SÁLUA OLIVEIRA CALIL DE PAULA AVALIAÇÃO DE BIOMARCADORES MOLECULARES EM MULHERES COM CÂNCER DE OVÁRIO Dissertação apresentada ao Programa de Pós- Graduação em Ginecologia, Obstetrícia e Mastologia da Faculdade de Medicina de Botucatu-UNESP, para obtenção do título de Doutor. Orientador: Prof. Dr. Paulo Traiman Co-Orientadora: Dra. Andréa Teixeira de Carvalho Botucatu 2014 Dedicatória Dedicatória Dedicatória Aos meus pais, Grimaldo e Sandra, pela dedicação, amor e incentivo ao longo de minha vida. Vocês são os principais reponsáveis por todas as minhas conquistas! Ao meu esposo e amigo Paulo Vellozo, pelo carinho e compreensão. Aos meus irmãos Thales e Grimaldo pela amizade e carinho sempre presentes. Aos meus avós, pelo incentivo e admiração, À Gabi e Belinha, companhias inseparáveis! Agradecimento Especial Agradecimentos especiais: Agradecimento Especial Professor Paulo Traiman, Dra. Andréa Teixeira de Carvalho, Professor Agnaldo Lopes da Silva Filho, Agradeço pelos ensinamentos fundamentais que contribuíram de forma incomensurável para a minha formação. Obrigada pelo privilégio de poder compartilhar da sabedoria de vocês! Sou imensamente grata! Agradecimentos - Agradecimentos: Agradecimentos Agradeço ao Dr. João Oscar Falcão Júnior pela parceria e entusiasmo. Seus conhecimentos foram essencias ao nosso crescimento. Meu agradecimento especial ao Dr. Geraldo Henrique Ribeiro pelos ensinamentos cirúrgicos fundamentais à minha formação! Tenho profunda admiração pelo senhor. Aos amigos Gustavo e Elisa pelo apoio, cortesia e disponibilidade. Foram ótimos nossos momentos juntos. Aos amigos do Hospital Mater Dei e do Instituto Mário Penna pela grande oportunidade de aprendizado e crescimento. Aos profissionais do Centro de Pesquisas René Rachou (Fiocruz), pelo apoio logístico na captação, preparação e processamento do material coletado durante as operações. Também pela cortesia, disponibilidade e receptividade. Aos professores e colaboradores do Setor de Pós-Graduação da Faculdade de Medicina de Botucatu- UNESP. Epígrafe EPÍGRAFE Epígrafe “Viver Acordando com nova esperança Desejar que o novo dia seja diferente Que um olhar dure o tempo necessário Que surja mais um momento que eu possa eterniza-lo Que meu trabalho continue sendo feito do melhor modo Que o entardecer eu o veja mais inebriante Mais um dia de vida devida a Deus Viver...como se fosse o último dia.” “Se não puder ser um poeta, Seja uma poesia.” PTraiman Resumo Resumo Resumo Introdução: O câncer de ovário (CO) é a maior causa de morte por neoplasia ginecológica nos países desenvolvidos. Atualmente, um modelo dualístico de classificação foi proposto, sendo os tumores tipo I de baixo grau, indolentes e com mutações estáveis. Os tumores tipo II são de alto grau e mais agressivos. Além disso, na tentativa de se entender o processo de carcinogênese, vários biomarcadores têm sido estudados como as micropartículas (MPs), as citocinas e as quimiocinas. O objetivo desse estudo é a avaliação de fatores solúveis da resposta inflamatória (MPs, citocinas e quimiocinas) em mulheres com CO e compará-los com os níveis encontrados em mulheres sem malignidade e com parâmetros clínicos. Métodos: Avaliaram-se 26 mulheres com CO e 16 mulheres sem evidência de neoplasia maligna (grupo controle). Foram coletadas amostras de plasma e tecido tumoral. A avaliação dos fatores inflamatórios foi realizada por meio da dosagem de citocinas (IL-1- β, IL-2, IL-6, IL-10, IL-12, IL-17A, TNF e IFN-gama, e quimiocinas (CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5) e das micropartículas (neutrófilos, leucócitos, monócitos, eritrócitos, endotélio, plaquetas e linfócitos) por citometria de fluxo/CBA (Cytometric Bead Array). As diferenças entre os grupos foram avaliadas pelos testes Kruskal- Walis ou Mann-Whitney e a sobrevida por Cox Regression. As diferenças com valor de p<0,05 foram consideradas significativas.Resultados: Não houve diferença entre os grupos em relação à idade, paridade e menopausa. No grupo de mulheres com CO, 10 (38,5%) tinham estadio I/II e 16 (61,5%) tinham estadio III/IV. Em relação ao tipo tumoral, segundo a nova classificação, 8 (30.8%) eram tipo I e 18 foram tipo II (69.2%). Citorredução ótima foi obtida em 15 (57.7%) mulheres com CO. Os valores de CA 125 foram significativamente diferentes entre os grupos. Não houve óbito em mulheres com tumores tipo I. Em relação às análises das MPs, observou-se aumento desse biomarcador proveniente de leucócitos nas mulheres com CO em relação ao controle. Houve ainda aumento das MPs de leucócitos no tumores tipo II em relação ao tipo I. Os níveis de MPs de neutrófilos, assim como os valores de CA125, foram maiores nas mulheres que tiveram citorredução ótima. Observou-se aumento dos níveis de IL-1- β, IL-6, TNF, IL-12, IFN-gama, IL-10, IL-17, IL-2, CCL-2, CCL-5, CXCL-8, CCL-9 e CXCL10 nas mulheres com CO em relação ao controle. Não houve diferenças estatísticas dos níveis de citocinas/quimiocinas em relação ao tipo tumoral I ou II. As mulheres com tumor tipo II que tiveram citorredução ótima, apresentaram maior expressão de IL-1- β e IL-12, além de aumento dos valores de CA-125. Os níveis de IL-1, IL- 12 e CXCL-8 também foram maiores naquelas com estadio avançado e que foram submetidas à cirurgia ótima. Conclusão: Mulheres com CO apresentaram aumento das frequências de MPs, citocinas e quimiocinas em relação àquelas sem neoplasia maligna. As MPs provenientes de neutrófilos, as citocinas IL-1- β e IL-12 e a quimiocina CXCL-8 possivelmente correlacionam-se com resposta ao tratamento em mulheres com CO. Palavras-chave: Câncer de ovário, resposta inflamatória, citocinas, quimiocinas e micropartículas. Summary Summary Summary Introduction : Ovarian cancer ( OC) is the leading cause of death from gynecological cancer in developed countries . Currently , a dualistic classification model was proposed .Type I tumors are low-grade , indolent and have stable mutations . Type II tumors are high grade and more aggressive . Moreover, in an attempt to understand the process of carcinogenesis , several biomarkers have been studied as microparticles (MPs ) , cytokines and chemokines. The purpose of the study was to evaluate the levels of circulating soluble biomarkers-microparticles, cytokines and chemokines- to characterize the pro-inflammatory/modulatory immune response in women with OC. The correlation between the biomarker levels and the clinico-pathological parameters were also analyzed. Methods : We evaluated 26 women with OC and 16 women without evidence of malignancy ( control group ) . Plasma samples and tumor tissue were collected . The assessment of inflammatory markers was performed by measurement of cytokine (IL -1- β , IL-2 , IL-6 , IL-10 , IL-12 , IL -17A , TNF and IFN –gamma), chemokines ( CXCL8 , CXCL -9 , CXCL 10 , CCL 2 , CCL5 ) and microparticles (neutrophils , leukocytes, monocytes, erythrocytes, endothelium , platelets and lymphocytes) by flow cytometry / CBA ( cytometric Bead Array) . Differences among groups were evaluated by Kruskal - Wallis or Mann - Whitney and survival by Cox Regression . Differences with p < 0.05 were considered significant. Results: There was no difference between groups regarding age , parity and menopause. In the group of women with OC , 10 ( 38.5 % ) were stage I / II and 16 ( 61.5 % ) were stage III / IV . Concerning tumor type , according to the new classification , 8 (30.8 % ) were type I and 18 ( 69.2 % ) were type II . Optimal cytoreduction was achieved in 15 ( 57.7 % ) women with OC . CA 125 values were significantly different between groups . There were no deaths in women with type I tumors. Stratifying by groups, an increased frequence of leukocytes microparticles (LMP) was observed in women with OC compared to control group. There was also an increase of LMPs in type II tumors compared to type I. Neutrophils derived microparticles (NMPs), as well as the values of CA125 were higher in women who were optimally debulked . A higher frequence of IL - 1- β , IL- 6, TNF , IL -12, IFN- gamma, IL -10, IL -17 , IL-2 , CCL -2 , CCL- 5, CXCL -8 , CCL - 9 and CXCL10 were detected in women with OC compared to control . We did not observe any statistical difference beteween cytokines / chemokines and tumor type I or II . Women with type II tumor who were optimally debulked showed higher expression of IL - 1 - β and IL -12 , and increased levels of CA -125 . The levels of IL-1, IL-12 and CXCL -8 were also higher in advanced disease with optimal surgical treatment. Conclusion : Women with OC showed increased frequencies of MPs , cytokines and chemokines compared to those without malignancy . MPs from neutrophils, IL-1 - β and IL-12 cytokine and chemokine CXCL -8 possibly correlate with response to treatment in women with OC. Keywords : Ovarian cancer , inflammatory response , cytokines , chemokines and microparticles . Símbolos, siglas e abreviaturas Símbolos, siglas e abreviaturas Símbolos, siglas e abreviaturas Ac Anticorpo CBA CCL Do Inglês: Cytometric Bead Array Quimiocina CO Câncer de Ovário CEP Comitê de Ética em Pesquisa cm Centímetro CMF CXCL Citometria de Fluxo Quimiocina EDTA-K3 Ácido Etileno-diamino Tetracético et al. E outro(s), e outra(s)
 FACS Do Inglês: Fluorescence-Activated Cell Sorting FIGO Federação Internacional de Ginecologia e Obstetrícia FITC Isoticianato de Fluoresceína FSC Do Inglês: Forward Scatter IFN-gama Interferon Gama IL Citocina IL-1- β INCA Instituto Nacional de Câncer

 μL Microlitro mL MPs OMS Mililitro Micropartículas Organização Mundial de Saúde NK Do Inglês: Natural Killer nm Nanômetro PE Ficoeritrina TNF Fator de Necrose Tumoral Alfa VEGF Fator de Crescimento Vascular Endotelial TCLE Termo de Consentimento Livre e Esclarecido Sumário SUMÁRIO Sumário 1. INTRODUÇÃO............................................................................................... 20 1.1. Câncer de Ovário........................................................................................ 21 1.2. Inflamação e câncer de ovário.................................................................... 25 1.2.1. Micropartículas ....................................................................................... 27 1.3. Justificativa do estudo................................................................................ 30 2. OBJETIVOS................................................................................................. 31 2.1 Objetivo Geral.............................................................................................. 32 2.2 Objetivos Específicos................................................................................. 32 3. MATERIAL E MÉTODOS.............................................................................. 33 3.1. Pacientes e espécimes......................................................................... 34 3.1.1 Critérios de Inclusão........................................................................... 35 3.2.Métodos................................................................................................. 36 3.2.1 Técnica Cirúrgica………………………………………………………… 36 3.2.2 Avaliação Histológica e Imunohistoquímica………………….. 36 3.2.3 Detecção do nível de citocinas/quimiocinas plasmáticas por citometria de fluxo.............................................................................................. 37 3.2.4 – Aquisição e análise dos dados de citocinas e quimiocinas plasmáticas por citometria de fluxo.................................................................... 40 3.2.5. Obtenção do plasma livre de plaquetas............................................ 