Development and validation a Nomogram for identify malignant pleural effusion

doi:10.21203/rs.3.rs-3096312/v1

Download PDF

Research Article

Development and validation a Nomogram for identify malignant pleural effusion

https://doi.org/10.21203/rs.3.rs-3096312/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Malignant pleural effusion (MPE) is a common complication in cancer patients, indicating the presence of pleural metastasis. However, the ability to identify MPE clinically is still limited. The aim of this study was to develop a novel nomogram-based model for predicting MPE.

Methods

Between July 2020 and May 2022, a total of 428 patients with pleural effusion (PE) were consecutively enrolled. Demographic data, laboratory test results, and pathological parameters were collected. The LASSO regression method was used to select potential variables, and a multivariate logistic regression method was employed to construct a nomogram. Internal validation was performed using a bootstrapping method, and the nomogram's performance was evaluated based on calibration, discrimination, and clinical utility.

Results

Out of the 428 patients with PE, 217 (50.7%) were diagnosed with MPE. A diagnostic model was established using least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression, which identified four variables: serum carcinoembryonic antigen (sCEA), serum neuron-specific enolase (sNSE), pleural carcinoembryonic antigen (pCEA), and pleural lactate dehydrogenase (pLDH). The internal validation of the model showed an area under the curve (AUC) of 0.894 (95% CI: 0.864–0.934). The model was well-calibrated, and decision curve analysis (DCA) indicated that using the proposed nomogram to predict MPE would obtain a net benefit if the threshold probability of MPE was between 5% and 95%.

Conclusion

This study aimed to construct a nomogram that includes four demographic and clinical characteristics of patients with PE. The nomogram can be highly beneficial in distinguishing between MPE and benign pleural effusion (BPE).

pleural effusion

parameter

nomogram

metastasis

diagnostic

Pleural effusion (PE) is a common clinical condition that can be classified as benign or malignant depending on the underlying cause. It occurs when the peritoneum or protein balance is affected by conditions such as heart failure, kidney failure, severe pneumonia, or hypoproteinemia, resulting in the accumulation of pathological fluid in the pleural cavity. Malignant pleural effusion (MPE) is caused by cancer, accounting for almost 10% of all cases, and is associated with several types of tumors, including lung cancer, breast cancer, lymphoma, ovarian cancer, gastric cancer, sarcoma, and colon cancer. In some cases, MPE can present significant clinical challenges. Therefore, distinguishing between benign pleural effusion (BPE) and MPE is essential not only for diagnosis but also for developing an effective treatment plan.

Accurate diagnosis of BPE and MPE requires a thorough analysis of PE fluid, including cytological examination, thoracoscopic surgery, and evaluation of biochemical indicators such as total protein levels, (1) various tumor markers, (2, 3) and the patients' symptoms. Despite being the gold standard for identifying MPE, cytological analysis has limitations, with a positive rate of only 30–50%. (4) Although the positive rate of pleural biopsy is high, its invasive nature makes it difficult for patients to accept. Additionally, the procedures required for a definitive diagnosis may cause a delay that could potentially hinder the effective management of the disease and its outcomes. (5) Recently, in their study, Ting Wang et al. (6) identified S100A2 as a potential biomarker for distinguishing between MPE and tuberculous pleural effusion (TPE). However, the current assay for this biomarker is not widely available, particularly in hospitals located in developing countries. Therefore, there is a need to develop a more accessible, cost-effective, and minimally invasive method for accurately diagnosing the cause of PE. In this study, we aimed to develop a diagnostic model to efficiently identify MPE. Our objective was to investigate the diagnostic potential of readily available demographic features and laboratory indicators in patients with PE.

Patient selection and ethics

This retrospective cohort study examines patients who were diagnosed with PE at the Affiliated Huaian No.1 People’s Hospital of Nanjing Medical University from July 2020 to May 2022. The study recruited patients who met the following criteria: (a) newly diagnosed with PE and (b) underwent diagnostic thoracentesis. Patients with bloody PE due to trauma or surgery, or with PE of unknown etiology, were excluded from the study. The differentiation between BPE and MPE was determined by pathologists and clinicians.

A total of 471 participants were initially enrolled in the study, with 43 subjects being excluded. The patient selection process is illustrated in Fig. 1. The institutional ethics committee of the hospital (KY-2023-023-01) approved this study, and since patient information was kept anonymous, signed informed consent was waived.

