A novel model to predict cancer-specific survival in patients with early-stage uterine papillary serous carcinoma (UPSC)

The development and external validation of a web-based nomogram for predicting overall survival with Ewing sarcoma in children

Background

Ewing sarcoma remains the second most prevalent primary aggressive bone tumor in teens and young adults. The aim of our study was to develop and validate a web-based nomogram to predict the overall survival for Ewing sarcoma in children.

Methods

A total of 698 patients, with 640 cases from the Surveillance, Epidemiology, and End Results (the training set) and 58 cases (the external validation set), were included in this study. Cox analyses were carried out to determine the independent prognostic indicators, which were further included to establish a web-based nomogram. The predictive abilities were tested through the concordance index, calibration curve, decision curve analysis, and area under the receiver operating characteristic curve.

Results

As suggested by univariate and multivariate Cox analyses, age, primary site, tumor size, metastasis stage (M stage), and chemotherapy were included as the independent predictive variables. The area under the receiver operating characteristic curve values, calibration curves, concordance index, and decision curve analysis from training and validation groups suggested the model has great clinical applications.

Conclusion

We developed a convenient and precise web-based nomogram to evaluate overall survival for Ewing sarcoma in children. The application of this nomogram would assist physicians and patients in making decisions.

Application of machine learning approaches in predicting clinical outcomes in older adults - a systematic review and meta-analysis

BACKGROUND

Machine learning-based prediction models have the potential to have a considerable positive impact on geriatric care.

DESIGN

Systematic review and meta-analyses.

PARTICIPANTS

Older adults (≥ 65 years) in any setting.

INTERVENTION

Machine learning models for predicting clinical outcomes in older adults were evaluated. A random-effects meta-analysis was conducted in two grouped cohorts, where the predictive models were compared based on their performance in predicting mortality i) under and including 6 months ii) over 6 months.

OUTCOME MEASURES

Studies were grouped into two groups by the clinical outcome, and the models were compared based on the area under the receiver operating characteristic curve metric.

RESULTS

Thirty-seven studies that satisfied the systematic review criteria were appraised, and eight studies predicting a mortality outcome were included in the meta-analyses. We could only pool studies by mortality as there were inconsistent definitions and sparse data to pool studies for other clinical outcomes. The area under the receiver operating characteristic curve from the meta-analysis yielded a summary estimate of 0.80 (95% CI: 0.76 - 0.84) for mortality within 6 months and 0.81 (95% CI: 0.76 - 0.86) for mortality over 6 months, signifying good discriminatory power.

CONCLUSION

The meta-analysis indicates that machine learning models display good discriminatory power in predicting mortality. However, more large-scale validation studies are necessary. As electronic healthcare databases grow larger and more comprehensive, the available computational power increases and machine learning models become more sophisticated; there should be an effort to integrate these models into a larger research setting to predict various clinical outcomes.

Development and Validation of a Prognostic Model for Overall Survival in Patients with Primary Pelvis and Spine Osteosarcoma: A Population-Based Study and External Validation

Background: Primary pelvis and spine osteosarcoma (PSOS) is a specific type of osteosarcoma that is difficult to treat and has a poor prognosis. In recent years, the research on osteosarcoma has been increasing, but there have been few studies on PSOS; in particular, there have been a lack of analyses with a large sample size. This study aimed to construct and validate a model to predict the overall survival (OS) of PSOS patients, as currently there are no tools available for assessing their prognosis. Methods: Data including demographic information, clinical characteristics, and follow-up information on patients with PSOS were collected from the Surveillance, Epidemiology, and End Results (SEER) database, as well as from the Spine Tumor Center of Changzheng Hospital. Variable selection was achieved through a backward procedure based on the Akaike Information Criterion (AIC). Prognostic factors were identified by univariate and multivariate Cox analysis. A nomogram was further constructed for the estimation of 1-, 3-, and 5-year OS. Calibration plots, the concordance index (C-index), and the receiver operating characteristic (ROC) were used to evaluate the prediction model. Results: In total, 83 PSOS patients and 90 PSOS patients were separately collected from the SEER database and Changzheng Hospital. In the SEER cohort, liver metastasis, lung metastasis, and chemotherapy were recognized as independent prognostic factors for OS (p < 0.05) and were incorporated to construct the initial nomogram. However, the initial nomogram showed poor predictive accuracy in internal and external validation. Then, we shifted our focus to the Changzheng data. Lung metastasis involving segments, Eastern Cooperative Oncology Group (ECOG) performance score, alkaline phosphatase (ALP) level, and en bloc resection were ultimately identified as independent prognostic factors for OS (p < 0.05) and were further incorporated to construct the current nomogram, of which the bias-corrected C-index was 0.834 (0.824–0.856). The areas under the ROC curves (AUCs) of the current nomogram regarding 1-, 3-, and 5-year OS probabilities were 0.93, 0.96, and 0.92, respectively. Conclusion: We have developed a predictive model with satisfactory performance and clinical practicability, enabling effective prediction of the OS of PSOS patients and aiding clinicians in decision-making.

