A BAYESIAN APPROACH TO DOSE-RESPONSE ASSESSMENT AND DRUG-DRUG INTERACTION ANALYSIS: APPLICATION TO IN VITRO STUDIES

The considerable search for synergistic agents in cancer research is motivated by the therapeutic benefits achieved by combining anti-cancer agents. Synergistic agents make it possible to reduce dosage while maintaining or enhancing a desired effect. Other favorable outcomes of synergistic agents include reduction in toxicity and minimizing or delaying drug resistance. Dose-response assessment and drug-drug interaction analysis play an important part in the drug discovery process, however analysis are often poorly done. This dissertation is an effort to notably improve dose-response assessment and drug-drug interaction analysis. The most commonly used method in published analysis is the Median-Effect Principle/Combination Index method (Chou and Talalay, 1984). The Median-Effect Principle/Combination Index method leads to inefficiency by ignoring important sources of variation inherent in dose-response data and discarding data points that do not fit the Median-Effect Principle. Previous work has shown that the conventional method yields a high rate of false positives (Boik, Boik, Newman, 2008; Hennessey, Rosner, Bast, Chen, 2010) and, in some cases, low power to detect synergy. There is a great need for improving the current methodology. We developed a Bayesian framework for dose-response modeling and drug-drug interaction analysis. First, we developed a hierarchical meta-regression dose-response model that accounts for various sources of variation and uncertainty and allows one to incorporate knowledge from prior studies into the current analysis, thus offering a more efficient and reliable inference. Second, in the case that parametric dose-response models do not fit the data, we developed a practical and flexible nonparametric regression method for meta-analysis of independently repeated dose-response experiments. Third, and lastly, we developed a method, based on Loewe additivity that allows one to quantitatively assess interaction between two agents combined at a fixed dose ratio. The proposed method makes a comprehensive and honest account of uncertainty within drug interaction assessment. Extensive simulation studies show that the novel methodology improves the screening process of effective/synergistic agents and reduces the incidence of type I error. We consider an ovarian cancer cell line study that investigates the combined effect of DNA methylation inhibitors and histone deacetylation inhibitors in human ovarian cancer cell lines. The hypothesis is that the combination of DNA methylation inhibitors and histone deacetylation inhibitors will enhance antiproliferative activity in human ovarian cancer cell lines compared to treatment with each inhibitor alone. By applying the proposed Bayesian methodology, in vitro synergy was declared for DNA methylation inhibitor, 5-AZA-2'-deoxycytidine combined with one histone deacetylation inhibitor, suberoylanilide hydroxamic acid or trichostatin A in the cell lines HEY and SKOV3. This suggests potential new epigenetic therapies in cell growth inhibition of ovarian cancer cells.

DOSE RESPONSE EFFECT OF Paracoccidioides brasiliensis IN AN EXPERIMENTAL MODEL OF ARTHRITIS

Paracoccidioidomycosis (PCM) is caused by the dimorphic fungus Paracoccidioides brasiliensis (Pb) and corresponds to prevalent systemic mycosis in Latin America. The aim of the present work was to evaluate the dose response effect of the fungal yeast phase for the standardization of an experimental model of septic arthritis. The experiments were performed with groups of 14 rats that received doses of 103, 104 or 105 P. brasiliensis (Pb18) cells. The fungi were injected in 50 µL of phosphate-buffered saline (PBS) directly into the knee joints of the animals. The following parameters were analyzed in this work: the formation of swelling in knees infused with yeast cells and the radiological and anatomopathological alterations, besides antibody titer by ELISA. After 15 days of infection, signs of inflammation were evident. At 45 days, some features of damage and necrosis were observed in the articular cartilage. The systemic dissemination of the fungus was observed in 11% of the inoculated animals, and it was concluded that the experimental model is able to mimic articular PCM in humans and that the dose of 105 yeast cells can be used as standard in this model.

