NETosis in pulmonary pleomorphic carcinoma

Abstract

Pulmonary pleomorphic carcinoma (PC) is a rare non‐small‐cell lung carcinoma (NSCLC) with a poor prognosis, characterized by tumor necrosis (TN). NETosis is a form of neutrophil‐specific cell death, which is morphologically characterized by prominent neutrophil infiltration and cell detritus in the necrotic foci. Seventy‐six patients with pulmonary PC who underwent complete resection were enrolled. Tumor necrosis was evaluated using digitally scanned resected specimens. The regions of NETosis were quantified using citrullinated histone H3 (citH3)‐ and myeloperoxidase‐positive regions. We examined the association between the NETosis area and the prognostic outcomes and assessed the correlation between the NETosis area and systemic inflammation. Tumor necrosis was observed in 70 patients (92%). In all the cases, the TN region was accompanied by a citH3‐positive region. The patients with high NETosis area (n = 54) had significantly shorter overall survival than those with low NETosis area (n = 16) (p = 0.013). Furthermore, a high NETosis area was an independent poor prognostic factor in the multivariate analyses. Systemic inflammatory markers, including C‐reactive protein (CRP), CRP‐to‐albumin ratio, and neutrophil‐to‐lymphocyte ratio, were significantly higher in patients with high NETosis area than in those with low NETosis area. Furthermore, the levels of these inflammatory markers were significantly decreased postsurgery. This study shows that in surgically resected pulmonary PC, patients with high NETosis areas have higher systemic inflammation and worse prognosis.

NETosis was observed in 92% of cases of pulmonary pleomorphic carcinoma (PC). The high NETosis area was associated with higher systemic inflammation and worse prognosis in PC.

Abbreviations

citH3

citrullinated histone H3

CRP

C‐reactive protein

CRP/Alb

CRP‐to‐albumin ratio

DN

dirty necrosis

EMT

epithelial–mesenchymal transition

MPO

myeloperoxidase

NET

neutrophil extracellular trap

NLR

neutrophil‐to‐lymphocyte ratio

NR

not reached

NSCLC

non‐small‐cell lung carcinoma

OS

overall survival

PC

pleomorphic carcinoma

RCC

renal cell carcinoma

RFS

relapse‐free survival

ROC

receiver operating characteristic

TN

tumor necrosis

1. INTRODUCTION

Pulmonary pleomorphic carcinoma is a rare NSCLC that accounts for 2%–3% of all NSCLCs in surgical series. ¹ , ² , ³ Pulmonary pleomorphic carcinomas comprise ≥10% of spindle cells and/or giant cells mixed with adenocarcinoma, squamous cell carcinoma, and large cell carcinoma. ⁴ , ⁵ , ⁶ Clinically, PC is characterized by an aggressive behavior, causing a poor prognosis. ⁷ , ⁸ , ⁹

Tumor necrosis is caused by chronic ischemia within tumors due to vascular collapse, high interstitial pressure, and/or rapid tumor growth that exceeds the blood supply. ¹⁰ Furthermore, presence of TN is reportedly associated with a poor prognosis in lung cancer. In stage I lung adenocarcinoma, presence of TN is associated with an increased risk of metastasis. ¹¹ , ¹² , ¹³ In pulmonary PC, massive TN (necrosis/tumors >25%) has been reported as an independent prognostic factor for disease‐free survival as well as OS. ³

In patients with TN, DN is currently attracting a lot of attention. Dirty necrosis is characterized by neutrophil infiltration and cell detritus ¹⁴ ; studies report that DN is an indicator of poor prognosis in RCC and metastatic lung cancer of the colon and rectum. ¹⁵ , ¹⁶ NETosis is a form of a neutrophil‐specific cell death associated with protection against infection. ¹⁷ , ¹⁸ During this process, activated neutrophils shed their DNA and intracellular contents, such as MPO, a web‐like structure that forms NETs. Recently, many researchers have highlighted the role of NETs in tumor progression and metastasis, ¹⁹ , ²⁰ , ²¹ and the role of NETosis is gaining attention in the field of oncology. Additionally, we recently reported a close association between DN and NETosis in RCC and metastatic lung cancer of the colon and rectum. ¹⁵ , ¹⁶ , ²² Furthermore, DN and NETosis are also associated with systemic inflammation. ¹⁵ , ²²

The purpose of this study was to evaluate the prognostic significance of NETosis in patients with pulmonary PC. We also examined the association between NETosis and systemic inflammation.

