Antonio Eleuteri

Estimating prognosis for survival after treatment of choroidal melanoma

Choroidal melanoma is fatal in about 50% of patients. This is because of metastatic disease, which usually involves the liver. Kaplan-Meier survival curves based only on tumor size and extent do not give a true indication of prognosis. This is because the survival prognosis of choroidal melanoma correlates not only with clinical stage but also with histologic grade, genetic type, and competing causes of death. We have developed an online tool that predicts survival using all these data also taking normal life-expectancy into account. The estimated prognosis is accurate enough to be relevant to individual patients. Such personalized prognostication improves the well-being of patients having an excellent survival probability, not least because it spares them from unnecessary screening tests. Such screening can be targeted at high-risk patients, so that metastases are detected sooner, thereby enhancing any opportunities for treatment. Concerns about psychological harm have proved exaggerated. At least in Britain, patients want to know their prognosis, even if this is poor. The ability to select patients with a high risk of metastasis improves prospects for randomised studies evaluating systemic adjuvant therapy aimed at preventing or delaying metastatic disease. Furthermore, categorization of tissue samples according to survival prognosis enables laboratory studies to be undertaken without waiting many years for survival to be measured. As a result of advances in histologic and genetic studies, biopsy techniques and statistics, prognostication has become established as a routine procedure in our clinical practice, thereby enhancing the care of patients with uveal melanoma.

Artificial Neural Networks Estimating Survival Probability after Treatment of Choroidal Melanoma

PURPOSEnTo describe neural networks predicting survival from choroidal melanoma (i.e., any uveal melanoma involving choroid) and to demonstrate the value of entering age, sex, clinical stage, cytogenetic type, and histologic grade into the predictive model.nnnDESIGNnNonrandomized case series.nnnPARTICIPANTSnPatients resident in mainland Britain treated by the first author for choroidal melanoma between 1984 and 2006.nnnMETHODSnA conditional hazard estimating neural network (CHENN) was trained according to the Bayesian formalism with a training set of 1780 patients and evaluated with a test set of another 874 patients. Conditional hazard estimating neural network-generated survival curves were compared with those obtained with Kaplan-Meier analyses. A second model was created with information on chromosome 3 loss, using training and test sets of 211 and 140 patients, respectively.nnnMAIN OUTCOME MEASURESnComparison of CHENN survival curves with Kaplan-Meier analyses. Representative results showing all-cause survival and inferred melanoma-specific mortality, according to age, sex, clinical stage, cytogenetic type, and histologic grade.nnnRESULTSnThe predictive model plotted a survival curve with 95% credibility intervals for patients with melanoma according to relevant risk factors: age, sex, largest basal tumor diameter, ciliary body involvement, extraocular extension, tumor cell type, closed loops, mitotic rate, and chromosome 3 loss (i.e., monosomy 3). A survival curve for the age-matched general population of the same sex allowed estimation of the melanoma-related mortality. All-cause survival curves generated by the CHENN matched those produced with Kaplan-Meier analysis (Kolmogorov-Smirnov, P<0.05). In older patients, however, the estimated melanoma-related mortality was lower with the CHENN, which accounted for competing risks, unlike Kaplan-Meier analysis. Largest basal tumor diameter was most predictive of mortality in tumors showing histologic and cytogenetic features of high-grade malignancy. Ciliary body involvement and extraocular extension lost significance when cytogenetic and histologic data were included in the model. Patients with a monosomy 3 melanoma of a particular size were predicted to have shorter survival if their tumor showed epithelioid cells and closed loops.nnnCONCLUSIONSnEstimation of survival prognosis in patients with choroidal melanoma requires multivariate assessment of age, sex, clinical tumor stage, cytogenetic melanoma type, and histologic grade of malignancy.

Partial Logistic Artificial Neural Network for Competing Risks Regularized With Automatic Relevance Determination

Time-to-event analysis is important in a wide range of applications from clinical prognosis to risk modeling for credit scoring and insurance. In risk modeling, it is sometimes required to make a simultaneous assessment of the hazard arising from two or more mutually exclusive factors. This paper applies to an existing neural network model for competing risks (PLANNCR), a Bayesian regularization with the standard approximation of the evidence to implement automatic relevance determination (PLANNCR-ARD). The theoretical framework for the model is described and its application is illustrated with reference to local and distal recurrence of breast cancer, using the data set of Veronesi (1995).

