Precise and automated lung cancer cell classification using deep neural network with multiscale features and model distillation.

Lung diseases globally impose a significant pathological burden and mortality rate, particularly the differential diagnosis between adenocarcinoma, squamous cell carcinoma, and small cell lung carcinoma, which is paramount in determining optimal treatment strategies and improving clinical prognoses. Faced with the challenge of improving diagnostic precision and stability, this study has developed an innovative deep learning-based model. This model employs a Feature Pyramid Network (FPN) and Squeeze-and-Excitation (SE) modules combined with a Residual Network (ResNet18), to enhance the processing capabilities for complex images and conduct multi-scale analysis of each channel's importance in classifying lung cancer. Moreover, the performance of the model is further enhanced by employing knowledge distillation from larger teacher models to more compact student models. Subjected to rigorous five-fold cross-validation, our model outperforms existing models on all performance metrics, exhibiting exceptional diagnostic accuracy. Ablation studies on various model components have verified that each addition effectively improves model performance, achieving an average accuracy of 98.84% and a Matthews Correlation Coefficient (MCC) of 98.83%. Collectively, the results indicate that our model significantly improves the accuracy of disease diagnosis, providing physicians with more precise clinical decision-making support.

Aspartate Transaminase-to-Albumin Ratio (ATAR), a Novel Prognostic Index, Predicts Outcomes in Patients with Small-Cell Lung Cancer.

BACKGROUND: Small cell lung cancer (SCLC) is characterized by high invasion rates, rapid progression, and poor prognoses. Thus, identifying SCLC patients at high risk of progression and death is critical to improve long-term survival. In this study, the aspartate transaminase-to-albumin ratio (ATAR) was examined as a prognostic factor for SCLC patients. METHODS: We screened 196 SCLC patients from December 2013 to September 2022 at the Sichuan Cancer Hospital. The data was collected from patients' medical information as well as from their blood results during diagnosis. Using the Youden index as a cutoff value, patients were divided into high-risk（> 0.54) and low-risk (≤ 0.54) ATAR groups. We analyzed the prognostic factors for overall survival (OS) using the Kaplan-Meier method, univariate and multivariate analyses, Cox regression, and the C-index. RESULTS: There were 109 (55.6%) smokers among the patients, and the median OS was 17.55 months. The Kaplan-Meier analysis indicated that patients with high-risk ATAR had significantly lower OS (p < 0.0001). A multivariate analysis demonstrated that elevated ATAR is an independent adverse predictor of OS (p < 0.001, HR = 1.907). Our study found that ATAR is an independent predictor of survival outcomes in SCLC, which was superior to ALB, PNI, and SII in predicting outcomes in low-risk and high-risk groups (all p < 0.05). Models combining ATAR with ALB, PNI, and SII showed more powerful prognostic value than their corresponding original models. Moreover, the prognostic indicator ATAR can significantly stratify stage I - II and III - IV SCLC patients (p ＜ 0.05). CONCLUSIONS: Peripheral blood ATAR prognostic index can be used as an independent predictor of SCLC patients before treatment.

A Multiplexed Approach to Assess Small Cell Lung Cancer Subtype Heterogeneity in Primary and Patient-Derived Tumor Samples.

Unlocking the heterogeneity of cancers is crucial for developing therapeutic approaches that effectively eradicate disease. As our understanding of markers specific to cancer subclones or subtypes expands, there is a growing demand for advanced technologies that enable the simultaneous investigation of multiple targets within an individual tumor sample. Indeed, multiplex approaches offer distinct benefits, particularly when tumor specimens are small and scarce. Here we describe the utility of two fluorescence-based multiplex approaches; fluorescent Western blots, and multiplex immunohistochemistry (Opal™) staining to interrogate heterogeneity, using small cell lung cancer as an example. Critically, the coupling of Opal™ staining with advanced image quantitation, permits the dissection of cancer cell phenotypes at a single cell level. These approaches can be applied to patient biopsies and/or patient-derived xenograft (PDX) models and serve as powerful methodologies for assessing tumor cell heterogeneity in response to therapy or between metastatic lesions across diverse tissue sites.

