estudo do papel da via de sinalizaÇÃo bdnf/trkb na...
TRANSCRIPT
UNIVERSIDADE ESTADUAL DE CAMPINAS
FACULDADE DE ODONTOLOGIA DE PIRACICABA
JULIANA KERN DE MORAES
ESTUDO DO PAPEL DA VIA DE SINALIZAÇÃO
BDNF/TrkB NA CARCINOGÊNESE BUCAL
STUDY OF ROLE OF BDNF/TrkB SIGNALING
PATHWAY IN BUCAL CARCINOGENESIS
Piracicaba
2019
JULIANA KERN DE MORAES
ESTUDO DO PAPEL DA VIA DE SINALIZAÇÃO
BDNF/TrkB NA CARCINOGÊNESE BUCAL
STUDY OF ROLE OF BDNF/TrkB SIGNALING
PATHWAY IN BUCAL CARCINOGENESIS
Tese apresentada à Faculdade de Odontologia de Piracicaba da Universidade Estadual de Campinas como parte dos requisitos exigidos para a obtenção do título de Doutora em Estomatopatologia, na Área de Patologia.
Thesis presented to the Piracicaba Dental
School of the University of Campinas in
partial fulfillment of the requirements for
the degree of Doctor in stomatopathology,
in pathology área.
Orientadora: Profª Drª Manoela Domingues Martins
ESTE EXEMPLAR CORRESPONDE À VERSÃO FINAL DA TESE
DEFENDIDA PELA ALUNA JULIANA KERN DE MORAES, E
ORIENTADA PELA PROFª DRª. MANOELA DOMINGUES MARTINS.
Piracicaba
2019
AGRADECIMENTOS
À Reitoria da Universidade Estadual de Campinas (UNICAMP), na
pessoa do Reitor Marcelo Knobel;
À Pró-reitoria da Pós-graduação, à Profª Drª Nancy Lopes Garcia;
À Pró-reitoria de Pesquisa, na pessoa do Prof. Dr. Munir Salomão Skaf;
À presidente da Pós-graduação Odontologia da Unicamp, Profa. Dra.
Karina Gonzales Silvério Ruiz;
À Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
(CAPES) – Governo Federal;
Ao Coordenador do Programa de Pós-graduação em Estomatopatologia
da FOP-UNICAMP, Prof. Dr. MÁRCIO AJUDARTE LOPES, Coordenador, e Prof.
Dr. ALAN ROGER DOS SANTOS SILVA, ex-Coordenador do Programa;
À equipe de professores de Pós-Graduação do Programa de
Estomatopatologia da Faculdade de Odontologia de Piracicaba:
Profª Drª Adriana Franco Paes Leme
Profª Drª Albina Messias De Almeida Milani Altemani
Prof. Dr.Edgard Graner
Prof. Dr.Helder Antonio Rebêlo Pontes
Prof. Dr.Jacks Jorge Junior
Prof. Dr.Mario Fernando De Goes
Prof. Dr.Oslei Paes De Almeida
Prof. Dr.Pablo Agustin Vargas
Prof. Dr.Ricardo Della Coletta
Profª Drª Ana Carolina Prado Ribeiro e Silva
Prof. Dr.Fábio Ramôa Pires
Prof. Dr.Felipe Paiva Fonseca
Profª Drª Fernanda Viviane Mariano Brum Corrêa
Prof. Dr.Hercílio Martelli Júnior
Prof. Dr.Luiz Paulo Kowalski
Profª Drª Lynne Bingle
Profª Drª Maria Elvira Pizzigatti Corrêa
À equipe da Faculdade de Odontologia da Universidade Federal do Rio
Grande do Sul (UFRGS), principalmente ao Prof. Dr. Pantelis Varvaki Rhados;
À equipe do Laboratório de Neurobiologia e Câncer e Laboratório de
Patologia do Hospital de Clínicas de Porto Alegre-RS; principalmente aos caros:
Prof. Dr. Rafael Roesler;
Profª Drª Caroline Brunetto de Farias;
Drª Emily Ferreira Salles Pilar;
Flávia Rejane Giusti;
Drª Mariane da Cunha Jaeger;
À equipe do Instituto do Câncer Infanto Juvenil (ICI) localizado em Porto
Alegre-RS.
Aos colegas de pós-graduação, principalmente Drª Vivian Petersen
Wagner e Gleyson Kleber do Amaral-Silva.
DEDICATÓRIA À PROFª DRª MANOELA DOMINGUES MARTINS
À minha querida orientadora...
Diz em um texto de autoria desconhecida que, há quatro leis da
espiritualidade, e destas frases gosto de lembrar em vários passos da minha
existência.
“A primeira diz: “A pessoa que vem é a pessoa certa”.
Ninguém entra em nossas vidas por acaso. Todas as pessoas ao nosso redor,
interagindo com a gente, têm algo para nos fazer aprender e avançar em cada
situação.
A segunda lei diz: “Aconteceu a única coisa que poderia ter acontecido”.
Nada, absolutamente nada do que acontece em nossas vidas poderia ter sido de
outra forma. Mesmo o menor detalhe.
A terceira diz: “Toda vez que você iniciar é o momento certo”.
Tudo começa na hora certa, nem antes nem depois. Quando estamos prontos para
iniciar algo novo em nossas vidas, é que as coisas acontecem.
E a quarta e última afirma: “Quando algo termina, termina”.
Com estas palavras, por mim consideradas sábias, deixo aqui meu imenso
agradecimento à professora Manoela Martins por ter sido a pessoa certa a entrar na
minha vida no momento certo. Não poderia ter sido diferente.
Muito aprendi com esta força de movimento em forma de mulher, que faz girar
conhecimento, batalhas contínuas em sua vida, fazendo vitórias com suas próprias
mãos e guardando o cansaço para traz todos os dias para seguir a sua busca.
Esta mulher que abraça todas as oportunidades, cria oportunidades para seus
pupilos e que faz com que tudo seja adicionado de amizade e companheirismo.
Esta mesma mulher que cuida de sua família com coragem e não se entrega à
nenhuma adversidade. Não mostra fraqueza em nenhuma situação e sempre mostra
que ser mulher não é empecilho para nada colocando-se em posições jamais
almejadas por muitas.
Só tenho a agradecer o exemplo de mulher que foste para mim neste período, o
exemplo de mestre e de ser humano.
Muito obrigada, à minha orientadora de vida!
