Search

Background: Bone sarcomas often present late with advanced stage at diagnosis and an according, varying short-term survival. In 2016, Nandra et al. generated a Bayesian belief network model for 1-year survival in patients with bone sarcomas. The purpose of this study is: (1) to externally validate the prior 1-year Bayesian belief network prediction model for survival in patients with bone sarcomas and (2) to develop a gradient boosting machine model using Nandra et al.’s cohort and evaluate whether the gradient boosting machine model outperforms the Bayesian belief network model when externally validated in an independent Danish population cohort. Material and Methods: The training cohort comprised 3493 patients newly diagnosed with bone sarcoma from the institutional prospectively maintained database at the Royal Orthopaedic Hospital, Birmingham, UK. The validation cohort comprised 771 patients with newly diagnosed bone sarcoma included from the Danish Sarcoma Registry during January 1, 2000–June 22, 2016. We performed area under receiver operator characteristic curve analysis, Brier score and decision curve analysis to evaluate the predictive performance of the models. Results: External validation of the Bayesian belief network 1-year prediction model demonstrated an area under receiver operator characteristic curve of 68% (95% confidence interval, 62%-73%). Area under receiver operator characteristic curve of the gradient boosting machine model demonstrated: 75% (95% confidence interval: 70%-80%), overall model performance by the Brier score was 0.09 (95% confidence interval: 0.077–0.11) and decision curve analysis demonstrated a positive net benefit for threshold probabilities above 0.5. External validation of the developed gradient boosting machine model demonstrated an area under receiver operator characteristic curve of 63% (95% confidence interval: 57%-68%), and the Brier score was 0.14 (95% confidence interval: 0.12–0.16). Conclusion: External validation of the 1-year Bayesian belief network survival model yielded a poor outcome based on a Danish population cohort validation. We successfully developed a gradient boosting machine 1-year survival model. The gradient boosting machine did not outperform the Bayesian belief network model based on external validation in a Danish population-based cohort.

Background and purpose — The use of point-of-care or local investigations before referral to specialist sarcoma centers as part of a fast-track diagnostic pathway varies, and may affect the time to diagnosis. We wanted to investigate differences in time intervals and proportion of malignancy in patients who were referred after initial diagnostic investigations were performed locally and in patients who were referred without these investigations. Patients and methods — We included 545 consecutive patients who were referred to Aarhus Sarcoma Center for suspected musculoskeletal sarcoma. Data on time intervals and investigations performed were collected from questionnaires and patient records. Patients who were referred from outside Aarhus uptake area after initial MRI/CT or histology performed locally were compared with patients who were referred from Aarhus uptake area without these investigations. Results — The median total interval from first symptom to diagnosis was 166 days for outside patients referred with MRI/CT or histology, which was 91 (95% CI: 76–106) days longer than for local patients who were referred without MRI/CT or histology. Comparing the same groups, the median diagnostic interval was 41 (95% CI: 30–51) days longer for outside patients including both primary care and hospital intervals. Both the proportion of malignancies (38% vs. 14%) and the proportion of sarcomas (24% vs. 7%) were higher in the outside group referred with MRI/CT or histology than in the local group without MRI/CT or histology. Interpretation — Pre-referral investigations at a local hospital increased the diagnostic interval by at least 1 month for 50% of the patients, and the proportion of malignancy was more than doubled—to almost 40%. If investigations are to be performed before referral to a sarcoma center, they should be part of the fast-track pathway in order to ensure timely diagnosis.

INTRODUCTION:Sarcoma is a rare type of cancer. The incidence increases with age and elderly patients may have comorbidity that affects the prognosis. The aim of this study was to describe the type and prevalence of comorbidity in a nationwide population-based study in Denmark from 2000-2013 and to analyse the impact of the different comorbidities on mortality. MATERIAL AND METHODS:The Danish Sarcoma Registry is a national clinical database containing all patients with sarcoma in the extremities or trunk wall from 2000 and onwards. By linking data to other registries, we were able to get patient information on an individual level including date and cause of death as well as the comorbidity type up to 10 years prior to the sarcoma diagnosis. Based on diseases in the Charlson Comorbidity Index, we pooled the patients into six categories: no comorbidity, cardiopulmonary disease, gastrointestinal disease, neurovascular disease, malignant neoplasms, and miscellaneous (diabetes, renal and connective tissue diseases). 2167 patients were included. RESULTS:The prevalence of comorbidity was 20%. For patients with localized disease, comorbidity increased the disease-specific mortality significantly (HR 1.70 (95% CI 1.36-2.13)). For patients with metastatic disease at the time of diagnosis, comorbidity did not affect the disease-specific mortality (HR 1.05 (95% CI 0.78-1.42)). The presence of another cancer diagnosis within 10 years prior to the sarcoma diagnosis was the only significant independent prognostic factor of disease-specific mortality with an increase of 66% in mortality rate compared to patients with no comorbidity (HR 1,66 (95% CI 1.22-2.25)). CONCLUSION:Comorbidity is a strong independent prognostic factor of mortality in patients with localized disease. This study emphasizes the need for optimizing the general health of comorbid patients in order to achieve a survival benefit from treatment of patients with localized disease, as this is potentially modifiable.