Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Prognostic implications of prostaglandin E-major urinary metabolite in resected non-small-cell lung cancer

Prognostic implications of prostaglandin E-major urinary metabolite in resected non-small-cell... OBJECTIVES: Cyclooxygenase-2-derived prostaglandin E2 (PGE2) is highly involved in the promotion of cancer progression. The end product of this pathway, PGE-major urinary metabolite (PGE-MUM), is a stable metabolite of PGE2 that can be assessed non-invasively and repeatedly in urine samples. The aim of this study was to assess the dynamic changes in perioperative PGE-MUM levels and their prog- nostic significance in non-small-cell lung cancer (NSCLC). METHODS: Between December 2012 and March 2017, 211 patients who underwent complete resection for NSCLC were analysed pro- spectively. PGE-MUM levels in 2 spot urine samples taken 1 or 2 days preoperatively and 3–6 weeks postoperatively were measured using a radioimmunoassay kit. RESULTS: Elevated preoperative PGE-MUM levels were associated with tumour size, pleural invasion and advanced stage. Multivariable analysis revealed that age, pleural invasion, lymph node metastasis and postoperative PGE-MUM levels were independent prognostic fac- tors. In matched pre- and postoperative urine samples obtained from patients who are eligible for adjuvant chemotherapy, an increase in V C The Author(s) 2023. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. THORACIC 2 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery PGE-MUM levels following resection was an independent prognostic factor (hazard ratio 3.017, P = 0.005). Adjuvant chemotherapy im- proved survival in patients with increased PGE-MUM levels after resection (5-year overall survival, 79.0 vs 50.4%, P = 0.027), whereas sur- vival benefit was not observed in those with decreased PGE-MUM levels (5-year overall survival, 82.1 vs 82.3%, P = 0.442). CONCLUSIONS: Increased preoperative PGE-MUM levels can reflect tumour progression and postoperative PGE-MUM levels are a prom- ising biomarker for survival after complete resection in patients with NSCLC. Perioperative changes in PGE-MUM levels may aid in deter- mining the optimal eligibility for adjuvant chemotherapy. Keywords: Prostaglandin E-major urinary metabolite • Non-small-cell lung cancer • Surgery • Biomarker • Adjuvant chemotherapy ABBREVIATIONS unclear. In the present study, we prospectively investigated the prognostic impact of perioperative PGE-MUM levels in patients AIC Akaike information criterion with resected non-small-cell lung cancer (NSCLC) and discussed BIC Bayesian information criterion the significance thereof as a clinical biomarker. CI Confidence interval COX Cyclooxygenase PATIENTS AND METHODS COPD Chronic obstructive pulmonary disease HR Hazard ratio Ethical statement MRD Minimal residual disease NSAIDs Non-steroidal anti-inflammatory drugs The ethics committees of Kitasato University (B12-119) and NSCLC Non-small-cell lung cancer Teikyo University (Teirin-19-243) approved this study, which was OS Overall survival conducted in accordance with the Declaration of Helsinki. PGE2 Prostaglandin E2 Informed consent was obtained from all participants. PGE-MUM Prostaglandin E-major urinary metabolite Collection of samples and data INTRODUCTION In total, 373 patients who underwent complete resection for pri- Lung cancer is the leading cause of cancer-related death world- mary NSCLC at Kitasato University Hospital, Japan, between December 2012 and March 2017 were prospectively enrolled in wide. Although active and passive smoking are the primary risk the present study. Patients with carcinoma in situ, non-curative re- factors for lung cancer, several occupational and environmental section, synchronous or metachronous (within 5 years) malignan- factors and pre-existing lung diseases such as chronic obstructive cies, clinically or pathologically confirmed distant metastasis or pulmonary disease (COPD) and idiopathic pulmonary fibrosis are pleural dissemination were excluded. Comorbidities were scored also involved [1]. Under these conditions, many environmental using Charlson comorbidity index [10]. All resected tumours were factors and cancer risk factors are associated with chronic inflam- histologically diagnosed using the WHO classification and staged mation. Inflammatory responses play integral roles at different in accordance with the seventh edition of the TNM classification stages of tumour development, including initiation, promotion, staging manual [11, 12]. Adjuvant chemotherapy was considered invasion and metastasis and also affect immune surveillance and for patients with pathologic stage IA (tumour size >2 cm), IB–IIIA responses to therapy [2]. NSCLC according to the Japanese guideline [13]. In inflammatory and tumourigenic signalling, prostaglandin E2 Two spot urine samples were obtained from patients with pri- (PGE2) is synthesized from arachidonic acid via a cyclooxygenase mary lung cancer in the morning at 1 or 2 days preoperatively (COX)-catalyzed reaction [3]. Since the lungs play an important and 3–6 weeks postoperatively. In patients who received adjuvant role in both the production and metabolism of PGE2, PGE2 levels chemotherapy, postoperative urine samples were obtained be- in clinical specimens are possible indicators of inflammation- fore the initiation of chemotherapy. All samples were centrifuged associated lung disorders [4]. Furthermore, COX-2-derived PGE2 at 1000g for 10 min, and the supernatants were stored at -20 C is reportedly the most significant prostaglandin involved in can- until analysis. cer progression [3, 5]. Although previous studies investigating the The prognostic impact of absolute pre- and postoperative association between COX-2/PGE2 activity in tumour tissue and PGE-MUM levels and perioperative changes of PGE-MUM levels tumour progression have been conducted, accurately measuring were investigated separately. Therefore, patients whose only pre- its expression in tissue samples has several limitations, such as its operative urine samples were available for PGE-MUM analysis rapid degradation, invasiveness of sampling and tumour hetero- were included in the present study. Among all cases, only geneity [6, 7]. patients whose matched pre- and postoperative PGE-MUM levels Prostaglandin E-major urinary metabolite (PGE-MUM) is a stable were available were included in the analysis of perioperative end product of PGE2 metabolism and can be measured in urine changes in PGE-MUM levels. samples [8]. Its measurement allows for non-invasive longitudinal evaluation and can reflect systemic COX-2/PGE2 activity. A recent Measurement of urinary prostaglandin E-major study demonstrated that PGE-MUM levels were significantly ele- vated in patients with lung adenocarcinoma and tended to be pos- urinary metabolite levels itively correlated with cancer progression [9]. However, dynamic changes in PGE-MUM during the perioperative period have not Each urine sample was tested individually using the PGE-MUM been evaluated to date, and their prognostic significance remains assay. A radioimmunoassay kit (Institute of Isotopes Co., Ltd, M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 3 Budapest, Hungary) was used to measure the PGE-MUM levels. Briefly, the synthesized level of bicyclic PGE-MUM was measured Table 1: Clinicopathologic characteristics of patients with using a competitive assay after alkaline treatment was performed, non-small-cell lung cancer (N = 211) followed by neutralization of the urine samples. PGE-MUM levels were normalized to the concentration of urinary creatinine Variables Value, n (%) (expressed as lg/g Cr) because PGE-MUM concentration Age (years) 67.4 [SD] 67.5 (9.7) depends on urinary volume. Sex Male 130 (62) Female 81 (38) Statistical analysis Smoking history Current or former 134 (64) Pre- and postoperative PGE-MUM levels were compared based Never 77 (36) Smoking index, mean 541.4 on clinicopathological characteristics. All continuous variables NSAIDs administration identified with abnormal distribution by the Shapiro–Wilk test Yes 178 (84) were analysed with Mann–Whitney U-test. Multivariable Cox re- No 33 (16) gression models were used to examine prognostic factors. Charlson comorbidity index Survival was calculated using the Kaplan–Meier method, and dif- 0 117 (55) 1 68 (32) ferences were determined by means of log-rank analysis. The > _2 26 (12) variables with a univariable P-value of <0.20 were included in COPD the multivariable model. Multicollinearity between variables in Yes 90 (43) the multivariable model was assessed using a variance inflation No 121 (57) IPF factor. The goodness of fit of the models was evaluated using the Yes 15 (7) Akaike information