Application Value of CYFRA21-1 Combined with NSE, CEA, and SCC-Ag in Lung Cancer.

BACKGROUND: This study aims to investigate the application value of serum cytokeratin 19 fragment (CYFRA21-1) combined with nerve-specific enolase (NSE), carcinoembryonic antigen (CEA), and squamous cell carcinoma antigen (SCC-Ag) in the diagnosis of lung cancer (LC). METHODS: A total of 831 cases of LC, 360 cases of benign lung disease (BLD) and 102 healthy controls, were enrolled. The data were processed using SPSS, GraphPad Prism, and MedCalc software. RESULTS: The tumor marker (TM) levels in the LC and BLD groups were significantly higher than those in the control group; the CYFRA21-1, NSE, and CEA levels in the patients with LC were higher than in those with BLD. In particular, the increase was predominantly observed for the levels of CEA and CYFRA21-1 in adenocarcinoma (LUAD), CYFRA21-1 and SCC-Ag in squamous cell carcinoma (LUSC), and NSE in small cell carcinoma (SCLC). The CYFRA21-1, NSE, and CEA levels were significantly higher in stage IV than in other stages in LC. Univariate binary logistic analysis showed that increased levels of all four TMs were risk factors for BLD and LC. The area under the curve (AUC) of CYFRA21-1 was most effective in distinguishing patients with BLD or LC from the controls and in distinguishing patients with BLD and LC. The AUCs of combined CYFRA21-1, NSE, and CEA were increased to 0.755, 0.922, and 0.783, respectively, with no significant difference with the AUC of the four combined tests. In the histological classification, the best predictors were CEA, for LUAD, CYFRA21-1 for LUSC, and NSE for SCLC. Moreover, the expression levels of CYFRA21-1, NSE, and CEA significantly decreased after each treatment course. CONCLUSIONS: The combined assay of CYFRA21-1, NSE, and CEA addresses the aspects of accuracy, sensitivity, specificity, and economic cost and should be considered as a potential diagnostic test in LC.

Small-Cell Lung Cancer: Is Liquid Biopsy a New Tool Able to Predict the Efficacy of Immunotherapy?

Small-cell lung cancer (SCLC) cases represent approximately 15% of all lung cancer cases, remaining a recalcitrant malignancy with poor survival and few treatment options. In the last few years, the addition of immunotherapy to chemotherapy improved clinical outcomes compared to chemotherapy alone, resulting in the current standard of care for SCLC. However, the advantage of immunotherapy only applies to a few SCLC patients, and predictive biomarkers selection are lacking for SCLC. In particular, due to some features of SCLC, such as high heterogeneity, elevated cell plasticity, and low-quality tissue samples, SCLC biopsies cannot be used as biomarkers. Therefore, the characterization of the tumor and, subsequently, the selection of an appropriate therapeutic combination may benefit greatly from liquid biopsy. Soluble factors, circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), and extracellular vesicles (EVs) are now useful tools in the characterization of SCLC. This review summarizes the most recent data on biomarkers detectable with liquid biopsy, emphasizing their role in supporting tumor detection and their potential role in SCLC treatment choice.

Real-life clinical management patterns in extensive-stage small cell lung cancer across France: a multi-method study.

