Impact of neoadjuvant therapy followed by laparoscopic radical gastrectomy with D2 lymph node dissection in Western population: A multi‐institutional propensity score‐matched study

Abstract Background and Objectives In the setting of a minimally invasive approach, we aimed to compare short and long‐term postoperative outcomes of patients treated with neoadjuvant therapy (NAT) + surgery or upfront surgery in Western population. Methods All consecutive patients from six Italian and one Serbian center with locally advanced gastric cancer who had undergone laparoscopic gastrectomy with D2 lymph node dissection were selected between 2005 and 2019. After propensity score‐matching, postoperative morbidity and oncologic outcomes were investigated. Results After matching, 97 patients were allocated in each cohort with a mean age of 69.4 and 70.5 years. The two groups showed no difference in operative details except for a higher conversion rate in the NAT group (p = 0.038). The overall postoperative complications rate significantly differed between NAT + surgery (38.1%) and US (21.6%) group (p = 0.019). NAT was found to be related to a higher risk of postoperative morbidity in patients older than 60 years old (p = 0.013) but not in patients younger (p = 0.620). Conversely, no difference in overall survival (p = 0.41) and disease‐free‐survival (p = 0.34) was found between groups. Conclusions NAT appears to be related to a higher postoperative complication rate and equivalent oncological outcomes when compared with surgery alone. However, poor short‐term outcomes are more evident in patients over 60 years old receiving NAT.

postoperative complications rate significantly differed between NAT + surgery (38.1%) and US (21.6%) group (p = 0.019). NAT was found to be related to a higher risk of postoperative morbidity in patients older than 60 years old (p = 0.013) but not in patients younger (p = 0.620). Conversely, no difference in overall survival (p = 0.41) and disease-free-survival (p = 0.34) was found between groups.
Conclusions: NAT appears to be related to a higher postoperative complication rate and equivalent oncological outcomes when compared with surgery alone. However, poor short-term outcomes are more evident in patients over 60 years old receiving NAT.

| INTRODUCTION
Gastric cancer is one of the most common cancers in the world 1,2 and it is the third primary cause of cancer-related death around the world afflicting more than 400 000 patients each year in China 3 and more than 10 000 cancer-related deaths in the United States in 2017. 4 According to Western guidelines patients with locally advanced gastric cancer (LAGC) should undergo radical gastrectomy with D2 lymph node dissection. 5,6 Additionally, neoadjuvant therapy (NAT) with a perioperative regimen is recommended for patients with more than or equal to Stage IB resectable gastric cancer. 7 Benefits in terms of survival of this multimodal treatment are mainly supported by MAGIC 8 and the FNCLCC-FFCD 9 trials that introduced NAT in the current clinical practice as a standard of care for LAGC in Western countries.
However, some issues have arisen from these studies such as the inclusion of patients with lower esophagus or esophagogastric junction cancer in the analysis and an inadequate lymph node dissection in most cases. Over time, a perioperative regimen for treatment of LAGC was established as a procedural reference model within this setting. 10,11 Although it appears that NAT can be administered without increasing postoperative morbidity compared with gastrectomy alone, no definitive conclusions can be drawn 12,13 and furthermore, results from the CRITICS trial show that incomplete preoperative NAT due mainly to toxicity is an independent risk factor in developing postoperative complications. 14 In Eastern Asia, upfront surgery (US) is still recommended as primary treatment and health insurance in Japan and South Korea does not support neoadjuvant treatment for surgically resectable LAGC, 10 despite clinical evidence on the use of NAT are establishing in Japan. 15,16 Along with the controversial management of LAGC, minimally invasive surgery is yet another variable that needs to be evaluated.
The adoption of a laparoscopic approach in the treatment of LAGC demonstrated beneficial short-term outcomes and comparable longterm outcomes over open surgery in multicenter Asian randomized controlled trials (RCTs). [17][18][19][20][21] Laparoscopic gastrectomy (LG) is also gaining consensus in Europe 22,23 and United States 24,25 with satisfactory results in terms of oncological quality and postoperative morbidity. However, LG cannot be considered the gold standard for LAGC surgery and it remains limited to referral Centers with experience in the field.
As evidence on both NAT and LG have increased and they are recognized as promising treatment strategies in West, the role of NAT in patients exclusively undergoing laparoscopic radical gastrectomy with D2 lymph node dissection should be investigated.
The objective of this study is to compare short and long-term outcomes of patients receiving NAT before surgery with those receiving surgery alone through a "real-word" retrospective analysis from Western Centers with proven experience in laparoscopic gastric cancer surgery.