criterion (AIC) and the Bayesian information No 196 (93) criterion (BIC). The AIC and BIC are model selection criteria and Tumour size, mean (SD) 33.3 (22.5) calculated as follows: AIC = -2 log (L)+ 2k, BIC = -2 log (L)+ k log Histology Adenocarcinoma 166 (79) (n). In this formula, n is the number of samples, L is the likelihood Squamous cell carcinoma 27 (13) for the mode and k is the number of parameters in the model Large cell neuroendocrine carcinoma 8 (4) [14]. When variables are selected in multivariable analysis, the Others 10 (5) number of samples changes owing to missing values. Since the Pleural invasion Yes 66 (31) normally used AIC can only be used when the number of sam- No 145 (69) ples is the same, we statistically selected a model based on the Pathologic N status BIC that can be compared even if the number of samples is dif- N0 172 (82) ferent. Missing data ware handled by listwise deletion. A P-value N1 15 (7) N2 24 (11) of <0.05 was considered to indicate statistical significance. Pathologic stage Statistical analyses were performed using JMP Pro version 15.0 IA 91 (43) (SAS Institute, Cary, NC, USA). IB 50 (24) IIA 20 (10) IIB 24 (11) IIIA 26 (12) RESULTS Adjuvant chemotherapy Yes 66 (31) Patient characteristics No 145 (69) COPD: chronic obstructive pulmonary disease; IPF: idiopathic pulmonary The clinicopathological data of 369 consecutive patients with fibrosis; NSAIDs: non-steroidal anti-inflammatory drugs; SD: standard NSCLC included in the present study were retrieved from medi- deviation. cal records. Among them, 157 patients who were treated with laxatives were excluded because laxative administration induces PGE2 production [15]. Accordingly, the relevant spot urine PGE-MUM levels were measured in the remaining 211 patients Adjuvant chemotherapy was administered to 66 (31%) patients. with NSCLC (130 men and 81 women). The clinicopathological Chemotherapy regimens included platinum doublet in 41 characteristics of these 211 patients are presented in Table 1. patients and tegafur-uracil in 25. None of the patients presented with inflammatory bowel dis- eases, acute cardiovascular problems and severe renal dysfunc- Prostaglandin E-major urinary metabolite levels in tion in which urine PGE-MUM levels were positively affected. Of pre- and postoperative urinary samples the 211 patients, 90 (43%) and 15 (7%) had COPD and idiopathic pulmonary fibrosis, respectively. A total of 178 patients (84.3%) PGE-MUM levels were evaluated using spot urine samples received non-steroidal anti-inflammatory drugs (NSAIDs) during obtained pre-(n = 211) and postoperatively (n = 196). Fifteen a postoperative course. Adenocarcinoma was the most common patients had only preoperative urine samples available for lung cancer (79%), followed by squamous cell carcinoma (13%), PGE-MUM analysis. The average PGE-MUM levels were 19.8 and large-cell neuroendocrine carcinoma (4%) and others (5%). In to- 17.9 mg/g Cr in pre- and postoperative samples, respectively. tal, 141 (67%) patients had pathologic stage I disease (91 with These levels were elevated compared with those in healthy vol- stage IA and 50 with stage IB), while 70 had more advanced dis- ease (20, 24 and 26 with stages IIA, IIB and IIIA, respectively). unteers (15.4 mg/g Cr) [9]. The pre- and postoperative PGE-MUM THORACIC 4 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery levels in the clinical samples are summarized in Table 2. PGE-MUM as a prognostic factor, we compared the prognosis of Preoperative PGE-MUM levels were associated with sex, smoking patients with high PGE-MUM levels with that of patients with low history and the presence of COPD but not histological cancer PGE-MUM levels. The levels were divided based on mean values, type, NSAIDs administration and comorbidity score. Patients 19.6 mg/g Cr in preoperative samples and 17.9 mg/g Cr in postop- erative samples. Patients with high PGE-MUM levels before and with large tumours (> _33 mm, P = 0.002), tumour with pleural inva- after resection exhibited significantly worse overall survival (OS) sion (P = 0.0002) and advanced-stage disease (P = 0.0006) exhib- (5-year OS, 82.3 vs 69.8%, P = 0.018 and 5-year OS, 85.7 vs 66.7%, ited significantly increased preoperative PGE-MUM levels, P = 0.001, respectively) (Fig. 1A and B). This trend was observed suggesting that PGE-MUM before tumour resection reflects tu- regardless of receipt of NSAIDs (Supplementary Material, Fig. S1) mour burden and progression. However, absolute postoperative However, despite the close association with tumour progression PGE-MUM levels were not associated with any clinical or patho- and negative prognostic impact in the univariable analysis, pre- logical characteristics including NSAIDs administration. operative PGE-MUM was not an independent prognostic factor in the multivariable analysis for OS (Table 3). Notably, Prognostic effect of prostaglandin E-major urinary postoperatively, the PGE-MUM level remained a significant prog- metabolite in non-small-cell lung cancer nostic factor [hazard ratio (HR), 6.462; 95% confidence interval (CI): 1.284–32.49, P = 0.024] as did age (HR 3.083, 95% CI: The median follow-up duration following resection was 1.076–8.836, P = 0.036), pleural invasion (HR 2.269, 95% CI: 51.4 months. A total of 37 patients experienced recurrence at dis- 1.165–4.383, P = 0.015) and lymph node metastasis (HR 2.216, tant sites, while 38 experienced loco-regional recurrence. Among 95% CI: 1.101–4.459, P = 0.026), suggesting that preoperative the 211 patients, 47 died during the follow-up period, 26 of lung PGE-MUM could be a predictive marker for tumour progression cancer and 21 of other causes. To investigate the role of at the time of surgery but does not affect prognosis Table 2: Pre- and postoperative prostaglandin E-major urinary metabolite levels in patients with non-small-cell lung cancer Variables Before surgery After surgery N Average PGE-MUM P-Value N Average PGE-MUM P-Value (mg/g Cr) (mg/g Cr) Sex Male 130 22.2 0.006 120 18.2 0.62 Female 81 15.9 76 17.4 Smoking history Current or former 134 22.4 0.002 124 17.9 0.95 Never 77 15.3 72 18.0 NSAIDs administration Yes 178 19.5 0.55 167 18.1 0.65 No 33 21.3 29 17.0 Charlson comorbidity index 0 117 20.7 0.57 112 18.0 0.52 1 68 18.1 59 16.9 > _2 26 19.8 25 19.7 COPD Yes 90 23.1 0.01 83 19.3 0.15 No 121 17.3 113 16.9 IPF Yes 15 25.9 0.13 15 23.1 0.07 No 196 19.3 181 17.5 Histology Adenocarcinoma 166 19.1 0.23 155 17.9 0.94 Non-adenocarcinoma 45 22.4 41 18.0 Tumour size > _33 mm 76 24.2 0.002 70 18.4 0.60 <33 mm 135 17.3 126 17.6 Pleural invasion Yes 66 25.8 0.0002 61 18.3 0.75 No 145 17.0 135 17.7 Lymph node metastasis Yes 39 22.1 0.33 37 16.5 0.39 No 172 19.3 159 18.2 Pathologic stage Stage I 141 17.1 0.0002 131 18.3 0.78 Stage II 44 19.7 40 16.9 Stage III 26 28.4 25 17.5 COPD: chronic obstructive pulmonary disease; IPF: idiopathic pulmonary fibrosis; NSAIDs: non-steroidal anti-inflammatory drugs; PGE-MUM: prostaglandin E-major urinary metabolite. M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 5 Figure 1: (A) Overall survival curve according to preoperative PGE-MUM levels. (B) Overall survival curve according to postoperative PGE-MUM levels. PGE-MUM lev- els were divided based on mean values (19.6 and 17.9 mg/g Cr in pre- and postoperative samples, respectively). PGE-MUM: prostaglandin E-major urinary metabolite levels after resection was an independent prognostic factor (HR, 2.272; 95% CI: 1.038–4.766, P = 0.039) along with pleural invasion, Table 3: Multivariable Cox regression analysis of factors lymph node metastasis and adjuvant chemotherapy (Table 4). In influencing overall survival in resected non-small-cell lung patients with increased PGE-MUM levels after resection, adjuvant cancer chemotherapy significantly improved survival (5-year OS, 79.0 vs 50.4%, P = 0.027) (Fig. 3A). However, in patients with decreased Variable Hazard 95% confidence P-Value ratio interval PGE-MUM levels after resection, there were no significant differ- ences in OS between patients who received adjuvant chemother- Lower Upper apy and those who did not (5-year OS, 82.1 vs 82.3%, P = 0.442) Age (years) (> _65/<65) 3.083 1.076 8.836 0.036 (Fig. 3B). Sex (male/female) 1.608 0.771 3.352 0.205 Tumour size (> _33 mm/<33 mm) 1.399 0.720 2.719 0.321 Pleural invasion (present/absent) 2.269 1.165 4.383 0.015 LN metastasis (present/absent) 2.216 1.101 4.459 0.026 DISCUSSION Preoperative PGE-MUM (mg/g Cr) 1.820 0.135 24.47 0.651 Postoperative PGE-MUM (mg/g Cr) 6.462 1.284 32.49 0.024 In the present study, we demonstrated that elevated preoperative LN: lymph node; PGE-MUM: prostaglandin E-major urinary metabolite. PGE-MUM levels were associated with cancer advancement at the time of surgery and postoperative PGE-MUM