BACKGROUND: We designed this study based on both a physician practice survey and real-world patient data to: (1) evaluate clinical management practices in extensive-stage small cell lung cancer (ES-SCLC) among medical centers located across France; and (2) describe first-line treatment patterns among patients with ES-SCLC following the introduction of immunotherapy into clinical practice. METHODS: A 50-item questionnaire was completed by physicians from 45 medical centers specialized in SCLC management. Responses were collected from June 2022 to January 2023. The survey questions addressed diagnostic workup of ES-SCLC, chemoimmunotherapy in first-line and second-line settings, and use of prophylactic cranial irradiation (PCI) and radiotherapy. In parallel, using a chart review approach, we retrospectively analyzed aggregated information from 548 adults with confirmed ES-SCLC receiving first-line treatment in the same centers. RESULTS: In ES-SCLC, treatment planning is based on chest computed tomography (CT) (as declared by 100% of surveyed centers). Mean time between diagnosis and treatment initiation was 2-7 days, as declared by 82% of centers. For detection of brain metastases, the most common imaging test was brain CT (84%). The main exclusion criteria for first-line immunotherapy in the centers were autoimmune disease (87%), corticosteroid therapy (69%), interstitial lung disease (69%), and performance status ≥ 2 (69%). Overall, 53% and 36% of centers considered that patients are chemotherapy-sensitive if they relapse within ≥ 3 months or ≥ 6 months after first-line chemoimmunotherapy, respectively. Among the 548 analyzed patients, 409 (75%) received chemoimmunotherapy as a first-line treatment, 374 (91%) of whom received carboplatin plus etoposide and 35 (9%) cisplatin plus etoposide. Overall, 340/548 patients (62%) received maintenance immunotherapy. Most patients (68%) did not receive radiotherapy or PCI. CONCLUSIONS: There is an overall alignment of practices reflecting recent clinical guidelines among medical centers managing ES-SCLC across France, and a high prescription rate of immunotherapy in the first-line setting.

Abnormal expression of LCA and CD43 in SCLC: a rare case report and brief literature review.

BACKGROUND: To present an unusual case of abnormal LCA expression and CD43 in SCLC and to review the reported literature to avoid potential diagnostic pitfalls. CASE PRESENTATION: A 73-year-old male patient suffered from persistent back pain for more than one month. MRI revealed a compression fracture of the L1-L5 vertebra. A CT scan revealed multiple nodules and masses at the left root of the neck, lung hilum and mediastinum, and multiple areas of bony destruction of the ribs. Histology of the tumor revealed that small and round cells were arranged in nests with areas of necrosis. The tumor cells were round to ovoid with scant cytoplasm and indistinct cell borders. The nuclear chromatin was finely granular, and the nucleoli were absent or inconspicuous. Immunohistochemically, the tumor cells were positive for cytokeratin, TTF-1, POU2F3, LCA, and CD43. CONCLUSION: This report highlights a potential diagnostic pitfall in the diagnosis of SCLC, urges pathologists to exercise caution in cases of LCA and CD43 positivity and illustrates the need for further immunohistochemical studies to avoid misdiagnosis.

SOX-1 antibodies positive Lambert-Eaton myasthenic syndrome with occult small cell lung cancer: A case report.

Lambert-Eaton myasthenic syndrome (LEMS) is a rare paraneoplastic neurological syndrome of the neuromuscular transmission. The symptoms often progress slowly and can be misdiagnosed in early stage. Seropositive SOX-1 antibodies are support for the diagnosis of LEMS and have high specificity for small cell lung cancer (SCLC). In this paper, we report a case of a 56-year-old man with smoking history who was admitted to hospital with progressive muscle weakness of the proximal legs. LEMS was diagnosed by repetitive nerve stimulation (RNS) testing and seropositive SOX-1 antibodies. Primary screening with chest computed tomography (CT) and integrated PET/CT did not reveal any tumor. After continuous follow-up, SCLC was found by chest CT and confirmed with pathological examination 10 months after the diagnosis of LEMS. Long-term follow-up and screening for occult SCLC in LEMS patients with positive SOX-1 antibodies are very important.

Novel electrochemical platform based on C3N4-graphene composite for the detection of neuron-specific enolase as a biomarker for lung cancer.