| Study population
A retrospective review of seven institutional databases was conducted to identify all patients who underwent LG for LAGC (Stages II and III) with curative intent from January 2005 to December 2019.
Tumor stages followed the 8th edition of American Joint Committee on Cancer (AJCC) TNM staging system for gastric cancer. 26

| Data collection and outcome measures
Baseline patient characteristics, intraoperative factors, and pathological tumor data were evaluated including gender, age, BMI, ASA status, tumor stage, type of surgery, operative time, blood loss, conversion rate, and intraoperative complications. Postoperative outcomes were evaluated in terms of complications and survival.
Postoperative complications occurred at any time during recovery or in the first 30 days after surgery were categorized based on Clavien-Dindo classification system. 27 Anastomotic leakage was evaluated in accordance with the definition and grading system of the UK Surgical Infection Study Group. 28 Postoperative mortality was defined as death by any cause within the first 30 or 90 days after surgery or at any time during a hospital stay.
Oncological outcomes in terms of disease-free survival (DFS) and overall survival (OS) were reported for each group. Locoregional recurrence was defined as recurred carcinoma of the remnant gastric pouch or at anastomosis site or within the lymphatic drainage area of the region of the primary tumor, confirmed by CT scan and/or pathological examination. Distant metastases were defined as recurrent tumors in the peritoneum, liver, nonregional lymph nodes, or outside the abdominal cavity such as lung, bones, or other sites.
Follow-up after surgery included physical examination every 3-6 months for the first 2 years and every 6-12 months for years 3-5. CT scan was performed every 6-12 months for the first 2 years, then annually for up to 5 years.

| Statistical analysis
Preliminary analysis concerned the treatment of missing data. Very little data were missing for BMI (0.8% of the total) and time to first flatus (3% of the total), which were considered as missing at random.
Missing value imputation was performed using the k nearest neighbors (KNN) algorithm, 29 using the Euclidean distance as a distance metric in the multidimensional space.

| RESULTS
The overall study population consisted of 366 patients. Before propensity score matching, statistically significant difference in stage tumor (p = 0.001) was recorded between NAT and US group (  We reported outcomes of a "real-world" retrospective analysis including patients with Stage II-III gastric cancer from seven institutions with a high level of experience in advanced laparoscopic surgery. We previously presented short and long-term outcomes from the same series of over 300 patients who underwent LG. 31  We hypothesize that several factors could affect outcomes after NAT as they relate to toxicity, 33 a worse nutritional status, sarcopenia, 34 and neutropenia. 35 However, our findings differ from those of others western 36,37 and eastern [38][39][40][41] studies claiming that the administration of NAT is not associated with a greater risk of postoperative morbidity when compared to the United States. In the present study, two main aspects need to be considered foremost of which is a preoperative difference among population studies. Patients with a mean age of 70 years old were included in the present study and it is reasonable to assume that their outcomes would be different when compared to patients younger than 10-15 years 37-40 or 20 years. 41 In fact, we demonstrated that the difference in complication rate between NAT and US group is relevant over 60 years old (p = 0.013) but not in younger patients (p = 0.620). Although it has been previously demonstrated that age is a significant predictor of postoperative complications, [42][43][44] it is still controversial if NAT could negatively affect short-term outcomes in older patients as reported by Fujitani et al. 45 or not. 46 A second aspect to consider is the minimally invasive approach for curative gastrectomy. In a comparative series of patients treated with gastrectomy and D2 lymphadenectomy, Wu et al. 38 observed a more intraoperative blood loss over the surgery alone group (p < 0.04) because of NAT-induced fibrosis surrounding lymph nodes. This could explain the higher rate of conversion to open surgery that we reported in the NAT group. However, only one international propensity scorematched study compared the outcomes of patients who received LG with D2 lymphadenectomy with or without NAT. 47 Although comparable postoperative complication rates were demonstrated between groups, age more than or equal to 60 was identified as a risk factor (OR 21.338; p < 0.001). Additionally, the remarkable difference in preoperative ASA score (ASA III = 2.3%) with the present study (ASA III-IV = 30%) could indirectly affect outcomes.
Based on this evidence, it can be assumed that age is not the only factor influencing postoperative morbidity, but also epidemiological data, were too small to distinguish between no effect and a small effect.
Conversely, our findings are consistent with more recent studies 36,37 demonstrating that NAT had not statistically significant effect on survival rates when compared with surgery alone in patients with LAGC and, based on these results, this appears to be confirmed even when a minimally invasive approach is adopted. Furthermore, even in the case of gastric signet ring cell carcinoma which is characterized by a worse prognosis, NAT does not appear to provide survival benefits compared with primary surgery. 49-51 Therefore, with regard to the lack of difference in OS and DFS that we found between groups, long-term outcomes seem absolutely justifiable after an adequate literature analysis, despite the need for welldesigned prospective trials to drawn definitive conclusions.
The current study has several limitations. In addition to its retrospective design, patients considered fit for surgery could more likely have been selected for the US group and we are aware that this cannot be overcome with propensity score-matched analysis. Data used was collected from different institutions over a 15-year period, and therefore the study suffers from a heterogeneity of the neoadjuvant regimens with no patients who received neoadjuvant therapy until 2010. Furthermore, the study provides neither the evaluation of NAT-related toxicity nor a preoperative assessment of patients' conditions including morbidity and frailty index or performance status and sarcopenia evaluation.

| CONCLUSIONS
This study suggests that NAT prior LG is related to a higher postoperative complication rate compared with surgery alone in patients with LAGC. The negative effect of NAT on postoperative morbidity is more evident in patients above 60 years old. NAT had a nonsignificant impact on DFS and OS when curative resection with D2 lymph node dissection was carried out. Further studies with more selective patient recruitment need to be conducted to define the real advantages of NAT before surgery in gastric cancer exclusively and to better understand the impact on postoperative and intraoperative complications in the patient population that undergoes laparoscopic gastrectomy with D2 dissection.

ACKNOWLEDGMENT
This study received no funding.