RESUMO
Os carcinomas de células escamosas orais (CCEO) constituem um dos seis tumores
malignos mais comuns no mundo. A quimioterapia utilizada como terapia
neoadjuvante e adjuvante à radioterapia ou cirurgia tem mostrado benefícios
modestos na sobrevida relacionada ao tratamento de CCEO em estágio avançado
quando comparada à sua utilização no tratamento de outras formas de câncer. O
Fator Neurotrófico Derivado do Cérebro (BDNF) é um fator de crescimento membro
da família de neurotrofinas que foi primeiramente conhecido como responsável por
sustentar o crescimento, a função e a plasticidade das células neurais. O BDNF
exerce os seus efeitos ligando-se ao receptor de tirosina quinase B (TrkB). Foi
relatado que o eixo BDNF / TrkB está superexpresso e associado a um mau
prognóstico, crescimento tumoral, invasão e metástase; transição epitelial-
mesenquimal (EMT) e resistência à quimioterapia em diferentes malignidades
neurais e não neurais humanas. Os objetivos do presente estudo são revisar a
literatura sobre o papel da via BDNF / TrkB em CCEO e tumores glandulares e além
disto, avaliar a expressão de BDNF, seu receptor (TrkB) e dois alvos dessa via, Akt
e proteína ribossômica S6 (RPS6), em mucosa oral normal (MON), leucoplasia oral
(LO) e carcinoma de células escamosas oral (CCEO), e correlacionar essa
expressão com os desfechos dos pacientes com CCEO, senescência celular e perfil
tronco. Dez casos de MON, 33 LO e 72 CCEO primários foram incluídos. Os casos
de CCEO foram montados em microarranjos teciduais (TMAs). Análises imuno-
histoquímicas para BDNF, TrkB, p-TrkB, p-Akt e p-RPS6 foram realizadas em todos
os casos. A senescência celular e o perfil de células tronco do CCEO foram
avaliados através da expressão imuno-histoquímica do p16 e BMI-1,
respectivamente. As lâminas foram digitalizadas em imagens de alta resolução e a
expressão imuno-histoquímica foi quantificada por meio de análise digital. Os valores
de corte foram determinados para avaliar a associação de cada marcador com
características clínico-patológicas. CCEO apresentou expressão aumentada da via
BDNF / TrkB / Akt em relação ao MON e LO. O CCEO diagnosticado em estágios
clínicos avançados apresentou uma regulação positiva do BDNF e de p-TrkB. Além
disso, a superexpressão de BDNF e p-Akt foi identificada como preditora de
sobrevida pobre, com razões de risco de 2,83 e 2,19, respectivamente. O aumento
no perfil de células tronco foi correlacionado com uma diminuição na expressão de p-
TrkB e p-Akt. Em conjunto, esses dados sugerem que a via BDNF / TrkB / Akt está
envolvida na aquisição do fenótipo maligno durante a carcinogênese oral e influência
o prognóstico do CCEO. Pacientes com aumento da expressão dessas proteínas
foram mais propensos a experimentar menor tempo de sobrevida. Este estudo
confirma que o BDNF / TrkB é um alvo terapêutico atraente para o carcinoma
espinocelular oral e em relação aos tumores de glândulas salivares, estudos
demonstraram que a via BDNF / TrkB está significativamente associada à invasão
perineural, migração, indução de transição epitelial-mesenquimal e metástase.
Sendo assim um possível alvo de interesse.
Palavras-chave: câncer bucal, BDNF,TrkB, Akt
ABSTRACT
Oral Squamous Cell Carcinomas (OSCC) are one of the six most common
malignancies in the world. Chemotherapy used as neoadjuvant therapy and
adjuvant to radiotherapy or surgery has shown modest survival benefits related to
the treatment of advanced HNSCC when compared to its use in the treatment of
other forms of cancer. Brain-derived neurotrophic factor (BDNF) is a growth factor
member of the neurotrophin family that was first known to be responsible for
sustaining the growth, function, and plasticity of neural cells. BDNF exerts its effects
by binding to the receptor of tyrosine kinase B (TrkB). It has been reported that the
BDNF / TrkB axis is overexpressed and associated with poor prognosis, tumor
growth, invasion and metastasis; epithelial-mesenchymal transition (EMT) and
resistance to chemotherapy in different human neural and non-neural malignancies.
The aim of the present study was to review the literature on the role of the BDNF /
TrkB signaling pathway in OSCC and glandular tumors and in addition to evaluate
the expression of BDNF, its receptor (TrkB) and two targets of this pathway, Akt
and ribosomal protein S6 (RPS6 ), in normal oral mucosa (NOM), oral leukoplakia
(OL) and oral squamous cell carcinoma (OSCC), and to correlate this expression
with the clinical outcomes of the patients with OSCC, cell senescence and stem cell
profile. Ten cases of NOM, 24 OL and 72 primary OSCC were included. The OSCC
cases were assembled in tissue microarrays (TMAs). Immunohistochemical
analyses for BDNF, TrkB, p-TrkB, p-Akt and p-RPS6 were performed in all cases.
Cell senescence and OSCC stem cell profile were assessed by
immunohistochemical expression of p16 and BMI-1, respectively. The slides were
scanned in high resolution images and the immunohistochemical expression was
quantified by means of digital analysis. Cut-off values were determined to evaluate
the association of each marker with clinical-pathological characteristics. OSCC
showed increased expression of the BDNF / TrkB / Akt pathway relative to NOM
and OL. The OSCC diagnosed in advanced clinical stages presented a positive
regulation of BDNF and p-TrkB. In addition, overexpression of BDNF and p-Akt was
identified as a predictor of poor survival, with risk ratios of 2.83 and 2.19,
respectively. The increase in the stem cell profile was correlated to a decrease in
the expression of p-TrkB and p-Akt. Together, these data suggest that the BDNF /
TrkB / Akt signaling pathway is involved in the acquisition of the malignant
phenotype during oral carcinogenesis and influences the prognosis of OSCC.
Patients with increased expression of these proteins were more likely to experience
shorter survival times. This study confirms that BDNF / TrkB is an attractive
therapeutic target for oral squamous cell carcinomas and concerning salivary gland
tumors, studies have demonstrated that the BDNF / TrkB pathway is significantly
associated with perineural invasion, migration, induction of epithelial-mesenchymal
transition, and metastasis. Thus being a possible target of interest.
Keywords: oral squamous cell carcinoma, BDNF, TrkB, Akt
SUMÁRIO
1 INTRODUÇÃO 14
2 ARTIGOS
22
2.1 Uncovering the role of brain-derived neurotrophic factor/tyrosine kinase receptor B signaling in head and neck malignancies
22
2.2 Activation of BDNF/TrkB/Akt axis may be involved in the acquisition of the malignant phenotype in potentially malignant disorders and predicts poor prognosis of oral squamous cell carcinoma
38
3 CONCLUSÃO 70
REFERÊNCIAS 71
ANEXOS 76
ANEXO 1 APROVAÇÃO COMITÊ DE ÉTICA EM PESQUISA 76
ANEXO 2 RELATÓRIO ORIGINALIDADE 82
14
1 INTRODUÇÃO
O câncer de cabeça e pescoço pode desenvolver-se em diversas
regiões anatômicas como faringe, hipofaringe, laringe, tonsilas palatinas, cavidades
nasais e seios paranasais, glândulas como as parótidas e cavidade oral (Marur et
al. 2016). O câncer oral (CO) representa um importante problema de saúde com
cerca de 354.860 casos novos estimados no mundo em 2018 (Globocan, 2018).
Segundo a Agência Internacional de Pesquisa em Câncer (IARC-França), o CO
abrange as regiões dos lábios, língua (porção móvel), gengivas, mucosa jugal,
palato, soalho bucal, algumas glândulas salivares. A incidência é maior em homens
do que em mulheres e afeta na maior parte dos casos indivíduos com faixa etária
entre cinquenta e sessenta anos de idade (Le Campion et al. 2016; Siegel et al.
2017). Índia, seguida pela China, Estados Unidos da América, Paquistão e
Bangladesh mostram a maior incidência de CO no mundo. O Brasil se encontra na
sexta posição. Tratando-se de continentes, a Ásia abriga as maiores incidências de
CO em ambos os sexos, seguida pela Europa e América (América Latina, Caribe e
América do Norte) (IARC). No Brasil, o Instituto Nacional de Câncer José Alencar
Gomes da Silva (INCA) estimou para o ano de 2018 (biênio 2018–2019), 11.200
novos casos de CO em homens e 3.500 em mulheres. Desta forma, o CO ocupa a
5º posição entre todos os tipos de cânceres em homens e 12º em mulheres. Novos
casos prospectivos, para ambos os sexos e todas as idades, aumentarão de
363.626 em 2020 para 450.870 em 2030 (Globocan, 2018).
Aproximadamente 90%–95% dos cânceres orais são histologicamente
carcinomas de células escamosas (CEC) e acometem principalmente lábio inferior,
língua e soalho bucal. Diversos fatores têm sido descritos como de risco para o
desenvolvimento de carcinomas de células escamosas oral (CCEO) e parecem agir
conjuntamente principalmente aumentando a taxa de mutações nas células.
