Evaluation of Inflammatory Biomarkers in Parotid Tumors

Introduction

Salivary gland neoplasms account for approximately 4% of all head and neck neoplasms and 1% to 5% of head and neck cancers and frequently occur in the parotid gland [1–5]. These rare tumors consist of a heterogeneous group of malignant and benign entities with various histological subtypes. The histopathological classification frequently changes with the identification of new tumor subtypes [6,7]. Accurate diagnosis of salivary gland neoplasms is clinically crucial, as misdiagnosis can lead to inappropriate treatment strategies, ranging from overtreatment of benign conditions to undertreatment of malignancies, which can result in adverse patient outcomes. Thus, improving diagnostic accuracy is vital for optimizing patient care.

Clinical evaluation, imaging methods, and fine needle aspiration biopsy are the most common preliminary diagnostic methods prior to treatment. However, histopathological examination of the tumor provides the final diagnosis [8,9]. Differentiating between benign-malignant, low-grade, high-grade, and hematological diseases before surgery will prevent over- and under-treatment. The evaluation of inflammatory biomarkers in parotid tumors might be a helpful tool among all the efforts to increase the diagnostic accuracy before the histopathological examination and surgery [10,11].

In recent years, growing evidence has shown the significant role of inflammation in cancer development and prognosis. Several studies have revealed the use of inflammatory markers like the neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), systemic immune inflammation index (SII), and systemic inflammation response index (SIRI) in predicting prognosis and distinction between the benign and malignant nature of tumors [12–19]. These biomarkers are easily calculated using complete blood count parameters, making them accessible tools. However, there is a knowledge gap regarding their diagnostic utility in parotid gland tumors. While some studies suggest that inflammatory markers can enhance diagnostic accuracy, conflicting evidence exists, and the specific role of the inflammatory markers in differential diagnosis has yet to be fully elucidated [20].

The aim of this retrospective study is to evaluate the use of inflammatory markers in the differential diagnosis of benign and malignant parotid gland tumors. By addressing the current limitations in diagnostic techniques and exploring the potential of inflammatory biomarkers, we seek to contribute to closing the knowledge gap and provide a clearer understanding of their role in clinical practice. This focus aligns with the broader objective of improving preoperative diagnostic strategies to ensure appropriate treatment planning and better patient outcomes.

Material and Methods

STUDY DESIGN:

This is a retrospective cohort study, which was approved by the Institutional Clinical Research Ethics Committee. The study was conducted in compliance with the Helsinki Declaration. The study size was determined based on the availability of eligible cases during the study period.

PATIENTS:

Patients who underwent parotidectomy at a single center between 2019 and 2023 were included in the study. Patients with primary parotid tumors (either benign or malignant) were recruited if their preoperative complete blood count and postoperative pathology results were available in our data system. A total of 240 patients were initially identified, and those meeting the inclusion criteria were retained after the application of the exclusion criteria, which were patients with non-tumor pathological findings (eg, reactive lymph node, chronic inflammation), skin tumor invasion or metastatic parotid tumors, acute or chronic infectious disease, inflammatory, rheumatological diseases, hematological disorders, history of corticosteroid therapy or chronic renal insufficiency, and a history of malignant disease.

DATA COLLECTION:

The data for the patients were obtained from the computer-based hospital system between January 2024 and February 2024. Demographic information, pathological results, and tumor sizes were recorded. Postoperative pathology results were considered for the determination of tumor sizes by using standardized histopathological methods. In multifocal tumors, the sizes of all tumors were measured, and their combined size was documented. The data obtained from the patients were evaluated according to the histopathological diagnosis as benign and malignant tumors. Benign tumors were divided as pleomorphic adenoma, Warthin tumor, and others, and malignant tumors were low grade, high grade, and lymphomas. High-grade mucoepidermoid carcinoma, squamous cell carcinoma, salivary duct carcinoma, and adenocarcinoma (not otherwise specified) were considered as high grade, and other malignant pathological subtypes were considered as low grade [4].

