Abstract
Background/Aim: Soft-tissue tumors are difficult to differentiate as benign or malignant. Immune markers, such as the neutrophil–lymphocyte ratio (NLR), monocyte–lymphocyte ratio (MLR), platelet–lymphocyte ratio (PLR), and absolute lymphocyte count (ALC) in serum, have been reported to be useful in the diagnosis and predicting prognosis of several malignancies. We investigated the diagnostic value of these immune markers in differentiating soft-tissue tumors. Patients and Methods: A total of 692 patients who underwent biopsy or surgery of soft-tissue tumors were included and divided into benign tumor, low-grade malignancy, or high-grade malignancy groups. Immune markers were calculated from the preoperative blood tests and compared between the groups. A receiver operating curve (ROC) analysis was conducted between the benign disease group and a combination of the groups with malignancy to determine which immune marker had the most diagnostic value. Results: NLR and MLR were significantly different between the three groups with benign disease having the lowest value and high-grade malignancies the highest. Benign disease was also associated with lower PLR and higher ALC. There was no difference between the low- and high-grade malignancies in PLR and ALC. From the ROC analysis, NLR had the highest area under the curve (AUC) value of 0.773 out of the four markers. When limited to small tumors (≤30 mm), NLR had the highest AUC value of 0.729. Conclusion: The NLR showed the highest diagnostic value, although the diagnostic ability was not adequately high to differentiate benign and malignant soft-tissue tumors alone. NLR may serve as diagnostic support in combination with clinical history, physical findings, and tumor-imaging results.
Soft-tissue sarcomas are rare neoplasms with an estimated incidence rate of six per 100,000 persons per year, whereas benign soft-tissue tumors are relatively common. Therefore, most soft-tissue tumors observed in clinical practice are benign. Although a relatively rapid growth, tumor size >5 cm, and deep location generally correlate with a diagnosis of malignancy, there is a relatively high probability of cases being exceptions to this (1). Imaging procedures, such as computed tomography and magnetic resonance imaging, do not allow reliable differentiation between benign and malignant tumors except in very typical cases, such as lipoma (2). Currently, there are no useful serum markers for diagnosing soft-tissue sarcomas unlike for tumor markers for carcinomas in various organs.
Owing to the difficulty in differentiating between benign and malignant soft-tissue tumors, unplanned excision of soft-tissue sarcomas is common, with an incidence reported to be as high as 40% (3). Unplanned excision of soft-tissue sarcomas means patients undergo additional invasive surgeries leading to a worse functional prognosis (4). Furthermore, some reports have indicated worsening recurrence rates and progression-free survival due to such unplanned excisions (5, 6). A simple method to differentiate between benign and malignant soft-tissue tumors would provide a breakthrough to reduce the number of such cases, especially in general practice.
Studies focusing on the role of inflammation in carcinogenic progression are not new and Virchow hypothesized that the origin of cancer was at sites of chronic inflammation (7, 8). Recently, various factors in the tumor microenvironment are known to induce angiogenesis and recruit various immune cells including neutrophils and monocytes into the tumor niche which are transformed into tumor-associated neutrophils or tumor-associated monocytes that participate in tumorigenesis (9). Focusing on these immune cells, immune markers in serum, such as the ratios of neutrophils to lymphocytes, monocytes to lymphocytes, and platelets to lymphocytes, have been reported to be useful in the diagnosis and prognosis of various diseases, including cancer (10-12). In musculoskeletal oncology, some studies have reported that such immune markers may be considered prognostic markers for patients with soft-tissue sarcomas (13, 14). However, there is no evidence indicating that these markers are useful in identifying malignancies among soft-tissue tumors.
In the present study, we examined whether serum levels of various immune markers at the time of initial diagnosis were useful for differentiating benign from malignant soft-tissue tumors. Furthermore, we investigated which of these immune markers had the highest diagnostic value.
