Search NASA⌕ Search

SEARCH · Search NASA

Results for “Multivariate Analysis”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

A comparative study of multivariable robustness analysis methods as applied to integrated flight and propulsion control

Three multivariable robustness analysis methods are compared and contrasted. The focus of the analysis is on system stability and performance robustness to uncertainty in the coupling dynamics between two interacting subsystems. Of particular interest is interacting airframe and engine subsystems, and an example airframe/engine vehicle configuration is utilized in the demonstration of these approaches. The singular value (SV) and structured singular value (SSV) analysis methods are compared to a method especially well suited for analysis of robustness to uncertainties in subsystem interactions. This approach is referred to here as the interacting subsystem (IS) analysis method. This method has been used previously to analyze airframe/engine systems, emphasizing the study of stability robustness. However, performance robustness is also investigated here, and a new measure of allowable uncertainty for acceptable performance robustness is introduced. The IS methodology does not require plant uncertainty models to measure the robustness of the system, and is shown to yield valuable information regarding the effects of subsystem interactions. In contrast, the SV and SSV methods allow for the evaluation of the robustness of the system to particular models of uncertainty, and do not directly indicate how the airframe (engine) subsystem interacts with the engine (airframe) subsystem.

Schierman, John D.↗
DOE OSTI2018

Flexible classification with spatial quantile comparison and novel statistical features applied to spent nuclear fuel analysis

Multivariate classification algorithms are a common tool in contemporary data science that have begun to be applied to nuclear forensic analyses. Using a multivariate generalization of quantile–quantile plots for comparing unknown statistical distributions, a new model-free multivariate classifier called the Quantile–Quantile Comparator has been developed and tested on the analysis of simulated irradiated nuclear fuel. A method to estimate the sample-specific probability of having made an incorrect classification decision by analyzing the distribution of quantile comparisons has been developed, and methods have been developed to analyze the partition of the known data library used to analyze unknown data. By analyzing this partition, the analyst may identify classes that are likely to be confused, identify test data that are inconsistent with the data library and return a “None-of-the-Above” classification decision, and optimize the class boundaries. In this paper, the development of these capabilities is described, and a case study using real experimental assay data collected from nuclear fuel extracted from the newly-released In conclusion, Spent Fuel Isotopic Composition Database to demonstrate the efficacy of these capabilities is presented.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS1977

Alteration mapping at Goldfield, Nevada, by cluster and discriminant analysis of LANDSAT digital data

The ability of Landsat multispectral digital data to differentiate among 62 combinations of rock and alteration types at the Goldfield mining district of Western Nevada was investigated by using statistical techniques of cluster and discriminant analysis. Multivariate discriminant analysis was not effective in classifying each of the 62 groups, with classification results essentially the same whether data of four channels alone or combined with six ratios of channels were used. Bivariate plots of group means revealed a cluster of three groups including mill tailings, basalt and all other rock and alteration types. Automatic hierarchical clustering based on the fourth dimensional Mahalanobis distance between group means of 30 groups having five or more samples was performed. The results of the cluster analysis revealed hierarchies of mill tailings vs. natural materials, basalt vs. non-basalt, highly reflectant rocks vs. other rocks and exclusively unaltered rocks vs. predominantly altered rocks. The hierarchies were used to determine the order in which sets of multiple discriminant analyses were to be performed and the resulting discriminant functions were used to produce a map of geology and alteration which has an overall accuracy of 70 percent for discriminating exclusively altered rocks from predominantly altered rocks.

Ballew, G.↗
NASA NTRS1977

Alteration mapping at Goldfield, Nevada, by cluster and discriminant analysis of Landsat digital data

The ability of Landsat multispectral digital data to differentiate among 62 combinations of rock and alteration types at the Goldfield mining district of Western Nevada was investigated by using statistical techniques of cluster and discriminant analysis. Multivariate discriminant analysis was not effective in classifying each of the 62 groups, with classification results essentially the same whether data of four channels alone or combined with six ratios of channels were used. Bivariate plots of group means revealed a cluster of three groups including mill tailings, basalt and all other rock and alteration types. Automatic hierarchical clustering based on the fourth dimensional Mahalanobis distance between group means of 30 groups having five or more samples was performed using Johnson's HICLUS program. The results of the cluster analysis revealed hierarchies of mill tailings vs. natural materials, basalt vs. non-basalt, highly reflectant rocks vs. other rocks and exclusively unaltered rocks vs. predominantly altered rocks. The hierarchies were used to determine the order in which sets of multiple discriminant analyses were to be performed and the resulting discriminant functions were used to produce a map of geology and alteration which has an overall accuracy of 70 percent for discriminating exclusively altered rocks from predominantly altered rocks.