2. MATERIALS AND METHODS

2.1. Case selection

Between January 2002 and December 2020, a total of 109 consecutive patients with pulmonary PC who underwent complete surgical resection at the National Cancer Center Hospital East were enrolled in this study. Patients who underwent segmentectomy or wedge resection (n = 8), incomplete resection (n = 3), received preoperative chemotherapy or thoracic radiation (n = 5), clinical or pathological diagnosis of stage IV (n = 3) or synchronous multiple primary lung cancer (n = 9), died within 1 month after surgery (n = 3), or tumors comprising only spindle cells or giant cell carcinomas without an NSCLC component (n = 2) were excluded.

Therefore, we retrospectively reviewed 76 patients (Figure S1). All specimens were collected after comprehensive written informed consent had been obtained from the patients. This study was approved by the Institutional Review Board of the National Cancer Center (IRB number: 2021‐451) and was undertaken following the guidelines of the Declaration of Helsinki.

2.2. Clinicopathologic characteristics

For each patient, the clinicopathologic characteristics collected from medical records included age, sex, smoking status, preoperative chemotherapy history, tumor diameter, single/multiple metastases, OS, and RFS. The pathologic stage was determined based on the ninth edition of the UICC TNM classification.

2.3. Histopathological evaluation

All sectioned specimens were fixed with 10% formalin by infusion through the bronchial tree and were embedded in paraffin. The tumors were sliced into 5 mm thick slices and serial 4 μm sections were stained with H&E. Pathological features were evaluated by two pathologists (H.O. and G.I.). The H&E‐stained slides with the largest cross‐section of the tumor were selected. The slides were digitally scanned and analyzed using Aperio VERSA (Aperio ImageScope) to calculate the TN area (Figure S2). Tumor necrosis was defined as the presence of microscopic necrosis as described by Sengupta et al. ²³ Briefly, TN was histologically characterized by homogeneous clusters and sheets of dead cells. In addition, we classified TN into DN (+) and DN (−), as described previously ¹⁵ , ¹⁶ , ²² (Figure S3). Briefly, if TN with abundant inflammatory cells and cell detritus was observed in two or more different locations in the field view at 40× magnification, it was defined as DN (+) (Figure 1A). DN (−) was defined as a TN without DN features (Figure 1B). Patients with mixed DN (+) and DN (−) were defined as DN (+).

FIGURE 1.

Dirty necrosis (DN) (+) and DN (−) tumor necrosis (TN), and area of TN in pulmonary pleomorphic carcinoma (PC). (A) DN (+) with abundant neutrophils and cell detritus in pulmonary PC. Scale bars: 500 μm (left), 250 μm (middle), 50 μm (right). (B) DN (−) with only tumor cells showing coagulation necrosis in pulmonary PC. Scale bars: 500 μm (left), 125 μm (middle), 50 μm (right). (C) Area of the TN in pulmonary PC. The x‐ and y‐axis represent the patient number and area of TN, respectively.

2.4. Immunohistochemistry

Tumor necrosis samples were stained with an anti‐MPO Ab (A039829‐2; Dako) and an anti‐citH3 Ab (Ab5103; Abcam) to evaluate NETosis. Immunostaining was carried out according to the standard protocol of the Benchmark ULTRA Automated Staining System (Roche/Ventana Medical Systems).

2.5. Evaluation of immunohistochemistry

For citH3 and MPO, products showing positive images within the necrotic areas were identified, and their areas were measured and evaluated. A virtual slide system prepared using Aperio VERSA (Aperio ImageScope) was used to measure the area. If there are multiple areas of positive findings exist within the tumor, the areas are summed.