42 3.2.6. Obtenção das micropartículas (MPs)....................................................... 42 3.2.7. Identificação de micropartículas (MPs) pela Citometria de fluxo............. 42 Análise fenotípica das MPs por citometria de fluxo........................................... 42 4. REFERÊNCIAS............................................................................................ 45 5. ARTIGO I...................................................................................................... 49 6. ARTIGO II...................................................................................................... 72 7. CONCLUSÃO................................................................................................ 97 Sumário 8. ANEXOS…………………………………………………………………………… 99 ANEXO I: Estadiamento FIGO……………………………………………………. 97 ANEXO II - Parecer do Comitê de Ética em Pesquisa...................................... 99 ANEXO III: Termo de consentimento livre e esclarecido………………………... 101 Introdução 20 1. Introdução Introdução 21 1.1 Câncer de Ovário O Câncer de ovário é a maior causa de morte por neoplasia ginecológica nos países desenvolvidos. Segundo dados da Agência Internacional de Pesquisa em Câncer, foram estimados 238.719 casos com 151.905 mortes por câncer de ovário no mundo em 20121. A incidência no Brasil, conforme dados do Instituto Nacional do Câncer, é de 5.680 novos casos em 2014, correspondendo a 2,1% de todos os cânceres em mulheres no país2. A maioria dos tumores malignos de ovário é diagnosticada em estadios avançados devido a falta de sintomas específicos nos estadios iniciais. Não há método propedêutico efetivo para rastreamento. Os marcadores tumorais e os exames imaginológicos apresentam altas taxas de resultados falsos positivos, especialmente quando realizados na pré-menopausa, e não são custo efetivos3,4,5. Desse modo, a sobrevida em 5 anos é de aproximdamente 40– 50% 6. Atualmente, o câncer de ovário é considerado um grupo de patologias com diferenças clínicopatológicas significativas devido a grande heterogeneidade molecular e comportamento biológico distintos. Com o objetivo de se conhecer melhor a origem e fisiopatologia dessa neoplasia, um modelo dualístico de classificação foi proposto, com a divisão entre tumores tipo I e II. Os tumores tipo I são de baixo grau, originam-se geralmente de lesões precursoras e são diagnosticados frequentemente em estadios iniciais (Figura 1A). Apresentam curso indolente, sendo que a transformação para tumores de alto grau pode ocorrer. Essas lesões correspondem a 25% dos casos e são responsáveis por apenas 10% das mortes por câncer de ovário. Introdução 22 Os tumores tipo II são mais agressivos, de alto grau, evoluem com rápida progressão e geralmente são diagnosticados em estadios avançados. Esses tumores correspondem a 75% casos e são responsáveis por 90% das mortes por câncer de ovário 7 (Figura 1B). Os tumores tipo I são geneticamente mais estáveis e apresentam mutações específicas de acordo com o tipo histológico. Desse modo, as mutações de KRAS, BRAF, ERB2, PTEN, CTNNB1, PIK3-CA são frequentemente encontradas nessas lesões, sendo raras as mutações em TP53. Por outro lado, os tumores tipo II são geneticamente instáveis e com mutações em TP53 em mais de 80% dos casos7 (Figura 1C). Introdução 23 B C Figura 1: A. Histologia referente à tumor ovariano tipo I. B. Histologia referente a tumor ovariano tipo II. C. Imunohistoquímica referente à tumor ovariano tipo II. A Introdução 24 Estima-se que aproximadamente 90% das neoplasias malignas ovarianas derivam do epitélio celômico. Tradicionalmente, os subtipos histológicos mais comuns eram baseados em suas características morfológicas: seroso, mucinoso, células claras, endometrióide e tumor de células transicionais, com características morfológicas semelhantes aos epitélios da tuba uterina, do trato gastrointestinal ou da endocérvix, do endométrio e do trato urinário, respectivamente. No entanto, um ovário normal não apresenta os caracteres morfológicos que se assemelham a esses tumores. Sendo assim, foi proposto, segundo a nova teoria, que as neoplasias com fenótipo mulleriano (seroso, endometrióide e de células claras) fossem originadas de tecido semelhante ao mulleriano, e não ao mesotélio, como suposto anteriormente. Desse modo, o desenvolvimento do carcinoma seroso de ovário ocorreria por mutações em cistos de inclusão ovarianos provenientes da tuba uterina ou por exfoliação de células malignas da tuba com implante na superfície ovariana. Para os tumores endometrióides e de células claras, o mecanismo proposto seria de menstruação retrógrada com implante em superfície ovariana. Os tumores mucinosos e de Brenner possivelmente originar-se-iam de células transicionais da junção tubomesotelial. No entanto, nenhuma das teorias agrega adequadamente todos os aspectos da carcinogênese ovariana7. Esse modelo dualístico ressalta a heterogeneidade do carcinoma ovariano e seus diferentes tipos de apresentação, o que aponta para a impossibilidade de método único para rastreamento dessa patologia. Introdução 25 O tratamento do cancer de ovário baseia-se na cirurgia citorredutora associada à quimioterapia nos casos indicados. O estadiamento é cirúrgico segundo as normas da Federação Internacional de Ginecologia Obstetrícia (FIGO)8,9 (Anexo I).O volume de doença residual após cirurgia primária é importante fator prognóstico. O objetivo da citorredução inicial é a remoção da maior quantidade de tecido tumoral possível, assim como da doença metastática. Caso a ressecção de todos os sítios metastáticos não seja possível, o ideal é reduzir o volume tumoral a uma condição “ótima“10,11. Considera-se citorredução ótima aquela em que a doença residual após cirurgia seja ≤ 1 cm em seu maior diâmetro 10. Apesar da maioria das mulheres com doença avançada apresentar remissão após tratamento adequado com cirurgia e quimioterapia, 70 a 80% aproximadamente, irão recorrer em dois anos 12. 1.2 Inflamação e Câncer de ovário A importância da resposta inflamatória na etiologia e patogênese do câncer tem sido descrita desde o século XIX, através dos trabalhos de Rudolf Ludwig Karl Virchow, que sugeriu que o infiltrado “linforeticular“ refletia a origem do câncer em sítios de inflamação crônica (“Die Cellularpathologie in ihrer Begründung auf physiologische und pathologische Gewebelehre‖, 1858). No entanto, o efeito paradoxo do infiltrado inflamatório na contenção e na progressão tumoral ainda é pouco compreendido 13. Pressupõe-se que a avaliação do processo inflamatório no câncer de ovário seja de grande importância para o estudo do comportamento desse tumor em relação ao hospedeiro. Introdução 26 A maioria das células tumorais secreta citocinas e quimiocinas, especialmente como consequência às mutações oncogênicas e às alterações de sinalização no sítio tumoral. Essas moléculas são potentes mediadores e reguladores da inflamação. No entanto, não está claro quais expressões de citocinas e quimiocinas apresentam relevância na regulação desse sistema complexo nos tumores ovarianos 13. Sabe-se que o microambiente pró-inflamatório do câncer de ovário está clinicamente relacionado com a disseminação peritoneal tumoral, com a produção maciça de ascite e com altas taxas de mortalidade. Vários estudos demostram que as neoplasias malignas ovarianas expressam altos níveis de algumas citocinas como TNF e IL-6, indicando o potencial dessas moléculas como indutoras/reguladoras da inflamação nesses tumores14. O TNF, por exemplo, apresenta efeitos pró- e anti-tumorais, a depender da estimulação de seus receptores. Em geral, o aumento desta citocina nas neoplasias malignas está associado à toxicidade envolvendo o sistema imune inato e adaptativo. O TNF-α também está envolvido na iniciação e promoção tumoral, além da alteração da homeostase com estímulo à angiogênese. Essa citocina tem-se mostrado importante regulador da ação de quimiocinas, por meio da via de sinalização do fator nuclear-kB (NF-kB).15,16,17 . A IL-6 particularmente atua na promoção do crescimento e como fator anti-apoptótico, apresentando-se em altas concentrações no plasma de mulheres com doença avançada 18. Vários estudos demostram que as células inflamatórias e da resposta imune são capazes de secretar diversas citocinas e quimiocinas que acabam por recrutar mais células da resposta inflamatória para o microambiente tumoral, o que Introdução 27 propicia a proliferação e sobrevida dessas células tumorais geneticamente alteradas13,19. Portanto, os níveis de citocinas e quimiocinas podem representar um reflexo do estado de imunidade do hospedeiro e servir como biomarcadores para predizer o prognóstico dessas mulheres com câncer de ovário, além de se constituírem como ferramentas complementares potenciais na sua monitoração pós-terapêutica. 1.2.1. Micropartículas (MPs) As micropartículas foram inicialmente descritas em 1964, e têm sido muito estudadas desde então20. MPs são vesículas de diferentes tamanhos, variando de 100 a 1000 nm, liberadas pelas células em situações fisiológicas e patológicas (Figura 2). Apresentam como característica relevante a alta exposição de fosfatidilserina que é translocada da parte interna para parte externa da membrana dessas vesículas. Essas vesículas apresentam expressões fenotípicas similares às células de origem e carreiam diversas substâncias como proteínas, enzimas, receptores e fatores de crescimento, citocinas e quimiocinas, lipídios e acidos nucléicos ( mRNA, miRNA e DNA do genoma). Pesquisas recentes mostram que as MPs apresentam atividade biológica e função importante nos processos de comunicação celular, imunidade, apoptose e homeostase 21.  Introdução 28 Figura 2: Micrografia eletrônica de varredura de um isolado de sangue periférico (doador saudável) evidenciando massa de vesículas extracelulares (seta). Adaptado de: Ogorevc E, et al. The role of extracellular vesicles in phenotypic cancer transformation. Radiolo Oncol 2013; 47(3): 197-205. A secreção das MPs é facilmente detectada em células com fenótipo tumoral, quando comparada às células normais do organismo. Isso chama a atenção para uma possível atuação dessas vesículas na carcinogênese. As MPs podem atuar na progressão tumoral em diversos aspectos: - Contribuem para o escape à imunovigilância22. - Auxiliam o tumor a evadir a apoptose, evitando-se o acúmulo de caspase 3 intracelular, por meio da liberação de MPs contendo esta enzima23 . - Aumentam a capacidade de adesão e invasão celular24. Introdução 29 - Induzem a transformação fenotípica celular, por meio do transporte de material genético, enzimas, citocinas e quimiocinas25. - Promovem a progressão e invasão das células tumorais por meio da alteração e degradação da matriz extracellular26. - Promovem a resistência a drogas, pelo acúmulo de quimioterápico nessas vesículas e consequente eliminação pelas células tumorais 27 . - Promovem a indução à angiogênese, especialmente através do transporte de fatores pró-angiogênicos 28. Está claro que as MPs são capazes de modular direta ou indiretamente o comportamento das células ao seu redor, especialmente através do transporte de proteínas e ácidos nucléicos21 . A presença de MPs derivadas do câncer e de células de sua resposta inflamatória pode servir como nova fonte de informação sobre a fisiopatologia desta doença, assim como agir como possível biomarcador para rastreio, diagnóstico e prognóstico dos tumores de ovário. Introdução 30 1.3 Justificativa do estudo O conhecimento da carcinogênse da neoplasia ovariana é de grande importância para elucidação dos mecanismos envolvidos na origem e patogênese desse tumor. O câncer de ovário