INFLAMMATORY BIOMARKERS:

The absolute neutrophil, lymphocyte, monocyte, and platelet counts were obtained from the preoperative complete blood counts of each patient, which were taken approximately a week before the surgery. Quality control measures were implemented for complete blood count testing to ensure reliable results. NLR was defined as the absolute neutrophil count (N) divided by the absolute lymphocyte count (L), as NLR=N/L. PLR was calculated by dividing the absolute platelet count (P) by the absolute lymphocyte count, as PLR=P/L. SII was calculated by multiplying the absolute platelet count with the absolute neutrophil count and divided by the absolute lymphocyte count, as SII=P×N/L. The SIRI was calculated by multiplying the absolute neutrophil count with the monocyte count (M) and divided by the absolute lymphocyte count (L), as SIRI = N×M/L.

STATISTICAL ANALYSIS:

Collected data were analyzed using SPSS Statistics 26.0 (IBM Corp, Armonk, NY, USA). Descriptive statistics were presented as frequency and percentage (%) for categorical variables and mean, standard deviation, minimum, and maximum statistics for numerical variables. The Pearson chi-square test was used to evaluate the relationship between categorical variables. The distributions of the variables were investigated by performing visual and analytical methods (Kolmogorov-Simirnov and Shapiro-Wilk tests). Non-parametric tests were used for the comparison of mean values, due to a lack of normal distribution. Bonferroni correction was applied to control for multiple testing issues, particularly for pairwise comparisons. Receiver operating characteristic (ROC) analysis was performed, and the Youden index, sensivity, and specifity were considered to identify the optimal cutoff values for NLR, PLR, SII, and SIRI. The area under the ROC curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated to assess their ability to differentiate Warthin tumors from pleomorphic adenomas and malignat lesions from benign ones. The Spearman correlation analysis was performed to determine the correlation between the tumor size and the delinated data. P values less than or equal to 0.05 were considered statistically significant.

Results

A total of 206 patients were included in this study, of which 91 (44.17%) were men and 115 (55.83%) were women. The mean age of the patients was 48.75 years (range: 12–90, SD: 15.39). Patients were divided according to histopathological evaluation. Benign tumors were diagnosed in 177 (85.92%) patients, predominantly pleomorphic adenoma (47.09%) and Warthin tumor (30.1%). Multifocal occurrence detected in 6 (9.68%) out of 62 patients with Warthin tumor. Malignant tumors constituted 14.08% of all tumors; 13 were low grade (44.83%), 12 were high grade (41.38%), and 4 (13.79%) were lymphomas. Detailed histopathological subtypes of the tumors are given in Table 1.

While Warthin tumors were most common in men (75.81% vs 24.19%), low-grade malignant tumors were more common in women (76.92% vs 23.08%). While there was no significant difference in terms of sex between malignant and benign tumor groups, a statistically significant difference was found between benign, low-grade, and high-grade tumor groups (chi-square test, P=0.091 and P=0.048, respectively).

The mean age was higher in patients with Warthin and high-grade malignant tumors. There was no significant difference between the ages of patients with malignant and benign tumors (Mann-Whitney U, P>0.05). A statistically significant difference was found in age between all histological subgroups (Kruskal-Wallis, P<0.05). Demographic features according to the histopathological tumor subtypes are given in Table 2.

The NLR, PLR, SII, and SIRI values in different tumor groups are summarized in Tables 3–5. In all patients with malignant tumors, NLR, PLR, SII, and SIRI values (mean±SD) were 2.58±0.78, 136.11±45.09, 772.69±793.20, and 1.49±2.66, respectively; while, in patients with benign tumors, the NLR, PLR, SII, and SIRI values were 2.24±0.96, 119.64±40.81, 562.34±250.64, and 1.06±0.62, respectively. No statistically significant difference was found between the patients with malignant and benign tumors (Table 3). When malignant tumors were divided as low-grade and high-grade subtypes (excluding lymphomas), NLR values (mean±SD) were 2.28±1.78 and 3.07±2.61; PLR values (mean±SD) were 121.19±37.04 and 155.12±52.04; SII values (mean±SD) were 577.91±190.24 and 1092.70±1167.25; and SIRI values (mean±SD) were 1.05±0.60 and 2.18±4.09, respectively.