Conditional estimates for uterine serous cancer: Tools for survivorship counseling and planning

OBJECTIVES

Develop conditional survival and risk-assessment estimates for uterine serous carcinoma (USC) overall and stratified by stage as tools for annual survivorship counseling and care planning.

METHODS

Patients in the National Cancer Data Base diagnosed between 2004 and 2014 with stage I-IV USC were eligible. Individuals missing stage or survival data or with multiple malignancies were excluded. Five-year conditional survival was estimated using the stage-stratified Kaplan-Meier method annually during follow-up. A standardized mortality ratio (SMR) estimated the proportion of observed to expected deaths in the U.S. adjusted for year, age, and race. The relationships between prognostic factors and survival were studied using multivariate Cox modeling at diagnosis and conditioned on surviving 5-years.

RESULTS

There were 14,575 participants, including 43% with stage I, 8% with stage II, 29% with stage III, and 20% with stage IV USC. Five-year survival at diagnosis vs. after surviving 5-years was 52% vs. 75% overall, 77% vs. 81% for stage I, 57% vs. 72% for stage II, 40% vs. 66% for stage III, and 17% vs. 60% for stage IV USC, respectively (P < 0.0001). Incremental improvements in 5-year conditional survival and reductions in SMR tracked with annual follow-up and higher stage. The adjusted risk of death at diagnosis vs. after surviving 5-years was 1.15 vs. 1.40 per 5-year increase of age, 1.26 vs. 1.68 for Medicaid insurance, 3.92 vs. 2.48 for stage III disease, and 6.65 vs. 2.79 for stage IV disease, respectively (P < 0.0001).

CONCLUSION

In USC, the evolution of conditional survival permits annual reassessments of prognosis to tailor survivorship counseling and care planning.

A Competing Risk-based Prognostic Model to Predict Cancer-specific Death of Patients with Spinal and Pelvic Chondrosarcoma

STUDY DESIGN

Retrospective analysis.

OBJECTIVE

The aim of this study was to develop and validate a competing-risk-based prognostic model and a nomogram for predicting the three- and five-year probability of cancer-specific death (CSD) in patients with spinal and pelvic chondrosarcoma.

SUMMARY OF BACKGROUND DATA

The issue of competing risk has rarely been addressed and discussed in survival analysis of bone sarcoma. In addition, the Fine and Gray model, a more accurate method for survival analysis in the context of competing risk, has also been less reported in prognostic study of chondrosarcoma.

METHODS

A total of 623 patients with spinal or pelvic chondrosarcoma were identified from the SEER database and were divided into a training and a validation cohort. These two cohorts were used to develop and validate a prognostic model to predict the 3- and 5-year probability of CSD, considering non-CSD as competing risk. The C-index, calibration plot, and decision curve analysis were used to assess the predictive performance and clinical utility of the model.

RESULTS

Older age (subdistribution hazards ratio [SHR]: 1.02, 95% confidence interval [CI]: 1.01∼1.03; P = 0.013), high grade (SHR: 2.68, 95% CI: 1.80∼3.99; P < 0.001), regional involvement (SHR: 1.66, 95% CI: 1.06∼2.58; P = 0.026), distant metastasis (SHR: 5.18, 95% CI: 3.11∼8.62; P < 0.001) and radical resection (SHR: 0.38, 95% CI: 0.24∼0.60; P < 0.001) were significantly associated with the incidence of CSD. These factors were used to build a competing-risk-based model and a nomogram to predict CSD. The C-index, calibration plot, and decision curve analysis indicated that the nomogram performs well in predicting CSD and is suitable for clinical use.

CONCLUSION

A competing-risk based prognostic model is developed to predict the probability of CSD of patients with spinal and pelvic chondrosarcoma. This nomogram performs well and is suitable for clinical use.Level of Evidence: 4.

Identifying the Risk Factors and Estimating the Prognosis in Patients with Pelvis and Spine Ewing Sarcoma: A Population-Based Study

STUDY DESIGN

Retrospective analysis.

OBJECTIVE

The study was designed to: (1) figure out risk factors of metastasis; (2) explore prognostic factors and develop a nomogram for pelvis and spine Ewing sarcoma (PSES).

SUMMARY OF BACKGROUND DATA

Tools to predict survival of PSES are still insufficient. Nomogram has been widely developed in clinical oncology. Moreover, risk factors of PSES metastasis are still unclear.

METHODS

The data were collected and analyzed from the Surveillance, Epidemiology, and End Results (SEER) database. The optimal cutoff values of continuous variables were identified by X-tile software. The prognostic factors of survival were performed by Kaplan-Meier method and multivariate Cox proportional hazards modeling. Nomograms were further constructed for estimating 3- and 5-year cancer-specific survival (CSS) and overall survival (OS) by using R with rms package. Meanwhile, Pearson χ2 test or Fisher exact test, and logistic regression analysis were used to analyze the risk factors for the metastasis of PSES.

