Prognostic factors and nomogram for cancer-specific death in non small cell lung cancer with malignant pericardial effusion

Clinical features and prognostic nomogram development for cancer-specific death in patients with dual primary lung cancer: a population-based study from SEER database

OBJECTIVE

This study aimed to develop a concise and valid clinical prediction model to assess the survival prognostic risk of cancer-specific death in patients with dual primary lung cancer (DPLC).

DATA SOURCE

Surveillance, epidemiology, and end results (SEER) database.

DESIGN

A retrospective population-based study.

METHODS

Data of DPLC patients from the database from 1992 to 2020 were collected. The number of DPLC patients was determined based on the first primary LC (FPLC) and second primary LC (SPLC), and patients were randomly assigned to a training set and a testing set in a 7:3 ratio. The primary endpoint was cancer-specific survival (CSS). Kaplan-Meier survival analysis was performed to construct survival curves. Cox analysis and bilateral stepwise regression were used to analyze prognostic factors for cancer-specific death in patients and establish the nomogram. The discriminative ability of the nomogram was assayed by C-index and calibration curves, decision-making ability was assessed by decision curve analysis (DCA), and nomogram performance was measured by receiver operating characteristic (ROC) curves.

RESULTS

This study included 997 DPLC patients, divided into a training set (n = 698) and a testing set (n = 299) in a 7:3 ratio. Age, gender, histological type, surgery, chemotherapy, T stage, N stage, and tumor size were identified as risk factors affecting CSS in DPLC patients (P < 0.05) and were utilized to establish a nomogram. The C-index of the nomogram in the training set was 0.671, and the AUC values of ROC curves for 1-year, 3-year, and 5-year survival rates were 0.84, 0.78, and 0.74, respectively. The C-index of the testing set was 0.644, and the AUC values were 0.72, 0.74, and 0.75, respectively. Calibration curves for both sets were close to the diagonal line, indicating good predictive ability of the nomogram. DCA curves demonstrated the good decision-making ability of the nomogram.

CONCLUSION

This study revealed the clinical features of DPLC patients and developed an effective nomogram for predicting CSS, which can assist clinicians in making accurate and personalized clinical decisions regarding patient treatment.

Application of machine learning algorithms to predict lymph node metastasis in gastric neuroendocrine neoplasms

Background

Neuroendocrine neoplasms (NENs) are tumors that originate from secretory cells of the diffuse endocrine system and typically produce bioactive amines or peptide hormones. This paper describes the development and validation of a predictive model of the risk of lymph node metastasis among gastric NEN patients based on machine learning platform.

Methods

In this investigation, data from 1256 patients were used, of whom 119 patients from the First Affiliated Hospital of Soochow University in China and 1137 cases from the surveillance epidemiology and end results (SEER) database were combined. Six machine learning algorithms, including the logistic regression model (LR), random forest (RF), decision tree (DT), Naive Bayes (NB), support vector machine (SVM), and k-nearest neighbor algorithm (KNN), were used to build the predictive model. The performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity.

Results

Among the 1256 patients with gastric NENs, 276 patients (21.97 %) developed lymph node metastasis. T stage, tumor size, degree of differentiation, and sex were predictive factors of lymph node metastasis. The RF model achieved the best predictive performance among the six machine learning models, with an AUC, accuracy, sensitivity, and specificity of 0.81, 0.78, 0.76, and 0.82, respectively.

Conclusion

The RF model provided the best prediction and can help physicians determine the lymph node metastasis risk of gastric NEN patients to formulate individualized medical strategies.

Development and Validation of Prognostic Nomograms for Lung Squamous Cell Carcinoma With Brain Metastasis in Patients Aged 45 Years or Older: A Population-Based Study

PURPOSE

We aimed to establish nomograms to predict the survival in patients aged ≥45  years with lung squamous cell carcinoma and brain metastasis.

METHODS

We collected patients diagnosed as lung squamous cell carcinoma with brain metastasis aged ≥45 years between 2010 and 2019 from the Surveillance, Epidemiology, and End Results database. Prognostic factors were determined by the univariate and multivariate Cox regression analysis, and then the nomogram was constructed to predict cancer-specific survival and overall survival. Nomograms were evaluated by decision curve analysis, the area under the receiver operating characteristic curve, calibration plot, concordance index, and risk group stratification.