Fatores exógenos relacionados ao estilo de vida, especialmente o consumo de
tabaco e a sinergia tabaco e álcool, parecem ser particularmente importantes,
assim como, em alguns casos, a exposição à luz solar para lábio e o hábito de
mascar fumo como betel têm sido descritos como fatores associados em algumas
áreas do globo (D'Souza e Addepalli 2018; Cohen et al. 2018; Andisheh-Tadbir et
15
al. 2010). Fatores intrínsecos como a desnutrição, condição sistêmica, idade,
gênero, fator hereditário e genes oncogênicos são relatados (GALBIATTI et al.
2013).
O tratamento do carcinoma oral e de cabeça e pescoço envolve cirurgia,
radiação e quimioterapia em diversas combinações dependendo do estágio e sítio
primário. Aproximadamente 40% dos pacientes em estágios iniciais (I e II) são
tratados apenas com cirurgia ou radiação. Para a maioria dos pacientes com
estágios localmente avançados (estágio III e IV), o tratamento ressecável ou
irressecável envolve quimioradioterapia, com ou sem indução e quimioterapia como
terapia sequencial. A metástase é tratada com combinação quimioterápica ou
agente único. O tratamento para recorrências locais/regionais depende do local da
recorrência, carga tumoral e antecedentes terapêuticos e pode variar de
reintervenção cirúrgica/radiação ou nova irradiação com quimioterapia ou somente
quimioterapia. Todos estes tratamentos, de alguma forma causam algum tipo de
morbidade imediata ou tardia (Marur et al. 2016).
A cirurgia é utilizada apenas para tumores ressecáveis em que as
margens tumorais possam ser bem definidas e a função possa ser preservada. A
radioterapia é utilizada no tratamento de tumores localmente avançados,
empregada como adjuvante da cirurgia ou concomitante à quimioterapia. As doses
de radiação variam de 60 Gy a 70 Gy. Apesar da atual utilização de radioterapia
com modulação de intensidade, as glândulas salivares, músculos constritores da
faringe, e glândula tireoide são muitas vezes prejudicados, levando à xerostomia,
disfagia, aspiração crônica e hipotireoidismo. A quimioterapia é utilizada como
neoadjuvante e adjuvante da cirurgia e radioterapia. A alta dose de cisplatina
continua sendo o padrão radiossensibilizador no tratamento do carcinoma oral,
porém pequenos benefícios na sobrevida tem sido observados com a utilização
também de fluorouracil e composto de docetaxel, assim como medicamentos
alternativos para radiossensibilização dos anticorpos monoclonais de EGFR como
cetuximab e panitumumab têm sido estudados (Marur et al. 2016; Vidal et al. 2017).
Embora a detecção e o tratamento da maioria das neoplasias malignas
tenham melhorado nas últimas décadas, a alta taxa de mortalidade do carcinoma
16
de células escamosas oral (CCEO) permaneceu inalterada (Warnakulasuriya,
2009). Como resultado, 145.000 mortes atribuídas ao CCEO ocorrem por ano
(Globocan, 2018). Esses dados demonstram que as estratégias de prevenção não
são eficazes; o diagnóstico ainda é tardio e os tratamentos atuais têm efetividade
limitada (Winn et al. 2015; Jansen et al. 2015). Cerca de apenas 30% a 50% dos
pacientes portadores de CCEO estarão vivos após três anos, indicando uma taxa
de sobrevida global baixa, consideravelmente menor do que outros tipos de câncer
tais como colorretal, colo de útero e mama (Haddad, Shin 2008; Curado, Hashibe
2009). Usualmente os pacientes com CCEO experienciam recorrência loco-regional
ou metástases à distância (Scully, Bagan 2008; Curado, Hashibe 2009).
O desenvolvimento do CCEO pode ser precedido por alterações no
epitélio oral denominadas de desordens potencialmente malignas (DPM). Dentre
elas, as leucoplasias, são as mais frequentemente observadas na prática clínica
(Cabarcos et al, 2018). Em 1978, a Organização Mundial de Saúde definiu a
leucoplasia como uma “placa branca que não pode ser caracterizada clinicamente
ou histopatologicamente como qualquer outra condição” (Waal, 2014). Atualmente,
é definida como uma placa predominantemente branca de risco questionável,
excluindo outras doenças ou distúrbios conhecidos que não apresentam risco
aumentado de câncer (Warnakulasuriya et al. 2007). Microscopicamente as
leucoplasias podem apresentar alterações não displásicas (hiperceratose,
acantose) ou diferentes graus de displasia epitelial (Villa and Sonis, 2018). O risco
de malignização de uma leucoplasia é variável na literatura, estudos mostram que
a taxa de transformação maligna para câncer invasivo de displasia ou carcinoma in
situ pode variar de 5% a 36%. Hiperceratoses podem transformar-se em carcinoma
invasivo em até 11% a 30% dos casos. Existem alguns parâmetros relatados que
supostamente predizem a transformação. Esses parâmetros incluem câncer
previamente diagnosticado na região da cabeça e pescoço, idade avançada,
homens tabagistas, duração da leucoplasia, tipo clínico e tamanho da leucoplasia.
As leucoplasias homogêneas têm menor risco de transformação em comparação
com os casos não homogêneos assim como, leucoplasias maiores oferecem maior
risco. Também podem predizer malignização, leucoplasias presentes em mulheres
não tabagistas. A região anatômica também pode predizer a malignização sendo
17
que as lesões situadas em bordo de língua, ventre de língua, soalho de boca e
palato mole tem maior probabilidade de apresentarem displasia. A prevalência de
leucoplasia para todas as idades pode ser de 1%, mas é maior em adultos idosos.
A proporção homem-mulher varia em diferentes partes do mundo e fumar é o fator
etiológico mais comumente associado ao seu desenvolvimento (Waal, 2014; Villa
and Sonis, 2018). Uma classificação de três níveis de displasia é tradicionalmente
usada por patologistas; onde existe um distúrbio arquitetural limitado ao terço
inferior do epitélio com atipia citológica, chamada de discreta. Há também um tipo
que apresenta um distúrbio arquitetural que se estende até o terço médio do
epitélio (moderada) e a displasia intensa apresenta distúrbio arquitetural em mais
de dois terços de profundidade do epitélio com atipia celular (Dionne, et al 2015). A
evolução das leucoplasias para o CCEO ainda não é bem compreendida e foram
feitos poucos progressos no entendimento deste processo nas últimas décadas
(Villa and Sonis, 2018).
Um passo apontado como fundamental é identificar os fatores
moleculares que impulsionam o início e a progressão do CCEO, pois esses fatores
podem direcionar novas terapias (Villa and Sonis, 2018). Vários genes e proteínas
emergem como potenciais marcadores de displasia e transformação maligna de
leucoplasias. O gene p16 é apontado como um gene supressor tumoral do ciclo
celular normal, e é associado com a imortalidade celular em câncer. Além disso,
altos níveis de expressão suprabasal de p53 estão correlacionados com o risco de
transformação. No entanto, a presença de p53 sozinha não pode ser considerada
um determinante específico da transformação maligna e um relato recente sugere
que a superexpressão do p53, no contexto da perda de proteína p16 e da
superexpressão da proteína Ki-67, aumenta substancialmente o valor preditivo
positivo. Além disso, o valor negativo preditivo desses marcadores individuais pode
chegar a 100% de especificidade, indicando verdadeira utilidade clínica. A
imortalidade celular associada à displasia oral, está associada à perda da
expressão do receptor do ácido retinóico (RAR) -b e p16INK4a (Dionne et al. 2015).