ainda representa um desafio devido às limitações para rastreamento e opções terapêuticas pouco eficazes, o que implica em baixa sobrevida. Medir a concentração sérica de citocinas, quimiocinas e micropartículas em mulheres com essa neoplasia é uma maneira pouco invasiva e indireta de se avaliar a atividade tumoral, a resposta inflamatória/reguladora sistêmica associada e o papel do microambiente formado na resposta pró- e anti-tumoral do hospedeiro. Além disso, o estudo de biomarcadores da inflamação pode propiciar maior compreensão sobre o comportamento biológico desse tumor, assim como sinalizar para futuros biomarcadores de atividade da doença e de sua monitoração terapêutica. Objetivos 31 2. Objetivos Objetivos 32 2.1. Geral: Avaliação de biomarcadores moleculares da resposta inflamatória em mulheres com carcinoma epitelial de ovário 2.2. Específicos: • Dosar e comparar fatores solúveis da resposta inflamatória- micropartículas (neutrófilos, leucócitos, monócitos, eritrócitos, endotélio, plaquetas, linfócitos), citocinas e quimiocinas (IL-1-β, IL-2, IL-6 IL-10, IL-12, IL- 17a, TNF, IFN- gamma, CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5)- em soro de mulheres com câncer epitelial de ovário e grupo controle • Correlacionar fatores solúveis da resposta inflamatória- micropartículas (neutrófilos, leucócitos, monócitos, eritrócitos, endotélio, plaquetas, linfócitos), citocinas e quimiocinas (IL-1-β, IL-2, IL-6 IL-10, IL-12, IL- 17a, TNF, IFN- gamma, CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5) - com parâmetros clínico-patológicos em mulheres com câncer epitelial de ovário • Correlacionar os níveis séricos de micropartículas (neutrófilos, leucócitos, monócitos, eritrócitos, endotélio, plaquetas, linfócitos) com outros fatores da resposta inflamatória – citocinas e quimiocinas (IL-1-β, IL-2, IL-6 IL-10, IL-12, IL- 17a, TNF, IFN- gamma, CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5)- em mulheres com câncer epitelial de ovário Materiais e Métodos 33 3. Materiais e Métodos Materiais e Métodos 34 3.1. Pacientes e espécimes Foram avaliadas 26 mulheres com diagnóstico de câncer epitelial de ovário (CO) nos estadios I a IV (casos) e 16 mulheres submetidas a histerectomia abdominal total com ooforectomia para tratamento de doença ginecológica benigna (grupo controle), atendidas em uma instituição hospitalar do Estado de Minas Gerais, no período de junho de 2010 a Outubro de 2013, com diagnóstico histopatológico confirmado no pós operatório. Foram realizados exames clínico, ginecológico e ultrassonográfico em todas as mulheres. O estadiamento do CO, foi estabelecido após laparotomia segundo protocolo de classificação pela Federação Internacional de Ginecologia Obstetrícia (FIGO)9 (Anexo I). Todos os tecidos removidos foram examinados por um único patologista. O critério de citorredução ótima foi definido como a permanênia de doença residual < 1cm de diâmetro, após a cirurgia primária. Os níveis séricos do marcador tumoral CA-125 foram obtidos previamente à operação, em todas as mulheres com suspeita de CO, assim como todas as amostras de sangue para o estudo foram coletadas no per operatório, antes da indução anestésica. Para diferenciação entre os tumores tipo I e II, foram utilizados parâmetros clínicos, histológicos e marcação imunohistoquímica para proteína p53, produto do gene suppressor tumoral. Foram classificados como tipo I aqueles tumores de baixo grau, diagnosticados em estadios iniciais, com marcação imunohistoquímica negativa para p53. Foram considerados tumores Materiais e Métodos 35 tipo II aqueles de alto grau, geralmente diagnosticados em estadios avançados e com marcação imunohistoquímica positiva para p53. Houve registro de óbito em prontuário para todas aquelas mulheres que faleceram pelo tumor ovariano incialmente diagnosticado. O estudo foi aprovado pelo Comitê de Ética em Pesquisa (Anexo II). As mulheres selecionadas foram esclarecidas sobre o protocolo de pesquisa e assinaram o Termo de Consentimento Livre Esclarecido (ANEXO III). 3.1.1 Critérios de Inclusão  Mulheres com diagnóstico anatomopatológico de CO submtidas a tratamento cirúrgico (casos);  Mulheres submetidas a ooforectomia abdominal total com ooforectomia para tratamento de doença ginecológica benigna e sem evidências de neoplasias malignas (grupo controle);  Ausência de processo infeccioso agudo peritoneal evidente à laparotomia  Mulheres sem evidência de processo inflamatório agudo sistêmico  Mulheres sem tratamento prévio por quimio ou radioterapia  Mulheres que não utilizaram imunossupressores, corticosteroids e/ou antiinflamatórios não esteróides ha pelo menos 3 meses  Termo de consentimento livre e esclarecido assinado Materiais e Métodos 36 3.2 Métodos 3.2.1 Técnica Cirúrgica As mulheres com CO foram submetidas à laparotomia mediana ampla, inventário da cavidade peritoneal, lavado peritoneal, histerectomia total, salpingo-ooforectomia bilateral, omentectomia, linfadenectomia pélvica e para- aórtica quando indicada, além dos procedimentos necessários à realização de citorredução ótima, quando possível ( ressecção intestinal, hepatectomia parcial, apendicectomia, peritoniectomia, dentre outros). 3.2.2 Avaliação Histológica e Imunohistoquímica As peças cirúrgicas retiradas foram fixadas em formaldeído 10%. O material foi identificado e encaminhado para avaliação anatomo-patológica. As mostras fixadas em formaldeído foram desidratadas em etanol em concentrações crescentes, variando de 70 a 99% ( 60 min. cada); depois imersas em xilol (2x 60 min.) e em parafina 60o(2x 60 min.); e incluídas em bloco de parafina. As lâminas foram coradas pela técnica de hematoxilina eosina (HE), sendo o tipo histológico, o grau de diferenciação tumoral e a presença de metástases classificados de acordo com a FIGO9. Para análise imunohistoquímica, os blocos de parafina foram submetidos a cortes seriados de 4 µm de espessura, que foram colocados em lâminas de vidro salinizadas a 8%. Esses fragmentos foram submetidos à técnica imunohistoquímica segundo método da estreptavidina-biotina- Materiais e Métodos 37 peroxidase, utilizando-se o Kit Universal da Novocastra (Novostain Super ABC Kit, Universal, Novocastra Laboratories, Newcatle upon Tyne, UK). Após as reações as lâminas foram cobertas por lamínula, com auxílio de Permaunt (Fischer Scientific Company, Fair Lawn, NJ, USA). Foram utilizados anticorpos p53 na diluição 1:100 (Novocastra Laboratories, Newcatle upon Tyne, UK, clone DO7). Para revelação das reações foi utilizado o diamino-benzidina. As lâminas foram analisadas em microscópico de luz convencional. As células consideradas positivas foram aquelas coradas com marrom após as reações. As células foram contadas nas áreas em pequeno aumento que exibiam maior concentração de células positivas. Foram considerados positivos aquelas com mais de 10% de células marcadas 29. 3.2.3 Detecção do nível de citocinas/ quimiocinas plasmáticas por citometria de fluxo Para a determinação do nível de citocinas e quimiocinas plasmáticas amostras de sangue periférico foram coletadas a vácuo em tubos de 10 ml contendo heparina sódica (Vacutainer – BD, E.U.A.) no pré-operatório, antes da indução anestésica, e o sangue foi lentamente adicionado sobre uma camada de solução de histopaque 1077 (SIGMA, E.U.A), na proporção de 2:1, em tubos cônicos de poliestireno com capacidade para 50ml (Falcon – BD, E.U.A.). Os tubos foram centrigugados a 400g por 40 minutos a 18º C (Centrífuga Beckman Modelo j-6b, E.U.A) e após a centrifugação o plasma obtido foi aliquotado e mantido a -20oC até a realização dos experimentos. Materiais e Métodos 38 Os níveis plasmáticos de citocinas e quimiocinas foram quantificados utilizando-se o sistema Cytometric Bead Array (CBA) (Becton Dickinson-BD) (Figura 3), que emprega uma mistura de esferas de poliestireno, de intensidades de fluorescência discretas e distintas, recobertas com anticorpos específicos para as citocinas e quimiocinas humanas que são detectadas no canal de fluorescência 3 (FL-3). Essa metodologia permite a avaliação simultânea de várias citocinas e quimiocinas no mesmo ensaio, empregando pequenos volumes de amostras séricas. Figura 3: Representação esquemática da técnica de CBA. Nota: Adaptado do bulário do Cytometric Bead Array Flex Set System BD- Biosciences, 2004-2005. Com micro-esferas marcadas, foram adicionados aos anticorpos anti-citocinas/quimiocinas, amostras de soro. Em seguida, a amostra foi marcada por imunofluorescência (ficoeritrina) e analisada no citômetro de fluxo. Os resultados foram obtidos por software específico para CBA (FCAP Array TM Software, BD, Pharmingen, EUA). Os anticorpos utilizados foram anti: IL-1, IL-2, IL-6 IL-10, IL-12, IL- 17A, TNF, IFN- gamma, CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5. Ac: Anticorpos; PE: Ficoeritrina Materiais e Métodos 39 Neste estudo, a metodologia de CBA foi adaptada dos protocolos originais propostos por Chen et al. (1999), como descrito a seguir: Alíquotas de 25mL de plasma diluído 1:5 com diluente G (reagente do kit CBA), alíquotas de 25mL dos padrões de citocinas/ quimiocinas, submetidos à diluição seriada com diluente G (“Top Standard” – 2500 pg/mL, 1:2 – 1250 pg/mL, 1:4 – 625 pg/mL, 1:8 – 312.5 pg/mL, 1:16 – 156 pg/mL, 1:32 – 80 pg/mL, 1:64 – 40 pg/mL, 1:128 – 20 pg/mL e 1:256 – 10 pg/mL) e 25mL de diluente G apenas (Controle Negativo), foram transferidas para tubos de poliestireno de 5mL (Falcon – BD, E.U.A). Posteriormente, a cada tubo foram adicionados 15mL da mistura de esferas de captura, conjugadas com anticorpos monoclonais selecionados para a avaliação (anti- IL-1, IL-2, IL-6 IL- 10, IL-12, IL- 17a, TNF, IFN- gamma, CXCL8, CXCL-9, CXCL 10, CCL 2, CCL5), com subseqüente incubação por 90 minutos, à temperatura ambiente, ao abrigo da luz. Após a incubação, as esferas de captura foram lavadas com 500mL da solução F (“Wash buffer”, reagente do kit CBA), centrifugadas a 600g, por 10 minutos a 18oC e o sobrenadante foi cuidadosamente aspirado e descartado. As esferas foram então re-incubadas na presença de 20mL do reagente B, que corresponde a um coquetel de anticorpos monoclonais anti- citocinas/quimiocinas humanas, conjugados com o fluorocromo PE (FL-2) por 90 minutos, temperatura ambiente, ao abrigo da luz. Após incubação, as esferas de captura foram novamente lavadas com 500mL da solução F, centrifugadas a 600g, por 10 minutos a 18oC e o sobrenadante foi cuidadosamente aspirado e descartado. Após centrifugação, as esferas foram ressuspendidas em 250mL de reagente F e imediatamente analisadas no citômetro de fluxo. Materiais e Métodos 40 3.2.4 – Aquisição e análise dos dados de citocinas e quimiocinas plasmáticas por citometria de fluxo A aquisição dos dados obtidos foi realizada no citômetro de fluxo FACSCalibur®(BD). Embora as esferas fluorescentes presentes no kit CBA sejam projetadas para serem excitadas com o laser de argônio (488nm), com emissão em comprimento de onda correspondente ao parâmetro Fluorescência 3 (FL-3), elas também podem ser excitadas pelo “red diodo laser”, com emissão de fluorescência detectadas no canal Fluorescência 4 (FL-4). Esta possibilidade de leitura pode ser obtida durante o processo de aquisição através da utilização do “Dual Laser CBA Template”, que simplifica os ajustes do equipamento, reduzindo a necessidade de compensações da interferência que a fluorescência emitida pelas esferas possa exercer sobre a fluorescência inerente aos anticorpos monoclonais anti-citocinas/quimiocinas humanas, conjugados com o fluorocromo PE (FL-2). Após as etapas de marcação, um total de 1.800 eventos/região (R1) foram obtidos com base em