Patients and Methods
A retrospective, single-center study was conducted at our Institute. Patients who underwent a biopsy or surgery for musculoskeletal soft-tissue tumors at our Institute between January 2005 and December 2020 were included (n=778). Patients meeting any of the following criteria were excluded (n=86): autoimmune disorders, such as rheumatoid arthritis; other known malignancies; receiving glucocorticoid medication; and those who underwent unplanned excision outside our Institute. A total of 692 patients (309 males, 383 females; mean age=56.6 years) were included in the study.
The patients were divided into three groups: A benign tumor group, including patients with benign tumors according to the WHO Classification of Tumours (15); a low-grade malignancy group, including patients with intermediate tumors according to the WHO classification and malignant tumors which were histologically grade 1 in the French Fédération Nationale des Centres de Lutte Contre le Cancer (FNCLCC) grading system; and a high-grade malignancy group, including patients with malignant tumors which were histologically grade 2 or 3 in the FNCLCC grading system (16).
All data were retrospectively collected from the patients’ medical records. Patient demographics such as age, sex, medical history, and tumor characteristics (tumor histology, size, location, and depth) were collected as background data. Histological tumor diagnosis was made by a pathologist, in accordance with the WHO Classification of Tumours and the FNCLCC grading system. Tumor size was defined as the greatest length on preoperative or pre-biopsy magnetic resonance imaging images. Tumor depth was defined as ‘superficial’ when the tumor was located more superficially than the fascia. The patient demographics and tumor characteristics for all cases are given in Table I.
Background data for all patients and tumors.
Immune markers, the neutrophil–lymphocyte ratio (NLR), monocyte–lymphocyte ratio (MLR), platelet–lymphocyte ratio (PLR), and absolute lymphocyte count (ALC) were calculated from the results of the blood tests conducted during the initial visit. Blood counts were measured by an automated hematology analyzer (XN-3000; Sysmex Co., Kobe, Japan) automatically. The NLR, MLR, and PLR were calculated as the absolute count (n/μl) of neutrophils, monocytes, and platelets, respectively, divided by the absolute lymphocyte count (n/μl). Each immune marker was evaluated and compared between the groups. Since a previous report has shown that small soft-tissue sarcomas are more likely to undergo unplanned excision, it can be extrapolated that it is more difficult to differentiate between benign and malignant small soft-tissue tumors (4). Therefore, we specifically conducted this comparison in patients with a small tumor size (≤30 mm) (n=298). Patient demographics and tumor characteristics in small tumor cases are shown in Table II.
Background data for patients with small tumor (≤30 mm).
Statistical analysis was performed using EZR (version 1.50; R version 3.6.3), a statistical software based on R and R commanders (17). Data are presented as means±standard deviations or medians (minimum–maximum) for continuous variables, when necessary, and in actual numbers and percentages for categorical variables. The immune markers were compared among the three groups. For continuous variables showing a normal distribution, a one-way analysis of variance was performed. For continuous variables showing a non-normal distribution, non-parametric multiple comparisons were performed using the Kruskal–Wallis test. Fisher’s exact test or chi-square test were used to compare categorical data. For comparisons between two groups, post-hoc Bonferroni’s multiple comparison adjustment was performed. To determine the immune marker with the highest diagnostic value, a receiver operating curve (ROC) analysis was conducted for the benign disease group and a combination of low- and high-grade malignancy groups and groups were compared using the area under the curve (AUC) value. The cut-off values were determined using Youden’s index. Differences were considered statistically significant at p<0.05.
This research was conducted in accordance with the guidelines of the Declaration of Helsinki, and was approved by the Institutional Review Board of the Authors’ affiliated institutions (approval number: 2022-169). Informed consent was obtained from all patients and other relevant persons regarding publication.
Results
All tumor cases. While comparing the backgrounds, the group with high-grade malignancy showed a higher male predominance (68.1%) than the groups with benign disease (40.7%) or low-grade malignancy (42.9%), with no substantial difference between the latter two groups. The group with benign disease consisted of younger patients than did the groups with low- and high-grade malignancy while the latter two showed no significant difference between them (55.2±15.7, 62.0±17.5 and 64.2±15.7 years, respectively). The group with benign disease also included a higher number of cases with superficial tumors (34.5%) than the groups with low-grade (14.3%) and high-grade (21.3%) malignancy, with no substantial difference between the latter two. Tumor size was largest in the group with low-grade malignancy and the smallest in group with benign disease, showing a significantly varied range for each group [benign disease: 35 (2-250) mm, low-grade malignancy: 100 (20-320) mm, and high-grade malignancy: 60 (10-400) mm] (Table I).