2.6. Statistical analysis

Associations between variables and the presence of TN were analyzed using Fisher's exact test (for categorical variables) and the Mann–Whitney U‐test (for continuous variables compared between the two groups). The correlation between TN and NETosis areas was evaluated using the Spearman rank correlation coefficient, and differences in pre‐ or postoperative laboratory findings were evaluated using a paired t‐test. The Kaplan–Meier method and the log‐rank test were used to assess the statistical significance of the differences in outcomes. The influence of clinicopathologic variables on OS and recurrence was assessed using the Cox proportional hazards model. All statistical analyses were undertaken using EZR (Saitama Medical Center, Jichi Medical University), a graphical user interface for R (The R Foundation for Statistical Computing), ²⁴ and figures of graphs and scatter plots were generated using GraphPad Prism version 9.3.1 (GraphPad Software). Statistical significance was set at p < 0.05.

3. RESULTS

3.1. Patient characteristics

Of the 76 pulmonary PC cases, male predominance (n = 66, 87%), a high percentage of smokers (n = 68, 89%), and high frequency of pleural invasion (n = 47, 62%) and vascular invasion (n = 67, 88%) were the clinicopathologic characteristics (Table S1), which was similar to those found in previous studies. ³ , ²⁵ The number of PCs with spindle cell and giant cell carcinomas was 58 and 18, respectively. Thirty‐eight cases had adenocarcinoma as an NSCLC component, 13 had squamous cell carcinoma, 24 had large cell carcinoma, and 1 had adenosquamous carcinoma.

3.2. Area of necrosis in PC

Representative images of DN (+) and DN (−) are shown in Figure 1A,B. The median tumor area was 722 mm² (range, 94–5717 mm²), and the TN area was 72 mm² (range, 0–2173 mm²) (Figure 1C). Tumor necrosis was not detected in six patients. As shown in Figure 2A,B, the regions morphologically recognized as DN were almost identical to the citH3‐positive regions, indicating that DN was correlated with NETosis in lung PC. ²² In the remaining 70 patients, the median value of citH3‐positive area, which was recognized as a NETosis marker, was 20.0 mm² (range, 0.03–1515 mm²) (Figure 2A–C), and the median value of MPO‐positive area was 58 mm² (range, 0.20–2073 mm²) (Figure 2A,B,D). The citH3‐positive area and TN areas correlated with each other in all patients (r = 0.855) (Figure S4).

FIGURE 2.

Immunohistochemical analysis of citrullinated histone H3 (citH3) and myeloperoxidase (MPO), and area of citH3 and MPO. (A) Immunohistochemical expression of citH3 and MPO in NETosis (+). Scale bars, 250 μm. (B) Immunohistochemical expression of citH3 and MPO in NETosis (−). Scale bars, 250 μm. (C) Area of citH3 in pulmonary PC. The x‐ and y‐axis represent patient number and area of citH3, respectively. (D) Area of MPO in pulmonary PC. The x‐ and y‐axis represent patient number and area of MPO, respectively.

3.3. Overall survival and RFS based on TN area

The patients were divided into two groups according to the values determined by the ROC curve for OS. The patient characteristics in the TN area are shown in Table S2. The median follow‐up duration for this cohort was 73 months (range, 5.7–204 months). As shown in Figure 3A,B, OS and RFS were significantly worse in high TN area (≥685.0 mm², n = 11) than in low TN area (<685.0 mm², n = 59) (high TN area vs. low TN area; median OS,10.4 vs. NR, log‐rank p = 0.005; median RFS, 4.3 vs. NR, log‐rank p < 0.001). In the Cox hazard model analysis for OS, although the TN area was a significantly worse prognostic factor in the univariate analysis, it was not an independent prognostic factor in the multivariate analysis (Table S3).

FIGURE 3.

Overall survival (OS) and relapse‐free survival (RFS) curves for patients with tumor necrosis (TN) and NETosis. (A) OS curves for patients with TN. Black line represents the group with a low TN; red line represents the group with a high TN. (B) RFS curves for patients with TN. Black line represents the group with a low TN; red line represents the group with a high TN. (C) OS curves of patients with NETosis. Black line shows the group with low NETosis areas; red line shows the group with high NETosis areas. (D) RFS curves of patients with NETosis. Black line shows the group with low NETosis areas; red line shows the group with high NETosis areas.