Lung cancer remains the leading cause of cancer mortality worldwide. Small cell lung cancer (SCLC) accounts for 10-15% of cases and has an overall 5-years survival rate of only 15%. Neuron-specific enolase (NSE) has been identified as a useful biomarker for early SCLC diagnosis and therapeutic monitoring. This work reports an electrochemical immunosensing platform based on a graphene-graphitic carbon nitride (g-C3N4) nanocomposite for ultrasensitive NSE detection. The g-C3N4 nanosheets and graphene nanosheets were synthesized via liquid exfoliation and integrated through self-assembly to form the nanocomposite. This nanocomposite was used to modify screen-printed carbon electrodes followed by covalent immobilization of anti-NSE antibodies. The unique properties of the graphene-g-C3N4 composite facilitated efficient antibody loading while also enhancing electron transfer efficiency and electrochemical response. Systematic optimization of experimental parameters was performed. The immunosensor exhibited a wide linear detection range of 10 pg/mL to 100 ng/mL and low limit of detection of 3 pg/mL for NSE along with excellent selectivity against interferences. Real serum matrix analysis validated the applicability of the developed platform for sensitive and accurate NSE quantifica-tion at clinically relevant levels. This novel graphene-g-C3N4 nanocomposite based electro-chemical immunoassay demonstrates great promise for early diagnosis of SCLC.

All That Is Small Is Not a Small-Cell Carcinoma: Thoracic SMARCA4-Deficient Undifferentiated Tumors Masquerading as SCLC.

Small-cell lung carcinoma (SCLC) cell lines have been widely utilized as a preclinical model of this highly aggressive disease. However, since their creation decades ago, novel tumor entities have been defined that might clinicopathologically mimic SCLC, which notably includes thoracic SMARCA4-deficient undifferentiated tumor (SMARCA4-UT). Multiomic reassessment of the presumed SCLC cell lines with high YAP1 expression reveals that nearly all of these tumors represent unsuspected SMARCA4-UT. See related article by Ng et al., p. 1846.

Improvement of differential diagnosis of lung cancer by use of multiple protein tumor marker combinations.

BACKGROUND: Differential diagnosis of non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC) in hospitalized patients is crucial for appropriate treatment choice. OBJECTIVE: To investigate the relevance of serum tumor markers (STMs) and their combinations for the differentiation of NSCLC and SCLC subtypes. METHODS: Between 2000 and 2003, 10 established STMs were assessed retrospectively in 311 patients with NSCLC, 128 with SCLC prior systemic first-line therapy and 51 controls with benign lung diseases (BLD), by automatized electrochemiluminescence immunoassay technology. Receiver operating characteristic (ROC) curves and logistic regression analyses were used to evaluate the diagnostic efficacy of both individual and multiple STMs with corresponding sensitivities at 90% specificity. Standards for Reporting of Diagnostic Accuracy (STARD guidelines) were followed. RESULTS: CYFRA 21-1 (cytokeratin-19 fragment), CEA (carcinoembryonic antigen) and NSE (neuron specific enolase) were significantly higher in all lung cancers vs BLD, reaching AUCs of 0.81 (95% CI 0.76-0.87), 0.78 (0.73-0.84), and 0.88 (0.84-0.93), respectively. By the three marker combination, the discrimination between benign and all malignant cases was improved resulting in an AUC of 0.93 (95% CI 0.90-0.96). In NSCLC vs. BLD, CYFRA 21-1, CEA and NSE were best discriminative STMs, with AUCs of 0.86 (95% CI 0.81-0.91), 0.80 (0.74-0.85), and 0.85 (0.79-0.91). The three marker combination also improved the AUC: 0.92; 95% CI 0.89-0.96). In SCLC vs. BLD, ProGRP (pro-gastrin-releasing peptide) and NSE were best discriminative STMs, with AUCs of 0.89 (95% CI 0.84-0.94) and 0.96 (0.93-0.98), respectively, and slightly improved AUC of 0.97 (95% CI 0.95-0.99) when in combination. Finally, discrimination between SCLC and NSCLC was possible by ProGRP (AUC 0.86; 95% CI 0.81-0.91), NSE (AUC 0.83; 0.78-0.88) and CYFRA 21-1 (AUC 0.69; 0.64-0.75) and by the combination of the 3 STMs (AUC 0.93; 0.91-0.96), with a sensitivity of 88% at 90% specificity. CONCLUSIONS: The results confirm the power of STM combinations for the differential diagnosis of lung cancer from benign lesions and between histological lung cancer subtypes.

Construction of a Diagnostic Model for Small Cell Lung Cancer Combining Metabolomics and Integrated Machine Learning.

