External Validation of a Prognostic Model of Survival for Resected Typical Bronchial Carcinoids

Predictive Modeling of Long-Term Prognosis After Resection in Typical Pulmonary Carcinoid: A Machine Learning Perspective

Typical Pulmonary Carcinoid (TPC) is defined by its slow growth, frequently necessitating surgical intervention. Despite this, the long-term outcomes following tumor resection are not well understood. This study examined the factors impacting Overall Survival (OS) in patients with TPC, leveraging data from the Surveillance, Epidemiology, and End Results database spanning from 2000 to 2018. We employed Lasso-Cox analysis to identify prognostic features and developed various models using Random Forest, XGBoost, and Cox regression algorithms. Subsequently, we assessed model performance using metrics such as Area Under the Curve (AUC), calibration plot, Brier score, and Decision Curve Analysis (DCA). Among the 2687 patients, we identified five clinical features significantly affecting OS. Notably, the Random Forest model exhibited strong performance, achieving 5- and 7-year AUC values of 0.744/0.757 in the training set and 0.715/0.740 in the validation set, respectively, outperforming other models. Additionally, we developed a web-based platform aimed at facilitating easy access to the model. This study presents a machine learning model and a web-based support system for healthcare professionals, assisting in personalized treatment decisions for patients with TPC post-tumor resection.

NEP-Score Thresholds Predict Survival of Patients With Bronchial Carcinoids

Survival prognostic markers are extremely needed to better define therapeutic strategies in patients with bronchial carcinoids (BC). We aim to verify the applicability of the NEP-Score in a homogeneous BC cohort and identify a derivative prognostic marker, the NEP-Score at diagnosis (NEP-D) that does not consider new metastases during follow-up. Sixty-four patients (38 females, and 26 males, mean age at diagnosis 58.9 ± 1.7 years) with BC were retrospectively evaluated. NEP-Score was calculated at the end of follow-up (NEP-T). A derivative score, the NEP-Score at diagnosis (NEP-D) that does not consider new metastases during follow-up, was then assessed. Patients were subdivided according to their living status at the end of follow-up. A NEP-Score threshold was investigated to predict survival. Mean NEP-T and mean NEP-D were significantly lower in live patients at end of follow-up. A NEP-T cut-off >138 significantly predicts survival. Atypical BC relapsed more frequently than Typical BC. Male gender and previous malignancy were negative prognostic factors for survival. We confirmed NEP-Score applicability in BC and NEP-D utility, being the latter a simple, quick, and cheap prognostic score that can help clinicians in decision making. The identified NEP-D threshold can predict NEN aggressiveness and may be used to define the best personalized therapeutic strategy. In this context, a validation study is needed.

Endobronchial and surgical treatment of pulmonary carcinoid tumors: A systematic literature review

The treatment of pulmonary carcinoid has changed over the last decades. Although surgical resection is still the gold standard, minimally invasive endobronchial procedures have emerged as a parenchyma sparing alternative for tumors located in the central airways. This review was performed to identify the optimal treatment strategy for pulmonary carcinoid, with a particular focus on the feasibility and outcome of parenchyma sparing techniques versus surgical resection. A systematic review of the literature was carried out using MEDLINE, Embase and the Cochrane databases, based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement. Two separate searches of publications in endobronchial and surgical treatment in patients with pulmonary carcinoid, were performed. Outcomes were overall survival, disease free survival, recurrence rate, complications, quality of life, and healthcare costs. Combining the two main searches for endobronchial therapy and surgical therapy yielded 3111 records. Finally, 43 studies concerning surgical treatment and 9 studies related to endobronchial treatment for pulmonary carcinoid were included. Assessment of included studies showed that lymph node involvement, histological grade, tumor location and tumor diameter were identified as poor prognostic factors and seem to be important for patients with pulmonary carcinoid. For patients with a more favorable prognosis, tumor location and tumor diameter are important factors that can help decide on the optimal treatment strategy. Centrally located small intraluminal pulmonary carcinoids, without signs of metastasis can be treated with minimally invasive alternatives such as endobronchial treatment or parenchyma sparing surgical resection. Patients with parenchyma sparing resections should be followed with long term follow up to exclude recurrence of disease. In a multidisciplinary setting, it should be determined whether individual patients are eligible for parenchyma sparing procedures or anatomical resection. Overall evidence is of low quality and future studies should focus on prospective trials in the treatment of pulmonary carcinoid.