Data Collection

Prior to any treatment, including the administration of intravenous fluids, diuretics, or chemotherapy, all patients underwent thoracentesis and blood withdrawal in a fasting state. The patients' demographic characteristics and laboratory data were obtained by reviewing their medical records in a structured manner. The variables of patient data were collected as follows: age, gender, comorbidities (hypertension, diabetes - yes or no) and smoking history (yes or no). In addition, laboratory test parameters were recorded, including platelet (PLT) counts, hemoglobin (Hb), and albumin at admission, as well as serum and PE carcinoembryonic antigen (CEA), serum neuron-specific enolase (NSE), serum squamous-cell carcinoma antigen (SCC), PE lactate dehydrogenase (LDH), and total protein (TP).

Statistical Analysis

Descriptive statistics were utilized to summarize baseline characteristics. For intergroup comparisons, Pearson chi-squared tests or Fisher's exact tests were applied as appropriate. In cases of missing data, the multiple imputation technique was used for further estimation. Predictor selection and regularization were achieved using the least absolute shrinkage and selection operator (LASSO) regression method, followed by multivariate logistic regression analysis to establish the predictive model for distinguishing between BPE and MPE. A nomogram was constructed based on the model. The area under the curve (AUC) was calculated to assess the model's discrimination capacity, and internal validation was performed using bootstrapping with 500 iterations. (7) The calibration of the model was evaluated by using the Hosmer–Lemeshow test. Additionally, the clinical usefulness of the model was assessed through decision curve analysis (DCA). (8) Statistical analyses were performed using R software (version 3.6.3; https://www.rproject.org). The value of P < 0.05 was considered statistically significant.

Patient characteristics

This study collected clinical data from 471 patients with new-onset PE, but 43 were excluded for not meeting the criteria mentioned previously (Fig. 1). The remaining 428 patients were included, and their etiology classification summary statistics are presented in Table 1. The patients were initially divided into two groups based on disease type: benign and malignant. In a cohort of 211 patients diagnosed with BPE, pneumonia was the most common cause accounting for 39.3% of cases, followed by chronic obstructive pulmonary disease (COPD) at 11.8% and empyema at 8.1%. In the MPE group of 217 patients, the major cause of MPE was lung adenocarcinoma, accounting for 63.6% (138/217) of cases, followed by squamous cell carcinoma of the lung at 7.8% (17/217).

Table 1

The aetiology of PE in the patients enrolled in this study.
Aetiology	Benign group (n = 211)	Malignant group (n = 217)	Total (n = 428)
Pneumonia	83	0	83
COPD	25	0	25
Empyema	17	0	17
Tuberculous pleurisy	10	0	10
Miscellaneous	76	0	76
Adenocarcinoma of the lung	0	138	138
Squamous cell carcinoma of the lung	0	17	17
Esophageal cancer	0	15	15
Small cell carcinoma of the lung	0	11	11
Breast cancer	0	5	5
Other cancers	0	31	31
Abbreviations: COPD, chronic obstructive pulmonary disease.

Out of the 217 cases of MPE, 102 of them had pathological cells that tested positive. On the other hand, all 94 cases in the benign group had negative pathological cells. This means that the sensitivity of diagnosing MPE was 47.5% (103/217), the specificity was 100% (211/211), and the accuracy was 73.4% (314/428).

Baseline characteristics and risk factors for MPE

Out of the 428 participants in this study, 217 (50.7%) were found to be suffering from MPE. The biological characteristics of both cases and controls have been presented in Tables 2. Upon initial examination, it was observed that patients with MPE were predominantly female and had higher levels of serum carcinoembryonic antigen (sCEA), serum neuron-specific enolase (sNSE), pleural effusion carcinoembryonic antigen (pCEA), pleural effusion lactate dehydrogenase (pLDH), hypoalbuminemia, and anemia (P < 0.05).

