Effective diagnosis of Alzheimer's disease by means of large margin-based methodology.

BACKGROUND Functional brain images such as Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) have been widely used to guide the clinicians in the Alzheimer's Disease (AD) diagnosis. However, the subjectivity involved in their evaluation has favoured the development of Computer Aided Diagnosis (CAD) Systems. METHODS It is proposed a novel combination of feature extraction techniques to improve the diagnosis of AD. Firstly, Regions of Interest (ROIs) are selected by means of a t-test carried out on 3D Normalised Mean Square Error (NMSE) features restricted to be located within a predefined brain activation mask. In order to address the small sample-size problem, the dimension of the feature space was further reduced by: Large Margin Nearest Neighbours using a rectangular matrix (LMNN-RECT), Principal Component Analysis (PCA) or Partial Least Squares (PLS) (the two latter also analysed with a LMNN transformation). Regarding the classifiers, kernel Support Vector Machines (SVMs) and LMNN using Euclidean, Mahalanobis and Energy-based metrics were compared. RESULTS Several experiments were conducted in order to evaluate the proposed LMNN-based feature extraction algorithms and its benefits as: i) linear transformation of the PLS or PCA reduced data, ii) feature reduction technique, and iii) classifier (with Euclidean, Mahalanobis or Energy-based methodology). The system was evaluated by means of k-fold cross-validation yielding accuracy, sensitivity and specificity values of 92.78%, 91.07% and 95.12% (for SPECT) and 90.67%, 88% and 93.33% (for PET), respectively, when a NMSE-PLS-LMNN feature extraction method was used in combination with a SVM classifier, thus outperforming recently reported baseline methods. CONCLUSIONS All the proposed methods turned out to be a valid solution for the presented problem. One of the advances is the robustness of the LMNN algorithm that not only provides higher separation rate between the classes but it also makes (in combination with NMSE and PLS) this rate variation more stable. In addition, their generalization ability is another advance since several experiments were performed on two image modalities (SPECT and PET).

Comparison of drug survival rates for tumor necrosis factor antagonists in rheumatoid arthritis.

BACKGROUND Persistence of anti-tumor necrosis factor (TNF) therapy in rheumatoid arthritis (RA) is an overall marker of treatment success. OBJECTIVE To assess the survival of anti-TNF treatment and to define the potential predictors of drug discontinuation in RA, in order to verify the adequacy of current practices. DESIGN An observational, descriptive, longitudinal, retrospective study. SETTING The Hospital Clínico Universitario de Valladolid, Valladolid, Spain. PATIENTS RA patients treated with anti-TNF therapy between January 2011 and January 2012. MEASUREMENTS Demographic information and therapy assessments were gathered from medical and pharmaceutical records. Data is expressed as means (standard deviations) for quantitative variables and frequency distribution for qualitative variables. Kaplan-Meier survival analysis was used to assess persistence, and Cox multivariate regression models were used to assess potential predictors of treatment discontinuation. RESULTS In total, 126 treatment series with infliximab (n = 53), etanercept (n = 51) or adalimumab (n = 22) were administered to 91 patients. Infliximab has mostly been used as a first-line treatment, but it was the drug with the shortest time until a change of treatment. Significant predictors of drug survival were: age; the anti-TNF agent; and the previous response to an anti-TNF drug. LIMITATION The small sample size. CONCLUSION The overall efficacy of anti-TNF drugs diminishes with time, with infliximab having the shortest time until a change of treatment. The management of biologic therapy in patients with RA should be reconsidered in order to achieve disease control with a reduction in costs.

Combined low initial DNA damage and high radiation-induced apoptosis confers clinical resistance to long-term toxicity in breast cancer patients treated with high-dose radiotherapy

BACKGROUND. Either higher levels of initial DNA damage or lower levels of radiation-induced apoptosis in peripheral blood lymphocytes have been associated to increased risk for develop late radiation-induced toxicity. It has been recently published that these two predictive tests are inversely related. The aim of the present study was to investigate the combined role of both tests in relation to clinical radiation-induced toxicity in a set of breast cancer patients treated with high dose hyperfractionated radical radiotherapy. METHODS. Peripheral blood lymphocytes were taken from 26 consecutive patients with locally advanced breast carcinoma treated with high-dose hyperfractioned radical radiotherapy. Acute and late cutaneous and subcutaneous toxicity was evaluated using the Radiation Therapy Oncology Group morbidity scoring schema. The mean follow-up of survivors (n = 13) was 197.23 months. Radiosensitivity of lymphocytes was quantified as the initial number of DNA double-strand breaks induced per Gy and per DNA unit (200 Mbp). Radiation-induced apoptosis (RIA) at 1, 2 and 8 Gy was measured by flow cytometry using annexin V/propidium iodide. RESULTS. Mean DSB/Gy/DNA unit obtained was 1.70 ± 0.83 (range 0.63-4.08; median, 1.46). Radiation-induced apoptosis increased with radiation dose (median 12.36, 17.79 and 24.83 for 1, 2, and 8 Gy respectively). We observed that those "expected resistant patients" (DSB values lower than 1.78 DSB/Gy per 200 Mbp and RIA values over 9.58, 14.40 or 24.83 for 1, 2 and 8 Gy respectively) were at low risk of suffer severe subcutaneous late toxicity (HR 0.223, 95%CI 0.073-0.678, P = 0.008; HR 0.206, 95%CI 0.063-0.677, P = 0.009; HR 0.239, 95%CI 0.062-0.929, P = 0.039, for RIA at 1, 2 and 8 Gy respectively) in multivariate analysis. CONCLUSIONS. A radiation-resistant profile is proposed, where those patients who presented lower levels of initial DNA damage and higher levels of radiation induced apoptosis were at low risk of suffer severe subcutaneous late toxicity after clinical treatment at high radiation doses in our series. However, due to the small sample size, other prospective studies with higher number of patients are needed to validate these results.