Ballew, G.↗
DOE OSTI2013

The Multi-Isotope Process (MIP) Monitor Project: FY13 Final Report

The Multi-Isotope Process (MIP) Monitor provides an efficient approach to monitoring the process conditions in reprocessing facilities in support of the goal of “… (minimization of) the risks of nuclear proliferation and terrorism.” The MIP Monitor measures the distribution of the radioactive isotopes in product and waste streams of a nuclear reprocessing facility. These isotopes are monitored online by gamma spectrometry and compared, in near-real-time, to spectral patterns representing “normal” process conditions using multivariate analysis and pattern recognition algorithms. The combination of multivariate analysis and gamma spectroscopy allows us to detect small changes in the gamma spectrum, which may indicate changes in process conditions. By targeting multiple gamma-emitting indicator isotopes, the MIP Monitor approach is compatible with the use of small, portable, relatively high-resolution gamma detectors that may be easily deployed throughout an existing facility. The automated multivariate analysis can provide a level of data obscurity, giving a built-in information barrier to protect sensitive or proprietary operational data. Proof-of-concept simulations and experiments have been performed in previous years to demonstrate the validity of this tool in a laboratory setting for systems representing aqueous reprocessing facilities. However, pyroprocessing is emerging as an alternative to aqueous reprocessing techniques.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2019

Failure Analysis on Multivariate Time-series Data given Uncertain Labels

Machine failure analysis and detection is critical to today’s industrial society. Given a number of failure events on multivariate temporal data, the ability to identify 1) the discriminative patterns prior to the events and 2) the most relevant features to the failure has practical use for early warning and root cause analysis. However, since these patterns are not necessarily adjacent to the onset of failure in time, faulty labels are often with uncertainty, which makes traditional supervised detection methods inapplicable. To address the label uncertainty and learn the complicated correlation in multivariate time series, we design Failure Analysis on Multivariate Time-series Data (MAMT) that jointly selects the most failure-relevant features and time-instances by a novel dynamic and directional label diffusion process. Extensive experiments demonstrate that MAMT is more effective while more efficient than popular baselines.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Differentiation of hepatic abscess from metastasis on contrast-enhanced dynamic computed tomography in patients with a history of extrahepatic malignancy: emphasis on dynamic change of arterial rim enhancement

Objectives: The objective of the study is to identify computed tomography (CT) findings that differentiate hepatic abscess from hepatic metastasis in a patient with a history of extrahepatic malignancy. Materials and methods: This retrospective study included 30 patients with 93 hepatic abscesses and 40 patients with 125 hepatic metastases who had a history of extrahepatic malignancy and underwent contrast-enhanced dynamic CT with arterial phase (AP) and portal venous phase (PVP). The diagnosis of hepatic abscess and hepatic metastasis was made using pathological confirmation or clinical diagnosis. Margin, patchy parenchymal enhancement, arterial rim enhancement, dynamic change of arterial rim enhancement, size discrepancy of lesions between arterial and portal phases, bile duct dilatation, perilesional hyperemia, and perilesional low density were evaluated by two radiologists independently. Significant findings for differentiating two groups were identified at univariate and multivariate analysis with nomogram for predicting hepatic abscess. Interobserver agreement was also analyzed for each variable. Results: Multivariate analysis revealed that patchy parenchymal enhancement (P < 0.001), arterial rim enhancement persistent through PVP (P < 0.001), and perilesional hyperemia (P = 0.013) were independent significant findings to predict hepatic abscess than metastasis. Among them, arterial rim enhancement persistent through PVP showed a highest odds ratio (OR 33.73) on multivariate analysis and a highest predictor point on a nomogram for predicting hepatic abscess. When two of these three criteria were combined, 80.7% (75/93) of hepatic abscess were correctly identified, with a specificity of 85.6% (107/125). When all three criteria were satisfied, specificity was up to 100% (125/125). Conclusions: At contrast-enhanced dynamic CT, patchy parenchymal enhancement, arterial rim enhancement persistent through PVP, perilesional hyperemia, and their combinations may be reliable CT features for differentiating hepatic abscess from metastasis in patients with a history of primary extrahepatic malignancy.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