Table 2

Characteristics of the participants.
Variables	Total, N (%)	PE		P
Variables	Total, N (%)	Benign group, N (%)	Malignant group, N (%)	P
Total patients	428	211	217
Gender				0.004
Female	152(35.5%)	60 (28.4%)	92 (42.4%)
Male	276(64.5%)	151 (71.6%)	125 (57.6%)
Age, year				0.258
＜60	142 (33.2%)	64 (30.3%)	78 (35.9%)
≥ 60	286 (66.8%)	147 (69.7%)	139 (64.1%)
Smoking history				0.857
No	309 (72.2%)	151 (71.6%)	158 (72.8%)
Yes	119 (27.8%)	60 (28.4%)	59 (27.2%)
Hypertension				0.128
No	325 (75.9%)	153 (72.5%)	172 (79.3%)
Yes	103 (24.1%)	58 (27.5%)	45 (20.7%)
Diabetes				0.669
No	389 (90.9%)	190 (90.0%)	199 (91.7%)
Yes	39 (9.11%)	21 (9.95%)	18 (8.29%)
Anemia				< 0.001
No	218 (50.9%)	131 (62.1%)	87(40.1%)
Yes	210 (49.1%)	80 (37.9%)	130 (59.9%)
Thrombocytopenia				0.066
No	320 (74.8%)	149 (70.6%)	171 (78.8%)
Yes	108 (25.2%)	62 (29.4%)	46 (21.2%)
Hypoalbuminemia				0.001
No	94 (22.0%)	31 (14.7%)	63 (29.0%)
Yes	334 (78.0%)	180 (85.3%)	154 (71.0%)
sCEA, ug/L				< 0.001
≤ 5	268 (62.6%)	188 (89.1%)	80 (36.9%)
＞5	160 (37.4%)	23 (10.9%)	137 (63.1%)
sNSE, ug/L				< 0.001
≤ 16.3	242 (56.5%)	146 (69.2%)	96 (44.2%)
＞16.3	186 (43.5%)	65 (30.8%)	121 (55.8%)
sSCC, ug/L				0.461
≤ 2.7	388 (90.7%)	194 (91.9%)	194 (89.4%)
＞2.7	40 (9.35%)	17 (8.06%)	23 (10.6%)
pCEA, ug/L				< 0.001
≤ 30	282 (65.9%)	200 (94.8%)	82 (37.8%)
＞30	146 (34.1%)	11 (5.21%)	135 (62.2%)
pLDH, ug/L				< 0.001
≤ 246	178 (41.6%)	115 (54.5%)	63 (29.0%)
＞246	250 (58.4%)	96 (45.5%)	154 (71.0%)
pTP, g/L				1.000
63–82	9 (2.10%)	4 (1.90%)	5 (2.30%)
≤ 63 or ＞82	419 (97.9%)	207 (98.1%)	212 (97.7%)
Abbreviations: PE, pleural effusion; sCEA, serum carcinoembryonic antigen; sNSE, serum neuron-specific enolase; sSCC, serum squamous cell carcinoma antigen; pCEA, pleural carcinoembryonic antigen; pLDH, pleural lactate dehydrogenase; pTP, pleural effusion total protein.

Predictors Entering the Model

After performing the LASSO regression method (Fig. 2), four predictors were found to be associated with MPE: sCEA, sNSE, pCEA, and pLDH (Table 2) using the 'Backwald: Wald' multiple logistic regression model. These results were then used to create a nomogram that can predict the probability of MPE.

Establishment of the Model and Nomogram

A predictive model for MPE was developed using four predictors: sCEA, pCEA, pLDH, and sNSE. The coefficient of each predictor in the model was calculated as follows: sCEA (1.777), pCEA (2.703), pLDH (1.208), and sNSE (1.327). To provide physicians with a tool for individualized prediction of MPE, a nomogram was constructed using multivariable logistic regression (see Fig. 3). The formula based on the model was described as follows: logistic (risk score) = -2.55 + 1.777 × sCEA + 2.703 × pCEA + 1.208 × pLDH + 1.327 × sNSE (Table 3).

Table 3

Estimated odds ratios determined in a logistic regression model (Backwald: Wald).
characteristics	B	SE	OR	95% CI	Z	P
sCEA	1.777	0.333	5.911	3.104–11.530	5.329	< 0.001
pCEA	2.703	0.391	14.925	7.165–33.530	6.912	< 0.001
pLDH	1.208	0.287	3.348	1.931–5.959	4.217	< 0.001
sNSE	1.327	0.277	3.768	2.207–6.565	4.781	< 0.001
Abbreviations: OR, odds ratio; CI, confidence interval.

Multidimensional methods were employed to validate the precision of the nomogram. The nomogram's area under curve (AUC) was determined to be 0.894 (95% confidence interval [CI] = 0.865–0.925) using a bootstrap method with 500 resamples (Fig. 4A). The nomogram's predicted probabilities corresponded closely with the actual clinical outcomes, as shown in Fig. 4B. Additionally, the Hosmer-Lemeshow test yielded a P-value of 0.901, indicating that the model was an ideal fit. According to the DCA, our nomogram's MPE risk estimates had a higher net benefit compared to screen-none or screen-all strategies when the threshold probability of MPE was between 5% and 95%. (Fig. 4C).

