Complete hazard ranking to analyze right-censored data: An ALS survival study.
ABSTRACT: Survival analysis represents an important outcome measure in clinical research and clinical trials; further, survival ranking may offer additional advantages in clinical trials. In this study, we developed GuanRank, a non-parametric ranking-based technique to transform patients' survival data into a linear space of hazard ranks. The transformation enables the utilization of machine learning base-learners including Gaussian process regression, Lasso, and random forest on survival data. The method was submitted to the DREAM Amyotrophic Lateral Sclerosis (ALS) Stratification Challenge. Ranked first place, the model gave more accurate ranking predictions on the PRO-ACT ALS dataset in comparison to Cox proportional hazard model. By utilizing right-censored data in its training process, the method demonstrated its state-of-the-art predictive power in ALS survival ranking. Its feature selection identified multiple important factors, some of which conflicts with previous studies.
Project description:With thousands of pesticides registered by the United States Environmental Protection Agency, it not feasible to sample for all pesticides applied in agricultural communities. Hazard-ranking pesticides based on use, toxicity, and exposure potential can help prioritize community-specific pesticide hazards. This study applied hazard-ranking schemes for cancer, endocrine disruption, and reproductive/developmental toxicity in Yuma County, Arizona. An existing cancer hazard-ranking scheme was modified, and novel schemes for endocrine disruption and reproductive/developmental toxicity were developed to rank pesticide hazards. The hazard-ranking schemes accounted for pesticide use, toxicity, and exposure potential based on chemical properties of each pesticide. Pesticides were ranked as hazards with respect to each health effect, as well as overall chronic health effects. The highest hazard-ranked pesticides for overall chronic health effects were maneb, metam-sodium, trifluralin, pronamide, and bifenthrin. The relative pesticide rankings were unique for each health effect. The highest hazard-ranked pesticides differed from those most heavily applied, as well as from those previously detected in Yuma homes over a decade ago. The most hazardous pesticides for cancer in Yuma County, Arizona were also different from a previous hazard-ranking applied in California. Hazard-ranking schemes that take into account pesticide use, toxicity, and exposure potential can help prioritize pesticides of greatest health risk in agricultural communities. This study is the first to provide pesticide hazard-rankings for endocrine disruption and reproductive/developmental toxicity based on use, toxicity, and exposure potential. These hazard-ranking schemes can be applied to other agricultural communities for prioritizing community-specific pesticide hazards to target decreasing health risk.
Project description:The proportional hazards assumption in the commonly used Cox model for censored failure time data is often violated in scientific studies. Yang and Prentice (2005) proposed a novel semiparametric two-sample model that includes the proportional hazards model and the proportional odds model as sub-models, and accommodates crossing survival curves. The model leaves the baseline hazard unspecified and the two model parameters can be interpreted as the short-term and long-term hazard ratios. Inference procedures were developed based on a pseudo score approach. Although extension to accommodate covariates was mentioned, no formal procedures have been provided or proved. Furthermore, the pseudo score approach may not be asymptotically efficient. We study the extension of the short-term and long-term hazard ratio model of Yang and Prentice (2005) to accommodate potentially time-dependent covariates. We develop efficient likelihood-based estimation and inference procedures. The nonparametric maximum likelihood estimators are shown to be consistent, asymptotically normal, and asymptotically efficient. Extensive simulation studies demonstrate that the proposed methods perform well in practical settings. The proposed method successfully captured the phenomenon of crossing hazards in a cancer clinical trial and identified a genetic marker with significant long-term effect missed by using the proportional hazards model on age-at-onset of alcoholism in a genetic study.
Project description:Because of concerns about the safety of a growing number of engineered nanomaterials (ENM), it is necessary to develop high-throughput screening and in silico data transformation tools that can speed up in vitro hazard ranking. Here, we report the use of a multiparametric, automated screening assay that incorporates sublethal and lethal cellular injury responses to perform high-throughput analysis of a batch of commercial metal/metal oxide nanoparticles (NP) with the inclusion of a quantum dot (QD1). The responses chosen for tracking cellular injury through automated epifluorescence microscopy included ROS production, intracellular calcium flux, mitochondrial depolarization, and plasma membrane permeability. The z-score transformed high volume data set was used to construct heat maps for in vitro hazard ranking as well as showing the similarity patterns of NPs and response parameters through the use of self-organizing maps (SOM). Among the materials analyzed, QD1 and nano-ZnO showed the most prominent lethality, while Pt, Ag, SiO2, Al2O3, and Au triggered sublethal effects but without cytotoxicity. In order to compare the in vitro with the in vivo response outcomes in zebrafish embryos, NPs were used to assess their impact on mortality rate, hatching rate, cardiac rate, and morphological defects. While QDs, ZnO, and Ag induced morphological abnormalities or interfered in embryo hatching, Pt and Ag exerted inhibitory effects on cardiac rate. Ag toxicity in zebrafish differed from the in vitro results, which is congruent with this material's designation as extremely dangerous in the environment. Interestingly, while toxicity in the initially selected QD formulation was due to a solvent (toluene), supplementary testing of additional QDs selections yielded in vitro hazard profiling that reflect the release of chalcogenides. In conclusion, the use of a high-throughput screening, in silico data handling and zebrafish testing may constitute a paradigm for rapid and integrated ENM toxicological screening.