The Role of Postmastectomy Radiation Therapy After Neoadjuvant Chemotherapy in Clinical Stage II-III Breast Cancer Patients With pN0: A Multicenter, Retrospective Study (KROG 12-05)

Purpose: The purpose of this study was to investigate the role of postmastectomy radiation therapy (PMRT) after neoadjuvant chemotherapy (NAC) in clinical stage II-III breast cancer patients with pN0. Methods and Materials: We retrospectively identified 417 clinical stage II-III breast cancer patients who achieved an ypN0 at surgery after receiving NAC between 1998 and 2009. Of these, 151 patients underwent mastectomy after NAC. The effect of PMRT on disease-free survival (DFS), locoregional recurrence-free survival (LRRFS), and overall survival (OS) was evaluated by multivariate analysis including known prognostic factors using the Kaplan-Meier method and compared using the log–rank test and Cox proportional regression analysis. Results: Of the 151 patients who underwent mastectomy, 105 (69.5%) received PMRT and 46 patients (30.5%) did not. At a median follow-up of 59 months, 5 patients (3.3%) developed LRR (8 sites of recurrence) and 14 patients (9.3%) developed distant metastasis. The 5-year DFS, LRRFS, and OS rates were 91.2, 98.1, and 93.3% with PMRT and 83.0%, 92.3%, and 89.9% without PMRT, respectively (all P values not significant). By univariate analysis, only age (≤40 vs >40 years) was significantly associated with decreased DFS (P=.027). By multivariate analysis, age (≤40 vs >40 years) and pathologic T stage (0-is vs 1 vs 2-4) were significant prognostic factors affecting DFS (hazard ratio [HR] 0.353, 95% confidence interval [CI] 0.135-0.928, P=.035; HR 2.223, 95% CI 1.074-4.604, P=.031, respectively). PMRT showed no correlation with a difference in DFS, LRRFS, or OS by multivariate analysis. Conclusions: PMRT might not be necessary for pN0 patients after NAC, regardless of clinical stage. Prospective randomized clinical trial data are needed to assess whether PMRT can be safely omitted in pN0 patients after NAC and mastectomy for clinical stage II-III breast cancer.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2015

Relationship Between Low Hemoglobin Levels and Outcomes After Treatment With Radiation or Chemoradiation in Patients With Cervical Cancer: Has the Impact of Anemia Been Overstated?

Purpose: Previous reports have suggested that anemia increases rates of recurrence after radiation therapy for cervical cancer. However, these studies may not have fully corrected for confounding risk factors. Using a well-characterized cohort of cervical cancer patients, we examined the association between anemia and outcomes before and after the introduction of chemoradiation as standard of care. Methods and Materials: We reviewed the records of 2454 patients who underwent definitive radiation therapy from 1980 through 2011. Minimum hemoglobin level (Hgb{sub min}) was recorded for 2359 patients (96%). Endpoints included freedom from central recurrence (FFCR), freedom from distant metastasis (FFDM), and disease-specific survival (DSS). Results: For the entire cohort, hemoglobin concentrations of 9, 10, and 12 g/dL before and during radiation were all significantly associated with FFCR, FFDM, and DSS (all P<.001) on univariate analysis. However, on multivariate analysis, only Hgb{sub min} less than 10 g/dL during RT (RT-Hgb{sub <10}) remained significant, and it was correlated with lower DSS (P=.02, hazard ratio [HR] = 1.28) and FFDM (P=.03, HR = 1.33) but not with FFCR. In a subset analysis of patients receiving chemoradiation (n=678), RT-Hgb{sub <10} was associated only with DSS (P=.008, HR = 1.49), not with FFCR or FFDM. In this subgroup, despite an association between RT-Hgb{sub <10} and DSS, the use of transfusion was not correlated with benefit. Conclusions: No evidence was found supporting anemia as an independent predictor of central recurrence in patients treated with definitive radiation therapy with or without chemotherapy. Less emphasis on correcting anemia in cervical cancer patients may be warranted.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2012

Multivariate Statistical Analysis of Water Chemistry in Evaluating the Origin of Contamination in Many Devils Wash, Shiprock, New Mexico

This report evaluates the chemistry of seep water occurring in three desert drainages near Shiprock, New Mexico: Many Devils Wash, Salt Creek Wash, and Eagle Nest Arroyo. Through the use of geochemical plotting tools and multivariate statistical analysis techniques, analytical results of samples collected from the three drainages are compared with the groundwater chemistry at a former uranium mill in the Shiprock area (the Shiprock site), managed by the U.S. Department of Energy Office of Legacy Management. The objective of this study was to determine, based on the water chemistry of the samples, if statistically significant patterns or groupings are apparent between the sample populations and, if so, whether there are any reasonable explanations for those groupings.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2012

