2015 Marshall Urist Young Investigator Award: Prognostication in Patients With Long Bone Metastases: Does a Boosting Algorithm Improve Survival Estimates?

Background

Survival estimation guides surgical decision-making in metastatic bone disease. Traditionally, classic scoring systems, such as the Bauer score, provide survival estimates based on a summary score of prognostic factors. Identification of new factors might improve the accuracy of these models. Additionally, the use of different algorithms—nomograms or boosting algorithms—could further improve accuracy of prognostication relative to classic scoring systems. A nomogram is an extension of a classic scoring system and generates a more-individualized survival probability based on a patient’s set of characteristics using a figure. Boosting is a method that automatically trains to classify outcomes by applying classifiers (variables) in a sequential way and subsequently combines them. A boosting algorithm provides survival probabilities based on every possible combination of variables.

Questions/purposes

We wished to (1) assess factors independently associated with decreased survival in patients with metastatic long bone fractures and (2) compare the accuracy of a classic scoring system, nomogram, and boosting algorithms in predicting 30-, 90-, and 365-day survival.

Methods

We included all 927 patients in our retrospective study who underwent surgery for a metastatic long bone fracture at two institutions between January 1999 and December 2013. We included only the first procedure if patients underwent multiple surgical procedures or had more than one fracture. Median followup was 8 months (interquartile range, 3-25 months); 369 of 412 (90%) patients who where alive at 1 year were still in followup. Multivariable Cox regression analysis was used to identify clinical and laboratory factors independently associated with decreased survival. We created a classic scoring system, nomogram, and boosting algorithms based on identified variables. Accuracy of the algorithms was assessed using area under the curve analysis through fivefold cross validation.

Results

The following factors were associated with a decreased likelihood of survival after surgical treatment of a metastatic long bone fracture, after controlling for relevant confounding variables: older age (hazard ratio [HR], 1.0; 95% CI, 1.0–1.0; p < 0.001), additional comorbidity (HR, 1.2; 95% CI, 1.0–1.4; p = 0.034), BMI less than 18.5 kg/m² (HR, 2.0; 95% CI, 1.2–3.5; p = 0.011), tumor type with poor prognosis (HR, 1.8; 95% CI, 1.6–2.2; p < 0.001), multiple bone metastases (HR, 1.3; 95% CI, 1.1–1.6; p = 0.008), visceral metastases (HR, 1.6; 95% CI, 1.4–1.9; p < 0.001), and lower hemoglobin level (HR, 0.91; 95% CI, 0.87–0.96; p < 0.001). The survival estimates by the nomogram were moderately accurate for predicting 30-day (area under the curve [AUC], 0.72), 90-day (AUC, 0.75), and 365-day (AUC, 0.73) survival and remained stable after correcting for optimism through fivefold cross validation. Boosting algorithms were better predictors of survival on the training datasets, but decreased to a performance level comparable to the nomogram when applied on testing datasets for 30-day (AUC, 0.69), 90-day (AUC, 0.75), and 365-day (AUC, 0.72) survival prediction. Performance of the classic scoring system was lowest for all prediction periods.

Conclusions

Comorbidity status and BMI are newly identified factors associated with decreased survival and should be taken into account when estimating survival. Performance of the boosting algorithms and nomogram were comparable on the testing datasets. However, the nomogram is easier to apply and therefore more useful to aid surgical decision making in clinical practice.

Level of Evidence

Level III, prognostic study.

Electronic supplementary material

The online version of this article (doi:10.1007/s11999-015-4446-z) contains supplementary material, which is available to authorized users.     