RESULTS

In total, 2437 patients were included, with 1706 and 731 in the cohorts of training and validation, respectively. The age, N stage, T stage, liver metastasis, chemotherapy, bone metastasis, along with radiotherapy were significant in predicting the survival, and adopted for the establishment of nomograms. In the training and validation sets, the concordance index were .713(95%CI:0.699-.728) & .700(95%CI:0.677-.722) in predicting cancer-specific survival and .715(95%CI:0.701-.729) & .712(95%CI:0.690-.735) in predicting overall survival, respectively. Besides, the area under the receiver operating characteristic curve for predicting cancer-specific survival and overall survival in the training set were all >.7 at 1-, 2-, and 3- years. Calibration plots proved the survival predicted by nomograms were consistent with the actual values. decision curve analysis revealed better clinical validity of the nomogram in predicting cancer-specific survival and overall survival at 1-year than TNM staging. Patients were stratified into the high-/low-risk groups according to the optimal cutoff value of 100.21 for cancer-specific survival and 91.98 for overall survival. A web-based probability calculator was constructed finally.

CONCLUSION

Two nomograms were developed for the prognostic prediction of lung squamous cell carcinoma patients with brain metastasis aged ≥45 years, providing guidance for decision-making in clinical practice.

Would the Addition of Immunotherapy Impact the Prognosis of Patients With Malignant Pericardial Effusion?

Pericardial effusion is a common finding in advanced-stage lung cancer. The presence of malignant cells or drainage of exudate effusion in the pericardial space may cause symptoms of dyspnea, pleuritic chest pain, and syncope. In addition to the difficulty physicians face in the detection and diagnosis of malignant pericardial effusion, treatment may be challenging considering the cancer prognosis and cardiovascular stability of the patient. Despite the availability of several treatment modalities for malignant pericardial effusion, including chemotherapy and surgery, patients with lung cancer historically present with poor prognoses. In addition to lung adenocarcinoma with malignant pericardial effusion, this case was complicated by COVID-19 and malignancy-associated obstructive pneumonia. We present a case of a 64-year-old woman with advanced non-small cell lung carcinoma (NSCLC) with malignant pericardial effusion who, despite testing positive for COVID-19 and having obstructive pneumonia, had favorable outcomes following systemic therapy with combined chemo-immunotherapy.

Prognostic Nomogram and Therapeutic Option of Cancer-Specific Death in the Patients with Metachronous Second Primary Lung Cancer

With the increase of long-term primary lung cancer survivors, studies focused on metachronous second primary lung cancer (SPLC) have become very urgent. This study aimed to develop a prognostic nomogram and determine therapeutic options of cancer-specific death for patients with metachronous SPLC with and without the competing risk of other-specific death. Study population came from the SEER-18 database between 2006 and 2016. According to the clinical practice guideline of SPLC, the interval time of IPLC and metachronous SPLC was set to 4 years. We constructed nomograms with Lasso + Cox regression model and competing risk model to predict the prognosis and identify therapeutic options of metachronous SPLC patients with the assessment of model performance by the C-index, calibration plot, and decision curve analysis. In addition, two subgroup analyses stratified by histology and tumor size were used to better select therapeutic options for a certain population. 1300 patients with metachronous SPLC were incorporated in this study with 50.1% of the 5-year cumulative incidence in cancer-specific death. Compared with Lasso + Cox regression analysis, competing risk analysis harbored a higher C-index (0.811 vs. 0.76) and better net benefit in predicting cancer-specific death of metachronous SPLC. Two statistical analyses suggested that surgery alone was a preferentially therapeutic option of metachronous SPLC, whereas the effect of surgery + radiation in treating metachronous SPLC was similar to radiation alone. Subgroup analyses indicated that patients with metachronous SPLC were considered receiving different therapeutic options in different histology and tumor size but preferred to receive surgical treatment as the first choice. For primary lung cancer survivors, aggressive surgical treatment was the first-line selection of metachronous SPLC, followed by radiation alone, surgery + radiation, and no surgery + radiation.

Development and External Validation of a Nomogram for Predicting Cancer-Specific Survival of Non-Small Cell Lung Cancer Patients With Ipsilateral Pleural Dissemination

Background

Non-small-cell lung cancer (NSCLC) patients with ipsilateral pleural dissemination are defined as M1a in the eighth of American Joint Committee on Cancer (AJCC) TNM staging. We aimed to build a nomogram to predict lung cancer specific survival (LCSS) of NSCLC patients with ipsilateral pleural dissemination and to compare the impact of primary tumor resection (PTR) on LCSS among patients with different features.

Methods

A total of 3,918 NSCLC patients with ipsilateral pleural dissemination were identified from the Surveillance, Epidemiology, and End Results (SEER) database. We selected and integrated significant prognostic factors based on competing risk regression to build a nomogram. The model was subjected to internal validation within SEER cohort and external validation with the cohort of 97 patients from Peking University People’s Hospital.