gráficos de tamanho (FSC) versus granulosidade (SSC) (Figura 4A). Para a análise dos dados, inicialmente as microesferas conjugadas com anticorpos monoclonais de captura correspondentes a cada citocina/quimiocina foram segregadas em gráficos de distribuição puntual FL-4 x FL-2, onde as cinco esferas com intensidades de fluorescência distintas ocuparam posições específicas ao longo do eixo Y (FL-4). A análise do deslocamento das esferas ao longo do eixo X (FL-2) foi empregada como variável proporcional à concentração de cada citocina/quimiocina presente na amostra (Figura 4B e 4C). Para a obtenção dos resultados da análise quantitativa de citocinas/quimiocinas plasmáticas, uma curva padrão foi construída utilizando os dados dos padrões Materiais e Métodos 41 de citocinas/quimiocinas em concentrações conhecidas (10 pg/mL – 2500 pg/mL) e empregada para determinar as concentrações de cada citocina /quimiocina no plasma teste. Um modelo de ajustamento através da curva do 5º parâmetro logístico, que permite o ajuste da melhor curva não linear para dados detectáveis, foi utilizado. Dessa forma, foi possível extrapolar valores de intensidades de fluorescência de amostras que não caíam dentro dos limites da curva padrão. Figura 4: Exemplo de perfil dos níveis plasmáticos de quimiocinas através de microsesferas de captura de mulheres com câncer de ovário por citometria de fluxo. (A) Determinação da região da população de microesferas através de gráficos de distribuição puntual de tamanho (FSC) versus granulosidade (SSC). (B) Deslocamento das microesferas, ao longo do eixo X (FL-2) em gráficos de distribuição puntual FL-2 versus FL-4, proporcional à concentração de cada quimiocina presente nos plasmas avaliados. (C) Histograma das microesferas ao longo do eixo X (FL2) em gráficos de distribuição puntual Fluorescência 2 versus Fluorescência 3, proporcional à concentração de cada quimiocina presente nos plasmas avaliados. Figura 2: Perfil Materiais e Métodos 42 3.2.5. Obtenção do plasma livre de plaquetas Amostras de sangue total coletadas em tubos contendo o anticoagulante citrato de sódio serão submetidas à centrifugação (600 x g) por 15 minutos à TA para obtenção do plasma pobre em plaquetas (PPP). O PPP será novamente centrifugado por 3 min (13.000 x g) a 4oC para obtenção do plasma livre de plaquetas (PLP) que será armazenado a -70oC até o momento do uso. 3.2.6. Obtenção das micropartículas (MPs) A quantificação das MPs no plasma será realizada pela citometria de fluxo adaptada dos protocolos descritos anteriormente30. Cem microlitros do PLP serão diluídos em solução de PBS contendo citrato e heparina (1 µg/mL) (diluição de 1:3) e novamente centrifugados por 90 minutos a 14.000 x g a 15°C. O sedimento rico em MPs será ressuspenso no tampão de ligação à anexina 1X (BD Biosciences, Calipornia, US). As MPs serão quantificadas por citometria de fluxo através da calibração com “microbeads” fluorescentes (Spherotech Inc.Libertyville, Illinois, US) de tamanho definido (0,7 a 0,9 µm). Dez microlitros das “beads” serão adicionados à 100 µL de PBS 1X estéril. 3.2.7. Identificação de micropartículas (MPs) pela Citometria de fluxo Análise fenotípica das MPs por citometria de fluxo A maioria dos reagentes utilizados neste estudo será obtida da empresa BD Biosciences, exceto quando mencionado. A caracterização fenotípica das MPs para determinação de sua origem celular será realizada com a incubação Materiais e Métodos 43 de 100 µL do plasma contendo as MPs com os anticorpos específicos para: neutrófilos (CD66/PE), linfócitos T (CD3/PE), plaquetas (CD41a/PerCP), leucócitos (CD45/APC), monócitos (CD14/PerCP), eritrócitos (CD235a/PECy5) e células endoteliais (CD51/61/PE). Após a adição dos anticorpos, as amostras serão incubadas por 30 min. protegidas da luz. As MPs serão ressuspensas em 100 µL do tampão de ligação de Anexina V (BD Pharmingen) e, finalmente, 5 µL de Anexina V/FITC serão adicionados, já que a Anexina V reconhece resíduos de fosfatidilserina que geralmente estão presentes na superfície das MPs. As amostras serão levadas ao citômetro de fluxo LSR-Fortessa (Becton Dickson - BD, E.U.A.), onde cerca de 100.000 eventos serão obtidos em cada amostra, sendo pelo menos 2.000 eventos dentro da região específica para MPs. As análises serão feitas utilizando o software FlowJo (Tree Star), onde serão construídos dots plots de todos os marcadores utilizados versus Anexina/FITC, permitindo assim a identificação e quantificação de cada população de MPs específica. Como as MPs possuem um tamanho de cerca de 1 µm, utilizou-se microesferas fluorescentes de tamanho definido (0,7 - 0,9 µm) para delimitar a região (R1) correspondente às MPs. As MPs isoladas do plasma foram então definidas de acordo com o tamanho (FSC) e a granulosidade (SSC) (Figura 5A). Além disso, utilizou-se a proteína ligadora de fosfolípedes Anexina V (capaz de reconhecer fosfatidilserinas presentes na superfície das MPs) conjugada com fluoresceína - FITC. Materiais e Métodos 44 Figura 5: Representação esquemática da seleção de micropartículas em plasma livre de plaquetas. Foram utilizadas microesferas fluorescentes de tamanho definido (0,7 - 0,9 µm) para delimitar a região (R1) correspondente às MPs em gráficos de tamanho (FSC) versus granulosidade (SSC) (Figura 5A). Além disso, foi utilizada Anexina V conjugada com fluoresceína – FITC versus o marcador de origem das micropartículas. Os resultados foram expressos em frequência percentual do quadrante duplo-positivo referente a presença de micropartículas Anexina V+ e o marcador de população celular+. No exemplo, temos dois exemplos de resultados da combinação Anexina V-FITC x CD41- PerCP (marcador de plaquetas) (Figuras 5B e 5C, respectivamente). 0 10 20 30 40 50 0 2 4 6 Plaqueta Endotélio Leucócito **p= 0.0015 *p= 0.014 Monócito % d e M P s an ex in a V ⁺ e xp re ss an d o o s m ar ca d o re s ce lu la re s % d e M P s an exin a V ⁺ exp ressan d o o s m arcad o res celu lares A SS C- H FSC-H R1 0.7mm 0.9mm B C D C D 41 -P er C P C D 41 -P er C P Referências 45 ref Referências 46 4. Referências: 1. Estimated Caner Incidence, Mortality and Prevalence Worldwild in 2012. International Agency for Research on Cancer. http://globocan.iarc.fr/Pages/fact_sheets_population.aspx (acessado em 20/01/2014). 2. Estimativa de Câncer no Brasil 2014. Instituto Nacional do Câncer. http://www.inca.gov.br/estimativa (acessado em 20/01/2014). 3. Anthoulakis C, Nikoloudis N. Pelvic MRI as the "gold standard" in the subsequent evaluation of ultrasound-indeterminate adnexal lesions: A systematic review. Gynecol Oncol. 2013(13):1271-7. 4. Ueland FR, et al. Preoperative differentiation of malignant from benign ovarian tumors: the efficacy of morphplogy indexing and Doppler flow sonography. Gynecol Oncol 2003; 91: 46-50. 5. Jian-Lun Xu, et al. Longitudinal evaluation of CA-125 velocity and prediction of ovarian cancer. Gynecologic Oncology .2012; 125: 70–74. 6. Cancer Facts and Figures, 2013. American Cancer Society, Surveillance Research. 2013; p:1-64. 7. Kurman, R J; Shih. The Origin and Pathogenesis of Epithelial Ovarian Cancer- a Proposed Unifying Theory. Am J Surg Pathol. 2010 ; 34(3): 433–443. http://globocan.iarc.fr/Pages/fact_sheets_population.aspx http://www.inca.gov.br/estimativa http://www.ncbi.nlm.nih.gov/pubmed?term=Anthoulakis%20C%5BAuthor%5D&cauthor=true&cauthor_uid=24183731 http://www.ncbi.nlm.nih.gov/pubmed?term=Nikoloudis%20N%5BAuthor%5D&cauthor=true&cauthor_uid=24183731 Referências 47 8. A. P. Heintz, F. Odicino, P. Maisonneuve et al., ―Carcinoma of the ovary. FIGO 26th annual report on the results of treatment in gynecological cancer,‖ International Journal of Gynaecology and Obstetrics. 2006; 95:161–192. 9. Benedet, J.L., et al. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. FIGO Committee on Gynecologic Oncology. Int J. Gynaecol Obstet. 2000. 70 (2): 209-62. 10. Chang SJ et al. Survival impact of complete cytoreduction to no gross residual disease for advanced-stage ovarian cancer: A meta-analysis. Gynecologic Oncology. 2013; 130: 493–498. 11. Trimbus JP, et al. Impact of Adjuvant Chemotherapy and Surgical Staging in Early-Stage Ovarian Carcinoma: European Organisation for Research and Treatment of Cancer–Adjuvant ChemoTherapy in Ovarian Neoplasm Trial. J Natl Cancer Inst 2003;95: 113–25. 12. Harmandayan, G.Z. et al. Ovarian cancer patient surveillance after curative- intent initial treatment. Gynecologic Oncology .2011; 120: 205–208. 13. Visser, K.E., A. Eichten, and L.M.Coussens, Paradoxical roles of the immune system during cancer development. NatRev Cancer, 2006. 6(1): p.24-37. 14. Chen, Yu-Li, et al. Interferon-gamma in ascites could be a predictive biomarker of outcome in ovarian carcinoma. Gynecologic Oncology .2013; 131: 63–68. Referências 48 15. Wajant H. The role of TNF in cancer. Results Probl Cell Differ. 2009; 49:1-15. 16. Szlosarek PW, Grimshaw MJ, Kulbe H, Wilson JL, Wilbanks GD, et al. Expression and regulation of tumor necrosis factor alpha in normal and malignant ovarian epithelium. Mol Cancer Ther. 2006; 5: 382–390. 17. Aggarwal BB .Signalling pathways of the TNF superfamily: a double- edged sword. Nat Rev Immunol. 2003; 3: 745–756. 18. Pinciroli et al. An IL6-correlated signature in serous epithelial ovarian cancer associates with growth factor response. BMC Genomics 2013, 14:508. 19. Son DS, Kabir SM, Dong YL, Lee E, Adunyah SE. Inhibitory effect of tumor suppressor p53 on proinflammatory chemokine expression in ovarian cancer cells by reducing proteasomal degradation of IkB. PLoS One. 2012. 7:e51116. 20. E Chargaff, R. West. The biological significance of the thromboplastic protein of blood. Journal of Biological Chemistry. 1946; 166: 187-196. 21. Weng, X., et al. Membrane microparticles and diseases. European Review for Medical and Pharmacological Sciences. 2013; 17: 2420-2427. 22. Valenti, R. Tumor-released microvesicles as vehicles of immunosuppression. Cancer research. 2007; 67: 2912-2915. 23. Hussein, M.N., et al. Inhibition of microparticle release triggers endothelial cell apoptosis and detachment. Thrombosis and Haemostasis. 2007; 98:1096-1107. http://www.ncbi.nlm.nih.gov/pubmed?term=Wajant%20H%5BAuthor%5D&cauthor=true&cauthor_uid=19137269 Referências 49 24. Angelucci, A. et al. Vesicle associated urokinase plasminogen activator promotes invasion in prostate cancer cell lines. Clinical and Experimental Metastasis. 200; 18: 163-170. 25. Al-Nedawi, K. et al. Intercellular transfer of the oncogenic receptor EGFRvIII by microvesicles derived from tumours cells. Nature Cell Biology. 2008; 10: 619-624. 26. Graves,E.L. Pro invasive properties in ovarian cancer ascites-derived membrane vesicles. Cancer Research. 2004; 64: 7045-7049. 27. Shedden, K., et al. Expulsion of small molecules in vesicles shed by cancer cells: association with gene expression and chemosensitivity profiles. Cancer research. 2003; 63: 4331-4337. 28. Al-Nedawi, K. et al. Endothelial expression of autocrine VEGF upon the uptake of tumor-derived microvesicles containing oncogenic EGFR. Proceedings of the National Academy of Sciences of the United States of America. 2009; 106: 3794-3799. 29. Wang TY, et al. Histologic and immunophenotypic classification of cervical carcinomas by expression of the p53 homologue p63: a study of 250 cases. Hum Pathol 2001; 32 (5):479-486. 30. Campos et al. Augmented plasma microparticles during acute Plasmodium vivax infection. Malaria Journal 2010, 9:327 Article 1 50 Article 1 51 ARTICLE I Frequency of circulating microparticles in patients with ovarian cancer Article Type: Original article Keywords: ovarian cancer , inflammatory response, microparticles . 