Gastric cancer is among the commonly occurring cancers worldwide and is one of the leading causes of cancer-related deaths (1). Malnutrition is an important factor affecting the course of disease and treatment response in gastric cancer patients (2). Biomarkers such as the Prognostic Nutritional Index (PNI) are objective parameters used to assess the nutritional status of patients. PNI reflects the nutritional status of patients based on serum albumin and total lymphocyte count, and recent studies have demonstrated the prognostic and post-operative value of PNI in gastric cancer patients (3). Additionally, signet ring cell gastric cancer (SRCC) constitutes approximately 8%–30% of all gastric cancers and is a subtype with a more aggressive course and poor prognosis. The tendency of SRCC to occur at a younger age, the risk of late diagnosis, and its resistance to treatment necessitate special examination of this subtype (4).

Many patients undergoing gastrectomy for gastric carcinoma experience significant nutritional management challenges upon hospital discharge. The transition from structured hospital nutritional protocols to self-administered dietary care presents considerable difficulties and frequently results in compromised nutritional status. Investigational studies have demonstrated that dietary habit modification is particularly problematic when these individuals cohabitate with family members maintaining conventional nutritional regimens. Pre-operative dietary patterns likely persist into the post-operative period. Despite established evidence regarding the prognostic significance of nutritional status, these persistent behavioral patterns frequently result in patients reverting to previous suboptimal dietary regimens following surgical intervention (3).

It is well established that all cancer patients experience some degree of nutritional impairment, with this effect being particularly pronounced in patients with cancers of the upper digestive tract and pancreas, often due to mechanical impediments to food intake. The metabolic demands of cancer create a competition for nutrients between the tumor and the host, disrupting the normal functioning of the human body. Therefore, comprehensive nutritional assessment and intervention by dedicated nutritionists in oncology units are essential components of cancer care. These specialists can evaluate individual patient needs and establish appropriate nutritional treatment protocols tailored to each patient's specific requirements (5).

In this study, we aimed to investigate the effect of PNI on perioperative complications and prognosis. Additionally, we aimed to determine whether PNI varies according to tumor type by comparing it across histological subtypes.

Data from a total of 373 operated patients were retrospectively screened. Forty-two patients (11.2%) were excluded due to insufficient follow-up and laboratory data, 38 patients (10.2%) due to metastatic disease detected during operation, perioperative mortality, and R1–R2 resection, 31 patients (8.3%) due to ECOG score ≥3 or failure to complete neoadjuvant-adjuvant treatments for any reason, 24 patients (6.4%) due to inadequate lymph node dissection, and 18 patients (4.8%) due to history of a second malignancy and additional organ resection. A total of 220 patients met the inclusion criteria for the study (Figure 1).

According to the normality test, age, BMI, tumor size, number of lymph nodes removed, and PNI values were not normally distributed. The mean age of all patients was 60.63 ± 10.56, BMI was 24.49 ± 3.44, and PNI value was 47.15 ± 6.07.

Of the patients, 73.2% (n = 161) were male and 26.8% (n = 59) were female; 8.2% (n = 18) were ASA 1, 36.8% (n = 81) ASA 2, 51.8% (n = 114) ASA 3, and 3.2% (n = 7) ASA 4; 9.5% (n = 21) were stage 1, 26.4% (n = 58) stage 2, and 64.1% (n = 141) stage 3; 29.5% (n = 65) had hypertension (HT), 27.3% (n = 60) had diabetes mellitus (DM), 6.8% (n = 15) had chronic obstructive pulmonary disease, and 14.1% (n = 31) had coronary artery disease (CAD); 20.9% (n = 46) had SRCC, 12.7% (n = 28) had well-differentiated, 30.5% (n = 67) had moderately differentiated, 29.5% (n = 65) had poorly differentiated, and 6.4% (n = 14) had mucinous adenocarcinoma; 57.3% (n = 126) had PeNI, and 63.9% (n = 140) had LVI. The tumor was located in the upper 1/3 in 33.6% (n = 74) of the patients, in the middle 1/3 in 25.9% (n = 57), and in the lower 1/3 in 35.9% (n = 79). Total gastrectomy was performed in 64.1% (n = 141) and subtotal gastrectomy in 35.9% (n = 79) (Table 1).

In the ROC analysis performed to find the cut-off value for evaluating 5-year overall survival (OS) for PNI, the PNI value was found to be 46.2 (AUC = 0.673, 95%CI: 0.599–0.747, p < 0.001; sensitivity 78.8%, specificity 51.9%) (Figure 2).

The 5-year survival rate was found to be 30.0%. When comparing the survival of patients with PNI values less than 46.2 and those greater than 46.2, it was observed that survival was worse in patients with low nutritional index (log rank p = 0.001) (Figure 3).

