Development and Internal Validation of Machine Learning Algorithms for Preoperative Survival Prediction of Extremity Metastatic Disease.
- 作者列表："Thio QCBS","Karhade AV","Bindels BJJ","Ogink PT","Bramer JAM","Ferrone ML","Calderón SL","Raskin KA","Schwab JH
BACKGROUND:A preoperative estimation of survival is critical for deciding on the operative management of metastatic bone disease of the extremities. Several tools have been developed for this purpose, but there is room for improvement. Machine learning is an increasingly popular and flexible method of prediction model building based on a data set. It raises some skepticism, however, because of the complex structure of these models. QUESTIONS/PURPOSES:The purposes of this study were (1) to develop machine learning algorithms for 90-day and 1-year survival in patients who received surgical treatment for a bone metastasis of the extremity, and (2) to use these algorithms to identify those clinical factors (demographic, treatment related, or surgical) that are most closely associated with survival after surgery in these patients. METHODS:All 1090 patients who underwent surgical treatment for a long-bone metastasis at two institutions between 1999 and 2017 were included in this retrospective study. The median age of the patients in the cohort was 63 years (interquartile range [IQR] 54 to 72 years), 56% of patients (610 of 1090) were female, and the median BMI was 27 kg/m (IQR 23 to 30 kg/m). The most affected location was the femur (70%), followed by the humerus (22%). The most common primary tumors were breast (24%) and lung (23%). Intramedullary nailing was the most commonly performed type of surgery (58%), followed by endoprosthetic reconstruction (22%), and plate screw fixation (14%). Missing data were imputed using the missForest methods. Features were selected by random forest algorithms, and five different models were developed on the training set (80% of the data): stochastic gradient boosting, random forest, support vector machine, neural network, and penalized logistic regression. These models were chosen as a result of their classification capability in binary datasets. Model performance was assessed on both the training set and the validation set (20% of the data) by discrimination, calibration, and overall performance. RESULTS:We found no differences among the five models for discrimination, with an area under the curve ranging from 0.86 to 0.87. All models were well calibrated, with intercepts ranging from -0.03 to 0.08 and slopes ranging from 1.03 to 1.12. Brier scores ranged from 0.13 to 0.14. The stochastic gradient boosting model was chosen to be deployed as freely available web-based application and explanations on both a global and an individual level were provided. For 90-day survival, the three most important factors associated with poorer survivorship were lower albumin level, higher neutrophil-to-lymphocyte ratio, and rapid growth primary tumor. For 1-year survival, the three most important factors associated with poorer survivorship were lower albumin level, rapid growth primary tumor, and lower hemoglobin level. CONCLUSIONS:Although the final models must be externally validated, the algorithms showed good performance on internal validation. The final models have been incorporated into a freely accessible web application that can be found at https://sorg-apps.shinyapps.io/extremitymetssurvival/. Pending external validation, clinicians may use this tool to predict survival for their individual patients to help in shared treatment decision making. LEVEL OF EVIDENCE:Level III, therapeutic study.
背景: 术前估计生存期对于决定四肢转移性骨病的手术治疗至关重要。为此开发了几种工具，但还有改进的余地。机器学习是一种越来越流行和灵活的基于数据集的预测模型构建方法。然而，由于这些模型的复杂结构，它引起了一些怀疑。 问题/目的: 本研究的目的是 (1) 开发机器学习算法，用于接受四肢骨转移手术治疗的患者的 90 天和 1 年生存率，以及 (2) 使用这些算法来识别这些临床因素 (人口统计学、治疗相关或手术)这与这些患者手术后的生存率最密切相关。 方法: 回顾性研究纳入 1999 年至 1090 年间在两家机构接受长骨转移手术治疗的 2017 例患者。队列中患者的中位年龄为 63 岁 (四分位距 [IQR] 54 ~ 72 岁)，56% 的患者 (610 中的 1090) 为女性，BMI中位数为 27千克/m (IQR 23 ~ 30千克/m)。受累部位最多的是股骨 (70%)，其次是肱骨 (22%)。最常见的原发肿瘤是乳腺 (24%) 和肺 (23%)。髓内钉是最常见的手术类型 (58%)，其次是假体内重建 (22%) 和钢板螺钉固定 (14%)。使用missfrest方法对缺失数据进行插补。用随机森林算法选取特征，在训练集上 (80% 的数据) 开发了五种不同的模型: 随机梯度提升、随机森林、支持向量机、神经网络、和惩罚logistic回归。选择这些模型是由于它们在二进制数据集中的分类能力。通过区分度、校准和总体性能对训练集和验证集 (20% 的数据) 进行模型性能评估。 结果: 我们发现五种模型之间的判别没有差异，曲线下面积范围为 0.86-0.87。所有模型都经过良好的校准，截距范围为-0.03 至 0.08，斜率范围为 1.03 至 1.12。Brier评分范围为 0.13 ~ 0.14。选择随机梯度提升模型部署为免费提供的基于web的应用程序，并提供了全局和个体层面的解释。对于 90 天生存率，与较差生存率相关的三个最重要因素是白蛋白水平较低、中性粒细胞与淋巴细胞比值较高以及原发肿瘤生长迅速。对于 1 年生存率，与较差生存率相关的三个最重要因素是白蛋白水平较低、原发肿瘤生长迅速和血红蛋白水平较低。 结论: 尽管最终模型必须经过外部验证，但算法在内部验证上表现出良好的性能。最终的模型已被纳入一个可自由访问的web应用程序，可以在 https://sorg-apps.shinyapps.io/extremitymetssurvival/ 。在外部验证之前，临床医生可以使用该工具预测其个体患者的生存期，以帮助共享治疗决策。 证据级别: III级，治疗性研究。