Dose-response curve to soil applied herbicides and susceptibility evaluation of different amaranthus species using model identity

Greenhouse studies were conducted in 2008-2009 with the objective of adjusting dose-response curves of the main soil-applied herbicides currently used in cotton for the control of Amaranthus viridis, A. hybridus, A. spinosus, A. lividus, as well as comparing susceptibility among different species, using the identity test models. Thirty six individual experiments were simultaneously carried out in greenhouse, in a sandy clay loam soil (21% clay, 2.36% OM) combining increasing doses of the herbicides alachlor, clomazone, diuron, oxyfluorfen, pendimethalin, prometryn, S-metolachlor, and trifluralin applied to each species. Dose-response curves were adjusted for visual weed control at 28 days after herbicide application and doses required for 80% (C80) and 95% (C95) control were calculated. All herbicides, except clomazone and trifluralin, provided efficient control of most Amaranthus species, but substantial differences in susceptibility to herbicides were found. In general, A. lividus was the least sensitive species, whereas A. spinosus demonstrated the highest sensitivity to herbicides. Alachlor, diuron, oxyfluorfen, pendimethalin, S-metolachlor, and prometryn are efficient alternatives to control Amaranthus spp. in a range of doses that are currently lower than those recommended to cotton.

Dose-response resistance to HIV-1/MuLV pseudotype virus ex vivo in a hairpin ribozyme transgenic mouse model

We have investigated the efficacy of a hairpin ribozyme targeting the 5′ leader sequence of HIV-1 RNA in a transgenic model system. Primary spleen cells derived from transgenic or control mice were infected with HIV-1/MuLV pseudotype virus. A significantly reduced susceptibility to infection in ribozyme-expressing transgenic spleen cells (P = 0.01) was shown. Variation of transgene-expression levels between littermates revealed a dose response between ribozyme expression and viral resistance, with an estimated cut off value below 0.2 copies of hairpin ribozyme per cell. These findings open up possibilities for studies on ribozyme efficacy and anti-HIV-1 gene therapy.

Routines in a European aquaculture company and the development of a Flavobacterium psychrophilum dose response pre-model in adult Rainbow Trout

An internship in a European company dealing with aquaculture and biotechnology - AquaBioTech Group, Malta - was undertaken to complete the Master Degree of Science in Aquaculture of the School of Tourism and Maritime Technology of the Polytechnic Institute of Leiria. Biotechnology and aquaculture are two areas that have been synergistically used to contribute for the progress and improvement of fish production. The AquaBioTech Group is an example of a company able to integrate these areas to maximizing their services. Located in Mosta (Malta) the company operates in a sustainable way using Recirculation Aquaculture Systems (RAS) to maintain aquaculture species. In collaboration with several companies and institutions, the AquaBioTech Group is involved and supports the development of important international research projects. The present report focuses on two important parts of the internship performed during 6 months. Initially, it will cover the operation and constitution of the company, describing the routines and techniques acquired. Then, it will describe a pathology trial that forms the practical and scientific component of this report. Despite the limitation to describe some confidential assays, this trial consisted in the infection of Rainbow Trout (Oncorhynchus mykiss) with the bacterium Flavobacterium psychrophilum in order to evaluate the mortality rates over time. The internship served to solidify theoretical knowledge acquired during the academic training, develop professional skills and provide an understanding of jobs available on the market.

Biological Effects and Dose-Response Assessment of Diesel Exhaust Particles on In Vitro Early Embryo Development in Mice

An increased risk of early pregnancy loss in women briefly exposed to high levels of ambient particulate matter during the preconceptional period was recently observed. The effects of this exposure on early embryo development are unknown. This study was designed to assess the dose-response and biological effects of diesel exhaust particles (DEP) on in vitro embryo development using the in vitro fertilization (IVF) mouse model. Zygotes obtained from superovulated mice after IVF were randomly cultured in different DEP concentrations (0, 0.2, 2, and 20 mu g/cm(2)) for 5 days and observed for their capacity to attach and develop on a fibronectin matrix until day 8. Main outcome measures included blastocyst rates 96 and 120 h after insemination, hatching discriminatory score, total cell count, proportion of cell allocation to inner cell mass (ICM) and trophectoderm (TE), ICM morphology, attachment rate and outgrowth area, apoptosis and necrosis rates, and Oct-4 and Cdx-2 expression. Multivariate analysis showed a negative dose-dependent effect on early embryo development and hatching process, blastocyst cell allocation, and ICM morphology. Although blastocyst attachment and outgrowth were not affected by DEP, a significant impairment of ICM integrity was observed in day 8 blastocysts. Cell death through apoptosis was significantly higher after DEP exposure. Oct-4 expression and the Oct-4/Cdx-2 ratio were significantly decreased in day 5 blastocysts irrespective of DEP concentration. Results suggest that DEP appear to play an important role in disrupting cell lineage segregation and ICM morphological integrity even at lower concentrations, compromising future growth and viability of the blastocyst.