Atualmente existe uma vasta literatura relatando o papel dos fatores de
crescimento e seus receptores na ativação das vias de sinalização intracelulares
18
envolvidas na regulação da diferenciação, sobrevivência, metabolismo, mobilidade
e crescimento celular nos diferentes tipos de câncer. Isto porque, a desregulação
dos fatores de crescimento, de seus receptores e das diversas vias de sinalização
por eles estimuladas é responsável pela aquisição do fenótipo maligno, da
capacidade de invasão e metástase dos tumores malignos (Cohen et al. 2011;
Alyasiri et al. 2012; Thomaz et al. 2016). Estes estudos são fundamentais para
compreender a patogênese do câncer, identificação de marcadores biológicos de
progressão tumoral e para o desenvolvimento de terapias alvo que atuem na
regulação dessas vias.
Existem estudos com terapia alvo para câncer de boca e de cabeça e
pescoço direcionados para membrana celular, vias de sinalização, fatores de
crescimento e seus receptores como EGFR, HER2 e HER3, MET, entre outros
(Marur et al. 2016).
Neste sentido, a via do Fator Neurotrófico Derivado do Cérebro (BDNF,
do inglês Brain-derived Neurotrophic Factor) e seu receptor Trompomiosina
Quinase B (TrkB) vem sendo bastante estudada em câncer de pâncreas, pulmão,
colón, próstata, mama, bexiga, neuroblastoma, meduloblastoma, mieloma múltiplo,
entre outros (Lai et al. 2010; Roesler et al. 2011; de Farias et al. 2012; Yin et al.
2015; Thomaz et al.2016; Akil, Perraud et al. 2016).
O BDNF é um membro da família das neurotrofinas que são fatores de
crescimento que participam da diferenciação, proliferação e sobrevivência de
células neurais e da neurogênese nos sistemas nervosos central e periférico. Esta
proteína foi identificada e isolada pela primeira vez em 1982 por Barde et al. do
cérebro de porco como um fator de sobrevivência para neurônios sensoriais
embrionários de galinhas em cultura. As neurotrofinas medeiam seus efeitos
interagindo com receptores de superfície celular específicos, que são divididos em
duas classes de acordo com a afinidade de ligação para os fatores de crescimento.
O receptor de NGF de baixa afinidade (p75N~FR) é capaz de ligar todas as
neurotrofinas com afinidade semelhante, mas não serve como um receptor
funcional para a transdução de sinal. Com o passar do tempo, demonstrou-se que
19
a família de receptores de proteína tirosina quinase trk (trkA, trkB e trkC), constitui
o componente essencial dos receptores de alta afinidade(Yamamoto et al.1996).
Desde 1991, foi identificado como um ligante para o receptor de tirosina
quinase B (TrkB) (Barde et al. 1982; Roesler et al. 2011; Levi-Montalcini 1987). O
principal ligante para o BDNF são os receptores de membrana celular
trompomiosina quinase (Trk) (Schecterson, Bothwell, 2010; Akil, Perraud, et al.,
2016), em especial o TrkB. A ligação do receptor de BDNF no TrkB desencadeia a
sua dimerização através de mudanças conformacionais e autofosforilação de
resíduos de tirosina no ambiente intracelular (Liu, Chan, Ye, 2016). Em câncer, a
estimulação da via de sinalização do BDNF/TrkB promove a ativação de moléculas
de sinalização como Akt (Figura 1), STAT3, Src, ERK e MAPK resultando no
aumento da proliferação celular, resistência a apoptose, invasão, metástase e
resistência a quimioterapia. Desta forma, o aumento da expressão de BDNF/TrkB
tem sido associado com comportamento mais agressivo de diferentes tumores
malignos e pior prognóstico, devido ao seu papel nos processos de angiogênese,
invasão, metástase e a resistência à quimioterapia (Kupferman et al. 2010; Guo et
al. 2011; Roesler et al. 2011; Fujikawa et al. 2012; Okamura et al. 2012; Sasahira et
al. 2013; Jia et al. 2015; Kim et al. 2015). Alguns autores também relatam que, o
TrkB é uma proteína transmembrânica e que é codificada pelo proto-oncogene Trk
humano, o qual tem a capacidade para adquirir propriedades oncogênicas. Este
proto-oncogene, através de um rearranjo cromossômico adquire a capacidade de
oncogênese participando nos processos de transformações malignas.
20
Figura 1. Ilustração da via de sinalização BDNF/TrkB/PI3K/Akt/m-TOR.
Fonte: Autores
Em câncer de cabeça e pescoço, apesar de poucos estudos terem sido
realizados, o BDNF/TrkB aparece altamente expresso nas células tumorais (Zhu et
al. 2007; Yilmaz et al. 2010; Jia et al. 2015), além de ter demonstrado relação com
pior prognóstico, aumento da angiogênese e linfangiogênese (Sasahira et al. 2013),
e de participar do processo de transição epitelial mesenquimal (EMT), que é
21
considerado um processo biológico chave na invasão tumoral e metástase de
tumores epiteliais (Jia et al. 2015).
A modulação terapêutica da via BDNF/TrkB tem sido investigada com a
utilização de moléculas de baixo peso antagonistas de TrkB ou com anticorpos
monoclonais com ação no BDNF ou TrkB (Camoratto et al. 1997; Thiele et al. 2009;
Yilmaz et al. 2010; Odate et al. 2013; Tanaka et al. 2014). Ensaio clínico de fase I
utilizando inibidores de Trk já foram realizadas e ensaios de Fase II estão
atualmente em curso (Norris et al. 2011; Minturn et al. 2011). No entanto, a
compreensão do papel do BDNF/TrkB no câncer, em especial no CEC de cabeça e
pescoço, assim como os efeitos do bloqueio farmacológico da sinalização
BDNF/TrkB e sua ação nas diferentes vias de sinalização PI3k-Akt-mTOR precisam
ser elucidadas. Assim como, não há estudos que avaliam o papel desta via durante
todo o processo de carcinogênese bucal e sua relação com a evolução dos casos
necessita ser melhor elucidada. Com o intuito de melhor esclarecer o papel da via
BDNF/TrkB/Akt em câncer de boca, foi realizada uma revisão de literatura e uma
avaliação da imunomarcação em amostras histopatológicas.
22
2 ARTIGOS
2.1 Uncovering the role of brain-derived neurotrophic factor/tyrosine kinase
receptor B signaling in head and neck malignancies.
Juliana Kern de Moraes, Vivian Petersen Wagner, Felipe Paiva Fonseca, Pablo
Agustin Vargas, Caroline Brunetto de Farias, Rafael Roesler, Manoela Domingues
Martins
Artigo publicado: J Oral Pathol Med. 2018 Mar;47(3):221-227. doi:
10.1111/jop.12611.
ABSTRACT
Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family
of growth factors that was first known as responsible for sustain the growth,
function, and plasticity of neural cells. BDNF exerts its effects by binding to
the tyrosine kinase receptor B (TrkB). The BDNF/TrkB axis has been reported
to be overexpressed in several neurogenic and non-neurogenic tumors. Its higher
expression was associated with a poor prognosis to patients affected by different
human malignancies, tumor growth, invasion, and metastasis; epithelial-
mesenchymal transition and resistance to chemotherapy. BDNF/TrkB represent
promising targets to the development of novel anticancer therapies. Some
clinical trials are currently evaluating the efficacy of Trk protein-target drugs in
different types of solid tumors. To date, few groups have evaluated the
BDNF/TrkB pathway in head and neck malignancies. The aims of this study were
to review the literature concerning the role of BDNF/ TrkB activation in head and
neck squamous cell carcinoma and malignant salivary gland tumors and to
discuss future perspectives of BDNF/TrkB-target therapy.
KEYWORDS
adenoid cystic carcinoma, head and neck cancer, oral squamous cell carcinoma,
salivary gland cancer, signaling pathway
23
1 INTRODUCTION
In the past years, the paradigm for cancer therapy evolved from conventional
non-specific treatments to highly selective, mechanism-based drugs.1
Therefore, the identification of deregulated molecular networks in different types of
tumors became a primordial goal in the challenging battle to beat cancer.