levels were a promising biomarker for poor prognosis after complete resection independently, whereas postoperative PGE-MUM may be a in patients with NSCLC. These results may provide useful infor- prognostic marker in patients with resected NSCLC. mation for perioperative management including postoperative systemic therapy. Implications of perioperative changes in PGE-MUM is a urinary metabolite of PGE2 and represents the activity of the COX-2 pathway. COX-2 and COX-2-derived PGE2 prostaglandin E-major urinary metabolite levels are reportedly involved in the tumour initiation and proliferation and indication for adjuvant chemotherapy of cancer cells both in vitro and in vivo [5, 16]. Several studies per- formed in clinical settings have shown that COX-2 overexpres- Since high PGE-MUM levels after resection were associated with sion is associated with poor prognosis in solid tumours, including poor prognosis in patients with NSCLC, we investigated the asso- NSCLC [6, 17]. Furthermore, based on the hypothesis that sup- ciation between pre- and postoperative PGE-MUM changes and pressing the COX-2 pathway could reduce cancer aggression and prognosis and assessed possible indication for adjuvant therapies improve survival, prospective clinical studies have been con- based on PGE-MUM changes. Matched pre- and postoperative ducted to examine the effects of COX-2 inhibitors combined PGE-MUM levels were available for 113 of 127 patients who with standard chemotherapies. However, no randomized con- were considered eligible for adjuvant chemotherapy according trolled trial has demonstrated a therapeutic advantage for these to the clinical guideline. Of the 113 patients, 60 received adjuvant agents [7, 18–21]. Considerable limitations discussed in these chemotherapy and 53 did not. Fifty patients experienced ele- studies included optimal patient selection based on clinical bio- vated PGE-MUM levels after resection relative to preoperative markers and the methods used to detect the markers. In a phase levels, with an average increase of 6.45 mg/g Cr. A reduction was III study, CALGB 30801, the enrolled patients were selected based observed in 63 patients, with an average reduction of 12.76 mg/g on COX-2 expression assessed using immunohistochemistry, and Cr. The OS of patients with increased PGE-MUM levels after re- the COX-2 inhibitor celecoxib failed to improve prognosis in section was significantly worse than that of patients with de- creased PGE-MUM levels (5-year OS, 81.6 vs 65.5%, P = 0.04) NSCLC with high COX-2 expression. However, a cohort of (Fig. 2). In multivariable analysis for OS, an increase in PGE-MUM patients who could benefit from celecoxib was identified in THORACIC 6 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery Figure 2: Overall survival curve according to the change in postoperative PGE-MUM levels compared with preoperative PGE-MUM levels. PGE-MUM: prostaglandin E-major urinary metabolite for lung cancer reportedly affects survival negatively [23, 24]. As postoperative PGE-MUM was an independent prognostic factor Table 4: Multivariable Cox regression analysis of factors in our study regardless of tumour progression, prolonged exces- influencing overall survival in resected non-small-cell lung sive inflammation after surgery might facilitate the survival and cancer with matched pre- and postoperative prostaglandin development of microscopic tumour cells, which could subse- E-major urinary metabolite examination quently lead to tumour recurrence and poor prognosis. Another interpretation is the role of PGE-MUM as a biomarker for resid- Variable Hazard 95% confidence P-Value ratio interval ual tumour cells. Since preoperative PGE-MUM reflects tumour burden and high postoperative PGE-MUM could be associated Lower Upper with poor prognosis, the lack of reduction in PGE-MUM after Age (years) (> _65/<65) 3.672 1.076 12.534 0.038 surgery is assumed to represent the presence of microscopic re- Sex (male/female) 1.759 0.669 4.626 0.253 sidual tumour cells. Although inflammation can promote tumour Tumour size (mm) (> _33/<33) 1.605 0.738 3.489 0.232 Pleural invasion (present/absent) 3.453 1.532 7.782 0.003 progression, tumours can also induce inflammation by secreting LN metastasis (present/absent) 2.865 1.317 6.229 0.008 inflammatory cytokines into their microenvironment to maintain Increase of PGE-MUM after 2.272 1.038 4.766 0.039 favourable conditions for their growth. PGE2 can reportedly be resection (yes/no) secreted from tumour cells, thereby promoting immune evasion Adjuvant therapy (no/yes) 2.400 1.041 5.533 0.040 [25]. Thus, high postoperative PGE-MUM levels and subsequent LN: lymph node; PGE-MUM: prostaglandin E-major urinary metabolite. poor prognosis may be a consequence of MRD with activation of the COX-2/PGE2 pathway. With the recent exploration of biomarkers and development subset analysis using a urinary PGE2 metabolite. Correlation and of drugs for unresectable lung cancers, greater attention has agreement were poor between urinary PGE2 metabolite and been paid to perioperative treatment tailored based on bio- COX-2 staining by immunohistochemistry, and prognosis in markers. Although indications for adjuvant therapies are primar- NSCLC with elevated urinary PGE metabolite was improved by ily determined based on pathological stage, even tumours with celecoxib treatment [7]. Given the advantage of urinary samples, the same advanced stage and histologic type represent a highly which enable repeated real-time assessment and do not require heterogeneous population. This includes patients who do not consideration of tumour heterogeneity in biomarker expression, truly require adjuvant therapy. Recent studies attempting to PGE-MUM could be promising for evaluating the activity of the identify tumours with a high risk for postoperative recurrence COX-2 pathway and reflecting the exact biological features of and select optimal indications for adjuvant therapies using devel- tumours. oped detection methods for MRD after surgical resection have Despite evidence on the association between COX-2-derived been conducted [26–28]. Based on the presumption that sus- PGE2 and prognosis in malignant tumours, the dynamics and tained elevated PGE-MUM levels after resection represent the prognostic significance of systemic COX-2 activity during the presence of MRD, we investigated the association between pre- perioperative period remain unclear. The present study demon- and postoperative PGE-MUM changes and prognosis to assess strated that high PGE-MUM levels after resection were associated possible indication for adjuvant therapies based on PGE-MUM with poor prognosis in patients with NSCLC. This result could be levels. Multivariable analysis revealed that an increase in underpinned by 2 possible implications: the cancer-promoting PGE-MUM levels after resection remained a significant prognos- effect of inflammation and the presence of microscopic residual tic factor. Furthermore, adjuvant chemotherapy improved OS in tumour cells, called minimal residual disease (MRD). patients with increased PGE-MUM levels after resection, while Inflammation has several tumour-promoting effects, including the survival benefit from adjuvant chemotherapy was not ob- proliferation, angiogenesis, invasiveness and metastasis [22]. The served in patients with decreased PGE-MUM levels after resec- inflammatory burden in patients undergoing curative resection tion. These results may support a role of perioperative changes in M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 7 Figure 3: (A) Overall survival in patients with increased PGE-MUM levels after resection who received adjuvant chemotherapy and those who did not. (B) Overall survival in patients with decreased PGE-MUM levels after resection who received adjuvant chemotherapy and those who did not. PGE-MUM: prostaglandin E-major urinary metabolite. PGE-MUM levels as a clinical biomarker representing the pres- ACKNOWLEDGMENTS ence of MRD and provide useful information for determining the optimal eligibility for adjuvant therapy to improve the survival of We are extremely grateful to Professor Mutsunori Fujiwara (1947– patients with NSCLC. 