Prognostic Significance of Human Apurinic/Apyrimidinic Endonuclease (APE/Ref-1) Expression in Rectal Cancer Treated With Preoperative Radiochemotherapy

Purpose: Human apurinic endonuclease/redox factor 1 (APE/Ref-1) mediates repair of radiation-induced DNA lesions and regulates transcription via redox-based activation. We investigated the predictive and prognostic significance of APE/Ref-1 expression in pretreatment biopsy specimens in locally advanced rectal cancer (LARC) (cT3-T4 or N+). Methods and Materials: APE/Ref-1 expression was analyzed by immunohistochemistry in pretreatment biopsy specimens obtained from 83 patients with LARC. Patients received preoperative radiotherapy of 50.4 Gy in 28 fractions, combined with oral capecitabine and leucovorin chemotherapy, followed by curative surgery. The prognostic significance of various clinicopathologic characteristics, including APE/Ref-1 protein expression, was evaluated. Results: APE/Ref-1 was expressed in 97% of patient samples. Exclusive APE/Ref-1 nuclear staining was observed in 49 of 83 samples (59%), and mixed nuclear and cytoplasmic staining was observed in 31 samples (37%). APE/Ref-1 nuclear expression levels were low in 49 patients (59%) and high in 34 patients (41%). The level of APE/Ref-1 nuclear expression was not a prognostic factor for overall and disease-free survival. Cytoplasmic expression of APE/Ref-1 was a borderline-significant predictive factor for pathologic tumor response (p = 0.08) and a significant prognostic factor for disease-free survival, as shown by univariate analysis (p = 0.037). Multivariate analysis confirmed that cytoplasmic localization of APE/Ref-1 is a significant predictor of disease-free survival (hazard ratio, 0.45; p = 0.046). Conclusions: APE/Ref-1 was expressed in a majority of pretreatment biopsy specimens from patients with LARC. The level of APE/Ref-1 nuclear expression was not a significant predictive and prognostic factor; however, cytoplasmic localization of the protein was negatively associated with disease-free survival. These results indicate that cytoplasmic expression of APE/Ref-1 represents an adverse prognostic factor for LARC patients who receive preoperative radiochemotherapy.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2012

Dose-Volume Analysis of Predictors for Gastrointestinal Toxicity After Concurrent Full-Dose Gemcitabine and Radiotherapy for Locally Advanced Pancreatic Adenocarcinoma

Purpose: To identify dosimetric predictors for the development of gastrointestinal (GI) toxicity in patients with locally advanced pancreatic adenocarcinoma (LAPC) treated with concurrent full-dose gemcitabine and radiotherapy (GemRT). Methods and Materials: From June 2002 to June 2009, 46 LAPC patients treated with definitive GemRT were retrospectively analyzed. The stomach and duodenum were retrospectively contoured separately to determine their dose-volume histogram (DVH) parameters. GI toxicity was defined as Grade 3 or higher GI toxicity. The follow-up time was calculated from the start of RT to the date of death or last contact. Univariate analysis (UVA) and multivariate analysis (MVA) using Kaplan-Meier and Cox regression models were performed to identify risk factors associated with GI toxicity. The receiver operating characteristic curve and the area under the receiver operating characteristic curve (AUC) were used to determine the best DVH parameter to predict for GI toxicity. Results: Of the patients, 28 (61%) received concurrent gemcitabine alone, and 18 (39%) had concurrent gemcitabine with daily erlotinib. On UVA, only the V{sub 20Gy} to V{sub 35Gy} of duodenum were significantly associated with GI toxicity (all p {<=} 0.05). On MVA, the V{sub 25Gy} of duodenum and the use of erlotinib were independent risk factors for GI toxicity (p = 0.006 and 0.02, respectively). For the entire cohort, the V{sub 25Gy} of duodenum is the best predictor for GI toxicity (AUC = 0.717), and the 12-month GI toxicity rate was 8% vs. 48% for V{sub 25Gy} {<=} 45% and V{sub 25Gy} > 45%, respectively (p = 0.03). However, excluding the erlotinib group, the V{sub 35Gy} is the best predictor (AUC = 0.725), and the 12-month GI toxicity rate was 0% vs. 41% for V{sub 35Gy} {<=} 20% and V{sub 35Gy} > 20%, respectively (p = 0.04). Conclusions: DVH parameters of duodenum may predict Grade 3 GI toxicity after GemRT for LAPC. Concurrent use of erlotinib during GemRT may increase GI toxicity.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2012

