological response and DFS require deeper exploration.Lastly, the relatively brief follow-up period of our study necessitates additional research to substantiate the long-term prognostic implications.
In summary, our research presents evidence that pre-treatment HALP scores can prognosticate MPR and DFS in locally advanced ESCC patients receiving neoadjuvant chemoimmunotherapy.Future studies are imperative to not only validate our findings but also to ascertain the clinical utility of HALP in guiding therapeutic decisions and enhancing patient outcomes.
FIGURE 1 (A) LASSO coefficient profiles of the all indicators.(B) Ten-fold cross-validation for tuning parameter selection in the LASSO model.

2 FIGURE 3
FIGURE 2 Kaplan-Meier survival curves.(A) DFS of patients with different HALP levels.(B) DFS of patients according to ypT stage.(C) DFS of patients according to ypN stage.(D) DFS of patients according to ypTNM stage.(E) DFS of patients according to MPR. (F) DFS of patients according to pCR.DFS, disease-free survival.pCR, pathological complete response.MPR, major pathological response.HALP, hemoglobin, albumin, lymphocyte, and platelet.
thoracic esophagus.At diagnosis, 72.63% of the patients were at clinical stage III or IV, with only 27.37% diagnosed at stage II.68.42% of the patients received two cycles of neoadjuvant chemoimmunotherapy.Post-neoadjuvant therapy, pCR was achieved in 17.38% of patients, and MPR in 44.78%.Inflammation and nutrition scores were categorized into low (L) and high (H) groups based on predetermined cutoff values (Supplementary Table 1).This categorization resulted in 30 patients (15.79%) in the L-PLR group,160 patients (84.21%) in the H-PLR group;43 patients (22.63%) in the L-NLR group, 147 patients (77.37%) in the H-NLR group; 165 patients (86.84%) in the L-LMR group, 25 patients (13.16%) in the H-LMR group; 157 patients (82.63%) in the L-HALP group, 33 patients (17.37%) in the H-HALP group; 33 patients (17.37%) in the L-SII group, 157 patients (82.63%) in the H-SII group; 154 patients (81.05%) in the L-SIRI group, 36 patients (18.95%) in the H-SIRI group; and 157 patients (82.63%) in the L-PNI group, 33 patients (17.37%) in the H-PNI group.The clinical and pathological characteristics of LA-ESCC patients are summarized in Table 1.

TABLE 1
Clinical characteristics in 190 esophageal squamous cell carcinoma patients.

TABLE 2
Multivariable analysis of Predictors for MPR.

TABLE 3
Comparison of the clinical variables in esophageal squamous cell carcinoma grouped by HALP.

TABLE 3 Continued
ASA score, the American Association of Anesthesiologists (ASA) classifies patients according to their physical condition and surgical risk.BMI, body mass index (kg/m 2 ).pCR, pathological complete response.MPR, major pathological response.HALP, hemoglobin, albumin, lymphocyte, and platelet.CI: 0.111,0.976,p=0.058).Consequently, future studies with larger cohorts are imperative to reinforce these findings more robustly.Previous research indicates that ESCC patients achieving pathological remission post-neoadjuvant therapy substantially prolong survival

TABLE 4
Univariate and multivariate analyses of DFS.