3.4. Prognostic impact of NETosis area

All 70 patients with necrosis were classified as DN (+) according to the criteria (Figure S3). We also examined the NETosis area, which is the citH3‐positive area. Tumor necrosis patients were divided into two groups according to the value determined by the ROC curve for OS: high NETosis area, ≥2.6 mm²; or low NETosis area, <2.6 mm². Patient characteristics according to NETosis area are listed in Table 1. A large NETosis area (n = 54) was significantly associated with large tumor size, stage progression, and vascular invasion (<0.001, 0.017, and 0.013, respectively). A high NETosis area was associated with significantly worse OS and RFS than a low NETosis area (n = 16) (high NETosis area vs. low NETosis area; median OS, 81.7 vs. NR, log‐rank p = 0.013; median RFS, 27.0 vs. NR, log‐rank p = 0.017) (Figure 3C,D). In the univariate and multivariate analyses, the NETosis area was an independent worse prognostic factor for OS (Table 2). Furthermore, we also examined whether NETosis area (citH3⁺ area) and MPO⁺ area divided by tumor area are also prognostic factors (Figure S5). The NETosis area (citH3⁺ area)/tumor area was a significant prognostic factor in both univariate and multivariate analyses for continuous variables (Table S4). However, MPO⁺/tumor area was not significant (data not shown).

TABLE 1.

Association of NETosis area with clinicopathologic features in patients with pulmonary pleomorphic carcinoma (n = 70)

Features	Low NETosis area, n = 16	High NETosis area, n = 54	p‐value
Age, years; median (range)	62.5 (47–78)	68 (43–84)	0.109
Sex
Male	14 (87.5)	47 (87)	0.999
Female	2 (12.5)	7 (13)
Smoking status
Never	1 (6)	7 (13)	0.672
Current/former	15 (94)	47 (77)
Tumor size, cm; median (range)	2.5 (1.8–8.2)	5.0 (1.9–10.9)	<0.001
p‐T
1	5 (3)	7 (13)	0.113
2	7 (44)	16 (30)
3	2 (13)	20 (37)
4	2 (13)	11 (10)
p‐N
0	13 (81)	30 (56)	0.074
1	1 (6)	18 (34)
2	2 (13)	6 (10)
p‐Stage
1	9 (56)	10 (18)	0.017
2	3 (19)	23 (43)
3	4 (25)	21 (39)
NSCLC component
Adenocarcinoma	8 (50)	30 (56)	0.196
Squamous cell carcinoma	5 (31)	7 (13)
Large cell carcinoma	3 (19)	17 (31)
Adjuvant chemotherapy	3 (19)	12 (22)	0.999
Lymphatic invasion	0 (0)	8 (15)	0.184
Vascular invasion	11 (69)	51 (94)	0.013
Pleural invasion	8 (50)	36 (67)	0.251

Note: Data are shown as n (%), unless otherwise indicated.

Abbreviation: NSCLC, non‐small‐cell lung carcinoma.

TABLE 2.

Univariable and multivariable Cox hazard model analyses for overall survival in patients with pulmonary pleomorphic carcinoma (n = 70)

Features		Univariable			Multivariable
		HR	95% CI	p‐value	HR	95% CI	p‐value
Age (per +1 year old)		1.04	0.99–5.76	0.104	–	–	–
Sex	Female	Ref.	0.45–8.13	0.376	–	–	–
	Male	1.92
Smoking status	Never	Ref.	0.48–8.54	0.342	–	–	–
	Current/former	2.02
p‐T (per +1)		1.61	1.08–2.40	0.018	1.11	0.60–2.03	0.743
p‐N	N0–1	Ref.	0.98–6.12	0.055	1.62	0.55–4.77	0.379
	N2	2.45
p‐Stage	I–II	Ref.	1.26–5.76	0.011	1.74	0.50–6.02	0.385
	III	2.69
Adjuvant chemotherapy	No	Ref.	0.36–2.23	0.814	–	–	–
	Yes	0.9
Lymphatic invasion	Absent	Ref.	0.55–4.74	0.378	–	–	–
	Present	1.62
Vascular invasion	Absent	Ref.	0.73–41.39	0.097	–	–	–
	Present	5.51
Pleural invasion	Absent	Ref.	0.82–4.61	0.132	–	–	–
	Present	1.94
NETosis area	Low	Ref.	1.22–21.97	0.026	4.46	1.02–19.54	0.047
	High	5.18

Abbreviations: CI, confidence interval; HR, hazard ratio; Ref., reference.