RESULTS

A total of 371 patients were included in this study. The 3- and 5-year CSS and OS rate were 65.8 ± 2.6%, 55.2 ± 2.9% and 64.3 ± 2.6%, 54.1 ± 2.8%, respectively. The year of diagnosis, tumor size, and lymph node invasion were associated with metastasis of patients with PSES. A nomogram was developed based on identified factors including: age, tumor extent, tumor size, and primary site surgery. The concordance index (C-index) of CSS and OS were 0.680 and 0.679, respectively. The calibration plot showed the similar trend of 3-year, 5-year CSS, and OS of PSES patients between nomogram-based prediction and actual observation, respectively.

CONCLUSION

PSES patients with earlier diagnostic year (before 2010), larger tumor size (>59 mm), and lymph node invasion, are more likely to have metastasis. We developed a nomogram based on age, tumor extent, tumor size, and surgical treatments for determining the prognosis for patients with PSES, while more external patient cohorts are warranted for validation.Level of Evidence: 3.

Clinicopathological characteristics and prognosis of uterine serous carcinoma: A SEER program analysis of 1016 cases

AIM

To identify the clinicopathological features and survival outcomes of uterine serous carcinoma (USC) and the prognostic factors influencing survival.

METHODS

The retrospective, population-based cohort study enrolled patients with USC diagnosed between 2001 and 2015 for the Surveillance, Epidemiology, and End Results (SEER) program. Kaplan-Meier analysis was performed to identify survival outcomes, multivariable Cox regression models were used to determine the risk factors influencing the disease-specific survival (DSS) and overall survival (OS).

RESULTS

A total of 1016 patients with USC from the SEER database were enrolled. The median age at diagnosis was 65 years. The 5-year OS was 48.5%, and the 5-year DSS rates were 58.0%, respectively. In the univariate analyses, AJCC stage, SEER summary stage, number of lymph nodes resected, and adjuvant therapy were significant predictors for OS and DSS, while grade, was significant only for OS. Multivariate Cox regression models demonstrated that poor grade, stage III/IV, distant disease, the number of lymph nodes resected being <4 and no adjuvant treatment were independent risk factors for poor OS, while stage III/IV, regional or distant disease, the number of lymph nodes <4 and no adjuvant treatment were independent risk factors for poor DSS. Multivariate Cox regression models also identified that chemotherapy and combination therapies were the independent risk factors for improved OS and DSS of early-stage USC.

CONCLUSION

USC had a relatively poor prognosis compared with endometriod carcinoma. Moreover, advanced stage and fewer lymph nodes resected were independent negative prognostic factors for survival, while adjuvant therapy was significant for improved survival.

A novel model to predict cancer-specific survival in patients with early-stage uterine papillary serous carcinoma (UPSC)

OBJECTIVE

Stage I-II uterine papillary serous carcinoma (UPSC) has aggressive biological behavior and leads to poor prognosis. However, clinicopathologic risk factors to predict cancer-specific survival of patients with stage I-II UPSC were still unclear. This study was undertaken to develop a prediction model of survival in patients with early-stage UPSC.

METHODS

Using Surveillance, Epidemiology, and End Results (SEER) database, 964 patients were identified with International Federation of Gynecology and Obstetrics (FIGO) stage I-II UPSC who underwent at least hysterectomy between 2004 and 2015. By considering competing risk events for survival outcomes, we used proportional subdistribution hazards regression to compare cancer-specific death (CSD) for all patients. Based on the results of univariate and multivariate analysis, the variables were selected to construct a predictive model; and the prediction results of the model were visualized using a nomogram to predict the cancer-specific survival and the response to adjuvant chemotherapy and radiotherapy of stage I-II UPSC patients.

RESULTS

The median age of the cohort was 67 years. One hundred and sixty five patients (17.1%) died of UPSC (CSD), while 8.6% of the patients died from other causes (non-CSD). On multivariate analysis, age ≥ 67 (HR = 1.45, P = .021), tumor size ≥ 2 cm (HR = 1.81, P = .014) and >10 regional nodes removed (HR = 0.52, P = .002) were significantly associated with cumulative incidence of CSD. In the age ≥67 cohort, FIGO stage IB-II was a risk factor for CSD (HR = 1.83, P = .036), and >10 lymph nodes removed was a protective factor (HR = 0.50, P = .01). Both adjuvant chemotherapy combined with radiotherapy and adjuvant chemotherapy alone decreased CSD of patients with stage I-II UPSC older than 67 years (HR = 0.47, P = .022; HR = 0.52, P = .024, respectively). The prediction model had great risk stratification ability as the high-risk group had higher cumulative incidence of CSD than the low-risk group (P < .001). In the high-risk group, patients with post-operative adjuvant chemoradiotherapy had improved CSD compared with patients who did not receive radiotherapy nor chemotherapy (P = .037). However, there was no such benefit in the low-risk group.

CONCLUSION

Our prediction model of CSD based on proportional subdistribution hazards regression showed a good performance in predicting the cancer-specific survival of early-stage UPSC patients and contributed to guide clinical treatment decision, helping oncologists and patients with early-stage UPSC to decide whether to choose adjuvant therapy or not.