2021), Department of Pathology, Nissan Tamagawa Hospital, Tokyo, Japan, for his excellent advice regarding this study. We also greatly appreciate Fujirebio Inc., for the assistance in measurement of uri- Limitations nary PGE-MUM levels. We thank Editage (www.editage.com)for the English language editing. The present study had several limitations. First, the investigation was conducted at a single centre with a limited number of en- Funding rolled patients. Second, PGE-MUM levels may be affected by sev- eral inflammatory conditions. Our cohort did not include The authors did not receive any specific funding for this work. patients with inflammatory bowel disease, which is the most well-known disease affecting PGE-MUM. However, the possibility Conflict of interest: none declared. that other inflammatory factors may have influenced PGE-MUM levels cannot be ruled out. Third, PGE-MUM levels may be af- Data availability fected by several drugs. Patients who were administered laxatives were treated were excluded from our study. However, in clinical The data underlying this article will be shared on reasonable request practice, patients often need to be administered laxatives and to the corresponding author. other drugs that may affect PGE-MUM during the postoperative course. In fact, most patients (84.3%) received NSAIDs during Author contributions perioperative period in the present study. Although NSAIDs administration was not significantly associated with pre- and Masashi Mikubo: Conceptualization; Data curation; Formal analy- postoperative PGE-MUM levels in our cohorts, that might be be- sis; Investigation; Writing—original draft. Yukitoshi Satoh: cause of the small number of patients who did not receive Conceptualization; Data curation. Mototsugu Ono: Data curation. NSAIDs. Thus, further investigation that considers the effects of Dai Sonoda: Data curation. Shoko Hayashi: Data curation. these drugs is needed to apply our results to clinical practice. Masahito Naito: Data curation. Yoshio Matsui: Data curation. Kazu Shiomi: Data curation. Masaaki Matsuura: Formal analysis; Investigation; Methodology; Writing—review & editing. Satoru Ito: CONCLUSION Supervision; Writing—review & editing. The results of the present study indicate that elevated preopera- Reviewer information tive PGE-MUM levels are associated with tumour progression and that postoperative PGE-MUM levels are a promising bio- Interdisciplinary CardioVascular and Thoracic Surgery thanks David marker for survival in patients with resected NSCLC. Prolonged G. Healy, Niccolo Daddi and the other, anonymous reviewer(s) for elevation of PGE-MUM levels may aid in identifying patients who their contribution to the peer review process of this article. may benefit from adjuvant systemic therapy. REFERENCES SUPPLEMENTARY MATERIAL [1] Collins LG, Haines C, Perkel R, Enck RE. Lung cancer: diagnosis and man- Supplementary material is available at ICVTS online. agement. Am Fam Physician 2007;75:56–63. THORACIC 8 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery [2] Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and can- [16] Ruan D, So SP. Prostaglandin E2 produced by inducible COX-2 and cer. Cell 2010;140:883–99. mPGES-1 promoting cancer cell proliferation in vitro and in vivo. Life Sci [3] Aiad M, Tahir A, Fresco K, Prenatt Z, Ramos-Feliciano K, Walia J et al. 2014;116:43–50. Does the combined use of aspirin and immunotherapy result in better [17] Koki A, Khan NK, Woerner BM, Dannenberg AJ, Olson L, Seibert K et al. outcomes in non-small cell lung cancer than immunotherapy alone? Cyclooxygenase-2 in human pathological disease. Adv Exp Med Biol Cureus 2022;14:e25891. 2002;507:177–84. [4] Bozyk PD, Moore BB. Prostaglandin E2 and the pathogenesis of pulmo- [18] Edelman MJ, Watson D, Wang X, Morrison C, Kratzke RA, Jewell S et al. nary fibrosis. Am J Respir Cell Mol Biol 2011;45:445–52. Eicosanoid modulation in advanced lung cancer: cyclooxygenase-2 ex- [5] Elwakeel E, Bru ¨ ggemann M, Wagih J, Lityagina O, Elewa MAF, Han Y pression is a positive predictive factor for celecoxib + chemotherapy— et al Disruption of prostaglandin E2 signaling in cancer-associated fibro- Cancer and Leukemia Group B Trial 30203. J Clin Oncol 2008;26:848–55. blasts limits mammary carcinoma growth but promotes metastasis. [19] Groen HJ, Sietsma H, Vincent A, Hochstenbag MM, van Putten JW, van Cancer Res 2022;82:1380–95. den Berg A et al. Randomized, placebo-controlled phase III study of [6] Khuri FR, Wu H, Lee JJ, Kemp BL, Lotan R, Lippman SM et al. docetaxel plus carboplatin with celecoxib and cyclooxygenase-2 expres- Cyclooxygenase-2 overexpression is a marker of poor prognosis in stage sion as a biomarker for patients with advanced non-small-cell lung can- I non-small cell lung cancer. Clin Cancer Res 2001;7:861–7. cer: the NVALT-4 study. J Clin Oncol 2011;29:4320–6. [7] Edelman MJ, Wang X, Hodgson L, Cheney RT, Baggstrom MQ, Thomas [20] Gitlitz BJ, Bernstein E, Santos ES, Otterson GA, Milne G, Syto M et al. A SP et al.; Alliance for Clinical Trials in Oncology. Phase III randomized, randomized, placebo-controlled, multicenter, biomarker-selected, phase placebo-controlled, double-blind trial of celecoxib in addition to stan- 2 study of apricoxib in combination with erlotinib in patients with ad- dard chemotherapy for advanced non-small-cell lung cancer with vanced non-small-cell lung cancer. J Thorac Oncol 2014;9:577–82. cyclooxygenase-2 overexpression: CALGB 30801 (alliance). J Clin Oncol [21] Edelman MJ, Tan MT, Fidler MJ, Sanborn RE, Otterson G, Sequist LV et al 2017;35:2184–92. Randomized, double-blind, placebo-controlled, multicenter phase II [8] Wang D, DuBois RN. Urinary PGE-M: a promising cancer biomarker. study of the efficacy and safety of apricoxib in combination with either Cancer Prev Res (Phila) 2013;6:507–10. docetaxel or pemetrexed in patients with biomarker-selected non- [9] Kawamoto H, Hara H, Araya J, Ichikawa A, Fujita Y, Utsumi H et al. small-cell lung cancer. J Clin Oncol 2015;33:189–94. Prostaglandin E-major urinary metabolite (PGE-MUM) as a tumor [22] Borsig L, Wolf MJ, Roblek M, Lorentzen A, Heikenwalder M. Inflammatory marker for lung adenocarcinoma. Cancers 2019;11:768. chemokines and metastasis–tracing the accessory. Oncogene 2014;33: [10] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classi- 3217–24. fying prognostic comorbidity in longitudinal studies: development and [23] Templeton AJ, McNamara MG, Seruga B, Vera-Badillo FE, Aneja P, validation. J Chronic Dis 1987;40:373–83. Ocana A et al. Prognostic role of neutrophil-to-lymphocyte ratio in solid [11] Travis WD, Brambilla E, Burke AP, Marx A, AGeditors, N. WHO tumors: a systematic review and meta-analysis. J Natl Cancer Inst 2014; Classification of Tumours of the Lung, Pleura, Thymus and Heart. 4th 106:dju124. edn. Lyon, France: IARC Press, 2015. [24] Proctor MJ, Morrison DS, Talwar D, Balmer SM, O’Reilly DS, Foulis AK [12] Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D, Chansky et al An inflammation-based prognostic score (mGPS) predicts cancer K et al.; International Association for the Study of Lung Cancer survival independent of tumour site: a Glasgow Inflammation Outcome International Staging Committee and Participating Institutions. The Study. Br J Cancer 2011;104:726–34. International Association for the Study of Lung Cancer lung cancer stag- [25] Konishi Y, Ichise H, Watabe T, Oki C, Tsukiji S, Hamazaki Y et al. ing project: proposals regarding the clinical staging of small cell lung Intravital imaging identifies the VEGF-TXA(2) axis as a critical promoter cancer in the forthcoming (seventh) edition of the tumor, node, metas- of PGE(2) secretion from tumor cells and immune evasion. Cancer Res tasis classification for lung cancer. J Thorac Oncol 2007;2:1067–77. 2021;81:4124–32. [13] Japan Lung Cancer Society Guidelines for diagnosis and treatment of [26] Qiu B, Guo W, Zhang F, Lv F, Ji Y, Peng Y et al. Dynamic recurrence risk the lung cancer, 2021. https://www.haigan.gr.jp/guideline/2021/1/2/ and adjuvant chemotherapy benefit prediction by ctDNA in resected 210102040100.html#cq27 (29 December 2022, date last accessed). NSCLC. Nat Commun 2021;12:6770. [14] Hook EB, Regal RR. Validity of methods for model selection, weighting [27] Xia L, Mei J, Kang R, Deng S, Chen Y, Yang Y et al. Perioperative ctDNA- for model uncertainty, and small sample adjustment in capture- based molecular residual disease detection for non-small cell lung can- recapture estimation. Am J Epidemiol 1997;145:1138–44. cer: a prospective multicenter cohort study (LUNGCA-1). Clin Cancer [15] Fujiwara M, Okayasu I, Oritsu M, Komatsu J, Yoshitsugu M, Katoh Y et al. Res 2021;28:3308–3317. Significant increase in prostaglandin E-main urinary metabolite by laxa- [28] Gale D, Heider K, Ruiz-Valdepenas A, Hackinger S, Perry M, Marsico G tive administration: comparison with ulcerative colitis. Digestion 2000; et al Residual ctDNA after treatment predicts early relapse in patients 61:201–6. with early-stage non-small cell lung cancer. Ann Oncol 2022;33:500–10. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Interactive Cardiovascular and Thoracic Surgery Oxford University Press

Prognostic implications of prostaglandin E-major urinary metabolite in resected non-small-cell lung cancer

Loading next page...