3.5. Comparison of NETosis area and laboratory findings related to systemic inflammation

Regarding systemic inflammatory findings, preoperative inflammatory values were significantly higher in the laboratory findings (CRP, CRP/Alb, NLR, and platelet‐to‐lymphocyte ratio), both in the median and in the ROC values of high NETosis area (Table 3). Furthermore, as shown in Figure S6, a comparison of pre‐ and postoperative laboratory values indicated that three values (CRP, CRP/Alb, and NLR) significantly decreased postoperatively in the high NETosis area (p = 0.018, 0.023, and 0.004, respectively).

TABLE 3.

Association of NETosis area with systemic inflammation in patients with pulmonary pleomorphic carcinoma (n = 70)

(A)
Preoperative examination (median, range)	Low NETosis area (ROC), n = 16	High NETosis area (ROC), n = 54	p value
CRP (mg/dL)	0.12, 0.02–12.79	1.4, 0.02–23.88	0.002
CRP/albumin ratio	22, 5–4740	327, 5–11,940	0.001
Neutrophil/lymphocyte ratio	0.378, 0.214–1.043	0.272, 0.058–1.073	0.009
Platelet/lymphocyte ratio	147.2, 85.3–194.2	200, 27.6–544.2	0.019

(B)
Preoperative examination (median, range)	Low NETosis area (median), n = 35	High NETosis area (median), n = 35	p value
CRP (mg/dL)	0.28, 0.02–23.88	2.16, 0.07–21.25	0.003
CRP/albumin ratio	75, 5–11,940	587, 17–8500	0.003
Neutrophil/lymphocyte ratio	0.327, 0.079–1.073	0.254, 0.058–0.589	0.036
Platelet/lymphocyte ratio	167.2, 27.6–459.8	199.4, 85.5–544.2	0.121

Abbreviations: CRP, C‐reactive protein; ROC, receiver operating characteristic.

4. DISCUSSION

This is the first study to show that NETosis is frequently associated with lung PC. Moreover, we found an association between the prognostic impact and systemic inflammation in the high NETosis area in pulmonary PC. Patients with a high NETosis area had a significantly shorter OS than those with low NETosis area. Furthermore, a high NETosis area was an independent poor prognostic factor in the multivariate analyses. Systemic inflammatory markers were significantly higher in patients with high NETosis area than in those with a low NETosis area. This result is consistent with the results obtained for RCC. ²²

A NET is defined as the phenomenon of trapping pathogens by forming a mesh‐like structure. ¹⁷ In NETs, peptidyl arginine deiminase 4 (PAD4) is activated by calcium ion binding and converts histone arginine residues to citrulline residues. Thus, NETosis can be detected by screening for citH3. ²⁶ , ²⁷ The H&E images of DN (+) areas showed neutrophil infiltration and accumulation of cellular degradation products (Figure 1A) and were found to overlap with the citH3 and MPO areas (Figure 2A). Therefore, we focused on NETosis, which has been implicated in cancer progression and metastasis. ²⁸ Furthermore, patients with pulmonary PC and a larger NETosis area had a poorer prognosis. These findings are consistent with the results observed in other carcinomas such as RCC and metastatic lung cancer of the colon and rectum. ¹⁵ , ¹⁶ , ²² NETosis was clearly more frequent in pulmonary PC than in renal cancer or lung metastasis of colorectal cancer. This could be a characteristic feature of pulmonary PC. In summary, it should be suggested that the presence of NETosis is an indicator of poor prognosis for all organs.