METHODS:OBJECTIVE:Large inoperable sacral chordomas show unsatisfactory local control rates even when treated with high dose proton therapy (PT). The aim of this study is assessing feasibility and reporting early results of patients treated with PT and concomitant hyperthermia (HT). METHODS: :Patients had histologically proven unresectable sacral chordomas and received 70 Gy (relative biological effectiveness) in 2.5 Gy fractions with concomitant weekly HT. Toxicity was assessed according to CTCAE_v4. A volumetric tumor response analysis was performed. RESULTS: :Five patients were treated with the combined approach. Median baseline tumor volume was 735 cc (range, 369-1142). All patients completed PT and received a median of 5 HT sessions (range, 2-6). Median follow-up was 18 months (range, 9-26). The volumetric analysis showed an objective response of all tumors (median shrinkage 46%; range, 9-72). All patients experienced acute Grade 2-3 local pain. One patient presented with a late Grade 3 iliac fracture. CONCLUSION:Combining PT and HT in large inoperable sacral chordomas is feasible and causes acceptable toxicity. Volumetric analysis shows promising early results, warranting confirmation in the framework of a prospective trial. ADVANCES IN KNOWLEDGE: :This is an encouraging first report of the feasibility and early results of concomitant HT and PT in treating inoperable sacral chordoma.
METHODS:BACKGROUND:National guidelines recommend screening and treatment for cancer-related bone disease and continued monitoring of bone-modifying agents. It is unclear whether a standardized screening tool is utilized to identify eligible patients and ensure appropriate supportive care is implemented. The purpose of this study was to evaluate current prescribing practices and optimize management of bone-modifying agents. METHODS:A retrospective chart review was performed to identify patients who received hormone deprivation therapy or had bone metastases through Hematology/Oncology or Urology clinics from 1 November 2016 to 31 October 2017. The primary endpoints of this study were the incidence of completed baseline dual-energy X-ray absorptiometry (DEXA) scan for patients on hormone deprivation therapy and percent of patients started on a bone-modifying agent for the prevention of skeletal-related events secondary to bone metastasis. Secondary endpoints included percent of patients with dental examinations prior to initiation, adequate calcium and vitamin D supplementation, incidence of osteonecrosis of the jaw or flu-like symptoms and education, and percent of bisphosphonate doses appropriately adjusted based on renal function. RESULTS:A total of 375 patients were assessed for baseline DEXA scans and bone-modifying therapy. Of the 226 patients on hormone deprivation therapy, 111 (49%) patients were appropriately screened with a DEXA scan prior to initiation of hormone deprivation therapy. Among the 149 patients with bone metastases, only 94 (63.1%) patients were started on a bone-modifying agent. CONCLUSIONS:Opportunities have been identified to optimize management of patients with cancer-related bone disease. Implementation of standardized tools may increase the rate of appropriate screening and initiation of bone-modifying therapy when warranted.
METHODS:PURPOSE:Low skeletal muscle mass has been associated with poor prognosis in patients with advanced lung cancer. However, little is known about the relationship between skeletal muscle mass and overall survival in patients with bone metastases from lung cancer. The objective of the present study was to evaluate the prognostic value of low trunk muscle mass in predicting overall survival in these patients. METHODS:The data from 198 patients who were diagnosed with bone metastases from lung cancer from April 2009 to May 2017 were retrospectively reviewed. The areas of the psoas and paravertebral muscles were measured at the level of the third lumbar vertebra on computed tomography scans taken at the time nearest to the diagnosis of bone metastasis. Muscle area was evaluated for male and female cohorts separately using different cutoff points. Cox proportional hazards analysis was performed to evaluate the factors independently associated with overall survival. RESULTS:The overall survival of patients in the lowest quartile for psoas muscle area or paravertebral muscle area was significantly shorter than that of patients above the 25th percentile for muscle area (p < 0.001). Multivariate analyses showed that paravertebral muscle mass (hazard ratio, 1.73; 95% confidence interval, 1.17-2.56; p = 0.006), epidermal growth factor receptor-targeted therapy, and performance status were independent prognostic factors. CONCLUSIONS:Low paravertebral muscle mass was associated with shorter survival, independently of known prognostic factors.