In the examination conducted by dividing patients with PNI values less than 46.2 as poor and those with values greater than 46.2 as good nutritional status into two categories, it was observed that PNI was lower in the group where PeNI and LVI were more common (p values 0.005 and 0.032, respectively). Patients with SRCC had statistically lower PNI compared to other adenocarcinoma types (p = 0.001). Between the two groups, 5-year survival was statistically worse in the low PNI group (p < 0.001). There was no statistically significant difference between response to neoadjuvant therapy and PNI value in patients receiving neoadjuvant therapy (p = 0.095). When comparing complications according to the Clavien-Dindo classification, Grade 1 (23.4% vs. 16.7%), Grade 2 (16.0% vs. 7.9%), Grade 3 (11.7% vs. 4.8%), and Grade 4 complication rates were higher in the low PNI group, whereas patients without complications were higher among the high PNI group (44.7% vs. 69.0%) (p = 0.006) (Table 2).

In the logistic regression analysis based on 5-year survival, while chronic obstructive pulmonary disease (COPD), tumor stage, LVI, PeNI, histological type being SRCC, and low PNI values were significant according to univariate analysis, in multivariate analysis, tumor stage (for stage I p = 0.01, for stage 2 p = 0.034, and for stage 3 p = 0.002) and PNI value (p = 0.001) stood out as the only parameters affecting 5-year survival (Table 3).

Gastric cancer is a complex clinical condition requiring the development of individualized strategies in treatment approaches, especially considering the factor of malnutrition. This study demonstrates not only the value of the Prognostic Nutritional Index (PNI) in predicting postoperative complications and long-term survival in patients with gastric cancer but also highlights the differences among histological subtypes. Particularly, the association of signet ring cell carcinoma (SRCC) with lower PNI values compared to other adenocarcinoma subtypes emphasizes the necessity of closer monitoring of nutritional status in the treatment of this aggressive subtype. The results of the study indicate that preoperative PNI assessment could be a valuable parameter in predicting patients’ risk of postoperative complications and forecasting long-term survival, thus helping clinicians in planning patient-centered treatment approaches.

Nutritional status inevitably affects tumor patients’ prognosis through their resistance to the tumor's catabolic activity. An increasing number of studies show that patients’ basic nutritional status is associated with long-term prognosis (12). Inadequate nutrition, decreased immunity, and increased inflammation not only affect cancer patients’ response to treatment but may also increase the likelihood of recurrence and metastasis of malignant tumors (13). Based on this, the Prognostic Nutritional Index (PNI), which reflects both the immune status (lymphocytes) and nutritional reserves (albumin) of patients, provides information about the overall physiological condition of patients. This simplicity confers a significant advantage in terms of ease of use (14).

PNI was first reported by Onodera et al. in 1984 to have prognostic value in patients undergoing surgery for gastrointestinal cancer (15). Since then, numerous studies have been conducted on this topic. It was reported that preoperative PNI was a good prognostic indicator of hepatocellular carcinoma, gastric cancer, colorectal carcinoma, and pancreatic cancer (16). A 2024 meta-analysis examining 18,596 patients found a significant relationship between PNI and OS (17). In our study, consistent with the literature, low PNI was associated with poor prognosis, and in multivariate analysis, it was found to be a risk factor affecting survival on its own.

There are various studies investigating the cut-off value for PNI, but there is no consensus on this issue. Some studies accept 45 as the cut-off value, as stated in Onodera's study (18, 19). However, when the literature is examined, <43, <44.8, ≤49.2, <52, and <52.9 have been found to be associated with poor prognosis (17). In our study, similar to the value in the original study, <46.2 with sensitivity 78.8% and specificity 51.9% was found to be associated with poor prognosis.

An inverse relationship is observed between PNI and the pathological stage of the tumor (20). According to a meta-analysis conducted in 2016, it was stated that there is a correlation between low PNI values and advanced pathological stage. However, according to the same meta-analysis, a direct relationship between stage and PNI could not be demonstrated in 5 different studies examined. In our study, while a direct relationship was found between stage and OS, no statistically significant relationship was found between stage and PNI (19).

In addition, it has been shown that there is a relationship between LVI and PeNI and PNI; PeNI and LVI were more common in patients with low PNI (19, 21). In a similar study published in 2024, no statistically significant relationship was found between PeNI and LVI in patients with PNI > 39.8 (11). In our study, PeNI and LVI were found to be more common in patients with lower PNI. Additionally, in our study, PNI was found to be lower in patients diagnosed with gastric cancer with signet ring cell histology compared to the other group. In light of the studies examined, it is understood that there is a relationship between these parameters, which are related to tumor aggressiveness and have been proven by various studies to have a negative effect on survival, and PNI. Probably, PNI values are found to be lower because the catabolic process is more intense in tumors with more aggressive behavior (22, 23). Prospective studies with large patient populations are needed to resolve conflicting findings on this issue.