1. Introduction Ovarian cancer comprises a heterogeneous group of neoplasms and is the most lethal gynecologic malignancy in the developed countries. Early detection of disease remains problematic as there is no good screening test and there is a lack of a clearly defined precursor lesion. Thereby, the majority of women are initially diagnosed with tumor metastases beyond the ovary, which results in diminished chances of long-term survival1,2. Surgical cytoreduction and adjuvant chemotherapy are the hallmarks of management for advanced ovarian cancer3. Currently, the molecular events leading to the development of epithelial ovarian cancer are still unknown. Because of the great heterogeneity in molecular and biological status, ovarian cancer is actually many different diseases, which has different clinical outcomes, and may requires different treatments. Thus, it was proposed a new classification, categorizing ovarian cancer into type I and II tumors. Type I tumors are suggested to behave in an indolent manner, with a stable genome, although somatic mutations are frequently detected in a number of genes4. Type II tumors are proposed to be more aggressive and genetically highly instable; most of them have TP53 mutations, and almost half of the cases have mutation, hypermethylation, or dysfunction of BRCA1/25.These aggressive tumors account for 75% of all ovarian cancer, and are responsible for 90% of deaths from the disease6. Many researches have been conducted to identify a specific biomarker to the ovarian cancer. Therefore, it has been hypothesized that microparticles (MPs) play a fundamental role in the ovarian tumor microenvironment. They are phospholipid vesicles (< 1 μm) released from cells in response to a variety of stimuli, and they have been studied and gained attention due to the large amounts of them releasing from tumor cells. MPs are able to transport cell-specific surface antigens of their cell of origin, such antigens being used to identify the various subtypes of MPs7. It is postulated that MPs releasing from cells may alter the biological activity of the receiver cell, either by transferring receptors that can induce cell signalling, or by transferring second messengers. This binding of MP surface antigens to their specific receptor on cells allows intercellular signaling between distant cells, as well as other processes such as modulation of the immune response8. The purpose of this study was to measure the frequency of circulating MPs in patients with ovarian cancer, and to compare the results with a group of women without malignancy. The correlation with clinic-pathological parameters was also analyzed. Article 1 52 2. Materials and methods 2.1. Patients and specimens Patients who were undergoing evaluation for suspected primary ovarian cancer from June 2010 to October 2013 were recruited in a Brazilian clinical center. All patients underwent debulking surgery and received standard platinum-based chemotherapy when indicated. The control group selected presented no underlying malignancy or infection requiring benign gynecological surgery. Their medical records were obtained until October 2013. The Institutional Review Board approved the study protocol, and all patients provided informed signed consent. Blood sampling was carried intra- operatively, before the anesthetic induction. Flow cytometry was the approach to analyze MPs, cytokines/chemokines, as the method to assess the number and the phenotype of MPs and cytokines/chemokines9. Histological grading and disease staging were based on the International Federation of Gynecology and Obstetrics (FIGO) classification10,11. In this study, FIGO stage I/II and FIGO stage III/IV ovarian cancers were considered early and advanced disease, respectively. Other clinical information including age, surgical findings, pathological and immunohistochemical characteristics (i.e type I or II tumors), CA125- levels, and survival, was obtained from the medical records. The maximum diameter of the residual tumor after surgery was also retrieved. Optimal debulking surgery was defined as the maximum diameter of residual tumor ≤1 cm. Otherwise, the debulking surgery was considered sub-optimal. Article 1 53 2.2. Purification of MPs from plasma MPs were prepared as described elsewhere12,13,14. Briefly, Citrated blood (0.5 mL) was centrifuged at 1,500 × g for 15 min, and plasma was then cooled to -20°C before storage at -80°C. Samples were further centrifuged at 13,000 x g for 3 min to obtain platelet-free plasma. The latter was diluted 1:3 in citrated PBS containing heparin and centrifuged at 14,000 x g for 90 min at 15°C. The resultant MP pellet was then resuspended in 1× annexin V binding buffer (BD Biosciences, California, US). 2.3. Flow cytometry assays 2.3.1. Detection of plasmatic MPs Unless otherwise stated, all reagentes and mAbs used in the Flow cytometry experiments were provided from BD Biosciences. MPs isolated from plasma were gated (R1) based on their forward (FSC) and side (SSC) scatter distribution in density plots as compared to the distribution of synthetic 0.7 – 0.9 μm SPHERO™ Amino Fluorescent Particles (Spherotech Inc. Libertyville, Illinois, US). Taking into account the presence of phosphatidylserine (PS) residues in MPs surface, events present in R1 were accessed for their positive staining for annexin V (BD Bioscience) – a classical marker for microparticles – using PE-conjugated monoclonal antibodies (mAbs). Mouse IgG PE conjugated isotype control mAbs were used to properly place gates. Annexin V+ events gated on R2 region were further assessed for immunolabeling with FITC- http://www.malariajournal.com/sfx_links?ui=1475-2875-9-327&bibl=B18 Article 1 54 conjugated mAbs against the cell markers CD66 (neutrophils), CD41a (platelets), CD51 (endothelial cells), CD235a (erythrocytes), CD45 (leukocytes), CD3 (lymphocytes), and CD14 (monocytes) or the correspondent mouse IgG FITC-conjugated isotype control mABs. The samples were analyzed in a Flow Cytometry FACSCalibur (Becton-Dickinson, California, US). Over 100,000 events were acquired on each sample, to reach at least 2,000 events within the MPs gate. 2.4.Statistical Analyses Statistical analyses were performed using SPSS software version 13.0 (SPSS Inc., Chicago, IL). Shapiro–Wilk tests were used to verify if the variables were normally distributed. Non-parametric data were compared using the Kruskal–Wallis test followed by the Dunns post-hoc test. Comparison between 2 groups was done using the Mann–Whitney U test with Bonferroni's correction (non-normal data) or t-test (normal data). Normal data are presented as mean and standard deviation, while non-normal variables are presented as median and interquartile range (25th–75th percentiles). Correlations were analyzed using the Pearson or Spearman 2-sided test, if the data were parametric or non- parametri, respectively, and Pearson χ2 test was performed to frequency differences. Differences were considered significant when p<0.05. Article 1 55 3. Results Characteristics of the study population are summarized in Table 1. There were no differences between the groups regarding age and menopausal status (p= 0,824 and p= 0.055, respectively). In the group of patients with OC, 10 (38.5%) had stage I/II and 16 (61.5%) had stage III/IV ovarian cancer. Among these patients, 8 were type I (30.8%) and 18 were type II (69.2%) tumors. Optimal cytoreduction occurred in 15 (57.7%) patients with ovarian cancer. All women with type I tumors had optimal cytoreduction. However, just 7 (38.9%) patients with type II had residual disease ≤ 1 cm in maximal diameter. CA125 levels were significantly (p=0.024) different between the type I and II tumors. Death was not observed in patients with type I, whereas in type II, 6 cases (33.3%) were registered. Article 1 56 The phenotype of plasma circulating MPs was investigated with specific mAbs for cell markers (i.e. CD41, CD235, CD45, CD3, CD51, CD14 and CD66) that were used to discriminate the cellular sources of annexin V+ MPs. The percentages of annexin V+ MPs stained for each cell marker was compared between samples from OC patients and control group. Other clinic-pathological data were also analyzed. The frequency of leukocyte (p < 0.005) derived-microparticles (LMPs) was significantly increased in plasma samples from OC patients, whereas endotelial derived MPs (EMPs) were lower in OC as compared to control group (Figure 1.A). LPMs were even higher in type II tumors as compared to type I (Figure 1.B). Article 1 57 A B Figure 1: A. Percentage of circulating microparticles (MPs) in patients with ovarian cancer (OC) and control group according to their specific cellular origin. B. Percentage of circulating microparticles (MPs) in patients with type I and II tumors according to their specific cellular origin. Increased number of neutrophil derived MPs (NMPs) were observed in women with OC who underwent optimal cytoreductive surgery (Figure 2.A). Furthermore, the NMPs were statistically higher in those patients with CA125 levels greater than 600 U/mL (Figure 2.B) compared with those women with CA125 levels less than 600U/mL. Thus, NMPs were significantly increased in plasma samples from OC patients who were optimally cytoreducted, and had higher levels of CA125 biomarker, compared with those women with optimal operation and CA125 levels less than 600U/mL (Figure 2.C). Article 1 58 A B C Figure 2. A. Percentage of circulating microparticles (MPs) according to optimal cytoreductive surgery. B. Percentage of circulating MPs according to CA125 levels. C. Percentage of circulating MPs according to optimal cytoreductive surgery and CA125 levels. Article 1 59 Similarly, when we analyzed the association between MPs with other clinical data, such as stage and cytoreductive operation, we observed that LMPs were higher in patients with advanced stage OC as well as with residual disease > 1cm after surgery when compared with those in early stage. Accordingly, the levels of NMPs were statistically increased in women with stage III/IV who had optimal operation as compared with those presenting the same stage not well debulked. The frequency of lymphocyte, monocyte, erythrocyte and endothelial derived-microparticles were significantly increased in plasma samples from patients with advanced OC and residual disease after surgery lower than 1cm as compared with those with early stage disease and optimal surgical treatment (Figure 3). Figure 3. Comparison of circulating microparticles (MPs) from patients with ovarian cancer (OC) according to stage and optimal cytoreductive surgery. Article 1 60 4. Discussion Although the incidence of OC is low, it is the most lethal gynecologic malignancy and typically has a poor prognosis. Studies have focused on the microparticles released from tumor and from inflammatory cells to identify a biomarker for OC. MPs may represent a potential biomarker because they can be easily isolated from blood, and they have particular features that may correlate with stage and clinical data. The present study investigated serum levels of neutrophils, platelets, endothelial, erythrocytes, leukocytes, lymphocytes and monocytes derived- microparticles in women with OC and without malignancy. The frequency of LMPs was increased in women with OC as well as was even higher in type II tumors when compared with those without malignancy. It has been known that tumor microenvironment consists of a variable combination of tumor cells, stromal fibroblasts, endothelial cells and infiltrating leukocytes, such as neutrophils, macrophages, lymphocytes