Among these, growth factor signaling pathways, neurotrophins and their
receptors represent emerging targets for the development of novel anticancer
therapies.2
Brain-derived neurotrophic factor (BDNF) is a growth factor member of
a family of functionally and structurally related proteins called neurotrophins that
also includes nerve growth factor (NGF), neurotrophin 3 (NT-3), and
neurotrophin 4/5 (NT-4/5). BDNF is expressed mainly involved in the
differentiation of neurons, responsible for sustain the growth and function of
neuronal synapses, proliferation, plasticity, and survival of neural adulthood
cells in the central and peripheral nervous system. This protein was
identified and isolated for the first time in 1982 by Barde et al. from the pig
brain as a survival factor for embryonic sensory neurons of chickens in culture.
Since 1991, it was identified as a ligand for tyrosine receptor kinase B
(TrkB).2-4
Both BDNF and TrkB are expressed by T and B lymphocytes, platelets,
macrophages, vascular endothelial cells, epidermal, and other non-neural
cells.5 In oral cavity, BDNF is necessary for the development of the dental
papilla/pulp - beginning of the dental innervation; modulation of proliferation,
and differentiation of epithelial and mesenchymal cells; formation and
maintenance of the sensory apparatus of the tongue (morphogenesis of the
gustatory epithelium). BDNF-derived gustatory epithelium is required for gustatory
axons to correctly locate and innervate fungiform papillae with BDNF-
mediated target being restricted to a critical period of development.6
Current research has shown that BDNF/TrkB pathway facilitates tumor progression
stimulating invasion, metastasis, and angiogenesis. BDNF/TrkB system has also
24
been reported to be responsible for chemoresistance, through Akt/PI-3K and
MAPK signaling pathways, in selected models of cancer. It has been shown to be
overexpressed in various cancer types like head and neck squamous cell
carcinoma (HNSCC), pancreas, Wilms’ tumors, lung, breast, gastric, hepatocellu-
lar, and also myelomas and lymphoid cancers. In addition, in various types of
tumors the overexpression of these proteins was associated with more aggressive
behavior and poor prognosis.7-22 Our group was the first to identify and
explore the role of BDNF and TrkB in colorectal cancer 23,24 and has been
evaluating this pathway in other cancers such as acute leukemia,25 Ewing
sarcoma,26 medulloblastoma,27,28 and breast and gynecologic cancer.29
Brain-derived neurotrophic factor and its receptor represent promising
targets for the development of novel anticancer therapies.10,27,30 Several series of
Trk inhibitors with excellent in vitro therapeutic potential have been reported and
a number of compounds have gone into the clinic.30 Some undergoing
clinical trials are evaluating the efficacy of Trk protein-target drugs in different
types of solid tumors.31,32 The most favorable results concerning Trk inhibitors
were observed in tumors that harbor NTRK gene fusions.31 Trk inhibitors
Entrectinib and LOXO-101, for example, demonstrated sustained clinical responses
among patients with metastatic or unresectable solid tumors with NTRK mutations.31
To date, few groups have evaluated the BDNF/TrkB pathway in head and
neck malignancies. Therefore, in this study, we aimed to review the literature
concerning the role of BDNF/TrkB activation in these tumors, as well as to
discuss future perspectives of BDNF/ TrkB-target therapy.
2 BDNF / TrkB IN HEAD AND NECK SQUAMOUS CELL CARCINOMA
Head and neck cancer comprises a group of malignancies that
involve the lips, oral cavity, pharynx, and larynx, with squamous cell carcinoma
(HNSCC) representing the most prevalent microscopic subtype. More than half a
million cases of HNSCC have been estimated to occur in 2012 worldwide.33,34
Over the past few decades, significant advances have been achieved in our
understanding of HNSCC, nevertheless, no absolute gain was achieved in the
survival rates,35 and in 2012, approximately 325 000 deaths occurred due to
25
HNSCC.33 Different groups have explored the role of BDNF/TrkB pathway in
HNSCC in the last years. The findings are summarized in Table 1 and Figure 1.
Figure 2 illustrates the BDNF immunostaining in head and neck squamous cell
carcinoma as in the most prevalent salivary gland cancers that will be discussed
later on.
The study by Zhu et al.36 represents the earliest evidence that TrkB is
overexpressed in HNSCC. This study assessed TrkB levels in the total lysates of
23 HNSCC tissue samples and in 21 HNSCC cell lines. All samples (tissue and
cells) presented detectable levels of this protein while the normal oral mucosa was
absolutely negative. No association between BDNF stimulus and HNSCC cells
invasiveness was observed in this pioneer report.36 Kupferman and colleagues
published three important studies that demonstrated the association between
BDNF/TrkB pathway and HNSCC aggressive behavior.15,17,21 Their initial
results demonstrated increased expression of BDNF and TrkB in HNSCC tissue
samples and cell lines,15 corroborating with Zuh et al.36 Nevertheless, the
outcomes concerning cell invasive potential of Kupferman et al.15 diverged from
those initially reported in 2007.36 Cell migration and invasive capacity were
increased following BDNF stimulus.15 This disagreement might be explained by
the fact that Kupferman et al. analyzed the effect of a BDNF concentration
gradient in high TrkB-expressing cell lines. These authors also observed that
BDNF/TrkB activated STAT3, AKT, and MAPK as downstream signaling cascades.
TrkB knockdown inhibited HNSCC cell migration and invasion, but had no
impact on cellular growth in vitro. This finding suggested that this pathway
have an impact in the epithelial-mesenchymal transition (EMT), which was
confirmed by the spindle-like morphology observed in TrkB-overexpressing
cells accompanied by a decrease in N-cadherin and Twist expression. TrkB
downregulation abrogates tumor growth in orthotopically injected tongue tumors
and led to an increase in E- cadherin expression, evidencing the role of this
pathway in tumor growth and EMT in vivo. Based on these promising results,
this group developed two following studies to evaluate the association between
the BDNF/TrkB pathway and HNSCC progression and resistance to cisplatin.17,21
Ylmaz et al. observed that a small molecule inhibitor of the Trk receptor family
26
(AZ64) was capable to sig- nificantly suppress cell proliferation and motility.17
Cisplatin-based chemotherapy is currently the most common adjuvant
treatment offered for patients with HNSCC. Nevertheless, the clinical response is
quite unsatisfactory. Loco-regional failure rates can reach 65% and the 3-year
survival rate 37%.37 Therefore, it is of great importance to identify molecular
pathways involved with HNSCC resistance to cisplatin. Cisplatin-resistant
HNSCC cell lines overexpress BDNF and TrkB, and the pathway activation
significantly impacts the cisplatin-IC50.17,21 Yet, the dose-response curves to
AZ64 were similar between cisplatin-resistant and parental HNSCC cell
lines.17 The mechanism by which TrkB inhibition can impact cisplatin
resistance is rather associated to oncogenic signaling, once AZ64 was able to
inhibit STAT3 and Src.17
TABLE 1 Results from currently available studies that investigated the role of BDNF/TrkB in head and neck squamous cell carcinoma
Head and neck squamous cell carcinoma
Authors (year) Methods Main findings Zhu et al. (2007)36 HNSCC tissue samples TrkB is overexpressed in HNSCC tissues and cell lines compared to normal
and cell lines oral mucosa. BNDF stimulus was not associated with cell invasion in vitro. Kupferman et al. (2010)15 HNSCC tissue samples BDNF and TrkB are expressed in more than 50% of human HNSCC tumors.
and cell lines BDNF/TrkB activation in vitro upregulates cell migration, invasion, and EMT markers. TrkB inhibition suppresses tumor growth of orthotopic HNSCC tumor. Ylmaz et al. (2010)17 HNSCC tissue samples, TrkB was identified in 30% of tumor samples and in orthotopically implanted
cell lines, and orthotopic model HNSCC tumors. TrkB inhibition suppressed cell proliferation and migration and sensitize cells to cisplatin. Dud as et al. (2011)38 OSCC cell line co-cultured OSCC-associated fibroblasts produce BDNF, and TrkB receptor expression
with periodontal ligament fibroblasts was increased in OSCC co-cultured with fibroblasts. This interaction induced EMT in OSCC cells. Lee et al. (2012)21 HNSCC cell lines and mouse BDNF/TrkB activation is associated with HNSCC resistance to cisplatin by
tumor xenografts triggering Akt-dependent signaling pathways and anti-apoptotic proteins. Sasahira et al. (2012)39 OSCC tissue samples and cell lines TrkB is associated with proliferation, invasive capacity, and VEGF expression
in vitro and angiogenesis, lymphangiogenesis, and poor prognosis in vivo. Chodroff et al. (2016)40 Orthotopic tongue cancer model BDNF was significantly upregulated and its inhibition, using ANA-12, reversed
pain-like behaviors induced by the tumor.