Enhancing survival prognostication in patients with choroidal melanoma by integrating pathologic, clinical and genetic predictors of metastasis

Survival in choroidal melanoma was modelled using accelerated failure time models. We combined pathological, clinical and genetic data, using imputation techniques. Performance was assessed using n-fold cross-validation. Using data from 3653 patients, we generated two models; the first using clinical data only and the second using clinical and laboratory data. The c-index of discrimination was 0.75 for the clinical model and 0.79 for the laboratory model. Calibration showed good correlation between predicted and observed mortality (p-value: 0.699 for clinical model and 0.801 for laboratory model). We conclude that our model provides reasonably reliable prognosis relevant to individual patients.

Retrospective review of 50 eyes with long-term silicone oil tamponade for more than 12 months

BackgroundSilicone oil (SO) is an established tamponade in treating complex vitreoretinal diseases. Although SO is intended to be removed after several weeks to months, permanent SO might be unavoidable in a small subgroup of patients with an extremely complicated clinical course. The aim of this study is to describe the long-term effects of intraocular SO tamponade.MethodsThis retrospective study included 50 patients with intraocular SO (Oxane 5700 Bausch & Lomb, Kingston-upon-Thames, UK) for at least 12xa0months. The most common reasons for long-term SO tamponade were: retinal re-detachment (re-RD), proliferative diabetic retinopathy (PDR), ocular trauma, and persistent hypotony.ResultsMean age was 59.2u2009±u200918.4xa0years, and mean duration of silicone oil in the eye was 54.5u2009±u200958.6xa0months (median, 30xa0months). The average number of previous surgeries were 2.2u2009±u20091.5. Anatomic success was achieved in 37/50 (74%) of patients. Visual acuities (logMAR) were 1.8u2009±u20090.6, 1.6u2009±u20090.6, 2u2009±u20090.7 and intraocular pressures (mmHg) were 15.6u2009±u20097, 15.7u2009±u20095.5, 16.5u2009±u20097.1 at 3xa0months, 1xa0year and at last follow-up respectively. The main long-term silicone-oil-related complications observed were: band keratopathy (8%), corneal decompensation (12%), iris rubeosis (14%), and optic neuropathy (28%). Forty percent of patients achieved ambulatory vision in the SO-filled eye at final follow-up.ConclusionLong-term silicone oil can be a last-resort option in selected patients with severe vitreoretinal disease. Anterior and posterior segment complications did occur at significant rates. Forty percent of our patients maintained ambulatory vision. The actual number of patients that achieved satisfactory stereopsis and benefited functionally from long-term SO was much less [7/50 (14%)].

An Efficient Time-Varying Filter for Detrending and Bandwidth Limiting the Heart Rate Variability Tachogram without Resampling: MATLAB Open-Source Code and Internet Web-Based Implementation

The heart rate variability (HRV) signal derived from the ECG is a beat-to-beat record of RR intervals and is, as a time series, irregularly sampled. It is common engineering practice to resample this record, typically at 4u2009Hz, onto a regular time axis for analysis in advance of time domain filtering and spectral analysis based on the DFT. However, it is recognised that resampling introduces noise and frequency bias. The present work describes the implementation of a time-varying filter using a smoothing priors approach based on a Gaussian process model, which does not require data to be regular in time. Its output is directly compatible with the Lomb-Scargle algorithm for power density estimation. A web-based demonstration is available over the Internet for exemplar data. The MATLAB (MathWorks Inc.) code can be downloaded as open source.

A web-based tool for the assessment of discrimination and calibration properties of prognostic models

Prognostic models are developed to assist clinicians in making decisions regarding treatment and follow-up management. The accuracy of these models is often assessed either in terms of their discrimination performance or calibration but rarely both. In this paper, we describe the development of an online tool for discrimination using Harrell C index and calibration using a Hosmer-Lemeshow type analysis (http://clinengnhs.liv.ac.uk/AADP/AADP_Welcome.htm). We show examples of using the tool on real data. We highlight situations where the model performed well in terms of either discrimination or calibration but not both depending on the sample size of the test set. We conclude that prognostic models should be assessed both in terms of discrimination and calibration and that calibration analysis should be carried out numerically and graphically.