Table III shows the NLR significantly differed among the three groups. It was the highest in that with high-grade malignancy, being significantly higher than in low-grade malignancy and in benign disease. Similar results were observed for MLR. For PLR, the group with benign disease had significantly lower values than those with low- or high-grade malignancy, with similar values in the latter two groups. Similar results were observed for ALC, with low- and high-grade malignancy groups having similar values, and patients with benign disease having significantly higher values than those of the other groups.
Median values (range) of the immune markers for all cases in each group.
The results of the ROC analyses of all cases are shown in Figure 1. The cut-off values for NLR, MLR, PLR and ALC were 2.332, 0.197, 124.2 and 1,572, respectively, while the AUC values were 0.773, 0.704, 0.651 and 0.671, respectively. The sensitivity, specificity, positive likelihood ratios and negative likelihood ratios are shown in Table IV. From the AUC values, the NLR was found to have the highest diagnostic value among all markers.
Receiver operating curve analysis of all cases. Values inside the graphs represent the cut-off value (with sensitivity, and specificity). A: Neutrophil–lymphocyte ratio. B: Monocyte-lymphocyte ratio. C: Platelet-lymphocyte ratio. D: Absolute lymphocyte count.
Results of receiver operating characteristics analysis of marker utility for all cases.
Small tumor (≤30 mm) cases. In the analyses of cases with small tumors (≤30 mm), we found no significant difference in sex between the groups, although the group with high-grade malignancy tended to have a higher male predominance than the groups with benign disease or low-grade malignancy. There was no significant difference in age between the groups, although age tended to increase as the grade increased. There was no significant difference in tumor depth between the groups. Low-grade malignancies consisted of significantly larger tumors than in the group with benign tumors (Table II).
High-grade malignancies had the highest median value of NLR and the benign tumor group the lowest, and significance was observed only between these two groups. Regarding MLR, high-grade malignancies had the highest median value, although with no significant differences between groups. There was no significant difference between the groups in regard to PLR. Furthermore, there were no significant differences in ALC between the groups (Table V).
Median values of the immune markers for patients with small tumor (≤30 mm).
The results of the ROC analyses of small tumor cases are shown in Figure 2. The cut-off values for NLR, MLR, PLR, and ALC were 1.778, 0.212, 149.5 and 1,240, respectively, while the AUC values were 0.729, 0.592, 0.612 and 0.591, respectively. The sensitivity, specificity, positive likelihood ratios and negative likelihood ratios are shown in Table VI. Only the NLR had an AUC value of >0.7 and showed the highest diagnostic value among the four markers, similar to the study of all cases.
Receiver operating curve analysis of small tumor (≤30 mm) cases. Values inside the graphs represent the cut-off value (with sensitivity, and specificity). A: Neutrophil–lymphocyte ratio. B: Monocyte-lymphocyte ratio. C: Platelet-lymphocyte ratio. D: Absolute lymphocyte count.
Results of receiver operating characteristics analysis of marker utility for patients with small tumor (≤30 mm).
Discussion
The findings of the present study regarding the relationship between inflammation and malignancy are not new. It is well known that chronic inflammation plays an important role in neoplastic progression. Several types of malignancies arise from areas of inflammation (e.g., reflux esophagitis, esophageal cancer, chronic hepatitis, and hepatocellular carcinoma). Cancer promotes and sustains inflammation, leading to angiogenesis and the creation of a conducive environment for their growth (7, 9, 18).