Pleomorphic adenoma and Warthin tumor constituted the 2 major subgroups of benign tumors. In patients with pleomorphic adenoma, the NLR, PLR, SII, and SIRI values (mean±SD) were 2.16±0.95, 122.66±38.78, 544.83±225.12, and 0.95±0.59, respectively; whereas in the Warthin tumor group, the NLR, PLR, SII, and SIRI values (mean±SD) were 2.43±1.01, 111.30±43.35, 584.35±281.07, and 1.31±0.66, respectively.

A significantly elevated PLR value (mean±SD; 155.12±52.04) was detected in high-grade malignant tumors, compared with that in benign tumors (mean±SD; 119.64±40.81) (P<0.05). Also, increased NLR, SII, and SIRI values were observed in malignant tumors, especially in high-grade tumors; however, the differences were not statistically significant (Table 4). Evaluation of the most common benign tumors, pleomorphic adenoma and Warthin tumor, revealed significant differences in PLR (mean±SD; 122.66±38.78 vs 111.30±43.35; P<0.02) and SIRI (0.95±0.59 vs 1.31±0.66; P<0.001). NLR showed a nearly significant difference (2.16±0.95 vs 2.43±1.01; P=0.051), whereas no difference was observed in SII value (Table 5).

Using NLR, PLR, SII, and SIRI values, we performed ROC analysis to differentiate Warthin tumors from pleomorphic adenomas and malignant tumors from benign ones. The cutoff and AUC values, along with the diagnostic performance metrics, such as accuracy, PPV, NPV, and other characteristics, are summarized in Tables 6 and 7.

Tumor size ranged between 0.5 and 9 cm, with a mean of 2.03±1.44 cm. Warthin tumors (3.53±1.72 cm) and high-grade malignant tumors (3.34±1.78 cm) were the largest (Table 1). No significant difference was found in tumor size between benign and malignant tumors (Mann-Whitney U test, P=0.46). Moreover, the tumor size was not correlated with NLR and SII (Spearman test, P>0.05); however, a weak correlation was detected in PLR and SIRI values (Spearmen test r=−0.150 and r=0.175 respectively, P<0.05). These correlations suggest potential, albeit limited, clinical implications of biomarker levels in relation to tumor size.

Discussion

LIMITATIONS OF THE STUDY:

This study is a retrospective single center study and has a small sample size of malignant tumors. The data were evaluated within a limited period of time. Also, many clinical factors could influence inflammatory markers, beyond the primary disease we studied. Therefore, there is a possibility that the results can be affected in people who are prone to subclinical infections or inflammation. Longitudinal larger cohort studies that assess inflammatory markers at multiple time points before and after tumor resection could provide a more accurate picture of their role in diagnosis and prognosis.

Conclusions

In this study, higher NLR, PLR, SII, and SIRI values were detected in malignant parotid tumors than in benign tumors, although these differences are not statistically significant. While the levels of inflammatory parameters demonstrated varying patterns among high-grade, low-grade, and benign tumors, elevated PLR and SII values were detected as the most striking parameters, either found to be statistically significant (P=0.05) or close to significant (P=0.07), respectively. Also, PLR and SIRI values (P=0.02 and P<0.001) were significantly different between Warthin tumors and pleomorphic adenomas. Therefore, examination of these peripheral blood inflammatory markers could be a valuable tool for the assessment of parotid tumors prior to surgery. It is also clear that their cost-effectiveness and simplicity would generate an additional impact on their usage. On the other hand, these inflammatory markers might be influenced by various factors, potentially compromising their diagnostic accuracy.

In conclusion, evaluation of inflammatory biomarkers, such as NLR, PLR, SII, and SIRI, shows promise in diagnosing parotid tumors, particularly in distinguishing Warthin tumor and high-grade malignancies. PLR and SIRI values were significant in these tumor subtypes, indicating their potential utility alongside existing diagnostic methods. Further prospective studies with larger sample sizes are warranted to validate these findings and determine the optimal clinical use of inflammatory biomarkers in managing parotid gland tumors.