Made available in DSpace on 2014-10-09T12:48:28Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:57:28Z (GMT). No. of bitstreams: 1 09066.pdf: 3054181 bytes, checksum: 62e4252d9f76ef15308c15317e1fca18 (MD5) / Dissertacao (Mestrado profissionalizante em lasers em Odontologia) / IPEN/D-MPLO / Intituto de Pesquisas Energeticas e Nucleares, IPEN/CNEN-SP; Faculdade de Odontologia, Universidade de Sao Paulo

Les cellules ganglionnaires transfèrent l'information visuelle de l’œil au cerveau, sous une forme encore débattue. Leurs réponses aux stimuli visuels sont non-linéaires, corrélées entre neurones, et une partie de l'information est présente au niveau de la population seulement. J'étudie d'abord la structure des réponses de population. Les cellules du cortex sont influencées par l'activité globale des neurones avoisinants, mais ces interactions manquaient encore de modèle. Je décris un modèle de population qui reproduit le couplage entre neurones et activité globale. Je montre que les neurones de la rétine de salamandre dépendent de l'activité globale de manière surprenante. Je décris ensuite une méthode pour caractériser la sensibilité de populations de neurones de la rétine de rat à des perturbations d'un stimulus. J'utilise des expériences en boucle fermée pour explorer sélectivement l'espace des perturbations autour d'un stimulus donné. Je montre que les réponses à de petites perturbations peuvent être décrites par une linéarisation de leur probabilité. Leur sensibilité présente des signes de codage efficace. Enfin, je montre comment estimer la sensibilité des réponses d'une population de neurones à partir de leur structure. Je montre que les machines de Boltzmann restreintes (RBMs) sont des modèles précis des corrélations neurales. Pour mesurer le pouvoir de discrimination des neurones, je cherche une métrique neurale telle que les réponses à des stimuli différents soient éloignées, et celles à un même stimulus soient proches. Je montre que les RBMs fournissent des métriques qui surpassent les métriques classiques pour discriminer de petites perturbations du stimulus. / Ganglion cells form the output of the retina: they transfer visual information from the eye to the brain. How they represent information is still debated. Their responses to visual stimuli are highly nonlinear, exhibit strong correlations between neurons, and some information is only present at the population level. I first study the structure of population responses. Recent studies have shown that cortical cells are influenced by the summed activity of neighboring neurons. However, a model for these interactions was still lacking. I describe a model of population activity that reproduces the coupling between each cell and the population activity. Neurons in the salamander retina are found to depend in unexpected ways on the population activity. I then describe a method to characterize the sensitivity of rat retinal neurons to perturbations of a stimulus. Closed-loop experiments are used to explore selectively the space of perturbations around a given stimulus. I show that responses to small perturbations can be described by a local linearization of their probability, and that their sensitivity exhibits signatures of efficient coding. Finally, I show how the sensitivity of neural populations can be estimated from response structure. I show that Restricted Boltzmann Machines (RBMs) are accurate models of neural correlations. To measure the discrimination power of neural populations, I search for a neural metric such that responses to different stimuli are far apart and responses to the same stimulus are close. I show that RBMs provide such neural metrics, and outperform classical metrics at discriminating small stimulus perturbations.

Spectral domain optical coherence tomography (SD-OCT), a non-invasive imaging method, is based on an analysis of a near-infrared light deflected from tisssue layers, that provides detailed images of retinal structures. Nerve cells of the retina, that originate from neuroectoderm, reflect neurodegeneration of the central nervous system (CNS), as well as acute damage of nerve structures caused by optic neuritis. The dissertation first presents established imaging protocol and quality standards for SD-OCT imaging in multiple sclerosis (MS). In the following section we introduce SD-OCT as a biomarker in MS. In a multicentric cross-sectional study, we had shown, that a single time measurement of peripapillary retinal nerve fiber layer thickness (RNFL) has a predictive value for a risk of disease progression in the next five years. Patients with a thickness of RNFL in the lowest tercile of the studied population had a relative risk of disease progression 2x higher than patients in the highest tercile. The second presented study tests whether the history of optic neuritis (ON) in MS is a risk factor for neurodegeneration of RNFL in later years. The study confirmed that long term changes of RNFL thickness in eyes post-ON and in eyes with no history of ON are not different. Therefore, we conclude that both,...