Results

Age (P < 0.001), gender (P = 0.037), T stage (P = 0.002), N stage (P < 0.001), metastasis pattern (P = 0.005), chemotherapy (P < 0.001), and PTR (P < 0.001) were independent prognostic factors. The calibration curves presented a good consistency and the Harrell’s C-index of nomogram were 0.682 (95%CI: 0.673–0.691), 0.687 (95%CI: 0.670–0.704) and 0.667 (95%CI: 0.584–0.750) in training, internal, and external validation cohort, respectively. Interaction tests suggested a greater LCSS difference caused by PTR in patients without chemotherapy (P < 0.001).

Conclusions

We developed a nomogram based on competing risk regression to reliably predict prognosis of NSCLC patients with ipsilateral pleural dissemination and validated this nomogram in an external Chinese cohort. This novel nomogram might be a practical tool for clinicians to anticipate the 1-, 3- and 5-year LCSS for NSCLC patients with pleural dissemination. Subgroup analysis indicated that patients without chemotherapy could get more benefit from PTR. In order to assess the role of PTR in the management of M1a patients more accurately, further prospective study would be urgently required.

Case study-based systematic review of literature on lymphoma-associated cardiac tamponade

This study aimed to compile all the relevant studies of patients presenting with pericardial tamponade before or after diagnosis of lymphoma, describe the clinical presentations of patients with lymphoma and cardiac tamponade, and assess the difference in overall survival based on the timing of cardiac tamponade diagnosis.

A comprehensive search strategy was conducted in the following databases: PubMed and Cochrane Library, using the following keywords: Lymphoma AND Cardiac Tamponade. The criteria for eligibility included cases with a confirmed diagnosis of lymphoma and cardiac tamponade, human studies, and publications in English language. The statistical analysis was performed using IBM Statistical Package for Social Sciences (SPSS) version 20. We included 48 research articles (n = 52 cases) with adequate reporting of measured outcomes. The median age of the patients was 52 (9–94) years. Only 6 patients were noted to have primary cardiac lymphoma, while the majority of cases were considered to have secondary cardiac lymphoma (88.5%). According to the data on the type of lymphoma reported through cytology and immunohistochemistry, 49 patients were diagnosed with non-Hodgkin lymphoma, and of these cases the most common subtype was large B-cell lymphoma (42.9%). Overall, the average duration of illness was 14 ± 23 days. A total of 13 patients had distant heart sounds, 12 cases were noted to be hypotensive, and 13 subjects were found to have increased jugular venous pressure.

Our retrospective study demonstrated that most patients presented with pericardial tamponade after lymphoma diagnosis, and those were mostly secondary cardiac lymphoma of the non-Hodgkin type with large B-cell as the most common subtype. Dyspnoea, oedema, and constitutional symptoms were the most common presenting signs. The median overall survival of patients with lymphoma and cardiac tamponade is 4 months, with no significant difference in mortality in the presentation timing before and after the diagnosis of lymphoma.

Predict multicategory causes of death in lung cancer patients using clinicopathologic factors

BACKGROUND

Random forests (RF) is a widely used machine-learning algorithm, and outperforms many other machine learning algorithms in prediction-accuracy. But it is rarely used for predicting causes of death (COD) in cancer patients. On the other hand, multicategory COD are difficult to classify in lung cancer patients, largely because they have multiple labels (versus binary labels).

METHODS

We tuned RF algorithms to classify 5-category COD among the lung cancer patients in the surveillance, epidemiology and end results-18, whose lung cancers were diagnosed in 2004, for the completeness in their follow-up. The patients were randomly divided into training and validation sets (1:1 and 4:1 sample-splits). We compared the prediction accuracy of the tuned RF and multinomial logistic regression (MLR) models.

RESULTS

We included 42,257 qualified lung cancers in the database. The COD were lung cancer (72.41%), other causes or alive (14.43%), non-lung cancer (6.85%), cardiovascular disease (5.35%), and infection (0.96%). The tuned RF model with 300 iterations and 10 variables outperformed the MLR model (accuracy = 69.8% vs 64.6%, 1:1 sample-split), while 4:1 sample-split produced lower prediction-accuracy than 1:1 sample-split. The top-10 important factors in the RF model were sex, chemotherapy status, age (65+ vs < 65 years), radiotherapy status, nodal status, T category, histology type and laterality, all of which except T category and laterality were also important in MLR model.

CONCLUSION

We tuned RF models to predict 5-category CODs in lung cancer patients, and show RF outperforms MLR in prediction accuracy. We also identified the factors associated with these COD.