HNSCC, Head and neck squamous cell carcinoma; OSCC, Oral squamous cell carcinoma.
27
F I GU RE 1 Schematic illustration of the main findings regarding BDNF/TrkB pathway regulation and function in head and neck squamous cell carcinoma
Furthermore, BDNF stimulus increased MDR1 (multidrug resistance 1)
levels and anti-apoptotic protein (XiAP) levels, associated with a decrease in
pro-apoptotic protein levels (Bmi) in cisplatin-resistant HNSCC cell lines.
Interestingly, TrkB knockdown reduced BDNF expression, suggesting a possible
autocrine activation of the pathway involved with HNSCC resistance.
Tumor-associated fibroblasts appear to have a role in activation of
BDNF/TrkB pathway in HNSCC.38 This idea is supported by the study of Dudas
et al., in which periodontal ligament fibroblasts were co-cultured with oral
squamous cell carcinoma (OSCC) to obtain tumor-associated fibroblasts. Under
co-culture conditions, TrkB gene and protein expression increased significantly
in OSCC cells but remained null in tumor-associated fibroblasts. BDNF
28
expression, by the other side, was considerably higher in fibroblasts compared
to OSCC cell, and co-culture conditions significantly increased BDNF expression
in tumor-associated fibroblasts.38
FI GU RE 2 Representative images of BDNF immunostaining in head and
neck squamous cell carcinoma, adenoid cystic carcinoma, and
mucoepidermoid carcinoma (original magnification, 200x)
The association between BDNF/TrkB activation and neovascularization in
HNSCC was also previously evaluated. Sasahira et al., 2012, demonstrated
that TrkB knockdown in OSCC cell lines decreased vascular endothelial
growth factor (VEGF), VEGF-C, and VEGF-D levels, inhibited growth ability, by
activating caspase-3, and reduced the invasive capacity, by a decrease in MMP-2
and MMP-9 levels. TrkB expression in OSCC tissue samples was significantly
correlated with microvessel density, lymphatic vessel density, and poor disease-free
survival. This group was also responsible for demonstrating that RUNX3
regulates TrkB expression in OSCC cell lines.39
A recent study demonstrated an association between BDNF/TrkB
activation and pain in an orthotopic tongue cancer model.40 Initially, the authors
performed a molecular characterization of the neuronal responses through RT-
PCR array for several genes known to be involved in activation of the pain
29
pathway. Among more than 80 genes evaluated, BDNF levels presented the
highest difference between the control group and the group with orthotopic
tongue cancer. BDNF signaling inhibition, using the TrkB antagonist ANA-12,
reversed pain-like behaviors produced by the tumor.40
3 BDNF / TrkB IN MALIGNANT SALIVARY GLAND TUMORS
Salivary gland tumors (SGT) are a rare and heterogeneous group of
neoplasms responsible for 3%-5% of head and neck neoplasms.41,42 The
main treatment option is surgery, leading to significant morbidity and variable
survival rates. Thus, there is an urgent need to better understand the
mechanisms underlying the invasive behavior of these tumors in order to
develop novel therapeutic strategies. Despite the emerging interest in
BDNF/TrkB molecular signaling in tumor progression, drugs that inhibit this
pathway, its role and therapeutic potential in malignant SGT, have not been
thoroughly investigated. Three studies were found in literature involving the role
of BDNF/TrkB pathway in these tumors as demonstrated on Table 2. All of them
were performed in adenoid cystic carcinoma (ACC) that represents the most
prevalente malignant SGT.42 This tumor exhibits a slow growth pattern,
presence of perineural invasion (PNI), lymph node involvement, distal
metastasis, and an aggressive behavior.43,44 A prominent hallmark of ACC is
perineural invasion (PNI) that represents cancer cell invasion in, around,
and through the nerves. It is a key factor responsible for the incomplete
surgical resection. The mechanisms involved in the PNI of ACC need to be
elucidated and could be modulated by BDNF/TrkB axis and epithelial-
mesenchymal transition (EMT) process.45,46
Kovalski, Paulino (2002) were the first to study the expression of BDNF in
29 cases of primary ACC of the head and neck. All cases exhibited diffuse and
uniform cytoplasmic staining for BDNF in both the luminal and abluminal cells
regardless of perineural spread status or histologic grade. BDNF was not
expressed by other common salivary gland tumors, including those with similar
histology, such as polymorphous low-grade adenocarcinoma. For these authors,
BDNF may play a causative role in its predilection for perineural invasion and
subsequent protracted clinical course.47
30
Recent studies indicate that BDNF/TrkB play a pivotal role in the in
tumor cell migration and invasion via epithelial-mesenchymal transition (EMT)-like
changes, such as E-cadherin downregulation and N-cadherin upregulation in
some tumors. These characteristics are associated with tumor aggressiveness45.
In this approach, Jia et al. (2015) verifies the expression of BDNF, TrkB, and E-
cadherin (an EMT biomarker) in 76 primary adenoid cystic carcinoma (ACC)
specimens and 20 normal salivary gland tissues. The high expression of TrkB and
the low expression of E-cadherin were significantly correlated with the poor
prognosis of patients with ACC. Additionally, the biological role of the BDNF/TrkB
axis in the EMT progression of ACC was evaluated using ACC cell line. The
results of the in vitro assays demonstrated that rhBDNF treatment significantly
downregulated E-cadherin expression and upregulated N-cadherin expression in
SACC cells, which was accompanied by the phenotypic changes from epithelial
to mesenchymal morphology and by increased motility. Additionally, obstruction of
TrkB by its inhibitor, k252a, significantly hindered the progression of ACC cells
through EMT indicating that these targets may be exploited for clinical therapy.
TABLE 2 Results from currently available studies that investigated the role
of BDNF/TrkB in salivary gland cancer.
Table 2. Results from currently available studies that investigated the role of BDNF/TrkB in
salivary gland cancer.
Salivary gland tumors
Authors (year) Methods Main findings Kowalski and Paulino (2002)
47 HNACC tissue samples All HNACCs exhibited diffuse and uniform cytoplasmic staining for BDNF protein
expression in both the luminal and abluminal cells. BDNF may play a causative role
in its predilection for perineural invasion.
Jia et al. (2015)45 SACC cell lines and tissue Overexpression of TrkB and BDNF and downregulation of E-Cadherin were correlated
samples with the invasion and metastasis in SACC. High TrkB and low E-cadherin expression
were correlated with poor prognosis of patients with SACC. BDNF-mediated TrkB
activation contributes to the EMT progression and the poor prognosis in SACC.
Shan et al. (2016)46 Salivary adenoid cystic BDNF and TrkB were significantly overexpressed in SACC tissue samples with more
carcinoma cell lines and staining intensity around the peripheral nerve in the SACC tissues.
tissue samples Co-culture of Schwann cells with SACC cells promote increase in BDNF secretion,
increase of TrkB expression, induce modification in cell morphology accompanied
by conversion of EMT hallmarks (downregulation of E‑cadherin and upregulation
of N‑cadherin and vimentin), and increase cell motility.