A retinopatia diabética (RD) é uma complicação microvascular do diabetes melito, sendo a principal causa de cegueira adquirida. Fatores angiogênicos como o vascular endothelial growth factor (VEGF) estão envolvidos na patogênese da RD. O VEGF-A é uma citocina potente e multifuncional que atua através dos receptores VEGFR-1 e VEGFR-2 expressados no endotélio vascular causando aumento da permeabilidade vascular e estímulo à neovascularização em processos fisiológicos e patológicos. A expressão do VEGFR-1 é acentuada por hipóxia e, apesar da afinidade, apresenta fraca resposta ao VEGF enquanto o VEGFR-2 é o principal mediador mitogênico, angiogênico e do aumento da permeabilidade vascular. Alguns polimorfismos do VEGF têm sido estudados na suscetibilidade e risco de progressão da RD. Importante associação entre o polimorfismo -634C/G e a presença de RD é relatada principalmente em relação ao alelo C. A homozigose CC estaria relacionada à RDP e níveis sérico e vítreo aumentados de VEGF sugerindo que a presença do alelo C seja um fator de risco independente para RD. Os conhecimentos sobre o VEGF levaram ao desenvolvimento de agentes anti-VEGF (pegaptanibe, ranibizumabe e bevacizumabe) com objetivo de inibir a neovascularização patológica. A terapia anti-VEGF é uma realidade cujos resultados são cada vez mais promissores na prática médica do tratamento da RD. / Diabetic retinopathy (DR), a DM microvascular complication, is the leading cause of blindness. Angiogenic factors such as vascular endothelial growth factor (VEGF) are involved in the pathogenesis of DR. VEGF-A is a potent, multifunctional cytokine that acts through the receptors VEGFR-1 and VEGFR-2 expressed in the vascular endothelium and causing increased vascular permeability and neovascularization stimulation in both physiological and pathological processes. The expression of VEGFR-1 is upregulated by hypoxia and is less responsive to VEGF compared to VEGFR-2 which is the main mediator mitogenic, angiogenic, and increased vascular permeability. VEGF polymorphisms have been studied in DR susceptibility and progression. Significant association between the polymorphism -634C/G and the presence of DR is reported mainly in relation to allele C. The homozygous CC is associated to proliferative DR and to increased vitreous and serum levels of VEGF suggesting that the presence of the C allele is an independent risk factor for RD. The knowledgement of VEGF lead to the development of anti-VEGF drugs (pegaptanib, ranibizumab and bevacizumab) aiming to prevent pathological neovascularization. The anti-VEGF therapy is a reality in practice medical treatment of DR.

Everyday an enormous amount of information is stored, processed and transmitted digitally around the world. Neurally-inspired compression models have been rapidly developed and researched as a solution to image processing tasks and channel error correction control. This dissertation presents a deep neural network (DNN) for gray high-resolution image compression and a fault-tolerant transmission system with channel error-correction capabilities. A feed-forward DNN implemented with the Levenberg-Marguardt learning algorithm is proposed and implemented for image compression. I demonstrate experimentally that the DNN not only provides better quality reconstructed images but also requires less computational capacity as compared to DCT Zonal coding, DCT Threshold coding, Set Partitioning in Hierarchical Trees (SPIHT) and Gaussian Pyramid. An artificial neural network (ANN) with improved channel error-correction rate is also proposed. The experimental results indicate that the implemented artificial neural network provides a superior error-correction ability by transmitting binary images over the noisy channel using Hamming and Repeat-Accumulate coding. Meanwhile, the network’s storage requirement is 64 times less than the Hamming coding and 62 times less than the Repeat-Accumulate coding. Thumbnail images contain higher frequencies and much less redundancy, which makes them more difficult to compress compared to high-resolution images. Bottleneck autoencoders have been actively researched as a solution to image compression tasks. However, I observed that thumbnail images compressed at a 2:1 ratio through bottleneck autoencoders often exhibit subjectively low visual quality. In this dissertation, I compared bottleneck autoencoders with two sparse coding approaches. Either 50\% of the pixels are randomly removed or every other pixel is removed, each achieving a 2:1 compression ratio. In the subsequent decompression step, a sparse inference algorithm is used to in-paint the missing the pixel values. Compared to bottleneck autoencoders, I observed that sparse coding with a random dropout mask yields decompressed images that are superior based on subjective human perception yet inferior according to pixel-wise metrics of reconstruction quality, such as PSNR and SSIM. With a regular checkerboard mask, decompressed images were superior as assessed by both subjective and pixel-wise measures. I hypothesized that alternative feature-based measures of reconstruction quality would better support my subjective observations. To test this