Based on the results of these studies, several reports regarding immune markers and the prognosis of malignancies have been published. In a systematic review and meta-analysis of 15 studies comprising a total of 8,563 patients, Ethier et al. reported that an elevated NLR was associated with worse overall and disease-free survival in patients with estrogen- and human epidermal growth factor receptor 2-negative breast cancer (10). Regarding musculoskeletal tumors, Garcia-Ortega et al. reported that an NLR>3.5 was a reliable prognostic factor in 169 patients with synovial sarcoma (19). Liang et al. reported that the combination of NLR and PLR was an independent prognostic factor for overall and disease-free survival, while NLR or PLR alone was not associated with patients who had soft-tissue sarcoma (14). Sato et al. reported that pre-treatment NLR<3.0 was an independent factor predictive of better progression-free survival (20), and Araki et al. reported that the combination of high PLR and low neutrophil count was an independent indicator for metastasis in patients with osteosarcoma (21). On the other hand, Strong et al. reported that the NLR was not independently associated with overall survival in patients with localized soft-tissue sarcomas of the extremities, whereas NLR ≥4.5 correlated with other prognostic factors, including large tumor size and high histological grade, concluding that NLR remains a potential prognostic biomarker (22). In addition to studies using immune markers as pretreatment prognostic predictors, Kobayashi et al. reported that reduced NLR (<3.8) after pazopanib treatment was independently associated with prolonged progression-free and overall survival (23).
Although there are many reports on the use of immune markers as prognostic factors, reports on their usefulness as diagnostic markers for musculoskeletal tumors are rare. Yapar et al. reported that pretreatment NLR, PLR, and MLR had diagnostic usefulness in patients with osteosarcoma compared to that in healthy controls, with NLR showing the highest significance in terms of AUC (AUC=0.763) compared to that of other markers (24). They further reported that NLR and MLR had diagnostic utility in differentiating healthy individuals from patients with cartilage tumors (enchondroma or low-grade chondrosarcoma), although their results did not support the differentiation between patients with enchondroma and those with low-grade chondrosarcoma (25). To the best of our knowledge, in contrast to the many reports showing immune markers to have prognostic value in soft-tissue tumors, and the few reports that showing their diagnostic value in bone tumors, there are currently no reports regarding their diagnostic value in soft-tissue tumors. This is the first study to focus on the diagnostic value of NLR, MLR, PLR and ALC in patients with soft tissue tumors.
In the current study, we found some significant difference in the pre-treatment values of four different immune markers. Based on these results, we demonstrated that immune markers, particularly NLR, might be used to differentiate between benign and malignant soft-tissue tumors. The NLR was the only marker with an AUC value >0.7 for the ROC curve in cases overall, as well as in those with small tumors, indicating that NLR had the best diagnostic ability of the four immune markers evaluated in this study. However, the sensitivity, specificity, and AUC of the ROC curve for NLR were not adequately high (Table IV and Table VI). These results are insufficiently reliable to consider NLR as an independent diagnostic criterion. However, NLR may serve as diagnostic support when combined with clinical history, physical findings, and imaging findings.
Ariizumi et al. reported a study comparing serum test results, such as white blood cell count, absolute neutrophil count, absolute monocyte count, platelets and C-reactive protein, rather than immune markers, between patients with benign disease and those with malignant soft-tissue tumors. They noted significantly elevated levels of neutrophils and monocytes, and lower levels of lymphocytes in patients with malignant soft-tissue tumors, while platelet count did not differ significantly. They also concluded that neutrophil counts are useful prognostic indicators in patients with soft-tissue sarcomas (26). These results may support our findings that NLR had the best diagnostic value among the four immune markers evaluated in our study. Furthermore, previous reports have shown that NLR is a useful prognostic predictor of malignancy in patients with soft-tissue tumors, and these results suggest that the NLR is directly proportional to the grade of the soft-tissue tumor, indicative of NLR being a useful diagnostic marker. A meta-analysis of the cut-off value for NLR as a prognostic predictor in malignant soft-tissue tumors in previous studies reported variable results within a range of 1.0-5.0, with most values between 2.3 and 3.8 (27). The cut-off value for differentiating healthy individuals from patients with osteosarcoma or low-grade chondrosarcoma has been reported to be 2.45 or 2, respectively (24, 25). Similar results were observed in this study, and the cut-off value for differentiating malignancy in soft-tissue tumors was 2.33. Owing to the wide range of cut-off values for prognostic markers and the paucity of data on diagnostic markers, future large-scale studies are warranted to determine accurate cut-off values for NLR as a diagnostic and prognostic marker.