and other inflammatory cells, and they play a fundamental role in tumor modulation and progression15,16. Other important consideration is the association between MPs and inflammatory response17. The successful expansion of malignant tumours requires an active collaboration between malignant and inflammatory cells via heterotypic cellular interactions. Thus, MPs provide a relatively new route of Article 1 61 communication between cancer cells and various stromal cells infiltrating the tumour interstitium. The actual molecular composition of MPs varies depending on the mechanism of their formation as well as the type and functional state of their cellular origin. MPs are important carriers of membrane components or bioactive molecules, and they can mediate exchange of signaling proteins and genetic material, which altogether may support tumor growth and progression. A large quantity of MPs has been identified in fluids of patients with malignant pathologies. Thereby, these study findings corroborate with the relationship between the immune system and OC inflammation, especially in tumors with higher aggressiveness, as ovarian type II tumors. These results enhance the dualistic model that categorizes OC in two types. They suggest a difference in susceptibility to carcinogenesis in both tumors. On the other hand, we might not find any explanation for the presence of the decreased levels of endothelial derived microparticles in the OC group, unless the low number of patients. We observed an increased expression of NMPs in patients with higher levels of CA125. This is an interesting result, and could be correlated with the ability of the tumor to expresses biomarkers. Moreover, NMPs were statistically higher in women who had optimally citoreductive surgery, especially those with great levels of CA125. Some authors demonstrated that activated polymorphonuclear neutrophils release MPs at the time of degranulation. NMPs have recently been shown to inhibit the inflammatory properties of human monocyte-derived macrophages in vitro, which in the local context may Article 1 62 participate in the control of immune responses. These NMPs have the particularity to be involved very early in inflammation, which might be crucial for determining later aspects of the cascade responsible for acquired immunity, and may responsible for controlling the immune response as well18,19. The present study showed that elevated frequency of NMPs could be considered as a prognostic factor for optimal surgical treatment. Nevertheless, additional studies are necessary to support this hypothesis. Probably, cancer MPs contribute to hypercoagulability, as various types of cancer cells are known to secrete MPs carrying tissue factor VII20. As mentioned earlier, MPs can interact with the extracellular matrix by depositing paracrine information or facilitating matrix degradation, thereby creating paths of least resistance. Bothey conditions can induce the stromal microenvironment to promote angiogenesis, evasion of the immune response, tumor invasion, and, potentially, metastasis21. Therefore, it is expected to find out a substantial number of MPs in advanced disease. This study demonstrated that circulating leukocyte, lymphocyte, monocyte, erythrocyte and endothelial derived MPs were higher in patients in advanced stage disease with optimal cytoreductive operation compared to those treated in early ovarian cancer. This is an original study that evaluated potential biomarkers for OC. The analysis of MPs is an easy and less invasive way to measure the tumor activity. Futhermore, it may provide a better understanding of the complex environment for disease development, especially currently with the new model for classification of OC. Article 1 63 A potential limitation of the statistics analysis performed here was that multiple comparison were realized, aiming to find possible associations between MPs and clinical/biological parameters; however, it should be take account that these significant findings were hypothesis-generating, and, consequently, its represent a realistic interpretation of the facts. Moreover, the limited number of patients in the study is explained by the low prevalence of the disease. The median follow-up could also be higher. Additional approaches should also be explored to guarantee the ability to isolate and quantify ovarian cancer MPs from blood and other biological fluids. This paper describes the initial attempts to characterize MP phenotype features in ovarian cancer. It was shown here that plasma levels of LMPs were increased in ovarian cancer as in type II tumors. An significant expression of NMPs were associated with higher levels of CA125 as well as with optimal surgery. MPs may serve as a novel source of disease related information, and possibly as specific and identifiable cancer biomarkers that may prove useful for screening, diagnosis and treatment. The role of the immune and inflammatory response in the process of carcinogenesis represents a line of research with great potential in oncology. Article 1 64 Conflict of interest statement None of the authors has any conflicts of interest Acknowledgments This study was supported by Fundação Oswaldo Cruz (FIOCRUZ), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG). The authors also thank the program for technological development in tools for health-PDTIS-FIOCRUZ for the use of its facilities. OAMF and ATC thank CNPq for fellowships (PQ). Article 1 65 5. Referências: 1. American Cancer Society. Cancer Facts and Figures 2013. Atlanta: American Cancer Society; 2013. 2. Chan A, Gilks B, Kwon J, Tinker AV. New insights into the pathogenesis of ovarian carcinoma. Time to rethink ovarian cancer screening. Obstet Gynecol. 2012; 120: p.935-40. 3. Chang S J et al. Survival impact of complete cytoreduction to no gross residual disease for advanced-stage ovarian cancer: A meta-analysis. Gynecologic Oncology. 2013; 130: p.493–498. 4. Kurman RJ, Shih Ie M. Molecular pathogenesis and extraovarian origin of epithelial ovarian cancer—shifting the paradigm. Hum Pathol. 2011; 42:918–31. 5. Spellman PT, et al. Integrated genomic analyses of ovarian carcinoma. Nature 2011; 474:609–15. 6. Kurman RJ, Shih IeM. The Origin and Pathogenesis of Epithelial Ovarian Cancer – Proposed Inifying Theory. Am J Surg Pathol. 2010; 34(3):433-43. 7. Van Doormaal FF, Kleinjan A, Di Nisio M, Buller HR, Nieuwland R. Cell- derived microvesicles and cancer. Neth J Med 2009; 67(7):266–73.
 http://www.nature.com/nature/foxtrot/svc/authoremailform?doi=10.1038/nature10166&file=/nature/journal/v474/n7353/full/nature10166.html&title=Integrated+genomic+analyses+of+ovarian+carcinoma&author=P.+T.+Spellman Article 1 66 8. Leroyer AS, Anfosso F, Lacroix R, Sabatier F, Simoncini S, Njock SM, et al. Endothelial-derived microparticles: biological conveyors at the crossroad of inflammation, thrombosis and angiogenesis. Thromb Haemost 2010; 104(3): 456-463. 9. van der Heyde HC, Gramaglia I, Combes V, George TC, Grau GE. Flow cytometric analysis of microparticles. Methods Mol Biol. 2011; 699:337-54. 10. A. P. Heintz, F. Odicino, P. Maisonneuve et al., ―Carcinoma of the ovary. FIGO 26th annual report on the results of treatment in gynecological cancer,‖ International Journal of Gynaecology and Obstetrics. 2006; 95:161–192. 11. Benedet, J.L., et al. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. FIGO Committee on Gynecologic Oncology. Int J. Gynaecol Obstet 2000. 70 (2): 209-62. 12. Combes V, Taylor T, Juhan-Vague I, Mège J, Mwenechanya J, Tembo M, Grau G, Molyneux M. Circulating endothelial microparticles in malawian children with severe falciparum malaria complicated with coma. JAMA 2004, 291:2542-2544. 13. Faille D, Combes V, Mitchell A, Fontaine A, Juhan-Vague I, Alessi M, Chimini G, Fusaï T, Grau G. Platelet microparticles: a new player in malaria parasite cytoadherence to human brain endothelium. FASEB J 2009, 23:3449-3458. http://www.ncbi.nlm.nih.gov/pubmed?term=van%20der%20Heyde%20HC%5BAuthor%5D&cauthor=true&cauthor_uid=21116991 http://www.ncbi.nlm.nih.gov/pubmed?term=Gramaglia%20I%5BAuthor%5D&cauthor=true&cauthor_uid=21116991 http://www.ncbi.nlm.nih.gov/pubmed?term=Combes%20V%5BAuthor%5D&cauthor=true&cauthor_uid=21116991 http://www.ncbi.nlm.nih.gov/pubmed?term=George%20TC%5BAuthor%5D&cauthor=true&cauthor_uid=21116991 http://www.ncbi.nlm.nih.gov/pubmed?term=Grau%20GE%5BAuthor%5D&cauthor=true&cauthor_uid=21116991 http://www.malariajournal.com/sfx_links?ui=1475-2875-9-327&bibl=B17 Article 1 67 14. Campos et al. Augmented plasma microparticles during acute Plasmodium vivax infection. Malaria Journal 2010, 9:327 15. Visser, K.E., A. Eichten, and L.M.Coussens, Paradoxical roles of the immune system during cancer development. Nat Rev Cancer, 2006. 6(1): 24-37. 16. Chen, Yu-Li, et al. Interferon-gamma in ascites could be a predictive biomarker of outcome in ovarian carcinoma. Gynecologic Oncology. 2013; 131:63–68. 17. Weng, X., et al. Membrane microparticles and diseases. European Review for Medical and Pharmacological Sciences. 2013; 17(18): 2420-2427. 18. Eken C; Gasser O; et al. Polymorphonuclear Neutrophil-Derived Ectosomes Interfere with the Maturation of Monocyte-Derived Dendritic Cells . J Immunol. 2008; 180:817-824. 19. Falanga A, Tartari CJ, Marchetti M. Microparticles in tumor progression. Thromb Res. 2012;129(1):132-136. 20. Yokota N; Koizume S; Self-production of tissue factor-coagulation factor VII complex by ovarian cancer cells. British Journal of Cancer. 2009; 101, 2023 – 2029. 21. Giusti I; et al. Microvesicles as Potential Ovarian Cancer Biomarkers. BioMed Research International. 2012; 13: 1-12. Article 2 68 Article 2 69 ARTICLE II Circulating cytokine, chemokine and microparticles signatures in women with ovarian cancer Article Type: Original article Keywords: Ovarian cancer , inflammatory response, cytokines, chemokines, microparticles . 