BACKGROUND: To date, no study has systematically explored the potential role of serum metabolites and lipids in the diagnosis of small cell lung cancer (SCLC). Therefore, we aimed to conduct a case-cohort study that included 191 cases of SCLC, 91 patients with lung adenocarcinoma, 82 patients with squamous cell carcinoma, and 97 healthy controls. METHODS: Metabolomics and lipidomics were applied to analyze different metabolites and lipids in the serum of these patients. The SCLC diagnosis model (d-model) was constructed using an integrated machine learning technology and a training cohort (n = 323) and was validated in a testing cohort (n=138). RESULTS: Eight metabolites, including 1-mristoyl-sn-glycero-3-phosphocholine, 16b-hydroxyestradiol, 3-phosphoserine, cholesteryl sulfate, D-lyxose, dioctyl phthalate, DL-lactate and Leu-Phe, were successfully selected to distinguish SCLC from controls. The d-model was constructed based on these 8 metabolites and showed improved diagnostic performance for SCLC, with the area under curve (AUC) of 0.933 in the training cohort and 0.922 in the testing cohort. Importantly, the d-model still had an excellent diagnostic performance after adjusting the stage and related clinical variables and, combined with the progastrin-releasing peptide (ProGRP), showed the best diagnostic performance with 0.975 of AUC for limited-stage patients. CONCLUSION: This study is the first to analyze the difference between metabolomics and lipidomics and to construct a d-model to detect SCLC using integrated machine learning. This study may be of great significance for the screening and early diagnosis of SCLC patients.

Lack of clinical utility of serum macrophage migration inhibitory factor (MIF) for monitoring therapy response and estimating prognosis in advanced lung cancer.

BACKGROUND: Lung cancer is a major burden to global health and is still among the most frequent and most lethal malignant diseases. Macrophage migration inhibitory factor (MIF) is a proinflammatory cytokine involved in a variety of processes including tumorigenesis, formation of a tumor microenvironment and metastasis. It is therefore a potential prognostic biomarker in malignant diseases. OBJECTIVE: In this study, we investigated the applicability of MIF in serum samples as a biomarker in lung cancer. METHODS: In a retrospective approach, we analyzed the sera of 79 patients with non-small-cell lung cancer (NSCLC) and 14 patients with small-cell lung cancer (SCLC) before the start of chemotherapy, as well as before the second and third chemotherapy cycle, respectively. Serum MIF levels were measured using a sandwich immunoassay with a sulfo-tag-labelled detection antibody, while pro-gastrin releasing peptide (proGRP) levels were determined with an enzyme-linked immunosorbent assay. RESULTS: No difference in serum MIF levels between responders and non-responders to chemotherapy was observed at all time points, while proGRP levels were significantly lower in responders before the second chemotherapy cycle (p = 0.012). No differences in biomarker levels depending on the histopathological classification of NSCLC patients was found. Moreover, in ROC curve analyses MIF was not able to distinguish between responders and non-responders to therapy. proGRP could differentiate between responders and non-responders before the second chemotherapy cycle (p = 0.015) with sensitivities of 43% at 90% and 95% specificity, respectively. Likewise, proGRP yielded significantly longer survival times of patients with low proGRP concentrations before the second chemotherapy cycle (p = 0.015) in Kaplan-Meier analyses, yet MIF showed no significant differences in survival times at all time points. Comparison with the biomarkers CEA and CYFRA 21-1 in the same cohort showed that these established biomarkers clearly performed superior to MIF and proGRP. CONCLUSIONS: From the present results, there is no indication that serum MIF may serve as a biomarker in prognosis and monitoring of response to therapy in lung cancer. Limitations of this study include its retrospective design, the inclusion of a larger NSCLC and a smaller SCLC subgroup, the classical chemotherapeutic treatment, the use of a non-diagnostic immunoassay (RUO-test) for MIF measurement and the lack of a validation cohort. Strengths of the study are its highly standardized procedures concerning sample collection, preanalytic treatment, measurements and quality control of the laboratory assays.

Real-World comprehensive genomic profiling data for diagnostic clarity in pulmonary Large-Cell neuroendocrine carcinoma.