Models comparison

The results of the ROC curve analyses are presented in Fig. 5A. Delong tests were utilized to compare the discrimination of the nomogram with the variables included in the nomogram alone (AUC_nomogram = 0.894, AUC_sCEA =0.761, AUC_pCEA =0.785, AUC_pLDH =0.627, AUC_sNSE =0.625). The nomogram demonstrated superior discriminatory accuracy in identifying patients with MPE compared to any individual variables included in the nomogram (all P < 0.05).

DCAs were utilized to compare the clinical usefulness of the models. The results showed that the nomogram outperformed the models that only contained risk factors included in the nomogram in terms of overall net benefit, across a wide range of threshold probabilities (Fig. 5B).

In the present study, a nomogram was developed to predict MPE among patients with PE. The nomogram incorporated four variables, including sCEA, pCEA, pLDH, and sNSE. The nomogram showed good discriminatory ability, calibration, and clinical usefulness. MPE is a common complication associated with a wide variety of neoplasms, including lung, breast, lymphoma, ovarian, and gastric cancers. Jung (9) reported that MPE is a serious complication of advanced tumors, with relatively high morbidity and mortality rates. Early differentiation of benign from malignant conditions is crucial for selecting the appropriate treatment, for example, surgery can be an effective treatment option for malignant lung cancer that has not invaded the pleura, leading to significant improvements in survival time. Additionally, patients with BPE, such as TPE and parapneumonic effusions can be clinically cured if treated in time.

Traditionally, differential diagnosis can be made based on certain specific findings, such as pleural biopsy, diagnostic thoracentesis, and PE cytology. (10) Unfortunately, these methods have limitations, such as invasiveness and limited diagnostic efficacy. (11) In this study, the sensitivity of cytology was 56.7%, and this result was also unsatisfactory. Earlier studies focused on specific factors associated with MPE, such as single-cell analysis of diverse immunophenotypes,(12)they performed single-cell RNA sequencing on 62,382 cells from MPE patients induced by non-small-cell lung cancer to describe the composition, lineage, and functional states of infiltrating immune cells in MPE, or the landscape map of human MPE immune cells ,(13)which described the interplay in MPE among different Th cells, as well as Th cells and lung cancer cells or mesothelial cells. However, these studies mentioned above were not available in the public clinical setting although they might have high diagnostic value, for the new markers were inaccessible.

In clinical practice, relying on a single parameter to distinguish between two conditions has limited effectiveness due to low sensitivity or specificity. Instead, combining several markers to construct a mathematical model can significantly improve diagnostic accuracy. Previous studies have developed several models for predicting MPE, and it was pointed out by the authors that these models could improve the diagnosis of MPE. For example, Chen et al. (14) revealed that the combination of FTIR near-infrared spectroscopy (NIRS) and machine learning is an innovative, rapid, and convenient method for clinical PE classification. Wang et al. (15, 16) reported a classification model to identify BPE and MPE using a CT volume and its 3D PE mask as inputs based on artificial intelligence. However, these models mentioned above may have disadvantages of high cost, the need for additional examinations, and trained personnel. We established a 4-marker predictive model by multivariate regression analysis in the development set. The 4-markers involved in our model were easily obtained, and more importantly, the total cost of the indexes involved in the scoring model was acceptable. The diagnostic performance was evaluated as very good in the same set through the use of bootstrap resampling. To the best of our knowledge, there have been few studies aimed at developing a simple and user-friendly predictive model based on common clinical parameters for identifying MPE.