hypothesis, I fed thumbnail images processed using either bottleneck autoencoder or sparse coding using either checkerboard or random masks into a Deep Convolutional Neural Network (DCNN) classifier. Consistent, with my subjective observations, I discovered that sparse coding with checkerboard and random masks support on average 2.7\% and 1.6\% higher classification accuracy and 18.06\% and 3.74\% lower feature perceptual loss compared to bottleneck autoencoders, implying that sparse coding preserves more feature-based information. The optic nerve transmits visual information to the brain as trains of discrete events, a low-power, low-bandwidth communication channel also exploited by silicon retina cameras. Extracting high-fidelity visual input from retinal event trains is thus a key challenge for both computational neuroscience and neuromorphic engineering. % Here, we investigate whether sparse coding can enable the reconstruction of high-fidelity images and video from retinal event trains. Our approach is analogous to compressive sensing, in which only a random subset of pixels are transmitted and the missing information is estimated via inference. We employed a variant of the Locally Competitive Algorithm to infer sparse representations from retinal event trains, using a dictionary of convolutional features optimized via stochastic gradient descent and trained in an unsupervised manner using a local Hebbian learning rule with momentum. Static images, drawn from the CIFAR10 dataset, were passed to the input layer of an anatomically realistic retinal model and encoded as arrays of output spike trains arising from separate layers of integrate-and-fire neurons representing ON and OFF retinal ganglion cells. The spikes from each model ganglion cell were summed over a 32 msec time window, yielding a noisy rate-coded image. Analogous to how the primary visual cortex is postulated to infer features from noisy spike trains in the optic nerve, we inferred a higher-fidelity sparse reconstruction from the noisy rate-coded image using a convolutional dictionary trained on the original CIFAR10 database. Using a similar approach, we analyzed the asynchronous event trains from a silicon retina camera produced by self-motion through a laboratory environment. By training a dictionary of convolutional spatiotemporal features for simultaneously reconstructing differences of video frames (recorded at 22HZ and 5.56Hz) as well as discrete events generated by the silicon retina (binned at 484Hz and 278Hz), we were able to estimate high frame rate video from a low-power, low-bandwidth silicon retina camera.

Retinoblastoma (Rb) é o tumor ocular mais frequente na infância e cada grande Centro deve conhecer o perfil dos seus pacientes. Foi realizado um estudo do tipo coorte retrospectivo e incluiu pacientes com Rb atendidos entre 1983 e 2012 nos Serviços de Oncologia Pediátrica, Oftalmologia e Genética Médica do Hospital de Clínicas de Porto Alegre (HCPA). De um total de 165 registros no período foram efetivamente incluídos 140 pacientes, sendo 95,0% destes provenientes de municípios do Rio Grande do Sul. Os sinais mais frequentes ao diagnóstico foram: leucocoria (73,6%) e estrabismo (20,7%). Identificamos a seguinte distribuição: doença unilateral (65,0%), bilateral (32,9%) sendo 80,4% com doença multifocal (p=0,015), trilateral (2,1%). A idade média dos pacientes por ocasião dos primeiros sinais e sintomas foi de 18,1 meses [mediana=12,0] e a idade média ao diagnóstico foi 23,5 meses [mediana=16,5]. Cinquenta pacientes (35,7%) foram diagnosticados no 1º ano de vida. O tempo de diagnóstico médio da coorte foi 5,4 meses [mediana=3,0], (amplitude=0-77,0). A idade média aos primeiros sinais e sintomas do grupo com critérios de hereditariedade foi de 12,3 meses enquanto a do grupo não hereditário foi de 21,6 meses (p=0,001), enquanto a idade média ao diagnóstico foi de 15,9 meses vs. 28 meses, respectivamente (p<0,001). Entretanto não houve diferença na sobrevida entre esses subgrupos. O estadiamento ocular dos pacientes ao diagnóstico na sua maioria foi avançado (classificação de Reese V em 76,5%, Internacional D ou E em 78,1%), sendo que 35,2% dos unilaterais e 34,8% dos bilaterais já apresentavam doença extraocular em pelo menos um olho ao diagnóstico. Quinze pacientes (10,7%) tinham doença metastática ao diagnóstico. Em relação ao tratamento, diferentes modalidades foram utilizadas, sendo a maioria dos pacientes submetidos à cirurgia, sendo esta enucleação em 88,1% e exenteração em 11,9%. Uma parcela significativa dos pacientes foi tratada com quimioterapia sistêmica (57,1%) e/ou radioterapia (37,1%). Do total de pacientes recrutados, 131 (93,6%) permaneceram vinculados ao hospital até 2012 ou até o óbito. Destes, 32 (22,9%) recidivaram, resultando em 19 óbitos com 84,2% por progressão do Rb. Uma segunda neoplasia primária esteve presente em 4,3% (N=6) e dentre esses, um paciente teve uma terceira neoplasia primária. O tempo de seguimento médio foi 323,2 meses [300,3; 346,1]. As sobrevidas nos diferentes subgrupos foram as seguintes: sobrevida global 86,4%; no não metastático 92,0%; no metastático 40,0%; entre os intraoculares 94,0%; entre os