The Multi-Isotope Process (MIP) Monitor Project: FY12 Progress and Accomplishments

The Multi-Isotope Process (MIP) Monitor, being developed at Pacific Northwest National Laboratory (PNNL), provides an efficient approach to monitoring the process conditions in reprocessing facilities in support of the goal of "...(minimization of) the risks of nuclear proliferation and terrorism." The MIP Monitor measures distributions of a suite of indicator (radioactive) isotopes present within product and waste streams of a nuclear reprocessing facility. These indicator isotopes are monitored on-line by gamma spectrometry and compared, in near-real-time, to spectral patterns representing "normal" process conditions using multivariate pattern recognition software. The monitor utilizes this multivariate analysis and gamma spectroscopy of reprocessing streams to detect small changes in the gamma spectrum, which may indicate changes in process conditions. Multivariate analysis methods common in chemometrics, such as principal component analysis (PCA) and partial least squares regression (PLS), act as pattern recognition techniques, which can detect small deviations from the expected, nominal condition. By targeting multiple gamma-emitting indicator isotopes, the MIP Monitor approach is compatible with the use of small, portable, relatively high-resolution gamma detectors that may be easily deployed throughout an existing facility. The automated multivariate analysis can provide a level of data obscurity, giving a built-in information barrier to protect sensitive or proprietary operational data. Proof-of-concept simulations and experiments have been performed in previous years to demonstrate the validity of this tool in a laboratory setting. Development of the MIP Monitor approach continues to evaluate the efficacy of the monitor for automated, real-time or near-real-time application. This report details follow-on research and development efforts sponsored by the U.S. Department of Energy Fuel Cycle Research and Development related to the MIP Monitor for fiscal year 2012 (FY12).

fuel characterization↗
DOE OSTI2012

Importance of Extracranial Disease Status and Tumor Subtype for Patients Undergoing Radiosurgery for Breast Cancer Brain Metastases

Purpose: In this retrospective study, we report on outcomes and prognostic factors for patients treated with stereotactic radiosurgery (SRS) for breast cancer brain metastases. Methods and Materials: We identified 132 consecutive patients with breast cancer who were treated with SRS for brain metastases from January 2000 through June 2010. We retrospectively reviewed records of the 51 patients with adequate follow-up data who received SRS as part of the initial management of their brain metastases. Overall survival (OS) and time to central nervous system (CNS) progression from the date of SRS were calculated using the Kaplan-Meier method. Prognostic factors were evaluated using the Cox proportional hazards model. Results: Triple negative subtype was associated with CNS progression on univariate analysis (hazard ratio [HR] = 5.0, p = 0.008). On multivariate analysis, triple negative subtype (HR = 8.6, p = 0.001), Luminal B subtype (HR = 4.3, p = 0.03), and omission of whole-brain radiation therapy (HR = 3.7, p = 0.02) were associated with CNS progression. With respect to OS, Karnofsky Performance Status (KPS) {<=} 80% (HR = 2.0, p = 0.04) and progressive extracranial disease (HR = 3.1, p = 0.002) were significant on univariate analysis; KPS {<=} 80% (HR = 4.1, p = 0.0004), progressive extracranial disease (HR = 6.4, p < 0.0001), and triple negative subtype (HR = 2.9, p = 0.04) were significant on multivariate analysis. Although median survival times were consistent with those predicted by the breast cancer-specific Graded Prognostic Assessment (Breast-GPA) score, the addition of extracranial disease status further separated patient outcomes. Conclusions: Tumor subtype is associated with risk of CNS progression after SRS for breast cancer brain metastases. In addition to tumor subtype and KPS, which are incorporated into the Breast-GPA, progressive extracranial disease may be an important prognostic factor for OS.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
NASA NTRS1991

Software For Multivariable Frequency-Domain Analysis

FREQ (Multivariable Frequency Domain Singular Value Analysis Package) software package of subroutines performing frequency-domain analysis of: continuous- or discrete-multivariable linear systems; any continuous system for which one calculates transfer matrix at points on imaginary axis; or any discrete system for which one calculates transfer matrix at points on unit circle. Four different versions available. Single-precision brief version LAR-14119, single-precision complete version LAR-14120, double-precision brief version LAR-14121, and double-precision complete version LAR-14122. Written in ANSI standard FORTRAN 77.