The clinical diagnostic efficiency of sCEA, pCEA, pLDH, and sNSE in the diagnosis of MPE was reportedly satisfactory. (17, 18, 19, 20, 21) A large body of work on the function of tumor markers showed that these contributed to the identification of MPE.(22)However, sSCC levels were filtered out by LASSO regression analysis, which is considered superior to select predictors by univariate analysis. (23) The difference between our study and previous studies may be attributed to the different causes of MPE. SCC is a relatively specific tumor marker for squamous cell carcinoma of the lung. The results showed that the percentage of squamous cell carcinoma of the lung (7.83%) in our study was lower than that in previous studies (100.00%). (22) Regarding “sCEA” and “pCEA”, it has been reported that CEA was especially associated with adenocarcinoma of the lung, (24, 25) and pCEA was an effective indicator for identifying lung cancer-associated MPE among various tumor biomarkers. (26) In general, CEA is a broad-spectrum tumor marker that can indicate the presence of various tumors. It is a useful tumor marker for evaluating the effectiveness of treatment, disease progression, monitoring, and prognosis estimation of colorectal cancer, breast cancer, and lung cancer. However, its specificity and sensitivity are not strong, and its role in the early detection of tumors is not significant. Regarding the “pLDH” and “sNSE”, it has been reported that LDH levels were positively correlated with lymphoma-associated MPE, and providing a sensitivity of 76.00% and a specificity of 81.00% to differentiate lymphoma-associated MPE and TPE. (27) For “sNSE”, it has been reported that sNSE levels are positively associated with small-cell lung carcinoma (SCLC), with a sensitivity of 60.78% and a specificity of 81.11% for the diagnosis of SCLC. SCLC is characterized by a high proliferation rate, early metastases, and an exceptionally poor prognosis. (28) These four predictors are readily available clinically, and the nomogram demonstrated good discriminatory ability and calibration, with the DCA evaluation showing its clinical utility. As pleural biopsy is an invasive procedure that patients may find difficult to undergo, and the sensitivity of cytology examination after diagnostic thoracentesis is low, and multiple inspections are required, this cost-free nomogram may be useful in screening MPE in these patients with PE.

This study has several limitations. Firstly, it is important to note some markers that could potentially aid in malignant diagnosis, such as adenosine deaminase (ADA), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP), were not evaluated in this study due to the limited availability of patients who underwent these tests. Nevertheless, previous studies have shown that ADA and CRP showed no significant difference between PE and MPE.^[27] this is a single-center, retrospective study with a limited number of patients. As a result, the prevalence of PE observed in this study may not accurately reflect the prevalence of PE in the wider Chinese population. To validate this predictive model, further studies with larger sample sizes from multiple centers are required. It is important to note that if these predictive models are used in a different population, they should be reassessed.

Our study has successfully developed a predictive model consisting of four markers that is both user-friendly and reliable for classifying PE into subtypes, specifically malignant or benign. This breakthrough provides a valuable tool for medical professionals to accurately diagnose and treat patients with PE. The proposed mathematical model has the potential to serve as a valuable supplementary diagnostic tool in differentiating MPE from BPE in clinical settings.

Acknowledgements

Not applicable

Authors’ contributions

Jinling Ji and Lei Yan analyzed data and wrote the manuscript; Kai Wang and Kun Jiang performed the data analysis and some of the data collection; Chang Li performed most of the data collection, analysis and validation; Yuzhang Jiang designed and supervised the study. All authors have read and approved the final manuscript.