These results demonstrated that the BDNF/TrkB pathway mediated the
EMT pro- gression of ACC and emphasis that prevention of BDNF/TrkB signal- ing
31
should prove helpful as a novel strategy for the treatment of SACC.
Shan et al. (2016) investigated the relationship of Schwann cells (SC)
and EMT in ACC via the BDNF/TrkB axis. Mimicking the cross- talk between ACC
cells and SCs in the PNI process, the authors observed that the co-cultured
SCs with ACC cells secreted more BDNF. Also, co-culturing significantly
repressed the expression of E- cadherin, but promoted the expression of EMT
markers (N-cadherin and vimentin) in the SACC-83 cells. These effects were
significantly blocked by TrkB inhibitor (K252a). In 187 tissue samples of SACC,
the expression levels of TrkB and S100A4 were both significantly associated
with PNI, while E-cadherin was significantly inversely associated with PNI. These
results indicated that the interaction of SCs and SACC-83 cells mediated the
PNI process by inducing the EMT via the BDNF/TrkB axis. Targeting the
interaction between SACC cells and SCs by inhibition of BDNF/TrkB signaling
may be potential strategy for anti-PNI therapy in SACC.46
4 FUTURES PERSPECTIVES
Taken the evidences together, BDNF/TrkB-target therapy emerges as a
promising alternative to treat head and neck malignancies. Either TrkB
antagonists, or monoclonal antibodies against BDNF or TrkB, might increase
disease control and improve survival rates. The most favorable results are
usually observed in patients that harbor NTRK gene fusions.31 Stransky et al.,
(2014) in a study that evaluated the proportion of fusions involving a kinase in 20
types of solid tumors, identified NTRK2 (TrkB) fusions in HNSCC.48 However,
since this first repost, no further study was conducted concerning this subject
and the prevalence of NTRK2 mutations in head and neck malignancies
remains somewhat obscure.
Most drugs, such as Entrectinib and Cabozantinib, are multitarget and act
on all Trk receptors as well as other pathways. LOXO-101, by the other side, is
considered a specific drug once it inhibits only TrkA, B, and C.32 Hence, the
precise contribution of TrkB inhibition to the clinical response improvement
cannot be established with these drugs. This issue can be overcome by testing
highly specific drugs to TrkB. Cazorla et al., 2010, developed a highly potent and
32
selective TrkB inhibitor, cyclotraxin-B, able to inhibit the two active sites of TrkB
(p-Y816/PLCc and p-Y515/Shc), as well as downstream processes.49 The same
group also identified a low-molecular-weight TrkB ligand, ANA-12, which
selective inhibited processes down- stream of TrkB without altering TrkA and
TrkC functions.50 These discoveries expanded the horizons allowing the effects
of TrkB-specific inhibition to be analyzed. To date, only pre-clinical studies
evaluated the role of these drugs were conducted; however, the effects on
HNSCC and salivary gland cancer remains utterly unexplored.
5 CONCLUSIONS
Activation of BDNF/TrkB pathway in HNSCC and salivary gland cancer leads
to a more aggressive phenotype. The studies reviewed herein showed that
this pathway induces HNSCC cells proliferation and invasion, stimulates EMT
and angiogenesis, and is associated with increased cisplatin resistance.
Regarding ACC, the studies demonstrated that the BDNF/TrkB pathway is
significantly associated with perineural invasion, migration, induction of epithelial-
mesenchymal transition (EMT), and metastasis. Furthermore, the prognostic value
of TrkB expression was proved in both HNSCC and ACC. The association of
BDNF/TrkB activation and other important outcomes, by the other side, remains
completely obscure. The association of this pathway and cancer stem-cell
population, recognized by their role in tumor resistance, was never analyzed.
Besides, the prevalence of NTRK2 mutations in HNSCC and salivary gland cancer
must be further explored in new studies comprising representative samples.
Several studies analyzed the impact of TrkB inhibition through protein
knockdown. Inhibitory drugs with potential to be further transposed to clinical
trials need to be evaluated in vitro and in vivo in HNSCC and ACC. At least, the
role of BDNF/TrkB pathway in other important salivary gland cancers, such as
mucoepidermoid carcinoma and acinic cell carcinoma, has never been explored
and deserves a thorough study.
33
REFERENCES
1. Vanneman M, Dranoff G. Combining immunotherapy and targeted therapies
in cancer treatment. Nat Rev Cancer. 2012;12:237-251.
2. Roesler R, Farias CB, Abujamra AL, Brunetto AL, Schwartsmann G.
BDNF/TrkB signaling as an anti-tumor target. Expert Rev Anticancer Ther.
2011;11:1473-1475.
3. Levi-Montalcini R. The nerve growth factor 35 years later. Science
1987;237:1154-1162.
4. Barde Y, Edgar D, Thoenen H. Purification of a new neurotrophic factor
from mammalian brain. EMBO J. 1982;1:549-553.
5. Shibayama E, Koizumi H. Cellular localization of the Trk neurotrophin receptor
family in human non-neuronal tissues. Am J Pathol. 1996;148:1807-1818.
6. Ma L, Lopez GF, Krimm RF. Epithelial-Derived BDNF Is required for gustatory
neuron targeting during a critical developmental period. J Neurosci. 2009;29:3354-
3364.
7. Ricci A, Greco S, Mariotta S, et al. Neurotrophins and neurotrophin receptors
in human lung cancer. Am J Respir Cell Mol Biol. 2001;25:439-446.
8. Ho R, Eggert A, Hishiki T, et al. Resistance to chemotherapy mediated by
TrkB in neuroblastomas. Cancer Res. 2002;62:6462-6466.
9. Sclabas GM, Fujioka S, Schmidt C, et al. Overexpression of tropo- myosin-
related kinase B in metastatic human pancreatic cancer cells. Clin Cancer Res.
2005;11:440-449.
10. Desmet CJ, Peeper DS. The neurotrophic receptor TrkB: a drug target in anti-
cancer therapy? Cell Mol Life Sci. 2006;63:755-759.
11. Han L, Zhang Z, Qin W, Sun W. Neurotrophic receptor TrkB: is it a predictor of
poor prognosis for carcinoma patients? Med Hypotheses. 2007;68:407-409.
12. Fauchais AL, Lalloue F, Lise MC, et al. Role of endogenous brain- derived
neurotrophic factor and sortilin in B cell survival. J Immunol. 2008;181:3027-3038.
13. Thiele CJ, Li Z, McKee AE. On Trk - the TrkB signal transduction pathway is
an increasingly important target in cancer biology. Clin Cancer Res.
2009;15:5962-5967.
14. Zhang L, Hu Y, Sun CY, et al. Lentiviral shRNA silencing of BDNF inhibits
34
in vivo multiple myeloma growth and angiogenesis via down-regulated
stroma-derived VEGF expression in the bone marrow milieu. Cancer Sci.
2010;101:1117-1124.
15. Kupferman ME, Jiffar T, El-Naggar A, et al. TrkB induces EMT and has a
key role in invasion of head and neck squamous cell carcinoma. Oncogene.
2010;29:2047-2059.
16. Lai PC, Chiu TH, Huang YT. Overexpression of BDNF and TrkB in human
bladder cancer specimens. Oncol Rep. 2010;24:1265-1270.
17. Yilmaz T, Jiffar T, De la Garza G, et al. Therapeutic targeting of Trk
suppresses tumor proliferation and enhances cisplatin activity in HNSCC.
Cancer Biol Ther. 2010;10:644-653.
18. Bellanger C, Dubanet L, Lise MC, et al. Endogenous neurotrophins and Trk
signaling in diffuse large B cell lymphoma cell lines are involved in sensitivity
to rituximab-induced apoptosis. PLoS ONE. 2011;6:e27213.