BACKGROUND: Pulmonary large-cell neuroendocrine carcinoma (LCNEC) is an uncommon subtype of lung cancer believed to represent a spectrum of tumors sharing characteristics of both small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). Other groups have proposed genomic LCNEC subtypes, including small cell-like, non-small cell-like, and carcinoid-like subtypes. The primary goal of this study was to better define the NSCLC-like subtype with comprehensive genomic profiling (CGP). METHODS: An institutional database was queried to identify tissue specimens (TBx, N = 1,426) and liquid biopsies (LBx, N = 39) submitted for CGP during routine clinical care (8/2014 - 7/2023) with a disease ontology of LCNEC. TBx were profiled with FoundationOne® (F1) or F1CDx, using hybrid-capture technology to detect genomic alterations (GAs). RESULTS: 1,426 LCNEC samples were genomically profiled. The presence of RB1 and TP53 genomic alterations (GAs) were used to define a SCLC-like subtype (n = 557). A carcinoid-like group was defined by the presence of MEN1 mutation in the absence of TP53 GAs (n = 25). The remaining 844 samples were compared to the SCLC-like group and GAs enriched relative to the SCLC-like samples with a false discovery rate (FDR) < 0.0001 were used to define a NSCLC-like group. These NSCLC-like subtype-defining GAs included SMARCA4, KRAS, FGF3/4/19, STK11, CDKN2A/B, MTAP, and CCND1. Under this schema, 530 samples were classified as NSCLC-like and 314 remained unclassified. CONCLUSIONS: Large-scale CGP can better characterize biologically distinct molecular subtypes in LCNEC. Further studies to define how these molecular subtypes may help inform treatment decisions in this complex and challenging malignancy are warranted.

When chronic obstructive pulmonary disease meets small cell lung cancer: an unusual case report of rapid progression.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory disease and a risk factor for lung cancer. Small cell lung cancer is a neuroendocrine tumor with a high degree of malignancy and an overall five-year survival rate of less than 7%. CASES PRESENTATION: Herein, we report the case of an 68-year-old male presented to the respiratory department with cough, sputum, and dyspnea. He was diagnosed as community acquired pneumonia and treated with intravenous anti-infection. Previous pulmonary function was definitively diagnosed as COPD. About 7 months after discharge, the patient returned to the hospital for cough and dyspnea. After diagnosis of the tumor, cisplatin, etoposide and durvalumab were administered. Finally the patient died of respiratory failure approximately 9 months after his diagnosis. CONCLUSIONS: For COPD patients with immunocompromised manifestations, it is necessary to be alert to complications and shorten the follow-up interval of chest CT. COPD may accelerate the formation and progression of SCLC.

Hyperinflation and reduced diffusing capacity predict prognosis in SCLC: value of extended pre-therapeutic lung function testing.

BACKGROUND: Small cell lung cancer (SCLC) is characterized by aggressive growth and poor prognosis. Although SCLC affects nearly exclusively heavy smokers and leads to frequent respiratory symptoms, the impact of pre-therapeutic lung function testing in SCLC is sparely investigated until now. Therefore, we sought to examine whether we could find prognostic markers in pre-therapeutic lung function testing of SCLC patients. PATIENTS AND METHODS: We retrospectively analysed a cohort of 205 patients with the diagnosis of SCLC between 2010 and 2018. Pre-therapeutic values of spirometry, body plethysmography and measurement of diffusing capacity was extracted from patients' charts. Comparisons between groups were performed using the Mann-Whitney U-test or by chi-square tests as appropriate. Kaplan-Meier analyses and COX-regression models were performed to correlate lung function parameters with patients' outcome. RESULTS: Airway obstruction itself, or the diagnosis chronic obstructive pulmonary disease (COPD) based on GOLD definitions did not correlate with survival in SCLC patients. Hyperinflation measured by increased residual volume and residual volume to total lung capacity ratio (log-rank p < 0.001) and reduced diffusing capacity (log-rank p = 0.007) were associated with reduced survival. Furthermore, patients with hyperinflation as well as impairments in gas exchange representing an emphysematic phenotype had the worst outcome (log-rank p < 0.001). CONCLUSION: We recommend including body plethysmography and measurement of diffusing capacity in the pre-therapeutic assessment of SCLC patients. Our findings suggest that reduction of hyperinflation may lead to better outcome in SCLC patients. Thus, in addition to effective tumour therapy, adequate therapy of the comorbidity of COPD should also be provided. In particular, measures to reduce hyperinflation by means of dual bronchodilation as well as respiratory physiotherapy should be further assessed in this setting.