The European Society of Thoracic Surgeons (ESTS) lung neuroendocrine tumors (NETs) database

Histological characteristics and clinical behaviour define lung neuroendocrine tumors (NETs), which are classified into four groups: typical (TC) and atypical carcinoids (AC), large-cell neuroendocrine carcinomas (LCNCs) and small-cell lung cancers (SCLCs). Historically, outcome and treatment of these rare neoplasms have been based on small, usually mono-institutional clinical series. Furthermore, their rarity makes quite impossible to design randomised clinical trial to compare different treatments especially in unusual clinical presentations. In 2012, the European Society of Thoracic Surgeons (ESTS) launched a new working-group, specifically dedicated to lung NETs, with the aim to develop modern knowledge on biology and behaviour of these tumors, and to disseminate it within the scientific community. A dedicated retrospective database was at first developed and sent to all the ESTS centres interested to this project. More than 2,000 operated NETs cases have been rapidly collected, and they represented the clinical substrate of several published scientific studies. The retrospective data collection intrinsic limitations in term of patients' selection and treatment, along with the problem of possible missing data, were the reasons why the ESTS NETs working-group decided in 2015 to design and promote a new prospective database, employing the official ESTS platform. The aim of this review paper is to report the ESTS Lung NETs working-group history and to explain the architecture and use of the lung NETs databases.

Lung and Thymic Carcinoids

Carcinoids of the lung and thymus are rare thoracic cancers. In general, lung carcinoid tumors have a favorable prognosis, particularly when diagnosed at an early stage and treated with surgical resection. Thymic neuroendocrine tumors may be associated with multiple endocrine neoplasia-1 syndrome, tend to have a more aggressive natural history, and relatively frequently secrete ectopic adrenocorticotropic hormone.

Improvement in TNM staging of pulmonary neuroendocrine tumors requires histology and regrouping of tumor size

OBJECTIVE

Neuroendocrine tumors of the lung are currently staged with the 7th edition TNM non-small cell lung cancer staging system. This decision, based on data analysis without data on histology or disease-specific survival, makes its applicability limited. This study proposes a specific staging system for these tumors.

METHODS

We retrospectively analyzed 510 consecutive patients (female/male, 313/197; median age, 61 years; interquartile range, 51-70) undergoing lung resection for a primary neuroendocrine tumor between 2000 and 2015 in 8 centers. Multivariable analysis was performed using a Cox proportional hazard model to identify factors associated with disease-specific survival. A new staging system was proposed on the basis of the results of this analysis. Kaplan-Meier disease-specific survival was analyzed by stage using the proposed and the 7th TNM staging system.

RESULTS

Follow-up was completed in 490 of 510 patients at a median of 51 months (interquartile range, 18-99). Histology (G1-typical carcinoid vs G2-atypical carcinoid vs G3-large-cell neuroendocrine carcinoma) and pT were independently associated with survival, but pN was not. After regrouping histology and pT, we proposed the following staging system: IA (pT1-2G1), IB (pT3G1, pT1G2), IIA (pT4G1, pT2-3G2, pT1G3), IIB (pT4G2, pT2-3G3), and III (pT4G3). The 5-year survivals were 97.9%, 81.0%, 69.1%, 51.8%, and 0%, respectively. By using the 7th TNM, 5-year survivals were 95.0%, 92.3%, 67.7%, 70.9%, and 65.1% for stage IA, IB, IIA, IIB, and III, respectively.

CONCLUSIONS

Incorporating histology and regrouping tumor stage create a unique neuroendocrine tumor staging system that seems to predict survival better than the 7th TNM classification.