Armstrong, Ernest S.↗
DOE OSTI2012

Low energy electron/recoil discrimination for directional Dark Matter detection

Directional detection is a promising Dark Matter search strategy. Even though it could accommodate to a sizeable background contamination, electron/recoil discrimination remains a key and challenging issue as for direction-insensitive detectors. The measurement of the 3D track may be used to discriminate electrons from nuclear recoils. While a high rejection power is expected above 20 keV ionization, a dedicated data analysis is needed at low energy. After identifying discriminant observables, a multivariate analysis, namely a Boosted Decision Tree, is proposed, enabling an efficient event tagging for Dark Matter search. We show that it allows us to optimize rejection while keeping a rather high efficiency which is compulsory for rare event search.With respect to a sequential analysis, the rejection is about ∼ 20 times higher with a multivariate analysis, for the same Dark Matter exclusion limit.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Multivariate Frequency Analysis of Annual Maxima Suspended Sediment Concentrations and Floods in the Jinsha River, China

Streamflow plays a critical role in sediment transport. Traditional frequency analysis of annual maximal (AM) suspended sediment concentrations (SSCs) is usually based on univariate distribution of SSCs and may provide a limited assessment of SSC state, leading to an underestimate or overestimate of the corresponding probability. To address this issue, we propose a multivariate frequency analysis framework is proposed for extreme sediment events using copula functions that account for the joint behavior of SSC and flood flow (i.e., peak discharge and volume). The uncertainties associated with copula modeling are also discussed. The Jinsha River in China is used as a case study. The results show that the copula-based multivariate method provides a more comprehensive assessment of extreme SSCs and their recurrence intervals than univariate frequency analysis. The uncertainty of copula modeling decreases with longer observation time and increases with larger return periods. The uncertainty ranges of the most likely design quantiles vary with different copulas, and the uncertainty range for the best-fit copula is not the smallest. This study is useful for the risk assessment of extreme SSC events under flood conditions.

42 ENGINEERING↗
DOE OSTI2011

Determinants of Toxicity, Patterns of Failure, and Outcome Among Adult Patients With Soft Tissue Sarcomas of the Extremity and Superficial Trunk Treated With Greater Than Conventional Doses of Perioperative High-Dose-Rate Brachytherapy and External Beam Radiotherapy

Purpose: The present study was undertaken to determine factors predictive of toxicity, patterns of failure, and survival in 60 adult patients with soft tissue sarcomas of the extremity and superficial trunk treated with combined perioperative high-dose-rate brachytherapy and external beam radiotherapy. Methods and Materials: The patients were treated with surgical resection and perioperative high-dose-rate brachytherapy (16 or 24 Gy) for negative and close/microscopically positive resection margins, respectively. External beam radiotherapy (45 Gy) was added postoperatively to reach a 2-Gy equivalent dose of 62.9 and 72.3 Gy, respectively. Adjuvant chemotherapy with ifosfamide and doxorubicin was given to patients with advanced high-grade tumors. Results: Grade 3 toxic events were observed in 18 patients (30%) and Grade 4 events in 6 patients (10%). No Grade 5 events were observed. A location in the lower limb was significant for Grade 3 or greater toxic events on multivariate analysis (p = .013), and the tissue volume encompassed by the 150% isodose line showed a trend toward statistical significance (p = .086). The local control, locoregional control, and distant control rate at 9 years was 77.4%, 69.5%, and 63.8%, respectively. On multivariate analysis, microscopically involved margins correlated with local control (p = .036) and locoregional control (p = .007) and tumor size correlated with distant metastases (p = .004). The 9-year disease-free survival and overall survival rate was 47.0% and 61.5%, respectively. Multivariate analysis showed poorer disease-free survival rates for patients with tumors >6 cm (p = .005) and microscopically involved margins (p = .043), and overall survival rates decreased with increasing tumor size (p = .011). Conclusions: Grade 3 or greater wound complications can probably be decreased using meticulous treatment planning to decrease the tissue volume encompassed by the 150% isodose line, especially in lower limb locations. Microscopically involved margins remain a predictor of local and locoregional failure, despite radiation doses >70 Gy. Patients with tumors {>=}6 cm and microscopically involved margins are at high risk of treatment failure and death from the development of distant metastases.

62 RADIOLOGY AND NUCLEAR MEDICINE↗