Identification of a visualized web-based nomogram for overall survival prediction in patients with limited stage small cell lung cancer.

Small-cell lung cancer (SCLC) is an aggressive lung cancer subtype with an extremely poor prognosis. The 5-year survival rate for limited-stage (LS)-SCLC cancer is 10-13%, while the rate for extensive-stage SCLC cancer is only 1-2%. Given the crucial role of the tumor stage in the disease course, a well-constructed prognostic model is warranted for patients with LS-SCLC. The LS-SCLC patients' clinical data extracted from the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2018 were reviewed. A multivariable Cox regression approach was utilized to identify and integrate significant prognostic factors. Bootstrap resampling was used to validate the model internally. The Area Under Curve (AUC) and calibration curve evaluated the model's performance. A total of 5463 LS-SCLC patients' clinical data was collected from the database. Eight clinical parameters were identified as significant prognostic factors for LS-SCLC patients' OS. The predictive model achieved satisfactory discrimination capacity, with 1-, 2-, and 3-year AUC values of 0.91, 0.88, and 0.87 in the training cohort; and 0.87, 0.87, and 0.85 in the validation cohort. The calibration curve showed a good agreement with actual observations in survival rate probability. Further, substantial differences between survival curves of the different risk groups stratified by prognostic scores were observed. The nomogram was then deployed into a website server for ease of access. This study developed a nomogram and a web-based predictor for predicting the overall survival of patients with LS-SCLC, which may help physicians make personalized clinical decisions and treatment strategies.

Comparation between novel online models and the AJCC 8th TNM staging system in predicting cancer-specific and overall survival of small cell lung cancer.

Background: Most of previous studies on predictive models for patients with small cell lung cancer (SCLC) were single institutional studies or showed relatively low Harrell concordance index (C-index) values. To build an optimal nomogram, we collected clinicopathological characteristics of SCLC patients from Surveillance, Epidemiology, and End Results (SEER) database. Methods: 24,055 samples with SCLC from 2010 to 2016 in the SEER database were analyzed. The samples were grouped into derivation cohort (n=20,075) and external validation cohort (n=3,980) based on America's different geographic regions. Cox regression analyses were used to construct nomograms predicting cancer-specific survival (CSS) and overall survival (OS) using derivation cohort. The nomograms were internally validated by bootstrapping technique and externally validated by calibration plots. C-index was computed to compare the accuracy and discrimination power of our nomograms with the 8th of version AJCC TNM staging system and nomograms built in previous studies. Decision curve analysis (DCA) was applied to explore whether the nomograms had better clinical efficiency than the 8th version of AJCC TNM staging system. Results: Age, sex, race, marital status, primary site, differentiation, T classification, N classification, M classification, surgical type, lymph node ratio, radiotherapy, and chemotherapy were chosen as predictors of CSS and OS for SCLC by stepwise multivariable regression and were put into the nomograms. Internal and external validations confirmed the nomograms were accurate in prediction. C-indexes of the nomograms were relatively satisfactory in derivation cohort (CSS: 0.761, OS: 0.761) and external validation cohort (CSS: 0.764, OS: 0.764). The accuracy of the nomograms was superior to that of nomograms built in previous studies. DCA showed the nomograms conferred better clinical efficiency than 8th version of TNM staging system. Conclusions: We developed practical nomograms for CSS (https://guowei2020.shinyapps.io/DynNom-CSS-SCLC/) and OS (https://drboidedwater.shinyapps.io/DynNom-OS-SCLC/) prediction of SCLC patients which may facilitate clinicians in individualized therapeutics.