Currently, there is no clear explanation as to why NLR might have the highest diagnostic value in differentiating between benign and malignant soft-tissue tumors. We have made some hypotheses as to why NLR showed the highest diagnostic value. Firstly, neutrophils are reported to promote tumor progression and angiogenesis through the production of various cytokines, including tumor necrosis factor-α, interleukin-1, interleukin-6, and vascular endothelial growth factor. Furthermore, tumor-associated neutrophils may function as immune suppressors and have been shown to inhibit lymphocyte-mediated immune responses (9, 17, 28, 29). Secondly, lymphocytes, such as natural killer cells and activated-T cells, exhibit antitumor activity and protect the host from the emergence and progression of malignant neoplasms. They play a major role in cytotoxic cell death and cytokine production, thereby inhibiting tumor cell growth and migration (7, 30, 31). Thus, the NLR reflects both neutrophil-dependent tumorigenesis-promoting inflammation, and host cell-mediated immunity (18). When a malignant tumor is present, neutrophils can increase tumor growth with simultaneous lymphocyte suppression due to an increase in tumor-associated neutrophils. Therefore, we can hypothesize that the NLR increases in proportion to the malignancy of the tumor (tumor proliferative and migratory ability). Although it is unclear whether inflammation induces malignancy or occurs as a result of tumor growth, or both, as reported in many studies (7-9, 17), there is a strong association between inflammation and malignancy (7-9, 17). Reflecting this strong association, it has been reported that elevated inflammatory markers such as C-reactive protein and erythrocyte sedimentation rate are found in cases of malignancy (32, 33), supporting our findings of an elevated NLR in patients with malignant soft-tissue tumor.
This study had some limitations. Firstly, it was a retrospective study. Secondly, the sample size, especially of the malignant tumor groups, was not adequately high due to the rarity of the disease. Thirdly, although the tumors were grouped according to the histological grade of malignancy, the groups comprised a mixture of tumors with various histological types. Owing to the rarity of the disease, it was not possible to separately evaluate each histological type. Furthermore, some types of tumors that can be diagnosed easily through imaging and physical examination (e.g., lipoma) were also included in this study; hence, the patients included were not limited to those with difficult-to-diagnose tumors based on other findings. Finally, our analysis lacked data regarding healthy individuals; therefore, we were not able to compare the benign tumor group with a healthy control group. However, the absence of a healthy control group was considered acceptable because in actual clinical practice it is necessary to differentiate between benign disease and malignancy in patients who have already developed a soft-tissue tumor.
Conclusion
In conclusion, we found that immune marker values differed between malignant tumor groups and the benign tumor group, with NLR having the most significant difference. The NLR showed the greatest diagnostic value out of the four evaluated markers; however, the AUC value of the ROC curve was 0.7-0.8, which was not adequately high to allow differentiation between a benign or malignant tumor. Overall, we propose that NLR may serve as diagnostic support in combination with clinical history, physical findings and tumor imaging results.
Acknowledgements
The Authors express their gratitude to the cooperation provided by the members of the Department of Diagnostic Pathology and Department of Clinical Laboratory at our Institute.
Footnotes
Authors’ Contributions
Atsushi Mihara: Conceptualization; data curation; formal analysis; investigation; methodology; software; visualization; writing – original draft. Ryuta Iwanaga: Methodology; Writing –- review and editing. Kiminori Yukata: Methodology; supervision; validation; writing – review and editing. Kenzo Fujii: Validation; writing – review and editing. Keiichi Muramatsu: Supervision; writing – review and editing. Koichiro Ihara: Supervision; writing – review and editing. Takashi Sakai: Supervision; project administration; writing – review and editing.
Conflicts of Interest
The Authors declare that there are no conflicts of interest regarding this study.
Funding
There was no funding to conduct this study.
Data Availability
The data used to support the findings of this study are available at: https://doi.org/10.6084/m9.figshare.23036753.v1
- Received May 6, 2023.
- Revision received May 22, 2023.
- Accepted May 24, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