Project description:Approximately one million people in the UK are served by private water supplies (PWS) where main municipal water supply system connection is not practical or where PWS is the preferred option. Chronic exposure to contaminants in PWS may have adverse effects on health. South West England is an area with elevated arsenic concentrations in groundwater and over 9000 domestic dwellings here are supplied by PWS. There remains uncertainty as to the extent of the population exposed to arsenic (As), and the factors predicting such exposure. We describe a hazard assessment model based on simplified geology with the potential to predict exposure to As in PWS. Households with a recorded PWS in Cornwall were recruited to take part in a water sampling programme from 2011 to 2013. Bedrock geologies were aggregated and classified into nine Simplified Bedrock Geological Categories (SBGC), plus a cross-cutting "mineralized" area. PWS were sampled by random selection within SBGCs and some 508 households volunteered for the study. Transformations of the data were explored to estimate the distribution of As concentrations for PWS by SBGC. Using the distribution per SBGC, we predict the proportion of dwellings that would be affected by high concentrations and rank the geologies according to hazard. Within most SBGCs, As concentrations were found to have log-normal distributions. Across these areas, the proportion of dwellings predicted to have drinking water over the prescribed concentration value (PCV) for As ranged from 0% to 20%. From these results, a pilot predictive model was developed calculating the proportion of PWS above the PCV for As and hazard ranking supports local decision making and prioritization. With further development and testing, this can help local authorities predict the number of dwellings that might fail the PCV for As, based on bedrock geology. The model presented here for Cornwall could be applied in areas with similar geologies. Application of the method requires independent validation and further groundwater-derived PWS sampling on other geological formations.
Project description:Zebrafish is an aquatic organism that can be used for high content safety screening of engineered nanomaterials (ENMs). We demonstrate, for the first time, the use of high content bright-field and fluorescence-based imaging to compare the toxicological effect of transition metal oxide (CuO, ZnO, NiO, and Co(3)O(4)) nanoparticles in zebrafish embryos and larvae. High content bright-field imaging demonstrated potent and dose-dependent hatching interference in the embryos, with the exception of Co(3)O(4) which was relatively inert. We propose that the hatching interference was due to the shedding of Cu and Ni ions, compromising the activity of the hatching enzyme, ZHE1, similar to what we previously proposed for Zn(2+). This hypothesis is based on the presence of metal-sensitive histidines in the catalytic center of this enzyme. Co-introduction of a metal ion chelator, diethylene triamine pentaacetic acid (DTPA), reversed the hatching interference of Cu, Zn, and Ni. While neither the embryos nor larvae demonstrated morphological abnormalities, high content fluorescence-based imaging demonstrated that CuO, ZnO, and NiO could induce increased expression of the heat shock protein 70:enhanced green fluorescence protein (hsp70:eGFP) in transgenic zebrafish larvae. Induction of this response by CuO required a higher nanoparticle dose than the amount leading to hatching interference. This response was also DTPA-sensitive. We demonstrate that high content imaging of embryo development, morphological abnormalities, and HSP70 expression can be used for hazard ranking and determining the dose-response relationships leading to ENM effects on the development of the zebrafish embryo.
Project description:Hazard identification is the key step in risk assessment and management of manufactured nanomaterials (NM). However, the rapid commercialisation of nano-enabled products continues to out-pace the development of a prudent risk management mechanism that is widely accepted by the scientific community and enforced by regulators. However, a growing body of academic literature is developing promising quantitative methods. Two approaches have gained significant currency. Bayesian networks (BN) are a probabilistic, machine learning approach while the weight of evidence (WoE) statistical framework is based on expert elicitation. This comparative study investigates the efficacy of quantitative WoE and Bayesian methodologies in ranking the potential hazard of metal and metal-oxide NMs-TiO?, Ag, and ZnO. This research finds that hazard ranking is consistent for both risk assessment approaches. The BN and WoE models both utilize physico-chemical, toxicological, and study type data to infer the hazard potential. The BN exhibits more stability when the models are perturbed with new data. The BN has the significant advantage of self-learning with new data; however, this assumes all input data is equally valid. This research finds that a combination of WoE that would rank input data along with the BN is the optimal hazard assessment framework.