 
/lp/oxford-university-press/prognostic-implications-of-prostaglandin-e-major-urinary-metabolite-in-uOa0EmKot9

References (36)

Publisher
Oxford University Press
Copyright
© The Author(s) 2023. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery.
eISSN
1569-9285
DOI
10.1093/icvts/ivac291
Publisher site
See Article on Publisher Site

Abstract

OBJECTIVES: Cyclooxygenase-2-derived prostaglandin E2 (PGE2) is highly involved in the promotion of cancer progression. The end product of this pathway, PGE-major urinary metabolite (PGE-MUM), is a stable metabolite of PGE2 that can be assessed non-invasively and repeatedly in urine samples. The aim of this study was to assess the dynamic changes in perioperative PGE-MUM levels and their prog- nostic significance in non-small-cell lung cancer (NSCLC). METHODS: Between December 2012 and March 2017, 211 patients who underwent complete resection for NSCLC were analysed pro- spectively. PGE-MUM levels in 2 spot urine samples taken 1 or 2 days preoperatively and 3–6 weeks postoperatively were measured using a radioimmunoassay kit. RESULTS: Elevated preoperative PGE-MUM levels were associated with tumour size, pleural invasion and advanced stage. Multivariable analysis revealed that age, pleural invasion, lymph node metastasis and postoperative PGE-MUM levels were independent prognostic fac- tors. In matched pre- and postoperative urine samples obtained from patients who are eligible for adjuvant chemotherapy, an increase in V C The Author(s) 2023. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. THORACIC 2 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery PGE-MUM levels following resection was an independent prognostic factor (hazard ratio 3.017, P = 0.005). Adjuvant chemotherapy im- proved survival in patients with increased PGE-MUM levels after resection (5-year overall survival, 79.0 vs 50.4%, P = 0.027), whereas sur- vival benefit was not observed in those with decreased PGE-MUM levels (5-year overall survival, 82.1 vs 82.3%, P = 0.442). CONCLUSIONS: Increased preoperative PGE-MUM levels can reflect tumour progression and postoperative PGE-MUM levels are a prom- ising biomarker for survival after complete resection in patients with NSCLC. Perioperative changes in PGE-MUM levels may aid in deter- mining the optimal eligibility for adjuvant chemotherapy. Keywords: Prostaglandin E-major urinary metabolite • Non-small-cell lung cancer • Surgery • Biomarker • Adjuvant chemotherapy ABBREVIATIONS unclear. In the present study, we prospectively investigated the prognostic impact of perioperative PGE-MUM levels in patients AIC Akaike information criterion with resected non-small-cell lung cancer (NSCLC) and discussed BIC Bayesian information criterion the significance thereof as a clinical biomarker. CI Confidence interval COX Cyclooxygenase PATIENTS AND METHODS COPD Chronic obstructive pulmonary disease HR Hazard ratio Ethical statement MRD Minimal residual disease NSAIDs Non-steroidal anti-inflammatory drugs The ethics committees of Kitasato University (B12-119) and NSCLC Non-small-cell lung cancer Teikyo University (Teirin-19-243) approved this study, which was OS Overall survival conducted in accordance with the Declaration of Helsinki. PGE2 Prostaglandin E2 Informed consent was obtained from all participants. PGE-MUM Prostaglandin E-major urinary metabolite Collection of samples and data INTRODUCTION In total, 373 patients who underwent complete resection for pri- Lung cancer is the leading cause of cancer-related death world- mary NSCLC at Kitasato University Hospital, Japan, between December 2012 and March 2017 were prospectively enrolled in wide. Although active and passive smoking are the primary risk the present study. Patients with carcinoma in situ, non-curative re- factors for lung cancer, several occupational and environmental section, synchronous or metachronous (within 5 years) malignan- factors and pre-existing lung diseases such as chronic obstructive cies, clinically or pathologically confirmed distant metastasis or pulmonary disease (COPD) and idiopathic pulmonary fibrosis are pleural dissemination were excluded. Comorbidities were scored also involved [1]. Under these conditions, many environmental using Charlson comorbidity index [10]. All resected tumours were factors and cancer risk factors are associated with chronic inflam- histologically diagnosed using the WHO classification and staged mation. Inflammatory responses play integral roles at different in accordance with the seventh edition of the TNM classification stages of tumour development, including initiation, promotion, staging manual [11, 12]. Adjuvant chemotherapy was considered invasion and metastasis and also affect immune surveillance and for patients with pathologic stage IA (tumour size >2 cm), IB–IIIA responses to therapy [2]. NSCLC according to the Japanese guideline [13]. In inflammatory and tumourigenic signalling, prostaglandin E2 Two spot urine samples were obtained from patients with pri- (PGE2) is synthesized from arachidonic acid via a cyclooxygenase mary lung cancer in the morning at 1 or 2 days preoperatively (COX)-catalyzed reaction [3]. Since the lungs play an important and 3–6 weeks postoperatively. In patients who received adjuvant role in both the production and metabolism of PGE2, PGE2 levels chemotherapy, postoperative urine samples were obtained be- in clinical specimens are possible indicators of inflammation- fore the initiation of chemotherapy. All samples were centrifuged associated lung disorders [4]. Furthermore, COX-2-derived PGE2 at 1000g for 10 min, and the supernatants were stored at -20 C is reportedly the most significant prostaglandin involved in can- until analysis. cer progression [3, 5]. Although previous studies investigating the The prognostic impact of absolute pre- and postoperative association between COX-2/PGE2 activity in tumour tissue and PGE-MUM levels and perioperative changes of PGE-MUM levels tumour progression have been conducted, accurately measuring were investigated separately. Therefore, patients whose only pre- its expression in tissue samples has several limitations, such as its operative urine samples were available for PGE-MUM analysis rapid degradation, invasiveness of sampling and tumour hetero- were included in the present study. Among all cases, only geneity [6, 7]. patients whose matched pre- and postoperative PGE-MUM levels Prostaglandin E-major urinary metabolite (PGE-MUM) is a stable were available were included in the analysis of perioperative end product of PGE2 metabolism and can be measured in urine changes in PGE-MUM levels. samples [8]. Its measurement allows for non-invasive longitudinal evaluation and can reflect systemic COX-2/PGE2 activity. A recent Measurement of urinary prostaglandin E-major study demonstrated that PGE-MUM levels were significantly ele- vated in patients with lung adenocarcinoma and tended to be pos- urinary metabolite levels itively correlated with cancer progression [9]. However, dynamic changes in PGE-MUM during the perioperative period have not Each urine sample was tested individually using the PGE-MUM been evaluated to date, and their prognostic significance remains assay. A radioimmunoassay kit (Institute of Isotopes Co., Ltd, M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 3 Budapest, Hungary) was used to measure the PGE-MUM levels. Briefly, the synthesized level of bicyclic PGE-MUM was measured Table 1: Clinicopathologic characteristics of patients with using a competitive assay after alkaline treatment was performed, non-small-cell lung cancer (N = 211) followed by neutralization of the urine samples. PGE-MUM levels were normalized to the concentration of urinary creatinine Variables Value, n (%) (expressed as lg/g Cr) because PGE-MUM concentration Age (years) 67.4 [SD] 67.5 (9.7) depends on urinary volume. Sex Male 130 (62) Female 81 (38) Statistical analysis Smoking history Current or former 134 (64) Pre- and postoperative PGE-MUM levels were compared based Never 77 (36) Smoking index, mean 541.4 on clinicopathological characteristics. All continuous variables NSAIDs administration identified with abnormal distribution by the Shapiro–Wilk test Yes 178 (84) were analysed with Mann–Whitney U-test. Multivariable Cox re- No 33 (16) gression models were used to examine prognostic factors. Charlson comorbidity index Survival was calculated using the Kaplan–Meier method, and dif- 0 117 (55) 1 68 (32) ferences were determined by means of log-rank analysis. The > _2 26 (12) variables with a univariable P-value of <0.20 were included in COPD the multivariable model. Multicollinearity between variables in Yes 90 (43) the multivariable model was assessed using a variance inflation No 121 (57) IPF factor. The goodness of fit of the models