A peripheral dose calculation model for estimating second cancer risk after breast radiotherapy

AbstractBreast cancer is one of the most common cancers affecting one in eight women during their lives. Survival rates have increased steadily thanks to early diagnosis with mammography screening and more efficient treatment strategies. Post-operative radiation therapy is a standard of care in the management of breast cancer and has been shown to reduce efficiently both local recurrence rate and breast cancer mortality. Radiation therapy is however associated with some late effects for long-term survivors. Radiation-induced secondary cancer is a relatively rare but severe late effect of radiation therapy. Currently, radiotherapy plans are essentially optimized to maximize tumor control and minimize late deterministic effects (tissue reactions) that are mainly associated with high doses (» 1 Gy). With improved cure rates and new radiation therapy technologies, it is also important to evaluate and minimize secondary cancer risks for different treatment techniques. This is a particularly challenging task due to the large uncertainties in the dose-response relationship.In contrast with late deterministic effects, secondary cancers may be associated with much lower doses and therefore out-of-field doses (also called peripheral doses) that are typically inferior to 1 Gy need to be determined accurately. Out-of-field doses result from patient scatter and head scatter from the treatment unit. These doses are particularly challenging to compute and we characterized it by Monte Carlo (MC) calculation. A detailed MC model of the Siemens Primus linear accelerator has been thoroughly validated with measurements. We investigated the accuracy of such a model for retrospective dosimetry in epidemiological studies on secondary cancers. Considering that patients in such large studies could be treated on a variety of machines, we assessed the uncertainty in reconstructed peripheral dose due to the variability of peripheral dose among various linac geometries. For large open fields (> 10x10 cm2), the uncertainty would be less than 50%, but for small fields and wedged fields the uncertainty in reconstructed dose could rise up to a factor of 10. It was concluded that such a model could be used for conventional treatments using large open fields only.The MC model of the Siemens Primus linac was then used to compare out-of-field doses for different treatment techniques in a female whole-body CT-based phantom. Current techniques such as conformai wedged-based radiotherapy and hybrid IMRT were investigated and compared to older two-dimensional radiotherapy techniques. MC doses were also compared to those of a commercial Treatment Planning System (TPS). While the TPS is routinely used to determine the dose to the contralateral breast and the ipsilateral lung which are mostly out of the treatment fields, we have shown that these doses may be highly inaccurate depending on the treatment technique investigated. MC shows that hybrid IMRT is dosimetrically similar to three-dimensional wedge-based radiotherapy within the field, but offers substantially reduced doses to out-of-field healthy organs.Finally, many different approaches to risk estimations extracted from the literature were applied to the calculated MC dose distribution. Absolute risks varied substantially as did the ratio of risk between two treatment techniques, reflecting the large uncertainties involved with current risk models. Despite all these uncertainties, the hybrid IMRT investigated resulted in systematically lower cancer risks than any of the other treatment techniques. More epidemiological studies with accurate dosimetry are required in the future to construct robust risk models. In the meantime, any treatment strategy that reduces out-of-field doses to healthy organs should be investigated. Electron radiotherapy might offer interesting possibilities with this regard.RésuméLe cancer du sein affecte une femme sur huit au cours de sa vie. Grâce au dépistage précoce et à des thérapies de plus en plus efficaces, le taux de guérison a augmenté au cours du temps. La radiothérapie postopératoire joue un rôle important dans le traitement du cancer du sein en réduisant le taux de récidive et la mortalité. Malheureusement, la radiothérapie peut aussi induire des toxicités tardives chez les patients guéris. En particulier, les cancers secondaires radio-induits sont une complication rare mais sévère de la radiothérapie. En routine