extraoculares 68,5%; entre os unilaterais e bilaterais ambos com cerca de 88,0%; entre os trilaterais (N=3) todos foram a óbito; entre os unilaterais intraoculares 94,9% e extraoculares 75,0% e entre os bilaterais intraoculares 94,5% e extraoculares 68,4%. No nosso meio, o diagnóstico de Rb ainda é feito predominantemente em estadios avançados o que reduz a sobrevida dos pacientes e o índice de preservação do olho e da visão, além de aumentar a intensidade dos tratamentos realizados e consequentemente, toxicidade e efeitos tardios destes. Avaliações clínicas e oftalmológicas periódicas nos primeiros anos de vida da criança oferecem maior oportunidade de um diagnóstico precoce e o encaminhamento rápido à um Centro de Referência multidisciplinar que contemple cuidados terciários em Oftalmologia e Oncologia Pediátrica é fundamental. Existe grande necessidade de investimentos regionais que facilitem o acesso ao diagnóstico e tratamento do Rb, o tumor ocular mais frequente na infância. / Retinoblastoma (Rb) is the most frequent ocular tumor diagnosed in children and every pediatric hospital must be familiar with its clinical presentation and patient characteristics. A retrospective cohort study was undertaken, with patients diagnosed with retinoblastoma from 1983 until 2012, treated at the Pediatric Oncology Unit, Ophthalmology Unit, and Medical Genetics Unit of the Hospital de Clínicas de Porto Alegre (HCPA). Of a total of 165 registries during this time frame, 140 patients were included in this study, with 95% of them from the state of Rio Grande do Sul. The most frequent signs and symptoms at diagnosis were: leukocoria (73.6%) and strabismus (20.7%). The following distribution was identified: unilateral disease (65.0%), bilateral disease (32.9%), being 80.4% with multifocal disease, (P=0,015), and trilateral disease (2.1%). The average age of patients at the appearance of the first sign or symptom was 18.1 months [median=12.0] and the average age at diagnosis was 23.5 months [median=16.5]. Fifty patients (35.7%) were diagnosed during their first year of age. The average time to diagnosis was of 5.4 months [median=3.0], (amplitude=0-77.0). In the hereditary retinoblastoma group, the average age at the appearance of the first sign or symptom was 12.3 months, whereas the non-hereditary group presented the first sign or symptom on average at 21.6 months (P=0,001). The average age at diagnosis was 15.9 months vs. 28 months for the hereditary and non-hereditary patients, respectively (P<0.001). However, no significant difference in overall survival was found when both groups were compared. Ocular staging at diagnosis was, for the most part, advanced disease, (Reese V classification: 76.5%, Internacional Classification of Retinoblastoma D or E in 78.1% patients), being that 35.2% of cases were comprised of unilateral disease and 34.8% of patients with bilateral disease already presented with extraocular lesions in at least one eye at diagnosis. Fifteen patients (10.7%) presented with metastasis at diagnosis. With regards to treatment, differnet modalities were employed, being that most patients underwent surgery with enucleation in 88.1% and e exenteration in 11.9%. A significant number of patients received systemic chemotherapy (57.1%) and/or radiotherapy (37.1%). Of all patients included, 131 (93.6%) remained in follow up at the hospital until 2012 or until their demise. Of these patients, 32 (22.9%) relapsed, leading to 19 deaths, 84.2% of them due to disease progression. Secondary malignancies were present in 6 patients (4.3%) and, of these, one patient presented with two different secondary malignancies. The average time of patient follow up was 323.2 months [300.3; 346.1]. Overall survival was of 86.4%, with the following time frames among the different patient subgroups: 92.0% for non-metastatic patients, 40.0% for metastatic patients, intraoculares 94.0% for patients with intraocular disease, and 68.5% for patients with extraocular lesions. With regards to unilateral or bilateral disease, overall survival was of 88.0%; for patients with trilateral disease, (N=3) all patients expired. Survival of patients with unilateral and intraocular disease was of 94.9%; patients with unilateral and extraocular disease presented a overall survival of 75.0%. Patients with bilateral intraocular lesions overall survival was of 94.5%, whereas patients with bilateral and extraocular disease had an overall survival of 68.4%. In our setting, Rb diagnosis still occurs when the patients already manifest advanced disease, which reduces considerably their overall survival and preservation of the ocular globe and vision. Moreover, late diagnosis requires more agressive treatments, and consequently leads to more frequent toxicities and late side effects. Periodic clinical and ophthalmologic evaluations during the first years of a child's life offer a greater chance of early diagnosis and referral to a multidisciplinary pediatric oncology center, which is crucial for the patient’s well being. There is much need of further investments which facilitate patient access to diagnosis and treatment for Rb, which is the most common ocular tumor in children.