Funding

Not applicable

Availability of data and materials

The data sets analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of the affiliated Huaian No.1 People's Hospital of Nanjing Medical University (KY-2023-023-01). A waiver of consent was obtained from the Ethics Committee of the affiliated Huaian No.1 People's Hospital of Nanjing Medical University due to the retrospective nature of the study making consent impractical and contacting patients to obtain consent would pose a greater risk than the waiver.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Zhu S, Du L, Xu D, et al. Ascitic fluid total protein, a useful marker in non-portal hypertensive ascites. J Gastroenterol Hepatol, 2020, 35: 271-7.
Liang L, Fang J, Han X, et al. Prognostic Value of CEA, CA19-9, CA125, CA724, and CA242 in Serum and Ascites in Pseudomyxoma Peritonei. Front Oncol, 2021, 11: 594763.
Schuster-Little N , Fritz-Klaus R , Etzel M , et al. Affinity-free enrichment and mass spectrometry analysis of the ovarian cancer biomarker CA125 (MUC16) from patient-derived ascites. Analyst, 2021, 146: 85-94.
Alexandrakis MG, Moschandrea JA, Koulocheri SA, et al. Discrimination between malignant and nonmalignant ascites using serum and ascitic fluid proteins in a multivariate analysis model. Digestive diseases and sciences, 2000, 45: 500-8.
Kipps E, Tan DS, Kaye SB. Meeting the challenge of ascites in ovarian cancer: new avenues for therapy and research. Nat Rev Cancer, 2013, 13: 273-82.
Wang T, Wang N, Zhang L, et al. S100A2: A potential biomarker to differentiate malignant from tuberculous pleural effusion. Indian J Cancer, 2021, 58: 241-7.
Collins GS, Reitsma JB, Altman DG, et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): the TRIPOD statement. Ann Intern Med, 2015, 162: 55-63.
Van Calster B, Wynants L, Verbeek JFM, et al. Reporting and Interpreting Decision Curve Analysis: A Guide for Investigators. Eur Urol, 2018, 74(6): 796-804.
Jung Y, Kim EY. Artificial intelligence in differentiating malignant from benign pleural effusion: a step beyond conventional methods. Thorax, 2023, 78: 325.
Gayen S. Malignant Pleural Effusion: Presentation, Diagnosis, and Management. Am J Med, 2022, 135: 1188-92.
Chen J, Yang Y, Wang Z, et al. A Multicancer Malignant Pleural Effusion Diagnostic Test Using Hexokinase 2 and Single-Cell Sequencing. Clinical chemistry, 2022, 68: 680-90.
Huang ZY, Shao MM, Zhang JC, et al. Single-cell analysis of diverse immune phenotypes in malignant pleural effusion. Nat Commun, 2021, 12: 6690.
Yi FS, Zhai K, Shi HZ. Helper T cells in malignant pleural effusion. Cancer Lett, 2021, 500: 21-8.
Chen Z, Chen K, Lou Y, et al. Machine learning applied to near-infrared spectra for clinical pleural effusion classification. Scientific reports, 2021, 11: 9411.
Guo YY, Peng XL, Zhan N, et al. Development and validation a simple model for identify malignant ascites. Int J Med Sci, 2021, 18: 1966-74.
Wang S, Tan X, Li P, et al. Differentiation of malignant from benign pleural effusions based on artificial intelligence. Thorax, 2023, 78: 376-82.
Chian CF, Wu FP, Tsai CL, et al. Echogenic swirling pattern, carcinoembryonic antigen, and lactate dehydrogenase in the diagnosis of malignant pleural effusion. Scientific reports, 2022, 12: 4077.
Basso SMM, Del Conte A, Zuccon U, et al. Diagnostic Accuracy of Pleural Fluid Cytology, Carcinoembryonic Antigen and C-Reactive Protein Together in Patients With Pulmonary Metastases and Malignant Pleural Effusion. Anticancer Res, 2020, 40: 5877-81.
Chen S, Huang K, Zou L, et al. Diagnostic value of SHOX2, RASSF1A gene methylation combined with CEA level detection in malignant pleural effusion. BMC Pulm Med, 2023, 23: 160.
Cheng C, Yang Y, Yang W, et al. The diagnostic value of CEA for lung cancer-related malignant pleural effusion in China: a meta-analysis. Expert Rev Respir Med, 2022, 16: 99-108.
Zhu J, Feng M, Liang L, et al. Is neuron-specific enolase useful for diagnosing malignant pleural effusions? evidence from a validation study and meta-analysis. BMC cancer, 2017, 17: 590.
Liu L, Liu B, Zhu LL, et al. Clinical significance of CYFRA21-1, Scc-Ag and telomerase activity in serum and pleural effusion of patients with squamous-cell lung cancer. Bioanalysis, 2012, 4: 2367-74.
Balachandran VP, Gonen M, Smith JJ, et al. Nomograms in oncology: more than meets the eye. Lancet Oncol, 2015, 16: e173-80.
Zhang X, Li J, Wang Y, et al. A Diagnostic Model With IgM Autoantibodies and Carcinoembryonic Antigen for Early Detection of Lung Adenocarcinoma. Front Immunol, 2021, 12: 728853.
Huang WW, Tsao SM, Lai CL, et al. Diagnostic value of Her-2/neu, Cyfra 21-1, and carcinoembryonic antigen levels in malignant pleural effusions of lung adenocarcinoma. Pathology, 2010, 42: 224-8.
Zhang H, Li C, Hu F,et al. Auxiliary diagnostic value of tumor biomarkers in pleural fluid for lung cancer-associated malignant pleural effusion. Respir Res, 2020, 21: 284.
Kim CH, Oh HG, Lee SY, et al. Differential diagnosis between lymphoma-associated malignant pleural effusion and tuberculous pleural effusion. Ann Transl Med, 2019, 7: 373.
Rudin CM, Brambilla E, Faivre-Finn C,et al. Small-cell lung cancer. Nat Rev Dis Primers, 2021, 7: 3.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Development and validation a Nomogram for identify malignant pleural effusion

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials and methods

Patient selection and ethics

Data Collection

Statistical Analysis

Results

Patient characteristics

Baseline characteristics and risk factors for MPE

Predictors Entering the Model

Establishment of the Model and Nomogram

Models comparison

Discussion

Declarations

References

Additional Declarations

Status:

Version 1