In this study, a high NETosis area was associated with high levels of systemic inflammation, and a comparison of inflammatory markers before and after surgery showed that some markers were significantly lower after surgery in patients with a high NETosis area. Although few reports have investigated the effect of NETosis on systemic inflammation in cancer, some studies have suggested a relationship between NETosis and systemic inflammation, and that the inhibition of NET production also reduces systemic inflammation. ²² , ²⁹ , ³⁰ Despite the fact that our study does not directly prove the effect of NETosis on systemic inflammation, Kuroe et al. ²² and this study suggested this relationship by comparing laboratory findings related to systemic inflammation.

In the present study, all TNs in the PCs had DNs, that is, NETosis. The role of NETs in cancer is varied ³¹ ; in colorectal and head and neck cancers, they induce apoptosis as an antitumor factor, ³² , ³³ whereas in lung cancer, NETs play a protumor role and are involved in tumor proliferation and metastasis. ³⁴ Pandolfi et al. ³⁵ reported that NETosis also induces EMT; therefore, NETosis could consequently be more frequently observed in PCs with a sarcomatoid component that is closely associated with EMT. We then examined whether NETosis was found around the sarcomatous component or the NSCLC component, and whether there is a difference in the tumor cell component near NETosis between cases with high and low NETosis area. Of the 16 low NETosis cases, 10 had NSCLC components around the NETosis and 6 had spindle or giant cell components. Also, of the 54 high NETosis cases, 30 had NSCLC components around the NETosis and 24 had spindle or giant cell components. There was no significant difference in the tumor cell component around NETosis in the high and low NETosis area (Table S5). Therefore, we could not show the close association between NETosis and EMT in this study. However, Kuroe et al. reported that in RCC, the NETosis was significantly more frequent in RCC with sarcomatous differentiation. ²² It is possible that the biological characteristics of the cancer cells themselves that produce the sarcomatous component may be related to the development of NETosis. The underlying molecular mechanism of this issue would be considered in the future.

This study possesses certain limitations. First, as the current study did not include cases with advanced stages, the impact of NETosis in advanced stages is unclear. Second, this study was a single‐center, retrospective study; therefore, validation in larger cohorts with other institutions or a prospective study may be necessary.

In this study, NETosis is a unique subtype of cell death associated with systemic inflammation and a poor prognosis in surgically resected pulmonary PC. More meticulous observation is needed in cohorts with a poor prognosis and a large area of NETosis.

AUTHOR CONTRIBUTIONS

Hajime Oi: Conceptualization; data curation; formal analysis; visualization; writing – original draft. Tetsuro Taki: Conceptualization; investigation; methodology; project administration; resources; supervision. Takashi Kuroe: Investigation; resources. Naoya Sakamoto: Investigation; resources. Shingo Sakashita: Investigation; resources. Motohiro Kojima: Investigation; resources. Eri Sugiyama: Investigation; resources. Shigeki Umemura: Investigation; resources. Tetsuya Sakai: Investigation; resources. Hiroki Izumi: Investigation; resources. Yoshitaka Zenke: Investigation; resources. Shingo Matsumoto: Investigation; resources. Kiyotaka Yoh: Investigation; resources. Makoto Ishii: Investigation; resources. Masahiro Tsuboi: Investigation; resources. Koichi Goto: Investigation; resources. Genichiro Ishii: Conceptualization; investigation; methodology; project administration; resources; supervision.

FUNDING INFORMATION

This study was supported by Japan Society for the Promotion of Science (21H02931).

CONFLICT OF INTEREST STATEMENT

Dr. Genichiro Ishii and Dr. Koichi Goto are editorial board members of Cancer Science. The other authors have no conflicts of interest.

ETHICS STATEMENT

Approval of the research protocol by an institutional review board: The research was approved by the Institutional Review Board of the National Cancer Center (approval no. 2021‐451).

Informed consent: No personally identifiable information was included in the manuscript.

Registry and the registration no. of the study/trial: N/A.

Animal studies: N/A.

Supporting information

Appendix S1.

Oi H, Taki T, Kuroe T, et al. NETosis in pulmonary pleomorphic carcinoma. Cancer Sci. 2025;116:524‐532. doi: 10.1111/cas.16332