A retinopatia diabética (RD) é uma complicação microvascular do diabetes melito, sendo a principal causa de cegueira adquirida. Fatores angiogênicos como o vascular endothelial growth factor (VEGF) estão envolvidos na patogênese da RD. O VEGF-A é uma citocina potente e multifuncional que atua através dos receptores VEGFR-1 e VEGFR-2 expressados no endotélio vascular causando aumento da permeabilidade vascular e estímulo à neovascularização em processos fisiológicos e patológicos. A expressão do VEGFR-1 é acentuada por hipóxia e, apesar da afinidade, apresenta fraca resposta ao VEGF enquanto o VEGFR-2 é o principal mediador mitogênico, angiogênico e do aumento da permeabilidade vascular. Alguns polimorfismos do VEGF têm sido estudados na suscetibilidade e risco de progressão da RD. Importante associação entre o polimorfismo -634C/G e a presença de RD é relatada principalmente em relação ao alelo C. A homozigose CC estaria relacionada à RDP e níveis sérico e vítreo aumentados de VEGF sugerindo que a presença do alelo C seja um fator de risco independente para RD. Os conhecimentos sobre o VEGF levaram ao desenvolvimento de agentes anti-VEGF (pegaptanibe, ranibizumabe e bevacizumabe) com objetivo de inibir a neovascularização patológica. A terapia anti-VEGF é uma realidade cujos resultados são cada vez mais promissores na prática médica do tratamento da RD. / Diabetic retinopathy (DR), a DM microvascular complication, is the leading cause of blindness. Angiogenic factors such as vascular endothelial growth factor (VEGF) are involved in the pathogenesis of DR. VEGF-A is a potent, multifunctional cytokine that acts through the receptors VEGFR-1 and VEGFR-2 expressed in the vascular endothelium and causing increased vascular permeability and neovascularization stimulation in both physiological and pathological processes. The expression of VEGFR-1 is upregulated by hypoxia and is less responsive to VEGF compared to VEGFR-2 which is the main mediator mitogenic, angiogenic, and increased vascular permeability. VEGF polymorphisms have been studied in DR susceptibility and progression. Significant association between the polymorphism -634C/G and the presence of DR is reported mainly in relation to allele C. The homozygous CC is associated to proliferative DR and to increased vitreous and serum levels of VEGF suggesting that the presence of the C allele is an independent risk factor for RD. The knowledgement of VEGF lead to the development of anti-VEGF drugs (pegaptanib, ranibizumab and bevacizumab) aiming to prevent pathological neovascularization. The anti-VEGF therapy is a reality in practice medical treatment of DR.