1. Introduction Ovarian cancer (OC) is still the most frequent cause of death by gynecological malignancy for women. It has the highest fatality-to-case ratio for all gynecologic malignancies because more than two-thirds of patients have advanced disease at diagnosis. The lethal nature of this pathology is predominantly attributed to its ability to spread extensively with minimal clinical symptoms. In the United States, it is the tenth most common malignancy in women, and accounts for the fifth highest number of deaths, an estimated 14.030 deaths in 20131. OC accounts for about 2.1% of all cancers among women at Brazil, and around 5,680 new cases are stimated in 20142.. Despite advances in the treatment of advanced disease, the overall 5-year survival rate is approximately 40-50%3. Article 2 70 To date, all attempted OC screening strategies have failed, probably because this disease represents different kinds of cancers4. A new classification for OC in two different types of cancer has been introduced: Type I tumors involve low grade and indolent tumors. It accounts for many early stage cancers and are characterized by specific mutations including KRAS, BRAF, ERB2, CTNNB1, PTEN, PIK3K, ARID1A and PPP2R1A. Type I tumors rarely harbor TP53 mutations and are stable genetically 5. Type II ovarian tumors are considered the most frequently diagnosed, aggressive, genetically instable, and often disseminated kind of disease. Type II tumors have a very high frequency of TP53 mutations 6. Molecular biology of oncogenesis in ovarian cancer consists of multiple complex pathways. Previous research has focused on identifying prognostic markers for OC but few definitive results have been robustly validated 7. Evidence is mounting that an inflammatory process contributes to tumor growth and metastasis to the peritoneum in ovarian cancer. This process is facilitated by cross talk between tumor cells and the surrounding cellular stroma. Thereby, cellular microparticles (MPs) (< 1 μm) have been used as sensitive markers of diverse process, such as inflammation, coagulation, vascular function and apoptosis, especially in patients with cancer. They are membrane vesicles released from cells into the blood stream in response to a variety of stimuli, i.e. cell activation, stress, serine proteases, pro-inflammatory cytokines and apoptosis. Futhermore, MPs are important carriers of membrane components or bioactive molecules, and therefore they may contribute to tumour development8,9. Article 2 71 Accordingly, malignant cells and stromal cells communicate and exchange information by direct cell-to-cell contacts as well as release of signalling molecules, such as soluble factors10,11. Thereby, there is an association between MPs, cytokine and chemokine activity during inflammatory responses. Cytokines are small soluble polypeptides released from cells to regulate the activities of other cells via interactions with specific cytokine receptors expressed on their surface12. In OC, there is an increased expression of cytokines/chemokines and/or receptors13. They are able to enhance key aspects associated with tumor progression, such as proliferation, migration and survival, including chemo-resistance14. This can be due to direct effects on the cancer cells themselves, or via indirect effects on components of the immune system. The purpose of the study was to evaluate the levels of circulating soluble biomarkers-microparticles, cytokines and chemokines- to characterize the pro- inflammatory/modulatory immune response in women with OC. The correlation between the biomarker levels and the clinico-pathological parameters were also analyzed. Article 2 72 2. Patients, Materials and methods 2.1. Patients and specimens Patients who were undergoing evaluation for suspected primary ovarian cancer from June 2010 to October 2013 were recruited in a Brazilian clinical center. All patients underwent debulking surgery and received standard platinum-based chemotherapy when indicated. The control group was selected as with no underlying malignancy or infection requiring benign gynecological surgery. Their medical records were obtained until October 2013. The Institutional Review Board approved the study protocol, and all of the patients provided informed signed consent. Blood sampling was carried out intra- operatively, before the anesthetic induction. Flow cytometry was the approach to analyze MPs, cytokines/chemokines 15. Histological grading and disease staging were based on the International Federation of Gynecology and Obstetrics (FIGO) classification16,17. In this study, FIGO stage I/II and FIGO stage III/IV ovarian cancers were considered early and advanced disease, respectively. Other clinical information including age, surgical findings, pathological and immunohistochemical characteristics (i.e type I or II tumors), CA125- levels, and survival was obtained from the medical records. The maximum diameter of the residual tumor after surgery was also retrieved. Optimal debulking surgery was defined as the maximum diameter of residual tumor ≤1 cm. Otherwise, the debulking surgery was considered sub-optimal. Article 2 73 2.2. Purification of MPs from plasma MPs were prepared as described elsewhere 18,19,20. Briefly, citrated blood (0.5 mL) was centrifuged at 1,500 × g for 15 min, and plasma was then cooled to -20°C before storage at -80°C. Samples were further centrifuged at 13,000 x g for 3 min to obtain platelet-free plasma. The latter was diluted 1:3 in citrated PBS containing heparin, and centrifuged at 14,000 x g for 90 min at 15°C. The resultant MP pellet was then resuspended in 1× annexin V binding buffer (BD Biosciences, California, US). 2.3. Flow cytometric assays 2.3.1. Detection of plasma MPs Unless otherwise stated, all reagents and mAbs used in the flow cytometric experiments were from BD Biosciences. MPs isolated from plasma were gated (R1) based on their forward (FSC) and side (SSC) scatter distribution in density plots as compared to the distribution of synthetic 0.7 – 0.9 μm SPHERO™ Amino Fluorescent Particles (Spherotech Inc. Libertyville, Illinois, US). Taking into account the presence of phosphatidylserine (PS) residues in MPs surface, events present in R1 were accessed for their positive staining for annexin V (BD Bioscience) – a classical marker for microparticles – using PE-conjugated monoclonal antibodies (mAbs). Mouse IgG PE conjugated isotype control mAbs were used to properly define the gates. Annexin V+ events gated on R2 region were further assessed for immunolabeling with FITC- http://www.malariajournal.com/sfx_links?ui=1475-2875-9-327&bibl=B18 Article 2 74 conjugated mAbs against the cell markers CD66 (neutrophils), CD41a (platelets), CD51 (endothelial cells), CD235a (erythrocytes), CD45 (leukocytes), CD3 (lymphocytes), and CD14 (monocytes) or the correspondent mouse IgG FITC-conjugated isotype control mABs. The samples were analyzed in a Flow Cytometry FACSCalibur (Becton-Dickinson, California, US). Over 100,000 events were acquired on each sample, to reach at least 2,000 events within the MPs gate. 2.3.2. Detection of plasmatic cytokine/chemokine levels by cytometric bead array immunoassay (CBA) The secreted cytokine/chemokine analysis by flow cytometry is a methodology in which the simultaneous measurement of multiple biomarkers in a single sample is performed21,22,23. For plasma biomarkers determination, whole blood samples were collected using ethylenediamine tetraacetic acid (EDTA) as the anti-coagulant. Plasma was maintained at – 80°C in aliquots thawed just before use. The CBA immunoassay kit (Becton Dickinson Biosciences Pharmingen, San Diego, CA, USA) was used for quantitative analysis of plasma biomarkers levels. The CBA kit uses 7·5 μm polystyrene microbeads, consisting of distinct populations, unique on their type 3 fluorescence intensity (FL-3) each coupled to monoclonal antibody (MoAb) against one of the biomarkers IL-1, IL-2, IL-6 IL-10, IL-12, IL-17a, TNF, IFN- gamma, CXCL-8, CXCL-9, CXCL-10, CCL-2, CCL-5 to capture a given cytokine/chemokine. The captured cytokines/chemokines were then detected Article 2 75 via direct immunoassay using a cocktail of different MoAbs coupled to phycoerythrin (PE) (FL-2). Briefly, 25 μl of plasmas or standards (previously diluted in diluent G, as recommended by the manufacturer) were added to 15μl of a bead cocktail, and incubated for 90 min at room temperature in the dark. The biomarkers standard calibrator mixture was used for each assay. After incubation, the samples and standards were washed with 500 μl of wash buffer (supplied with CBA kit) and centrifuged at 600 g for 7 min at room temperature. Subsequently, 20 μl of detection cocktail consisting of six PE-conjugated MoAbs were added to each tube, and the mixture re-incubated for 90 min at room temperature in the dark. Following incubation, the samples and standards were washed again with 500 μl of wash buffer and centrifuged at 600 g for 7 min at room temperature to remove unbound detector reagent. After washing, 250 μl of wash buffer was added to each tube prior to data acquisition using a fluorescence activated cell sorter (FACS)Calibur® flow cytometer (Becton Dickinson). Although the fluorescently labelled particles in the BD CBA immunoassay are designed to be excited by the 488 nm laser common to all BD flow cytometers, they can also be excited by the red diode laser on dual-laser BD FACSCalibur instruments. The detection of particle emission on the FL-4 channel simplifies the instrument set-up procedure and reduces the need for fluorescence compensation. Thus, a total of 1,800 events/gate were acquired after proper set-up of a flow cytometer to measure forward (FSC) and side (SSC) light scatters, and dual-colour (FL-4 and FL-2) flow cytometric acquisition, using a dual-laser BD CBA template. Data analysis was performed using BD Bioscience CBA software. The results were expressed as pg/mL. Article 2 76 2.4. Statistical Analysis Statistical analyses were performed using SPSS software version 13.0 (SPSS Inc., Chicago, IL). Shapiro–Wilk tests were used to verify if the variables were normally distributed. Non-parametric data were compared using the Kruskal–Wallis test followed by the Dunns post-hoc test. Comparison between 2 groups was done using the Mann–Whitney U test with Bonferroni's correction (non-parametric data) or t-test (parametric data). Parametric data are presented as mean and standard deviation, while non-parametric variables are presented as median and interquartile range (25th–75th percentiles). Correlations were analyzed using the Pearson or Spearman 2-sided test, and Pearson χ2 test to frequency differences. Differences were considered significant when P<0.05. 2.4.1 Network Analyses Cytokine/chemokine/MPs networks were assembled to assess the association between the cytokine,chemokine and MPs each clinical group. Spearman’s correlation test was performed to assess the association between biomarker levels (pg/mL), and frequency of MPs (%). The positive and negative correlations were significant when the p<0.05. The correlation index (r) were used to categorize the correlation strength as negative (r<0), moderate (0.36>r<0.67) and strong (r>0.68). The Graphpad Prism 5.00 software (San Diego, USA) was used for correlation analyses, and the Cytoscape (version 2.8) used for composing networks of biomolecules interactions. Article 2 77 3. Results Characteristics of the study population are summarized in Table 1. There were no differences between the groups regarding age and menopausal status (p= 0,824 or p= 0.055, respectively). In the group of patients with OC, 10 (38.5%) had stage I/II and 16 (61.5%) had stage III/IV ovarian cancer. Among these patients, 8 were type I (30.8%) and 18 were type II (69.2%) tumors. Optimal cytoreduction occurred in 15 (57.7%) patients with ovarian cancer. All women with type I tumors had optimal cytoreduction. However, just 7 (38.9%) patients with type II had residual disease ≤ 1 cm in maximal diameter. CA125 levels were significantly (p=0.024) different between the type I and II tumors. Death was not observed in patients with type I, whereas in type II, 6 cases (33.3%) were registered. Article 2 78 A brief review of tumor microenvironment is that tumor has pro-inflammatory activity via pro-inflammatory cytokines. Moreover, the tumor microenvironment is regulated by modulatory cytokines. Both molecules lead to production of chemokines that attract more immune cells to the tumoral microenvironment. Therefore, higher levels of all pro-inflammatory (IL-1β, IL-6, TNF-α, IL-12p70, IFN-γ), modulatory cytokines (IL-2, IL-10, IL-17a), and chemokines (CCL-2, CCL-5, CXCL-8, CXCL-9 e CXCL-10) were found in patients with OC compared to control group (CN) (Figure 1). Accordingly, the association of ovarian cancer and pro-inflammatory activity was established. Article 2 79 Figure 1. Levels of pro-inflammatory (IL-1β, IL-6, TNF-α, IL-12p70, IFN-γ), regulatory cytokines (IL-2, IL-10, IL-17a), and chemokines (CCL-2, CCL5, CXCL8, CXCL9 E CXCL10) in ovarian cancer (OC) and control (CN) groups. The results are presented in a column chart format and are expressed as the median and interquartile range at pg/mL. Statistical differences between OC and CN were considered significant when p < 0.05. There were no differences in the plasma levels of cytocines and chemokines between type I and type II tumors (Figure 2A). However, when we analyzed the association between these biomarkers with the type of tumor and optimal cytoreductive surgery, we observed that the percentage of pro- inflammatory cytokines IL-1-β and IL-12p70 were statistically higher in patients with type II tumors who had residual disease after surgery <1cm compared to those type II tumors with great volume of residual disease after primary operation (Figure 2B). Article 2 80 A B Figure 2A. Levels of cytokines and chemokines in ovarian cancer patients accordint to the presence of type I or type II tumors. 