clinique, les plans de radiothérapie sont essentiellement optimisées pour un contrôle local le plus élevé possible tout en minimisant les réactions tissulaires tardives qui sont essentiellement associées avec des hautes doses (» 1 Gy). Toutefois, avec l'introduction de différentes nouvelles techniques et avec l'augmentation des taux de survie, il devient impératif d'évaluer et de minimiser les risques de cancer secondaire pour différentes techniques de traitement. Une telle évaluation du risque est une tâche ardue étant donné les nombreuses incertitudes liées à la relation dose-risque.Contrairement aux effets tissulaires, les cancers secondaires peuvent aussi être induits par des basses doses dans des organes qui se trouvent hors des champs d'irradiation. Ces organes reçoivent des doses périphériques typiquement inférieures à 1 Gy qui résultent du diffusé du patient et du diffusé de l'accélérateur. Ces doses sont difficiles à calculer précisément, mais les algorithmes Monte Carlo (MC) permettent de les estimer avec une bonne précision. Un modèle MC détaillé de l'accélérateur Primus de Siemens a été élaboré et validé avec des mesures. La précision de ce modèle a également été déterminée pour la reconstruction de dose en épidémiologie. Si on considère que les patients inclus dans de larges cohortes sont traités sur une variété de machines, l'incertitude dans la reconstruction de dose périphérique a été étudiée en fonction de la variabilité de la dose périphérique pour différents types d'accélérateurs. Pour de grands champs (> 10x10 cm ), l'incertitude est inférieure à 50%, mais pour de petits champs et des champs filtrés, l'incertitude de la dose peut monter jusqu'à un facteur 10. En conclusion, un tel modèle ne peut être utilisé que pour les traitements conventionnels utilisant des grands champs.Le modèle MC de l'accélérateur Primus a été utilisé ensuite pour déterminer la dose périphérique pour différentes techniques dans un fantôme corps entier basé sur des coupes CT d'une patiente. Les techniques actuelles utilisant des champs filtrés ou encore l'IMRT hybride ont été étudiées et comparées par rapport aux techniques plus anciennes. Les doses calculées par MC ont été comparées à celles obtenues d'un logiciel de planification commercial (TPS). Alors que le TPS est utilisé en routine pour déterminer la dose au sein contralatéral et au poumon ipsilatéral qui sont principalement hors des faisceaux, nous avons montré que ces doses peuvent être plus ou moins précises selon la technTque étudiée. Les calculs MC montrent que la technique IMRT est dosimétriquement équivalente à celle basée sur des champs filtrés à l'intérieur des champs de traitement, mais offre une réduction importante de la dose aux organes périphériques.Finalement différents modèles de risque ont été étudiés sur la base des distributions de dose calculées par MC. Les risques absolus et le rapport des risques entre deux techniques de traitement varient grandement, ce qui reflète les grandes incertitudes liées aux différents modèles de risque. Malgré ces incertitudes, on a pu montrer que la technique IMRT offrait une réduction du risque systématique par rapport aux autres techniques. En attendant des données épidémiologiques supplémentaires sur la relation dose-risque, toute technique offrant une réduction des doses périphériques aux organes sains mérite d'être étudiée. La radiothérapie avec des électrons offre à ce titre des possibilités intéressantes.

Fluoride dose response in pH-cycling models using bovine enamel

The aim of this study was to establish methodologies for verification of the fluoride solution dose-response relationship using bovine enamel and pH-cycling models. Six models of the cariogenic challenge were performed, varying the time of demineralization and pH, time of remineralization, composition of de- and remineralization solutions, frequency and time of application of treatment solutions and pH-cycling duration. For the evaluation of the fluoride effect on caries dynamics, two proposed models provided for improvement in standardization of methods leading to a higher level of precision, demonstrating a dose response between treatments with regard to surface microhardness and Delta Z. For the evaluation of the fluoride effect on enamel remineralization, the addition of fluoride to the de- and remineralization solutions and the reduction of frequency and time of application of fluoride solutions led to a more suitable pH-cycling model. Copyright (C) 2005 S. Karger AG, Basel.