External Validation of the Liverpool Uveal Melanoma Prognosticator Online

PurposenTo validate the Liverpool Uveal Melanoma Prognosticator Online (LUMPO) in a cohort of patients treated at the University of California-San Francisco (UCSF).nnnMethodsnA retrospective chart review was performed of 390 patients treated between 2002 and 2007 for choroidal melanoma at UCSF. Similar patients (n = 1175) treated at the Liverpool Ocular Oncology Centre (LOOC) were included in the study. The data were analyzed using the model previously developed for LUMPO, an online prognostication tool combining multiple prognostic factors. Main outcome measures included all-cause mortality and melanoma-specific mortality. Reliability of the survival estimates in each group of patients was indicated by the C-indices of discrimination and Hosmer-Lemeshow test.nnnResultsnPatients treated at UCSF tended to be younger with thicker tumors, and were more likely to receive proton beam radiotherapy as primary treatment compared to patients at LOOC. There were no significance differences with respect to ciliary body involvement, melanoma cytomorphology, and mitotic counts between the two groups. Death occurred in 140/390 (35%) patients from UCSF and 409/1175 (34%) patients from LOOC, with no difference in overall mortality by Kaplan-Meier analysis (log rank test, P = 0.503). For all-cause mortality and melanoma-specific mortality, the C-index of discrimination and Hosmer-Lemeshow test at 5 years after treatment indicated good discrimination performance of the model, with no statistically significant difference between observed and predicted survival.nnnConclusionsnDespite differences between the two cohorts, external validation in patients treated at UCSF indicates that LUMPO estimated the all-cause and melanoma-specific mortality well.

Continuous and Discrete Time Survival Analysis: Neural Network Approaches

In this paper we describe and compare two neural network models aimed at survival analysis modeling, based on formulations in continuous and discrete time. Learning in both models is approached in a Bayesian inference framework. We test the models on a real survival analysis problem, and we show that both models exhibit good discrimination and calibration capabilities. The C index of discrimination varied from 0.8 (SE=0.093) at year 1, to 0.75 (SE=0.034) at year 7 for the continuous time model; from 0.81 (SE=0.07) at year 1, to 0.75 (SE=0.033) at year 7 for the discrete time model. For both models the calibration was good (p<0.05) up to 7 years.

Ambulatory heart rate range predicts mode-specific mortality and hospitalisation in chronic heart failure

Objective We aimed to define the prognostic value of the heart rate range during a 24u2005h period in patients with chronic heart failure (CHF). Methods Prospective observational cohort study of 791 patients with CHF associated with left ventricular systolic dysfunction. Mode-specific mortality and hospitalisation were linked with ambulatory heart rate range (AHRR; calculated as maximum minus minimum heart rate using 24u2005h Holter monitor data, including paced and non-sinus complexes) in univariate and multivariate analyses. Findings were then corroborated in a validation cohort of 408 patients with CHF with preserved or reduced left ventricular ejection fraction. Results After a mean 4.1u2005years of follow-up, increasing AHRR was associated with reduced risk of all-cause, sudden, non-cardiovascular and progressive heart failure death in univariate analyses. After accounting for characteristics that differed between groups above and below median AHRR using multivariate analysis, AHRR remained strongly associated with all-cause mortality (HR 0.991/bpm increase in AHRR (95% CI 0.999 to 0.982); p=0.046). AHRR was not associated with the risk of any non-elective hospitalisation, but was associated with heart-failure-related hospitalisation. AHRR was modestly associated with the SD of normal-to-normal beats (R2=0.2; p<0.001) and with peak exercise-test heart rate (R2=0.33; p<0.001). Analysis of the validation cohort revealed AHRR to be associated with all-cause and mode-specific death as described in the derivation cohort. Conclusions AHRR is a novel and readily available prognosticator in patients with CHF, which may reflect autonomic tone and exercise capacity.