was evaluated using the Yes 15 (7) Akaike information criterion (AIC) and the Bayesian information No 196 (93) criterion (BIC). The AIC and BIC are model selection criteria and Tumour size, mean (SD) 33.3 (22.5) calculated as follows: AIC = -2 log (L)+ 2k, BIC = -2 log (L)+ k log Histology Adenocarcinoma 166 (79) (n). In this formula, n is the number of samples, L is the likelihood Squamous cell carcinoma 27 (13) for the mode and k is the number of parameters in the model Large cell neuroendocrine carcinoma 8 (4) [14]. When variables are selected in multivariable analysis, the Others 10 (5) number of samples changes owing to missing values. Since the Pleural invasion Yes 66 (31) normally used AIC can only be used when the number of sam- No 145 (69) ples is the same, we statistically selected a model based on the Pathologic N status BIC that can be compared even if the number of samples is dif- N0 172 (82) ferent. Missing data ware handled by listwise deletion. A P-value N1 15 (7) N2 24 (11) of <0.05 was considered to indicate statistical significance. Pathologic stage Statistical analyses were performed using JMP Pro version 15.0 IA 91 (43) (SAS Institute, Cary, NC, USA). IB 50 (24) IIA 20 (10) IIB 24 (11) IIIA 26 (12) RESULTS Adjuvant chemotherapy Yes 66 (31) Patient characteristics No 145 (69) COPD: chronic obstructive pulmonary disease; IPF: idiopathic pulmonary The clinicopathological data of 369 consecutive patients with fibrosis; NSAIDs: non-steroidal anti-inflammatory drugs; SD: standard NSCLC included in the present study were retrieved from medi- deviation. cal records. Among them, 157 patients who were treated with laxatives were excluded because laxative administration induces PGE2 production [15]. Accordingly, the relevant spot urine PGE-MUM levels were measured in the remaining 211 patients Adjuvant chemotherapy was administered to 66 (31%) patients. with NSCLC (130 men and 81 women). The clinicopathological Chemotherapy regimens included platinum doublet in 41 characteristics of these 211 patients are presented in Table 1. patients and tegafur-uracil in 25. None of the patients presented with inflammatory bowel dis- eases, acute cardiovascular problems and severe renal dysfunc- Prostaglandin E-major urinary metabolite levels in tion in which urine PGE-MUM levels were positively affected. Of pre- and postoperative urinary samples the 211 patients, 90 (43%) and 15 (7%) had COPD and idiopathic pulmonary fibrosis, respectively. A total of 178 patients (84.3%) PGE-MUM levels were evaluated using spot urine samples received non-steroidal anti-inflammatory drugs (NSAIDs) during obtained pre-(n = 211) and postoperatively (n = 196). Fifteen a postoperative course. Adenocarcinoma was the most common patients had only preoperative urine samples available for lung cancer (79%), followed by squamous cell carcinoma (13%), PGE-MUM analysis. The average PGE-MUM levels were 19.8 and large-cell neuroendocrine carcinoma (4%) and others (5%). In to- 17.9 mg/g Cr in pre- and postoperative samples, respectively. tal, 141 (67%) patients had pathologic stage I disease (91 with These levels were elevated compared with those in healthy vol- stage IA and 50 with stage IB), while 70 had more advanced dis- ease (20, 24 and 26 with stages IIA, IIB and IIIA, respectively). unteers (15.4 mg/g Cr) [9]. The pre- and postoperative PGE-MUM THORACIC 4 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery levels in the clinical samples are summarized in Table 2. PGE-MUM as a prognostic factor, we compared the prognosis of Preoperative PGE-MUM levels were associated with sex, smoking patients with high PGE-MUM levels with that of patients with low history and the presence of COPD but not histological cancer PGE-MUM levels. The levels were divided based on mean values, type, NSAIDs administration and comorbidity score. Patients 19.6 mg/g Cr in preoperative samples and 17.9 mg/g Cr in postop- erative samples. Patients with high PGE-MUM levels before and with large tumours (> _33 mm, P = 0.002), tumour with pleural inva- after resection exhibited significantly worse overall survival (OS) sion (P = 0.0002) and advanced-stage disease (P = 0.0006) exhib- (5-year OS, 82.3 vs 69.8%, P = 0.018 and 5-year OS, 85.7 vs 66.7%, ited significantly increased preoperative PGE-MUM levels, P = 0.001, respectively) (Fig. 1A and B). This trend was observed suggesting that PGE-MUM before tumour resection reflects tu- regardless of receipt of NSAIDs (Supplementary Material, Fig. S1) mour burden and progression. However, absolute postoperative However, despite the close association with tumour progression PGE-MUM levels were not associated with any clinical or patho- and negative prognostic impact in the univariable analysis, pre- logical characteristics including NSAIDs administration. operative PGE-MUM was not an independent prognostic factor in the multivariable analysis for OS (Table 3). Notably, Prognostic effect of prostaglandin E-major urinary postoperatively, the PGE-MUM level remained a significant prog- metabolite in non-small-cell lung cancer nostic factor [hazard ratio (HR), 6.462; 95% confidence interval (CI): 1.284–32.49, P = 0.024] as did age (HR 3.083, 95% CI: The median follow-up duration following resection was 1.076–8.836, P = 0.036), pleural invasion (HR 2.269, 95% CI: 51.4 months. A total of 37 patients experienced recurrence at dis- 1.165–4.383, P = 0.015) and lymph node metastasis (HR 2.216, tant sites, while 38 experienced loco-regional recurrence. Among 95% CI: 1.101–4.459, P = 0.026), suggesting that preoperative the 211 patients, 47 died during the follow-up period, 26 of lung PGE-MUM could be a predictive marker for tumour progression cancer and 21 of other causes. To investigate the role of at the time of surgery but does not affect prognosis Table 2: Pre- and postoperative prostaglandin E-major urinary metabolite levels in patients with non-small-cell lung cancer Variables Before surgery After surgery N Average PGE-MUM P-Value N Average PGE-MUM P-Value (mg/g Cr) (mg/g Cr) Sex Male 130 22.2 0.006 120 18.2 0.62 Female 81 15.9 76 17.4 Smoking history Current or former 134 22.4 0.002 124 17.9 0.95 Never 77 15.3 72 18.0 NSAIDs administration Yes 178 19.5 0.55 167 18.1 0.65 No 33 21.3 29 17.0 Charlson comorbidity index 0 117 20.7 0.57 112 18.0 0.52 1 68 18.1 59 16.9 > _2 26 19.8 25 19.7 COPD Yes 90 23.1 0.01 83 19.3 0.15 No 121 17.3 113 16.9 IPF Yes 15 25.9 0.13 15 23.1 0.07 No 196 19.3 181 17.5 Histology Adenocarcinoma 166 19.1 0.23 155 17.9 0.94 Non-adenocarcinoma 45 22.4 41 18.0 Tumour size > _33 mm 76 24.2 0.002 70 18.4 0.60 <33 mm 135 17.3 126 17.6 Pleural invasion Yes 66 25.8 0.0002 61 18.3 0.75 No 145 17.0 135 17.7 Lymph node metastasis Yes 39 22.1 0.33 37 16.5 0.39 No 172 19.3 159 18.2 Pathologic stage Stage I 141 17.1 0.0002 131 18.3 0.78 Stage II 44 19.7 40 16.9 Stage III 26 28.4 25 17.5 COPD: chronic obstructive pulmonary disease; IPF: idiopathic pulmonary fibrosis; NSAIDs: non-steroidal anti-inflammatory drugs; PGE-MUM: prostaglandin E-major urinary metabolite. M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 5 Figure 1: (A) Overall survival curve according to preoperative PGE-MUM levels. (B) Overall survival curve according to postoperative PGE-MUM levels. PGE-MUM lev- els were divided based on mean values (19.6 and 17.9 mg/g Cr in pre- and postoperative samples, respectively). PGE-MUM: prostaglandin E-major urinary metabolite levels after resection was an independent prognostic factor (HR, 2.272; 95% CI: 1.038–4.766, P = 0.039) along with pleural invasion, Table 3: Multivariable Cox regression analysis of factors lymph node metastasis and adjuvant chemotherapy (Table 4). In influencing overall survival in resected non-small-cell lung patients with increased PGE-MUM levels after resection, adjuvant cancer chemotherapy significantly improved survival (5-year OS, 79.0 vs 50.4%, P = 0.027) (Fig. 3A). However, in patients with decreased Variable Hazard 95% confidence P-Value ratio interval PGE-MUM levels after resection, there were no significant differ- ences in OS between patients who received adjuvant chemother- Lower Upper apy and those who did not (5-year OS, 82.1 vs 82.3%, P = 0.442) Age (years) (> _65/<65) 3.083 1.076 8.836 0.036 (Fig. 3B). Sex (male/female) 1.608 0.771 3.352 0.205 Tumour size (> _33 mm/<33 mm) 1.399 0.720 2.719 0.321 Pleural invasion (present/absent) 2.269 1.165 4.383 0.015 LN metastasis (present/absent) 2.216 1.101 4.459 0.026 DISCUSSION Preoperative PGE-MUM (mg/g Cr) 1.820 0.135 24.47 0.651 Postoperative PGE-MUM (mg/g Cr) 6.462 1.284 32.49 0.024 In the present study, we demonstrated that elevated preoperative LN: lymph node; PGE-MUM: prostaglandin E-major urinary metabolite. PGE-MUM levels were associated with cancer advancement at the time of surgery