In situ protocol for the detemination of dose-response effect of low-fluoride dentifrices on enamel remineralization

No in situ protocol has assessed the dose-response effects of fluoride dentifrices involving low-fluoride formulations. Objective: To assess the ability of an in situ remineralization model in determining dose-response effects of dentifrices containing low fluoride concentrations ([F]) on bovine enamel. Material and Methods: Volunteers wore palatal appliances containing demineralized enamel blocks and brushed their teeth and devices with the dentifrices supplied (double-blind, crossover protocol) separately for 3 and 7 days. Surface hardness (SH), integrated subsurface hardness (AKHN) and [F] in enamel were determined. Data were analyzed by ANOVA, Tukey's test and Pearson's correlation (p<0.05). Results: Dose-response relationships were verified between [F] in dentifrices and SH, AKHN and enamel [F]. Higher correlation coefficients between enamel [F] and SH and AKHN were obtained for the 3-day period. Significant differences in SH and AKHN were observed among all groups for the 3-day period, but not between 0-275, 275-550, and 550-1,100 mu g F/g dentifrices for the 7-day period, nor between 3- and 7-day periods for the 1,100 mu g F/g groups. Conclusions: Considering that the peak remineralization capacity of the conventional dentifrice (1,100 mu g F/g) was achieved in 3 days, this experimental period could be used in future studies assessing new dentifrice formulations, especially at low-fluoride concentrations.

Positioning and number of nutritional levels in dose-response trials to estimate the optimal-level and the adjustment of the models

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Improving the accuracy of radiation pneumonitis dose response models

The prognosis for lung cancer patients remains poor. Five year survival rates have been reported to be 15%. Studies have shown that dose escalation to the tumor can lead to better local control and subsequently better overall survival. However, dose to lung tumor is limited by normal tissue toxicity. The most prevalent thoracic toxicity is radiation pneumonitis. In order to determine a safe dose that can be delivered to the healthy lung, researchers have turned to mathematical models predicting the rate of radiation pneumonitis. However, these models rely on simple metrics based on the dose-volume histogram and are not yet accurate enough to be used for dose escalation trials. The purpose of this work was to improve the fit of predictive risk models for radiation pneumonitis and to show the dosimetric benefit of using the models to guide patient treatment planning. The study was divided into 3 specific aims. The first two specifics aims were focused on improving the fit of the predictive model. In Specific Aim 1 we incorporated information about the spatial location of the lung dose distribution into a predictive model. In Specific Aim 2 we incorporated ventilation-based functional information into a predictive pneumonitis model. In the third specific aim a proof of principle virtual simulation was performed where a model-determined limit was used to scale the prescription dose. The data showed that for our patient cohort, the fit of the model to the data was not improved by incorporating spatial information. Although we were not able to achieve a significant improvement in model fit using pre-treatment ventilation, we show some promising results indicating that ventilation imaging can provide useful information about lung function in lung cancer patients. The virtual simulation trial demonstrated that using a personalized lung dose limit derived from a predictive model will result in a different prescription than what was achieved with the clinically used plan; thus demonstrating the utility of a normal tissue toxicity model in personalizing the prescription dose.

CAUSAL INFERENCE WITH A CONTINUOUS TREATMENT AND OUTCOME: ALTERNATIVE ESTIMATORS FOR PARAMETRIC DOSE-RESPONSE FUNCTIONS WITH APPLICATIONS.