Retinoblastoma (Rb) é o tumor ocular mais frequente na infância e cada grande Centro deve conhecer o perfil dos seus pacientes. Foi realizado um estudo do tipo coorte retrospectivo e incluiu pacientes com Rb atendidos entre 1983 e 2012 nos Serviços de Oncologia Pediátrica, Oftalmologia e Genética Médica do Hospital de Clínicas de Porto Alegre (HCPA). De um total de 165 registros no período foram efetivamente incluídos 140 pacientes, sendo 95,0% destes provenientes de municípios do Rio Grande do Sul. Os sinais mais frequentes ao diagnóstico foram: leucocoria (73,6%) e estrabismo (20,7%). Identificamos a seguinte distribuição: doença unilateral (65,0%), bilateral (32,9%) sendo 80,4% com doença multifocal (p=0,015), trilateral (2,1%). A idade média dos pacientes por ocasião dos primeiros sinais e sintomas foi de 18,1 meses [mediana=12,0] e a idade média ao diagnóstico foi 23,5 meses [mediana=16,5]. Cinquenta pacientes (35,7%) foram diagnosticados no 1º ano de vida. O tempo de diagnóstico médio da coorte foi 5,4 meses [mediana=3,0], (amplitude=0-77,0). A idade média aos primeiros sinais e sintomas do grupo com critérios de hereditariedade foi de 12,3 meses enquanto a do grupo não hereditário foi de 21,6 meses (p=0,001), enquanto a idade média ao diagnóstico foi de 15,9 meses vs. 28 meses, respectivamente (p<0,001). Entretanto não houve diferença na sobrevida entre esses subgrupos. O estadiamento ocular dos pacientes ao diagnóstico na sua maioria foi avançado (classificação de Reese V em 76,5%, Internacional D ou E em 78,1%), sendo que 35,2% dos unilaterais e 34,8% dos bilaterais já apresentavam doença extraocular em pelo menos um olho ao diagnóstico. Quinze pacientes (10,7%) tinham doença metastática ao diagnóstico. Em relação ao tratamento, diferentes modalidades foram utilizadas, sendo a maioria dos pacientes submetidos à cirurgia, sendo esta enucleação em 88,1% e exenteração em 11,9%. Uma parcela significativa dos pacientes foi tratada com quimioterapia sistêmica (57,1%) e/ou radioterapia (37,1%). Do total de pacientes recrutados, 131 (93,6%) permaneceram vinculados ao hospital até 2012 ou até o óbito. Destes, 32 (22,9%) recidivaram, resultando em 19 óbitos com 84,2% por progressão do Rb. Uma segunda neoplasia primária esteve presente em 4,3% (N=6) e dentre esses, um paciente teve uma terceira neoplasia primária. O tempo de seguimento médio foi 323,2 meses [300,3; 346,1]. As sobrevidas nos diferentes subgrupos foram as seguintes: sobrevida global 86,4%; no não metastático 92,0%; no metastático 40,0%; entre os intraoculares 94,0%; entre os extraoculares 68,5%; entre os unilaterais e bilaterais ambos com cerca de 88,0%; entre os trilaterais (N=3) todos foram a óbito; entre os unilaterais intraoculares 94,9% e extraoculares 75,0% e entre os bilaterais intraoculares 94,5% e extraoculares 68,4%. No nosso meio, o diagnóstico de Rb ainda é feito predominantemente em estadios avançados o que reduz a sobrevida dos pacientes e o índice de preservação do olho e da visão, além de aumentar a intensidade dos tratamentos realizados e consequentemente, toxicidade e efeitos tardios destes. Avaliações clínicas e oftalmológicas periódicas nos primeiros anos de vida da criança oferecem maior oportunidade de um diagnóstico precoce e o encaminhamento rápido à um Centro de Referência multidisciplinar que contemple cuidados terciários em Oftalmologia e Oncologia Pediátrica é fundamental. Existe grande necessidade de investimentos regionais que facilitem o acesso ao diagnóstico e tratamento do Rb, o tumor ocular mais frequente na infância. / Retinoblastoma (Rb) is the most frequent ocular tumor diagnosed in children and every pediatric hospital must be familiar with its clinical presentation and patient characteristics. A retrospective cohort study was undertaken, with patients diagnosed with retinoblastoma from 1983 until 2012, treated at the Pediatric Oncology Unit, Ophthalmology Unit, and Medical Genetics Unit of the Hospital de Clínicas de Porto Alegre (HCPA). Of a total of 165 registries during this time frame, 140 patients were included in this study, with 95% of them from the state of Rio Grande do Sul. The most frequent signs and symptoms at diagnosis were: leukocoria (73.6%) and strabismus (20.7%). The following distribution was identified: unilateral disease (65.0%), bilateral disease (32.9%), being 80.4% with multifocal disease, (P=0,015), and trilateral disease (2.1%). The average age of patients at the appearance of the first sign or symptom was 18.1 months [median=12.0] and the average age at diagnosis was 23.5 months [median=16.5]. Fifty patients (35.7%) were diagnosed during their first year of age. The average time to diagnosis was of 5.4 months [median=3.0], (amplitude=0-77.0). In the hereditary retinoblastoma group, the average age at the appearance of the first sign or symptom was 12.3 months, whereas the non-hereditary