and postoperative PGE-MUM levels were a promising biomarker for poor prognosis after complete resection independently, whereas postoperative PGE-MUM may be a in patients with NSCLC. These results may provide useful infor- prognostic marker in patients with resected NSCLC. mation for perioperative management including postoperative systemic therapy. Implications of perioperative changes in PGE-MUM is a urinary metabolite of PGE2 and represents the activity of the COX-2 pathway. COX-2 and COX-2-derived PGE2 prostaglandin E-major urinary metabolite levels are reportedly involved in the tumour initiation and proliferation and indication for adjuvant chemotherapy of cancer cells both in vitro and in vivo [5, 16]. Several studies per- formed in clinical settings have shown that COX-2 overexpres- Since high PGE-MUM levels after resection were associated with sion is associated with poor prognosis in solid tumours, including poor prognosis in patients with NSCLC, we investigated the asso- NSCLC [6, 17]. Furthermore, based on the hypothesis that sup- ciation between pre- and postoperative PGE-MUM changes and pressing the COX-2 pathway could reduce cancer aggression and prognosis and assessed possible indication for adjuvant therapies improve survival, prospective clinical studies have been con- based on PGE-MUM changes. Matched pre- and postoperative ducted to examine the effects of COX-2 inhibitors combined PGE-MUM levels were available for 113 of 127 patients who with standard chemotherapies. However, no randomized con- were considered eligible for adjuvant chemotherapy according trolled trial has demonstrated a therapeutic advantage for these to the clinical guideline. Of the 113 patients, 60 received adjuvant agents [7, 18–21]. Considerable limitations discussed in these chemotherapy and 53 did not. Fifty patients experienced ele- studies included optimal patient selection based on clinical bio- vated PGE-MUM levels after resection relative to preoperative markers and the methods used to detect the markers. In a phase levels, with an average increase of 6.45 mg/g Cr. A reduction was III study, CALGB 30801, the enrolled patients were selected based observed in 63 patients, with an average reduction of 12.76 mg/g on COX-2 expression assessed using immunohistochemistry, and Cr. The OS of patients with increased PGE-MUM levels after re- the COX-2 inhibitor celecoxib failed to improve prognosis in section was significantly worse than that of patients with de- creased PGE-MUM levels (5-year OS, 81.6 vs 65.5%, P = 0.04) NSCLC with high COX-2 expression. However, a cohort of (Fig. 2). In multivariable analysis for OS, an increase in PGE-MUM patients who could benefit from celecoxib was identified in THORACIC 6 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery Figure 2: Overall survival curve according to the change in postoperative PGE-MUM levels compared with preoperative PGE-MUM levels. PGE-MUM: prostaglandin E-major urinary metabolite for lung cancer reportedly affects survival negatively [23, 24]. As postoperative PGE-MUM was an independent prognostic factor Table 4: Multivariable Cox regression analysis of factors in our study regardless of tumour progression, prolonged exces- influencing overall survival in resected non-small-cell lung sive inflammation after surgery might facilitate the survival and cancer with matched pre- and postoperative prostaglandin development of microscopic tumour cells, which could subse- E-major urinary metabolite examination quently lead to tumour recurrence and poor prognosis. Another interpretation is the role of PGE-MUM as a biomarker for resid- Variable Hazard 95% confidence P-Value ratio interval ual tumour cells. Since preoperative PGE-MUM reflects tumour burden and high postoperative PGE-MUM could be associated Lower Upper with poor prognosis, the lack of reduction in PGE-MUM after Age (years) (> _65/<65) 3.672 1.076 12.534 0.038 surgery is assumed to represent the presence of microscopic re- Sex (male/female) 1.759 0.669 4.626 0.253 sidual tumour cells. Although inflammation can promote tumour Tumour size (mm) (> _33/<33) 1.605 0.738 3.489 0.232 Pleural invasion (present/absent) 3.453 1.532 7.782 0.003 progression, tumours can also induce inflammation by secreting LN metastasis (present/absent) 2.865 1.317 6.229 0.008 inflammatory cytokines into their microenvironment to maintain Increase of PGE-MUM after 2.272 1.038 4.766 0.039 favourable conditions for their growth. PGE2 can reportedly be resection (yes/no) secreted from tumour cells, thereby promoting immune evasion Adjuvant therapy (no/yes) 2.400 1.041 5.533 0.040 [25]. Thus, high postoperative PGE-MUM levels and subsequent LN: lymph node; PGE-MUM: prostaglandin E-major urinary metabolite. poor prognosis may be a consequence of MRD with activation of the COX-2/PGE2 pathway. With the recent exploration of biomarkers and development subset analysis using a urinary PGE2 metabolite. Correlation and of drugs for unresectable lung cancers, greater attention has agreement were poor between urinary PGE2 metabolite and been paid to perioperative treatment tailored based on bio- COX-2 staining by immunohistochemistry, and prognosis in markers. Although indications for adjuvant therapies are primar- NSCLC with elevated urinary PGE metabolite was improved by ily determined based on pathological stage, even tumours with celecoxib treatment [7]. Given the advantage of urinary samples, the same advanced stage and histologic type represent a highly which enable repeated real-time assessment and do not require heterogeneous population. This includes patients who do not consideration of tumour heterogeneity in biomarker expression, truly require adjuvant therapy. Recent studies attempting to PGE-MUM could be promising for evaluating the activity of the identify tumours with a high risk for postoperative recurrence COX-2 pathway and reflecting the exact biological features of and select optimal indications for adjuvant therapies using devel- tumours. oped detection methods for MRD after surgical resection have Despite evidence on the association between COX-2-derived been conducted [26–28]. Based on the presumption that sus- PGE2 and prognosis in malignant tumours, the dynamics and tained elevated PGE-MUM levels after resection represent the prognostic significance of systemic COX-2 activity during the presence of MRD, we investigated the association between pre- perioperative period remain unclear. The present study demon- and postoperative PGE-MUM changes and prognosis to assess strated that high PGE-MUM levels after resection were associated possible indication for adjuvant therapies based on PGE-MUM with poor prognosis in patients with NSCLC. This result could be levels. Multivariable analysis revealed that an increase in underpinned by 2 possible implications: the cancer-promoting PGE-MUM levels after resection remained a significant prognos- effect of inflammation and the presence of microscopic residual tic factor. Furthermore, adjuvant chemotherapy improved OS in tumour cells, called minimal residual disease (MRD). patients with increased PGE-MUM levels after resection, while Inflammation has several tumour-promoting effects, including the survival benefit from adjuvant chemotherapy was not ob- proliferation, angiogenesis, invasiveness and metastasis [22]. The served in patients with decreased PGE-MUM levels after resec- inflammatory burden in patients undergoing curative resection tion. These results may support a role of perioperative changes in M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery 7 Figure 3: (A) Overall survival in patients with increased PGE-MUM levels after resection who received adjuvant chemotherapy and those who did not. (B) Overall survival in patients with decreased PGE-MUM levels after resection who received adjuvant chemotherapy and those who did not. PGE-MUM: prostaglandin E-major urinary metabolite. PGE-MUM levels as a clinical biomarker representing the pres- ACKNOWLEDGMENTS ence of MRD and provide useful information for determining the optimal eligibility for adjuvant therapy to improve the survival of We are extremely grateful to Professor Mutsunori Fujiwara (1947– patients with NSCLC. 