Project description:Within EU FP7 project NANOVALID, the (eco)toxicity of 7 well-characterized engineered nanomaterials (NMs) was evaluated by 15 bioassays in 4 laboratories. The highest tested nominal concentration of NMs was 100?mg/l. The panel of the bioassays yielded the following toxicity order: Ag?>?ZnO?>?CuO?>?TiO2?>?MWCNTs?>?SiO2?>?Au. Ag, ZnO and CuO proved very toxic in the majority of assays, assumingly due to dissolution. The latter was supported by the parallel analysis of the toxicity of respective soluble metal salts. The most sensitive tests/species were Daphnia magna (towards Ag NMs, 24-h EC50?=?0.003?mg Ag/l), algae Raphidocelis subcapitata (ZnO and CuO, 72-h EC50?=?0.14?mg Zn/l and 0.7?mg Cu/l, respectively) and murine fibroblasts BALB/3T3 (CuO, 48-h EC50?=?0.7?mg Cu/l). MWCNTs showed toxicity only towards rat alveolar macrophages (EC50?=?15.3?mg/l) assumingly due to high aspect ratio and TiO2 towards R. subcapitata (EC50?=?6.8?mg Ti/l) due to agglomeration of TiO2 and entrapment of algal cells. Finally, we constructed a decision tree to select the bioassays for hazard ranking of NMs. For NM testing, we recommend a multitrophic suite of 4 in vitro (eco)toxicity assays: 48-h D. magna immobilization (OECD202), 72-h R. subcapitata growth inhibition (OECD201), 30-min Vibrio fischeri bioluminescence inhibition (ISO2010) and 48-h murine fibroblast BALB/3T3 neutral red uptake in vitro (OECD129) representing crustaceans, algae, bacteria and mammalian cells, respectively. Notably, our results showed that these assays, standardized for toxicity evaluation of "regular" chemicals, proved efficient also for shortlisting of hazardous NMs. Additional assays are recommended for immunotoxicity evaluation of high aspect ratio NMs (such as MWCNTs).
Project description:Clinical trials often collect multiple outcomes on each patient, as the treatment may be expected to affect the patient on many dimensions. For example, a treatment for a neurological disease such as ALS is intended to impact several dimensions of neurological function as well as survival. The assessment of treatment on the basis of multiple outcomes is challenging, both in terms of selecting a test and interpreting the results. Several global tests have been proposed, and we provide a general approach to selecting and executing a global test. The tests require minimal parametric assumptions, are flexible about weighting of the various outcomes, and are appropriate even when some or all of the outcomes are censored. The test we propose is based on a simple scoring mechanism applied to each pair of subjects for each endpoint. The pairwise scores are then reduced to a summary score, and a rank-sum test is applied to the summary scores. This can be seen as a generalization of previously proposed nonparametric global tests (e.g., O'Brien, 1984). We discuss the choice of optimal weighting schemes based on power and relative importance of the outcomes. As the optimal weights are generally unknown in practice, we also propose an adaptive weighting scheme and evaluate its performance in simulations. We apply the methods to analyze the impact of a treatment on neurological function and death in an ALS trial.
Project description:The integration of rapid assays, large datasets, informatics, and modeling can overcome current barriers in understanding nanomaterial structure-toxicity relationships by providing a weight-of-the-evidence mechanism to generate hazard rankings for nanomaterials. Here, we present the use of a rapid, low-cost assay to perform screening-level toxicity evaluations of nanomaterials in vivo. Calculated EZ Metric scores, a combined measure of morbidity and mortality in developing embryonic zebrafish, were established at realistic exposure levels and used to develop a hazard ranking of diverse nanomaterial toxicity. Hazard ranking and clustering analysis of 68 diverse nanomaterials revealed distinct patterns of toxicity related to both the core composition and outermost surface chemistry of nanomaterials. The resulting clusters guided the development of a surface chemistry-based model of gold nanoparticle toxicity. Our findings suggest that risk assessments based on the size and core composition of nanomaterials alone may be wholly inappropriate, especially when considering complex engineered nanomaterials. Research should continue to focus on methodologies for determining nanomaterial hazard based on multiple sub-lethal responses following realistic, low-dose exposures, thus increasing the availability of quantitative measures of nanomaterial hazard to support the development of nanoparticle structure-activity relationships.
Project description:In vitro high throughput screening platforms based on mechanistic injury pathways are been used for hazard assessment of engineered nanomaterials (ENM). Toxicity screening and other in vitro nanotoxicology assessment efforts in essence compare and rank nanomaterials relative to each other. We hypothesize that this ranking of ENM is susceptible to dispersion and dosimetry protocols, which continue to be poorly standardized. Our objective was to quantitate the impact of dosimetry on toxicity ranking of ENM. A set of eight well-characterized and diverse low aspect ratio ENMs, were utilized. The recently developed in vitro dosimetry platform at Harvard, which includes preparation of fairly monodispersed suspensions, measurement of the effective density of formed agglomerates in culture media and fate and transport modeling was used for calculating the effective dose delivered to cells as a function of time. Changes in the dose-response relationships between the administered and delivered dose were investigated with two representative endpoints, cell viability and IL-8 production, in the human monocytic THP-1 cells. The slopes of administered/delivered dose-response relationships changed 1:4.94 times and were ENM-dependent. The overall relative ranking of ENM intrinsic toxicity also changed considerably, matching notably better the in vivo inflammation data (R(2?)=?0.97 versus 0.64). This standardized dispersion and dosimetry methodology presented here is generalizable to low aspect ratio ENMs. Our findings further reinforce the need to reanalyze and reinterpret in vitro ENM hazard ranking data published in the nanotoxicology literature in the light of dispersion and dosimetry considerations (or lack thereof) and to adopt these protocols in future in vitro nanotoxicology testing.