2.B Levels of cytokines and chemokines in ovarian cancer patients accordint to the presence of type I or type II tumors versus optimal cytoreductive surgery . The results are presented in a column chart format and are expressed as the median and interquartile range at pg/mL. Statistical differences were considered significant when p < 0.05. Article 2 81 Figure 3: Levels of serum biomarkers versus optimal cytoreductive surgery and CA125 levels. . The results are presented in a column chart format and are expressed as the median and interquartile range at pg/mL. Statistical differences were considered significant when p < 0.05. Stage disease were also analysed and we noted the association between higher levels of pro-inflammatory cytokines IL-1-β and IL-12p70 in patients with optimal cytoreductive surgery in advanced disease of OC compared with those at the same stage with residual tumor >1cm after operation (Figure 4). Other important finding was the higher concentration of chemokine CXCL-8 in patients with advanced disease who were optimally debulked compared with those with the same stage, but not well-debulked, and those with early OC (Figure 4). Article 2 82 Figure 4: Levels of serum biomarkers versus optimal cytoreductive surgery and stage. The results are presented in a column chart format and are expressed as the median and interquartile range at pg/mL. Statistical differences were considered significant when p < 0.05. In our recent paper, the phenotype of plasma circulating MPs was investigated, and specific mAbs for cell markers i.e. CD41, CD235, CD45, CD3, CD51, CD14 and CD66 were used to discriminate the cellular sources of annexin V+ MPs. The percentages of annexin V+ MPs stained for each cell marker was compared between samples from OC patients and control group. Other clinic-pathological data were also analyzed. Article 2 83 We observed that the frequency of leukocyte (LMPs) (p < 0.005) derived- microparticles were significantly increased in plasma samples from OC patients as compared with CN, and in patients with type II tumors as compared with type I tumors. Other association was the higher levels of neutrophil derived MPs (NMPs) in patients presenting optimal surgery and increased levels of CA125 as well (>600 U/mL) (See other findings in the first paper). To evaluate the supposed relationships among cytokines, chemokines and MPs in OC and CN group, using all data obtained in this study, we built biological networks, where the nodes represent the cytokines, chemokines and MPs evaluated, and the edges represent positive or negative correlations. We observed that there was a balance among cytokines, chemokines and MPs in the CN group. There were two networks to OC, the first one was composed of twelve cytokines/chemokines, and the other was composed of seven MPs. We found that the first network to OC was characterized by strong and moderate correlations between inflammatory/regulatory cytokines and chemokines, as we found mainly strong correlations in the other network among MPs. Noteworthy, there is a pro inflammatory microenvironment and a route of communication, with exchange of signaling factors, which altogether may support tumour growth and progression (Figure 5). Article 2 84 Figure 5. Biomarkers networks in CN group (A) and OC (B). Chemokine, cytokine and MPs nodes were assembled as well as the biomarkers correlation indexes amongst groups (negative ; moderate and strong positive correlation). Article 2 85 Thereby, we decided to create a cellular interaction network according to type I and II tumors. We found many moderate and strong correlations among cytokines, chemokines and MPs, and just one negative interconnection in the networks. Interestingly, we also observed a greater number of correlations in type II tumors as compared to type I. This result may reflect the differences between carcinogenesis in both tumors (Figure 6). Figure 6. Biomarkers networks in type I (A) and type II (B) tumors. Chemokine, cytokine and MPs nodes were assembled as well as the biomarkers correlation indexes amongst groups (negative ; moderate and strong positive correlation). Article 2 86 4. Discussion Studies on the correlation of clinical outcome for ovarian cancer with cytokines/ chemokines and MPs included together are rare. Hence, we performed the present study in two steps. In the first stage, neutrophils, platelets, endothelial, erythrocyte, leukocyte, lymphocyte and monocytes derived microparticles were analyzed in women with OC and in women without malignancy, and the results were described before. In the second step, several cytokines and chemokines were evaluated in the some patients, and then, they were related to clinical data, and also to a network analysis with MPs. A variety of cytokines, chemokines and MPs are produced in ovarian cancer by different cells accounting for a complex cell interaction and regulation of differentiation, activation, function and survival. The interaction between cytokines, chemokines, MPs, growth factors and their receptors forms a comprehensive network at the tumor site, which is primary responsible for overall tumor progression and spreading or induction of antitumor immune responses and tumor rejection24,25. In the present study, we found higher levels of all pro-inflammatory cytokines (IL-1-β, IL-6, TNF-α, IL-12p70, IFN-γ), regulatory cytokines (IL-2, IL- 10, IL-17a), and chemokines (CCL-2, CCL-5, CXCL-8, CXCL-9 e CXCL-10) in patients with OC compared to CN. These results suggest that there is a pro- inflammatory/regulatory environment for disease development. However, we did not observe any difference in these expressions between type I and type II tumors, and this could be attributed to the small number of patients. Article 2 87 When we stratified the patients in groups, we noted that the levels of pro- inflammatory cytokines IL-1-β and IL-12p70 were statistically higher in patients with type II tumors who were optimally debulked as compared to those with great volume of residual disease after surgery. These cytokines were equally increased in those patients with CA125 upper than 600U/mL and with optimal cytoreductive operation. Besides that, they were even higher in those women with advanced disease and optimal surgical treatment as compared to those not optimally debulked. IL-1-β is mainly produced by monocytes and macrophages in response to innate immunity stimulation, as happens in malignant pathologies. OC cells also produce IL-1 β, although activated immune cells remain the major source of this cytokine. The production of IL- 1β by ovarian cancer cells can enhance their invasion ability and stimulate the release of other cytokines, chemokines and proangiogenic factors such as vascular endothelial growth. Consequently, in OC, IL-1-β usually promotes tumor angiogenesis, invasiveness, and induces inflammatory mechanisms 26,27. Accordingly, it was expected to find higher levels of IL-1-β in women with aggressive tumor and advanced stage OC, especially in type II tumors. Despite this, complete citoreduction was possible in these patients. This it was may feasible due to the great expression of IL-12 in these groups. IL-12 has potent effects on innate and adaptive immunity. It can induce IFN-γ secretion by T cells and NK cells. IL-12 can augment the cytolytic activity of lymphocytes from patients with cancer. Furthermore, IL-12 increases the lysis of autologous tumor cells by tumor infiltrating lymphocytes from patients with ovarian cancer. In addition, it also enhances the antitumor activity Article 2 88 of NK cells 28,29. Some researchers have investigated the precise mechanisms involved in the antitumor activity of IL-12. The results of preclinical studies suggest several potential strategies for the use of IL-12 in cancer therapy 30. The group of patients categorized by stage and cytoreductive surgery had more expression of chemokine CXCL-8 as compared with those with advanced stage OC with optimally operation. In a general analysis, tumor cells can stimulate the innate immune cells activity, which is primary responsible for production of some cytokines as IL-1, IL-6 e TNF. These cytokines play an important role in immune responses, and they are able to stimulate the expression of chemokines, such as CXCL-8, CCL-2, and VEGF. These proinflammatory and proangiogenic factors synergistically interact with each other to create a tumor microenvironment. In addition, they promote a paracrine synergy with cancer cells to support malignant tumor progression. CXCL-8 is an important inflammatory mediator, and it can even stimulate the expression of IL- 12 31,32. An interesting matter was the supposed global relationships among cytokines, chemokines and MPs presented in clusters. We could observe strong and moderate correlations between cytokines and chemokines, with a major attention given to the chemokine CXCL-8 in patients with OC. In addition, it was demonstrated a strong correlation among almost all MPs in the malignancy group. Another noteworthy finding was the substantial differences between networks from type I and type II tumors. This result may reflect the discrepancies between carcinogenesis in both tumors. Our data suggest that Article 2 89 there is an interaction between these soluble factors that are crucial to tumor growth and may validate this network as a key therapeutic target in ovarian cancer. This study has some limitations. The statistics analysis performed was that multiple comparison were realized, aiming to find possible associations among cytokines, chemokines, MPs and clinical/biological parameters; however, it should be taken account in consideration that these significant findings were hypothesis-generating, and, consequently, its represent a realistic interpretation of the facts. Moreover, the limited number of patients in the study is explained by the low prevalence of the disease. The median follow-up could also be higher. The current study is a promising and original research. Our significant findings may indicate several implications, mainly in type I and type II ovarian tumors. We showed that women with OC have higher expression of pro- inflammatory/regulatory cytokines and chemokines. IL- 1β, IL-12, CXCL-8 may have a major role in the pathogenesis of this tumor. MPs may serve as a novel source of disease related information and possibly as a cancer biomarker. In addition, we demonstrated the strong interactions in networks among inflammatory and immune soluble factors of women with ovarian malignancy. The full understanding of the immunobiology of cancer immunosurveillance and immunoediting will hopefully stimulate development of more effective immunotherapeutic approaches to diagnostic, treat and prevent this tumor. Article 2 90 Conflict of interest statement None of the authors has any conflicts of interest Acknowledgments This study was supported by Fundação Oswaldo Cruz (FIOCRUZ), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pe