2021), Department of Pathology, Nissan Tamagawa Hospital, Tokyo, Japan, for his excellent advice regarding this study. We also greatly appreciate Fujirebio Inc., for the assistance in measurement of uri- Limitations nary PGE-MUM levels. We thank Editage (www.editage.com)for the English language editing. The present study had several limitations. First, the investigation was conducted at a single centre with a limited number of en- Funding rolled patients. Second, PGE-MUM levels may be affected by sev- eral inflammatory conditions. Our cohort did not include The authors did not receive any specific funding for this work. patients with inflammatory bowel disease, which is the most well-known disease affecting PGE-MUM. However, the possibility Conflict of interest: none declared. that other inflammatory factors may have influenced PGE-MUM levels cannot be ruled out. Third, PGE-MUM levels may be af- Data availability fected by several drugs. Patients who were administered laxatives were treated were excluded from our study. However, in clinical The data underlying this article will be shared on reasonable request practice, patients often need to be administered laxatives and to the corresponding author. other drugs that may affect PGE-MUM during the postoperative course. In fact, most patients (84.3%) received NSAIDs during Author contributions perioperative period in the present study. Although NSAIDs administration was not significantly associated with pre- and Masashi Mikubo: Conceptualization; Data curation; Formal analy- postoperative PGE-MUM levels in our cohorts, that might be be- sis; Investigation; Writing—original draft. Yukitoshi Satoh: cause of the small number of patients who did not receive Conceptualization; Data curation. Mototsugu Ono: Data curation. NSAIDs. Thus, further investigation that considers the effects of Dai Sonoda: Data curation. Shoko Hayashi: Data curation. these drugs is needed to apply our results to clinical practice. Masahito Naito: Data curation. Yoshio Matsui: Data curation. Kazu Shiomi: Data curation. Masaaki Matsuura: Formal analysis; Investigation; Methodology; Writing—review & editing. Satoru Ito: CONCLUSION Supervision; Writing—review & editing. The results of the present study indicate that elevated preopera- Reviewer information tive PGE-MUM levels are associated with tumour progression and that postoperative PGE-MUM levels are a promising bio- Interdisciplinary CardioVascular and Thoracic Surgery thanks David marker for survival in patients with resected NSCLC. Prolonged G. Healy, Niccolo Daddi and the other, anonymous reviewer(s) for elevation of PGE-MUM levels may aid in identifying patients who their contribution to the peer review process of this article. may benefit from adjuvant systemic therapy. REFERENCES SUPPLEMENTARY MATERIAL [1] Collins LG, Haines C, Perkel R, Enck RE. Lung cancer: diagnosis and man- Supplementary material is available at ICVTS online. agement. Am Fam Physician 2007;75:56–63. THORACIC 8 M. Mikubo et al. / Interdisciplinary CardioVascular and Thoracic Surgery [2] Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and can- [16] Ruan D, So SP. Prostaglandin E2 produced by inducible COX-2 and cer. Cell 2010;140:883–99. mPGES-1 promoting cancer cell proliferation in vitro and in vivo. Life Sci [3] Aiad M, Tahir A, Fresco K, Prenatt Z, Ramos-Feliciano K, Walia J et al. 2014;116:43–50. Does the combined use of aspirin and immunotherapy result in better [17] Koki A, Khan NK, Woerner BM, Dannenberg AJ, Olson L, Seibert K et al. outcomes in non-small cell lung cancer than immunotherapy alone? Cyclooxygenase-2 in human pathological disease. Adv Exp Med Biol Cureus 2022;14:e25891. 2002;507:177–84. [4] Bozyk PD, Moore BB. Prostaglandin E2 and the pathogenesis of pulmo- [18] Edelman MJ, Watson D, Wang X, Morrison C, Kratzke RA, Jewell S et al. nary fibrosis. Am J Respir Cell Mol Biol 2011;45:445–52. Eicosanoid modulation in advanced lung cancer: cyclooxygenase-2 ex- [5] Elwakeel E, Bru ¨ ggemann M, Wagih J, Lityagina O, Elewa MAF, Han Y pression is a positive predictive factor for celecoxib + chemotherapy— et al Disruption of prostaglandin E2 signaling in cancer-associated fibro- Cancer and Leukemia Group B Trial 30203. J Clin Oncol 2008;26:848–55. blasts limits mammary carcinoma growth but promotes metastasis. [19] Groen HJ, Sietsma H, Vincent A, Hochstenbag MM, van Putten JW, van Cancer Res 2022;82:1380–95. den Berg A et al. Randomized, placebo-controlled phase III study of [6] Khuri FR, Wu H, Lee JJ, Kemp BL, Lotan R, Lippman SM et al. docetaxel plus carboplatin with celecoxib and cyclooxygenase-2 expres- Cyclooxygenase-2 overexpression is a marker of poor prognosis in stage sion as a biomarker for patients with advanced non-small-cell lung can- I non-small cell lung cancer. Clin Cancer Res 2001;7:861–7. cer: the NVALT-4 study. J Clin Oncol 2011;29:4320–6. [7] Edelman MJ, Wang X, Hodgson L, Cheney RT, Baggstrom MQ, Thomas [20] Gitlitz BJ, Bernstein E, Santos ES, Otterson GA, Milne G, Syto M et al. A SP et al.; Alliance for Clinical Trials in Oncology. Phase III randomized, randomized, placebo-controlled, multicenter, biomarker-selected, phase placebo-controlled, double-blind trial of celecoxib in addition to stan- 2 study of apricoxib in combination with erlotinib in patients with ad- dard chemotherapy for advanced non-small-cell lung cancer with vanced non-small-cell lung cancer. J Thorac Oncol 2014;9:577–82. cyclooxygenase-2 overexpression: CALGB 30801 (alliance). J Clin Oncol [21] Edelman MJ, Tan MT, Fidler MJ, Sanborn RE, Otterson G, Sequist LV et al 2017;35:2184–92. Randomized, double-blind, placebo-controlled, multicenter phase II [8] Wang D, DuBois RN. Urinary PGE-M: a promising cancer biomarker. study of the efficacy and safety of apricoxib in combination with either Cancer Prev Res (Phila) 2013;6:507–10. docetaxel or pemetrexed in patients with biomarker-selected non- [9] Kawamoto H, Hara H, Araya J, Ichikawa A, Fujita Y, Utsumi H et al. small-cell lung cancer. J Clin Oncol 2015;33:189–94. Prostaglandin E-major urinary metabolite (PGE-MUM) as a tumor [22] Borsig L, Wolf MJ, Roblek M, Lorentzen A, Heikenwalder M. Inflammatory marker for lung adenocarcinoma. Cancers 2019;11:768. chemokines and metastasis–tracing the accessory. Oncogene 2014;33: [10] Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classi- 3217–24. fying prognostic comorbidity in longitudinal studies: development and [23] Templeton AJ, McNamara MG, Seruga B, Vera-Badillo FE, Aneja P, validation. J Chronic Dis 1987;40:373–83. Ocana A et al. Prognostic role of neutrophil-to-lymphocyte ratio in solid [11] Travis WD, Brambilla E, Burke AP, Marx A, AGeditors, N. WHO tumors: a systematic review and meta-analysis. J Natl Cancer Inst 2014; Classification of Tumours of the Lung, Pleura, Thymus and Heart. 4th 106:dju124. edn. Lyon, France: IARC Press, 2015. [24] Proctor MJ, Morrison DS, Talwar D, Balmer SM, O’Reilly DS, Foulis AK [12] Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D, Chansky et al An inflammation-based prognostic score (mGPS) predicts cancer K et al.; International Association for the Study of Lung Cancer survival independent of tumour site: a Glasgow Inflammation Outcome International Staging Committee and Participating Institutions. The Study. Br J Cancer 2011;104:726–34. International Association for the Study of Lung Cancer lung cancer stag- [25] Konishi Y, Ichise H, Watabe T, Oki C, Tsukiji S, Hamazaki Y et al. ing project: proposals regarding the clinical staging of small cell lung Intravital imaging identifies the VEGF-TXA(2) axis as a critical promoter cancer in the forthcoming (seventh) edition of the tumor, node, metas- of PGE(2) secretion from tumor cells and immune evasion. Cancer Res tasis classification for lung cancer. J Thorac Oncol 2007;2:1067–77. 2021;81:4124–32. [13] Japan Lung Cancer Society Guidelines for diagnosis and treatment of [26] Qiu B, Guo W, Zhang F, Lv F, Ji Y, Peng Y et al. Dynamic recurrence risk the lung cancer, 2021. https://www.haigan.gr.jp/guideline/2021/1/2/ and adjuvant chemotherapy benefit prediction by ctDNA in resected 210102040100.html#cq27 (29 December 2022, date last accessed). NSCLC. Nat Commun 2021;12:6770. [14] Hook EB, Regal RR. Validity of methods for model selection, weighting [27] Xia L, Mei J, Kang R, Deng S, Chen Y, Yang Y et al. Perioperative ctDNA- for model uncertainty, and small sample adjustment in capture- based molecular residual disease detection for non-small cell lung can- recapture estimation. Am J Epidemiol 1997;145:1138–44. cer: a prospective multicenter cohort study (LUNGCA-1). Clin Cancer [15] Fujiwara M, Okayasu I, Oritsu M, Komatsu J, Yoshitsugu M, Katoh Y et al. Res 2021;28:3308–3317. Significant increase in prostaglandin E-main urinary metabolite by laxa- [28] Gale D, Heider K, Ruiz-Valdepenas A, Hackinger S, Perry M, Marsico G tive administration: comparison with ulcerative colitis. Digestion 2000; et al Residual ctDNA after treatment predicts early relapse in patients 61:201–6. with early-stage non-small cell lung cancer. Ann Oncol 2022;33:500–10.

Journal

Interactive Cardiovascular and Thoracic SurgeryOxford University Press

Published: Jan 9, 2023

Keywords: Prostaglandin E-major urinary metabolite; Non-small-cell lung cancer; Surgery; Biomarker; Adjuvant chemotherapy

There are no references for this article.