group presented the first sign or symptom on average at 21.6 months (P=0,001). The average age at diagnosis was 15.9 months vs. 28 months for the hereditary and non-hereditary patients, respectively (P<0.001). However, no significant difference in overall survival was found when both groups were compared. Ocular staging at diagnosis was, for the most part, advanced disease, (Reese V classification: 76.5%, Internacional Classification of Retinoblastoma D or E in 78.1% patients), being that 35.2% of cases were comprised of unilateral disease and 34.8% of patients with bilateral disease already presented with extraocular lesions in at least one eye at diagnosis. Fifteen patients (10.7%) presented with metastasis at diagnosis. With regards to treatment, differnet modalities were employed, being that most patients underwent surgery with enucleation in 88.1% and e exenteration in 11.9%. A significant number of patients received systemic chemotherapy (57.1%) and/or radiotherapy (37.1%). Of all patients included, 131 (93.6%) remained in follow up at the hospital until 2012 or until their demise. Of these patients, 32 (22.9%) relapsed, leading to 19 deaths, 84.2% of them due to disease progression. Secondary malignancies were present in 6 patients (4.3%) and, of these, one patient presented with two different secondary malignancies. The average time of patient follow up was 323.2 months [300.3; 346.1]. Overall survival was of 86.4%, with the following time frames among the different patient subgroups: 92.0% for non-metastatic patients, 40.0% for metastatic patients, intraoculares 94.0% for patients with intraocular disease, and 68.5% for patients with extraocular lesions. With regards to unilateral or bilateral disease, overall survival was of 88.0%; for patients with trilateral disease, (N=3) all patients expired. Survival of patients with unilateral and intraocular disease was of 94.9%; patients with unilateral and extraocular disease presented a overall survival of 75.0%. Patients with bilateral intraocular lesions overall survival was of 94.5%, whereas patients with bilateral and extraocular disease had an overall survival of 68.4%. In our setting, Rb diagnosis still occurs when the patients already manifest advanced disease, which reduces considerably their overall survival and preservation of the ocular globe and vision. Moreover, late diagnosis requires more agressive treatments, and consequently leads to more frequent toxicities and late side effects. Periodic clinical and ophthalmologic evaluations during the first years of a child's life offer a greater chance of early diagnosis and referral to a multidisciplinary pediatric oncology center, which is crucial for the patient’s well being. There is much need of further investments which facilitate patient access to diagnosis and treatment for Rb, which is the most common ocular tumor in children.

Les maladies à polyglutamines (polyQ) sont des maladies neurodégénératives héréditaires dominantes causées par une expansion de CAG traduite en longue expansion de polyglutamine dans la protéine correspondante. Ces maladies comprennent l’ataxie spinocérébélleuse 7 (SCA7) et la maladie de Huntington (MH) causées par une expansion de polyQ dans les protéines ataxine-7 (ATXN7) et huntingtine (htt), respectivement. Les souris SCA7 et MH développent des rétinopathies similaires suggérant des pathomécanismes communs toujours inexpliqués. Durant ma thèse, j’ai trouvé qu’en réponse à la toxicité des polyQ, les photorécépteurs (PR) perdent leur différenciation alors que d’autres migrent ou meurent. De plus, cette mortalité cellulaire active la prolifération des cellules gliales de Müller et leur différenciation en PR. Récemment, j’ai trouvé que l’ATXN7 et la htt se trouvent dans le cil primaire et leur mutation mène à une perte des protéines endogènes des cils associée à des défauts du cil. / Polyglutamine (polyQ) disorders are dominantly inherited neurodegenerative disorders caused by the expansion of CAG repeats translated into long polyQ tracts in the corresponding proteins. These diseases include Spinocerebellar ataxia 7(SCA7) and Huntington’s Disease (HD), caused by polyQ expansion ataxin-7 (ATXN7) and huntingtin (htt), respectively. SCA7 and HD mouse models develop similar retinopathies suggesting common pathomechanisms. In my thesis, I found that, in response to polyQ toxicity, SCA7 photoreceptors (PR) undergo several cell fates ranging from their deconstruction, to their migration and their death. Moreover, this cell death activates the proliferation of Müller glial cells and their differentiation into PR like cells. The pathomechanisms underlying HD and SCA7 are still unknown. Recently, I found that ATXN7 and htt are localized to the PR cilia and that the mutant proteins lead to a progressive loss of the wild-type proteins that correlates with defects in the PR cilia.