Causal inference with a continuous treatment is a relatively under-explored problem. In this dissertation, we adopt the potential outcomes framework. Potential outcomes are responses that would be seen for a unit under all possible treatments. In an observational study where the treatment is continuous, the potential outcomes are an uncountably infinite set indexed by treatment dose. We parameterize this unobservable set as a linear combination of a finite number of basis functions whose coefficients vary across units. This leads to new techniques for estimating the population average dose-response function (ADRF). Some techniques require a model for the treatment assignment given covariates, some require a model for predicting the potential outcomes from covariates, and some require both. We develop these techniques using a framework of estimating functions, compare them to existing methods for continuous treatments, and simulate their performance in a population where the ADRF is linear and the models for the treatment and/or outcomes may be misspecified. We also extend the comparisons to a data set of lottery winners in Massachusetts. Next, we describe the methods and functions in the R package causaldrf using data from the National Medical Expenditure Survey (NMES) and Infant Health and Development Program (IHDP) as examples. Additionally, we analyze the National Growth and Health Study (NGHS) data set and deal with the issue of missing data. Lastly, we discuss future research goals and possible extensions.

Dose-response effects of systemic anandamide administration in mice sequentially submitted to the open field and elevated plus-maze tests

The endocannabinoid system is involved in the control of many physiological functions, including the control of emotional states. In rodents, previous exposure to an open field increases the anxiety-like behavior in the elevated plus-maze. Anxiolytic-like effects of pharmacological compounds that increase endocannabinoid levels have been well documented. However, these effects are more evident in animals with high anxiety levels. Several studies have described characteristic inverted U-shaped dose-response effects of drugs that modulate the endocannabinoid levels. However, there are no studies showing the effects of different doses of exogenous anandamide, an endocannabinoid, in animal models of anxiety. Thus, in the present study, we determined the dose-response effects of exogenous anandamide at doses of 0.01, 0.1, and 1.0 mg/kg in C57BL/6 mice (N = 10/group) sequentially submitted to the open field and elevated plus-maze. Anandamide was diluted in 0.9% saline, ethyl alcohol, Emulphor® (18:1:1) and administered ip (0.1 mL/10 g body weight); control animals received the same volume of anandamide vehicle. Anandamide at the dose of 0.1 mg/kg (but not of 0.01 or 1 mg/kg) increased (P < 0.05) the time spent and the distance covered in the central zone of the open field, as well as the exploration of the open arms of the elevated plus-maze. Thus, exogenous anandamide, like pharmacological compounds that increase endocannabinoid levels, promoted a characteristic inverted U-shaped dose-response effect in animal models of anxiety. Furthermore, anandamide (0.1 mg/kg) induced an anxiolytic-like effect in the elevated plus-maze (P < 0.05) after exposing the animals to the open field test.

The prednisolone suppression test in depression: Dose-response and changes with antidepressant treatment

Depressed patients have reduced glucocorticoid receptor (GR) function, as demonstrated by resistance to the suppressive effects of the synthetic glucocorticoid hormone, and GR agonist, dexamethasone. We have developed a suppressive test with prednisolone, a synthetic glucocorticoid that is similar to cortisol in its pharmacodynamics and pharmacokinetics, and binds to both the GR and the mineralocorticoid receptor (MR). We have found that depressed patients suppress normally to prednisolone, unless they are particularly non-responsive to treatment. In the present study, we evaluated 28 inpatients with treatment-resistant depression (TRD), and compared salivary cortisol secretion (at 0900 h, 1200 h and 1700 h) after placebo or after prednisolone (5 mg), before and after an inpatient treatment admission. Half of the patients (n = 14) reached treatment response. When comparing the assessment between admission and discharge, cortisol output after placebo fell (-26% of area under the curve; p = 0.024) while the output after prednisolone did not change. Moreover, there was no change in the response to prednisolone (percentage suppression) between admission at discharge, and this was not influenced by treatment response. Finally, we could confirm and extend our previously published data with prednisolone (5 mg), showing that depressed patients (n = 12) and controls (n = 12) suppressed equally to both 5 and 10 mg doses of prednisolone. This study suggests that the response to prednisolone is similar in depressed patients and controls at different doses of prednisolone, and does not change with symptomatic improvement. This is in contrast with findings, from us and others, using other measures of hypothalamic-pituitary-adrenal axis function, such as basal cortisol levels or the response to dexamethasone. Thus, we propose that the prednisolone suppression test may offer specific biological and clinical information, related to its action at both the GR and the MR. (C) 2010 Elsevier Ltd. All rights reserved.