Pulmonary function over 2 years in diabetic patients treated with prandial inhaled Technosphere Insulin or usual antidiabetes treatment: a randomized trial

Aims: Development of inhaled insulin has increased the need to understand its pulmonary safety. This study evaluated pulmonary function changes in diabetes patients receiving inhaled Technosphere Insulin (TI) or usual antidiabetes treatment (usual care). Methods: This randomized, open‐label study was conducted at 220 sites (25 July 2005 to 29 August 2008). Pulmonary function tests [forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC), total lung capacity (TLC) and lung diffusion capacity for carbon monoxide (DL_CO)] were prospectively followed over 2 years in patients with type 1 or type 2 diabetes receiving TI (n = 730) or usual care (n = 824), along with a cohort without diabetes not receiving any specific therapy (n = 145). Results: Baseline demographics and pulmonary function were similar between diabetes treatment groups. Lung function declined from baseline in all groups. TI was non‐inferior to usual care for mean change in FEV₁ from baseline to month 24 [mean (s.e.m.) 0.037 (0.0119) l; 95% CI 0.014 to 0.060] using mixed‐model repeated‐measure with a pre‐specified non‐inferiority margin of 50 ml/year. After a greater initial decline at month 3 with TI, rate of change (slope) in FEV₁, FVC and DL_CO (months 3–24) was not statistically different between treatment groups. TI was well tolerated; no serious safety concerns emerged. The most common respiratory event associated with TI was mild, transient cough, occurring within minutes of inhalation. Conclusions: Observed changes in lung function with TI were small, occurred early after therapy initiation, remained non‐progressive over 2 years and were unlikely to be clinically meaningful.     

曲马多治疗糖尿病性神经病疼痛的双盲随机试验

目的：本研究旨在评价曲马多治疗糖尿病性神经病的有效性和安全性。背景：糖尿病性神经病疼痛是糖尿病患者的主要症状，其治疗和其它的末梢神经病变一样，不能令人满意。曲马多是作用于中枢的镇痛剂，用于治疗中到重度疼痛。方法：多中心、门诊病例、随机、双盲、安慰剂做对照、设置平行组的研究，包括药效消退期（在此期间停用所有镇痛剂）和一个为期42天的双盲治疗期。总共有131名患糖尿病性神经病疼痛的患者接受曲马多（n=65）或安慰剂胶囊治疗（n=66），每日4次。42天的治疗结束后或治疗中断后比较曲马多组和安慰剂组平均疼痛强度评分，进行初级药效分析。次级药效分析为疼痛缓解评分和根据日常活动和睡眠状况作出生活质量评估。结果：曲马多平均剂量为210mg／天，治疗糖尿病性神经病疼痛较安慰剂有显著疗效（P〈0．001）。曲马多组患者生理（P=0．02）和社会功能（P=0．04）的评分显著优于安慰剂组。未发现对睡眠有显著改善作用。曲马多最常见的不良事件是恶心、便秘、头疼和嗜睡。结论：安慰剂做对照的实验结果表明曲马多在治疗糖尿病性神经病疼痛上是安全和有效的。     

A case of slipped capital femoral epiphysis results from hypophosphatemic rickets： treatment and midterm follow-up

In this case, a male patient presented with a clinically and radiographieally unstable slipped capital femoral epiphysis （SCFE） as well as slipped calcaneal epiphysis years. Subsequent thorough at the age of 23 work-up revealed that he had some features of rickets and labo- ratory test demonstrated he had hypophos- phatemia （2.3mg/dl）, normocalcemia, normal vi- tamin D metabolite levels, and secondary hy- perparathyroidism.     

Initial Evaluation of a 290-μm Diameter Subcutaneous Glucose Sensor: Glucose Monitoring with a Biocompatible, Flexible-Wire, Enzyme-Based Amperometric Microsensor in Diabetic and Nondiabetic Humans