Le chien présente de nombreuses maladies génétiques dont les rétinopathies auxquelles je me suis intéressée. Historiquement, l’homme a appliqué aux chiens une sélection artificielle extrêmement forte, créant ainsi près de 400 races répondant à des besoins spécifiques. En uniformisant ainsi de nombreux traits phénotypiques et comportementaux, l’homme a sélectionné les allèles recherchés mais a également concentré involontairement des allèles délétères responsables de maladies génétiques. Ces maladies étant pour beaucoup cliniquement et génétiquement similaires aux maladies humaines, le chien constitue alors un modèle de choix pour en rechercher les bases génétiques et développer de nouvelles thérapies pour un bénéfice mutuel pour l’homme et le chien. J’ai travaillé sur l’Atrophie Progressive de la Rétine (APR) dans deux races avec le concours du Dr Gilles Chaudieu et des clubs de race : le border collie et le berger picard. Concernant l’APR du border collie, les analyses réalisées avant mon arrivée avaient identifié un locus de 20Mb sur le chromosome X. J’ai tout d’abord mis à jour les données épidémiologiques et cliniques d’environ 500 chiens et poursuivi la collecte de prélèvements. Une analyse de liaison génétique pangénomique sur 130 chiens m’a permis de confirmer le mode de transmission et le locus. J’ai ensuite conduis plusieurs analyses génétiques (« homozygosity mapping », analyses des génotypes, séquençage de gènes candidats) sans pour autant réduire le locus ni identifier de mutation causale. Dans ce contexte et grâce à l’avènement des nouvelles technologies de séquençage, j’ai choisi de séquencer le génome complet de trois chiens atteints et deux porteurs apparentés. Ce travail m’a permis d’identifier 117variants dont 9 dans le locus d’intérêt, mais aucun dans des régions codantes. Je me suis donc focalisée sur les variants de régions régulatrices pour 13 gènes candidats. De plus, j’ai identifié cinq variants structuraux, encore en cours d’étude. Concernant l’APR du berger picard, nous avons relancé le projet par la collecte de prélèvements et la réalisation d’un pedigree de 154 chiens. Ce travail a amené à une collaboration tripartite internationale par le séquençage de deux chiens atteints et deux indemnes d’APR, dont les analyses sont en cours. Cette thèse m’a permis d’aborder la génétique d’une maladie génétique rare chez l’homme grâce à un modèle spontané original ; et même si ce travail n’a pas abouti à l’identification de mutations à ce jour, ces APR restent de bons modèles génétique et thérapeutique pour les rétinites pigmentaires humaines. / Dogs are affected by numerous genetic diseases including retinopathies, the subject of my thesis. Historically, humans have exerted a very strong artificial selection to dogs, thus creating some 400 breeds to perform specific tasks or to harbour specific traits. By such an homogenisation of phenotypic and behavioural traits, humans have selected desired alleles but also concentrated undesired deleterious alleles responsible of genetic diseases. These diseases are clinically and genetically similar to human genetic diseases, making dogs a model of choice to search for the genetic bases of such genetic diseases and to develop new therapies for a mutual benefit for dogs and humans. I worked on Progressive Retinal Atrophies (PRA) in two breeds, thanks to the contribution of the veterinary ophthalmologist Dr. Gilles Chaudieu and the breed clubs of border collies and berger picard. Concerning the border collie PRA, genetic analyses performed before my thesis allowed the identification of a locus of 20Mb on the X chromosome. I first completed and supplemented the epidemiological and clinical data for about 500 dogs and continued the sample and data collection. A first genetic linkage analysis on 130 dogs allowed to confirm the transmission mode and the locus. I then performed several genetic analyses (« homozigosity mapping », genotypes analyses, candidate gene sequencing) without significantly reducing the locus,neither finding the causal mutation, unfortunately. In this context and thanks to the development of new sequencing technologies, I chose to sequence the entire genome of three PRA affected border collies and two unaffected related dogs. This work allowed the identification of 117variants, 9 in the locus but none in coding regions. I thus focused my research on genetic variants from potential regulatory regions for 13 candidate genes. In addition, I identified five structural variants. The analysis of these variants is still ongoing. Concerning the berger picard PRA, we have re-initiated the project by collecting samples and designing a large pedigree of 154 dogs. This workled to an international collaboration by the sequencing of the entire genomes of 2 affected and 2 unaffected bergers picards for which the statistical analyses are ongoing. This thesis allowed me to work on the genetics of rare diseases in humans with a spontaneous and original animal model and even if thiswork has not yet reached the identification of the mutations, these two PRAs in these two breeds remain good genetic and therapeutic models for human RP RetinitisPigmentosa.