There are studies examining the relationship between pre-neoadjuvant nutritional status and response to neoadjuvant therapy in gastric cancer. In a study published in 2024, it was found that patients with low PNI values had worse responses to neoadjuvant therapy (24). In their article published in 2021, Meng F. et al. developed a scoring system combined with a systemic immune-inflammatory index in addition to PNI, and according to this, they state that patients with low PNI values respond worse to neoadjuvant therapy (25). In another study, the frequency of pathological complete response was found to be higher in non-metastatic gastric cancer patients with better immune-nutritional status, but no statistical difference was found (6.6% vs. 1.2%, P = 0.107) (26). In our study, regression was observed in 59.2% of those receiving neoadjuvant therapy in the high PNI group and in 44.2% of patients in the low PNI group, but the effect of PNI value calculated before neoadjuvant therapy on the response to neoadjuvant therapy was not statistically determined (p = 0.095). It is seen that nutritional status is related to the response to neoadjuvant therapy, but additional studies are needed to determine whether PNI values can be used to predict response to neoadjuvant therapy.

PNI value also has an effect on surgical complications in the postoperative period. In our study, complication rates according to the Clavien-Dindo classification were found to be significantly higher in the patient group with PNI < 46.2 (p = 0.006). In a meta-analysis published by Yang et al. in 2016, including 3396 gastric cancer patients, a strong association was demonstrated between low PNI values and postoperative complications (OR = 1.74, 95% CI = 1.41–2.16, p < 0.01). In this meta-analysis, pooled results from five different studies revealed that patients with poor nutritional status were more vulnerable to postoperative complications (19). In the current literature, no study reporting the absence of a relationship between PNI and postoperative complications has been found. Similarly, in our study, while Grade 1 (23.4% vs. 16.7%), Grade 2 (16.0% vs. 7.9%), Grade 3 (11.7% vs. 4.8%), and Grade 4 complication rates were higher in the low PNI group, the rate of patients without complications was higher in the high PNI group (44.7% vs. 69.0%). This suggests that well-nourished patients may be better able to tolerate the immunosuppression associated with circulating inflammatory cytokines that could be induced by post-operative complications (27). Therefore, when planning surgical treatment for gastric cancer patients, preoperative assessment of PNI value can be used as a valuable parameter in predicting the risk of postoperative complications and planning patient-specific approaches.

In our study, the PNI value in patients receiving neoadjuvant therapy is the PNI value calculated from blood taken before the start of neoadjuvant therapy. It may be thought that taking the PNI value before treatment in the group receiving neoadjuvant therapy and before operation in the group not receiving neoadjuvant therapy may harm homogeneity, but in patients who underwent surgery after neoadjuvant therapy, there was a relationship between the PNI value calculated before neoadjuvant chemotherapy and OS, while no relationship was found between the PNI value calculated in blood taken before surgery after neoadjuvant therapy and OS. The probable reason for this is the hematological toxicity of chemotherapeutic agents and bone marrow suppression. Therefore, it has been stated that it would be more appropriate to use the value taken before neoadjuvant therapy in order to use PNI as a prognostic indicator (28).

The clinical implications of our findings extend beyond mere prognostic assessment. Given that lower PNI values correlate with increased complications and reduced survival, the role of nutritional intervention becomes paramount. In the pre-neoadjuvant or preoperative phase, ensuring optimal nutritional status through daily nutritionist consultation and intervention is crucial. The availability of specialized immunonutrition formulations provides additional tools for optimizing patient outcomes. Our results reinforce the principle that improved nutritional status directly translates to better treatment outcomes, emphasizing the need for proactive nutritional management as an integral component of gastric cancer treatment protocols.

Future prospective studies should consider incorporating serial PNI measurements during the postoperative period to evaluate the dynamic changes in nutritional status and their impact on long-term outcomes. Such longitudinal assessment could provide valuable insights into the optimal timing and effectiveness of nutritional interventions.”

There are some limitations in our study. Primarily, our study is a retrospective study. Recurrence timing and disease-free survival analysis could not be performed because all data on patients’ adjuvant treatment protocols and durations and recurrence timing could not be fully accessed. However, in the survival analysis performed, only tumor-related deaths were referenced. The sensitivity and specificity of the detected cut-off value are not sufficient, but there is no study in the literature that finds a cut-off value with high sensitivity and specificity.

Due to sample size limitations and the retrospective nature of the study, propensity score matching analysis could not be performed to create matched cohorts based on PNI and tumor stage.

Despite these limitations, our study provides valuable information on 5-year OS, surgical complications, and nutritional status according to histological subtypes with PNI and contributes to the literature.

In conclusion, PNI is one of the parameters that can determine survival in gastric cancer. Major surgical complications are more common in patients with low PNI. It is observed that low PNI values are associated with SRCC, LVI, and PeNI, which have a worse prognosis. It is observed that PNI decreases directly proportional to tumor aggressiveness. However, it is evident that prospective studies with large patient populations are needed both to increase the reliability of this parameter and to find an optimal cut-off value. We believe that conducting prospective studies with large populations regarding parameters such as PNI and including these parameters in the evaluation is important for planning patient-specific treatment.