Results of the initial clinical evaluation in 20 human subjects of a subcutaneously implanted microsensor-based amperometrically glycemia-monitoring system, carried out between April 1994 and June 1995, are reported. The system was based on the electrical connection (“wiring”) of the reaction centers of glucose oxidase to a gold electrode and on elimination of the chemicals that interfere with glucose monitoring through their horseradish peroxidase-catalyzed oxidation by internally generated hydrogen peroxide. The sensor was finer than a 29-gauge needle and had no leachable components. Because of its high selectivity for glucose, the sensor output was virtually nil at zero glucose level. This enables prompt “one-point” in vivo calibration of the sensor with a single blood glucose sample. Microsensors were subcutaneously implanted in ten nondiabetic and ten insulin-dependent diabetes mellitus (IDDM) volunteers. All subjects underwent standard meal tests and intravenous glucose-tolerance tests (IVGTT) in addition to hourly plasma glucose measurements. The sensor signals were continuously recorded, and the glucose concentration estimates were derived by calibrating the sensor using a single blood sample (one-point calibration). Regression analysis revealed that the sensor-estimated glucose concentrations were linearly related to the plasma glucose concentrations (r² = 0.75) over a wide glucose concentration range (2–28 mmol/L) (sensor estimate = plasma * 0.96 + 0.26 mmol/L). The difference between the estimated and actual glucose concentration was −0.13 ± 0.23 mmol/L [mean ± 95% confidence interval (CI), n = 546], and 95% of the estimates fell in clinically acceptable zones of the Clarke error grid. The sensing delay time was 10.4 ± 2.3 min as measured by the IVGTT. The subjects reported no discomfort associated with wearing the sensors.     

Comparison of the subacute toxicity and efficacy of 3-hydroxypyridin-4- one iron chelators in overloaded and nonoverloaded mice

Five orally effective iron chelators of the 3-hydroxypyridin-4-one series have been administered intraperitoneally to iron-overloaded and nonoverloaded male mice at a dose of 200 mg/kg/24 h for a total of 60 days to investigate the effect on iron loading and toxicity. There was a significant reduction in hepatic iron at the end of the study in the iron-overloaded mice with all compounds studied using chemical iron quantitation (P less than .001) and with Perls' stain (P less than .01). Liver iron removal with the hydroxypyridinones ranged from 37% with CP20 to 63% with CP51, compared with 46% removal for desferrioxamine (DFO). There was no significant reduction in splenic or cardiac iron with any chelator. There were no deaths in iron-overloaded animals receiving any of the hydroxypyridin-4-ones, but significantly more deaths in the nonoverloaded groups as a whole (P less than .03). No weight loss was observed with any chelator. Significant reductions in hemoglobin and white cell count were observed with CP20(L1). No histologic abnormalities of kidney, spleen, bone marrow, or stifle joints were observed. Intracytoplasmic inclusion bodies were observed in the centrilobular hepatocytes of animals administered each of the hydroxypyridin-4-ones, while the DFO-treated and control groups showed no such changes.     

Detection of anti-HTLV-I Tax antibodies in HTLV-I enzyme-linked immunosorbent assay-negative individuals

The HTLV-I tax gene protein (Tax) is not packaged within the mature viral particle from which the proteins for the commercially available enzyme-linked immunosorbent assay (ELISA) are derived. Screening of 162 individuals within a cohort of white intravenous (IV) drug abusers, previously identified as having an increased incidence of HTLV-I infection, demonstrated that seven of them had antibodies to the HTLV-I Tax protein but tested negative in HTLV-I ELISAs and Western blots prepared from purified virion proteins. Three out of 35 individuals in other behaviorally defined high-risk groups also displayed this limited pattern of reactivity to HTLV-I proteins. The presence of the anti-HTLV- I p40/Tax antibodies was determined by radioimmunoprecipitation assay (RIPA), which also revealed low levels of anti-env reactivity. The specificity of the anti-p40 reactivity was confirmed on specific Tax ELISAs and Western blots prepared from recombinantly produced Tax. In vitro gene amplification by the polymerase chain reaction (PCR) was used to establish the presence of sequences homologous to HTLV-I proviral DNA in four/four of these HTLV-I ELISA negative, Tax ELISA/Tax western blot/RIPA positive individuals. These data suggest that the true incidence of HTLV-I infection within high-risk cohorts is greater than previously reported.     

Serving those who serve: Psychotherapy with military clients

As in the Vietnam era, the wars in Iraq and Afghanistan are not the focus of the nightly news. Nevertheless, there are countless numbers of individuals enlisting, serving, deploying, redeploying, and leaving military branches every day. These individuals are underserved in many ways, especially after they end their formal relationship with their service. The purpose of this introduction is to describe the ongoing concerns and psychological problems of our military personnel and the efforts, described by the authors of papers in this issue, to address these concerns and problems.