The BRAIN TRIAL: a randomised, placebo controlled trial of a Bradykinin B2 receptor antagonist (Anatibant) in patients with traumatic brain injury

BACKGROUND Cerebral oedema is associated with significant neurological damage in patients with traumatic brain injury. Bradykinin is an inflammatory mediator that may contribute to cerebral oedema by increasing the permeability of the blood-brain barrier. We evaluated the safety and effectiveness of the non-peptide bradykinin B2 receptor antagonist Anatibant in the treatment of patients with traumatic brain injury. During the course of the trial, funding was withdrawn by the sponsor. METHODS Adults with traumatic brain injury and a Glasgow Coma Scale score of 12 or less, who had a CT scan showing an intracranial abnormality consistent with trauma, and were within eight hours of their injury were randomly allocated to low, medium or high dose Anatibant or to placebo. Outcomes were Serious Adverse Events (SAE), mortality 15 days following injury and in-hospital morbidity assessed by the Glasgow Coma Scale (GCS), the Disability Rating Scale (DRS) and a modified version of the Oxford Handicap Scale (HIREOS). RESULTS 228 patients out of a planned sample size of 400 patients were randomised. The risk of experiencing one or more SAEs was 26.4% (43/163) in the combined Anatibant treated group, compared to 19.3% (11/57) in the placebo group (relative risk = 1.37; 95% CI 0.76 to 2.46). All cause mortality in the Anatibant treated group was 19% and in the placebo group 15.8% (relative risk 1.20, 95% CI 0.61 to 2.36). The mean GCS at discharge was 12.48 in the Anatibant treated group and 13.0 in the placebo group. Mean DRS was 11.18 Anatibant versus 9.73 placebo, and mean HIREOS was 3.94 Anatibant versus 3.54 placebo. The differences between the mean levels for GCS, DRS and HIREOS in the Anatibant and placebo groups, when adjusted for baseline GCS, showed a non-significant trend for worse outcomes in all three measures. CONCLUSION This trial did not reach the planned sample size of 400 patients and consequently, the study power to detect an increase in the risk of serious adverse events was reduced. This trial provides no reliable evidence of benefit or harm and a larger trial would be needed to establish safety and effectiveness. TRIAL REGISTRATION This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN23625128.

Efficacy and safety of acupuncture for the treatment of non-specific acute low back pain: a randomised controlled multicentre trial protocol [ISRCTN65814467]

BACKGROUND Low back pain and its associated incapacitating effects constitute an important healthcare and socioeconomic problem, as well as being one of the main causes of disability among adults of working age. The prevalence of non-specific low back pain is very high among the general population, and 60-70% of adults are believed to have suffered this problem at some time. Nevertheless, few randomised clinical trials have been made of the efficacy and efficiency of acupuncture with respect to acute low back pain. The present study is intended to assess the efficacy of acupuncture for acute low back pain in terms of the improvement reported on the Roland Morris Questionnaire (RMQ) on low back pain incapacity, to estimate the specific and non-specific effects produced by the technique, and to carry out a cost-effectiveness analysis. METHODS/DESIGN Randomised four-branch controlled multicentre prospective study made to compare semi-standardised real acupuncture, sham acupuncture (acupuncture at non-specific points), placebo acupuncture and conventional treatment. The patients are blinded to the real, sham and placebo acupuncture treatments. Patients in the sample present symptoms of non specific acute low back pain, with a case history of 2 weeks or less, and will be selected from working-age patients, whether in paid employment or not, referred by General Practitioners from Primary Healthcare Clinics to the four clinics participating in this study. In order to assess the primary and secondary result measures, the patients will be requested to fill in a questionnaire before the randomisation and again at 3, 12 and 48 weeks after starting the treatment. The primary result measure will be the clinical relevant improvement (CRI) at 3 weeks after randomisation. We define CRI as a reduction of 35% or more in the RMQ results. DISCUSSION This study is intended to obtain further evidence on the effectiveness of acupuncture on acute low back pain and to isolate the specific and non-specific effects of the treatment.