19. Lam CT, Yang ZF, Lau CK, et al. Brain-derived neurotrophic factor promotes
tumorigenesis via induction of neovascularization: implication in hepatocellular
carcinoma. Clin Cancer Res. 2011;17:3123-3133.
20. Vanhecke E, Adriaenssens E, Verbeke S, et al. Brain-derived neurotrophic
factor and neurotrophin-4/5 are expressed in breast cancer and can be targeted
to inhibit tumor cell survival. Clin Cancer Res. 2011;17:1741-1752.
21. Lee J, Jiffar T, Kupferman ME. A novel role for BDNF-TrkB in the regulation
of chemotherapy resistance in head and neck squamous cell carcinoma. PLoS
ONE. 2012;7:e30246.
22. Okugawa Y, Tanaka K, Inoue Y, et al. Brain-derived neurotrophic
factor/tropomyosin-related kinase B pathway in gastric cancer. Br J Cancer.
2013;108:121-130.
23. Brunetto de Farias C, Rosemberg DB, Heinen TE, et al. BDNF/TrkB content
and interaction with gastrin-releasing peptide receptor blockade in colorectal
cancer. Oncology. 2010;79:430-439.
24. Brunetto de Farias CB, Heinen TE, Dos Santos RP, et al. BDNF/TrkB
signaling protects HT-29 human colon cancer cells from EGFR inhibition.
Biochem Biophys Res Commun. 2012;425:328-332.
35
25. Portich JP, Gil MS, Dos Santos RP, et al. Low brain-derived neurotrophic
factor levels are associated with active disease and poor prognosis in
childhood acute leukemia. Cancer Biomark. 2016;17:347-352.
26. Heinen TE, Dos Santos RP, Da Rocha A, et al. Trk inhibition reduces cell
proliferation and potentiates the effects of chemotherapeutic agents in Ewing
sarcoma. Oncotarget. 2016;7:34860-34880.
27. Thomaz A, Jaeger M, Buendia M, et al. BDNF/TrkB signaling as a potential
novel target in pediatric brain tumors: anticancer activity of selective trkb
inhibition in medulloblastoma cells. J Mol Neurosci. 2016;59:326-333.
28. Schmidt AL, Brunetto de Farias CB, Abujamra AL, et al. BDNF and PDE4,
but not the GRPR, regulate viability of human medulloblastoma cells. J Mol
Neurosci. 2010;40:303-310.
29. Cornelio DB, Brunetto de Farias CB, Prusch DS, et al. Influence of GRPR
and BDNF/TrkB signaling on the viability of breast and gynecologic cancer cells.
Mol Clin Oncol. 2013;1:148-152.
30. Wang T, Yu D, Lamb ML. Trk kinase inhibitors as new treatments for cancer
and pain. Expert Opin Ther Pat. 2009;19:305-319.
31. Khotskaya YB, Holla VR, Farago AF, et al. Targeting TRK family proteins in
cancer. Pharmacol Ther. 2017;173:58-66.
32. Bailey JJ, Schirrmacher R, Farrell K, Bernard-Gauthier V. Tropomyosin
receptor kinase inhibitors: an updated patent review for 2010-2016 - Part II.
Expert Opin Ther Pat. 2017;8:1-19.
33. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and
mortality worldwide: sources, methods and major patterns in GLO- BOCAN 2012.
Int J Cancer. 2015;136:359-386.
34. Curado MP, Johnson NW, Kerr AR, et al. Oral and oropharynx cancer in
South America: incidence, mortality trends and gaps in public databases as
presented to the Global Oral Cancer Forum. Transl Res Oral Oncol. 2016;1:1-7.
35. Jemal A, Ward EM, Johnson CJ, et al. Annual report to the nation on the
status of cancer, 1975-2014, featuring survival. J Natl Cancer Inst. 2017;109:1-22.
36. Zhu L, Werner JA, Mandic R. Implications of tropomyosin-related kinase B
(TrkB) in head and neck cancer. Anticancer Res. 2007;27:3121-3126.
36
37. Seiwert TY, Salama JK, Vokes EE. The chemoradiation paradigm in head
and neck cancer. Nat Clin Pract Oncol. 2007;4:156-171.
38. Dud as J, Bitsche M, Schartinger V, et al. Fibroblasts produce brain- derived
neurotrophic factor and induce mesenchymal transition of oral tumor cells. Oral
Oncol. 2011;47:98-103.
39. Sasahira T, Ueda N, Yamamoto K, et al. Trks are novel oncogenes involved
in the induction of neovascularization, tumor progression, and nodal metastasis
in oral squamous cell carcinoma. Clin Exp Metastasis. 2013;30:165-176.
40. Chodroff L, Bendele M, Valenzuela V, Henry M, Ruparel S. EXPRESS: BDNF
signaling contributes to oral cancer pain in a preclinical orthotopic rodent model. Mol
Pain. 2016;12:1-17.
41. Tian Z, Li L, Wang L, Hu Y, Li J. Salivary gland neoplasms in oral and
maxillofacial regions: a 23-year retrospective study of 6982 cases in an eastern
Chinese population. Int J Oral Maxillofac Surg. 2010;39:235-242.
42. Vasconcelos AC, Nor F, Meurer L, et al. Clinicopathological analysis of
salivary gland tumors over a 15-year period. Braz Oral Res. 2016;30:1-7.
43. Coca-Pelaz A, Rodrigo JP, Bradley PJ. Adenoid cystic carcinoma of the
head and neck - an update. Oral Oncol. 2015;51:652-661.
44. Amit M, Binenbaum Y, Trejo-Leider L, et al. International collaborative
validation of intraneural invasion as a prognostic marker in adenoid cystic
carcinoma of the head and neck. Head Neck. 2015;37:1038-1045.
45. Jia S, Wang W, Hu Z, et al. BDNF mediated TrkB activation con- tributes to
the EMT progression and the poor prognosis in human salivary adenoid cystic
carcinoma. Oral Oncol. 2015;51:64-70.
46. Shan C, Wei J, Hou R, et al. Schwann cells promote EMT and the Schwann
like differentiation of salivary adenoid cystic carcinoma cells via the BDNF/TrkB
axis. Oncol Rep. 2016;35:427-435.
47. Kowalski PJ, Paulino AFG. Perineural invasion in adenoid cystic carcinoma:
its causation/promotion by brain-derived neurotrophic factor. Hum Pathol.
2002;33:933-936.
48. Stransky N, Cerami E, Schalm S, Kim JL, Lengauer C. The landscape of
kinase fusions in cancer. Nat Commun. 2014;5:4846.
37
49. Cazorla M, Jouvenceau A, Rose C, et al. Cyclotraxin-B, the first highly
potent and selective TrkB inhibitor, has anxiolytic properties in mice. PLoS ONE.
2010;5:e9777.
50. Cazorla M, Pre mont J, Mann A, et al. Identification of a low-molecular weight
TrkB antagonist with anxiolytic and antidepressant activity in mice. J Clin Invest.
2011;121:1846-1857.
38
2.2 Activation of BDNF/TrkB/Akt axis may be involved in the acquisition of the
malignant phenotype in potentially malignant disorders and predicts poor
prognosis of oral squamous cell carcinoma
Running title: BDNF/TrkB/Akt axis in oral carcinogenesis
* This manuscript was formatted in the norms of the Modern Pathology (ISSN: 0893-
3952, 2018 Impact factor: 6.655*).
Juliana Kern#1, Vivian Petersen Wagner#1, Felipe Paiva Fonseca2, Gleyson Kleber
do Amaral-Silva1, Caroline Brunetto de Farias3,4, Emily Ferreira Salles Pilar5, Lauro
Gregianin6, Rafael Roesler3,7, Pablo Agustin Vargas1, Manoela Domingues
Martins1,5,8.
1Department of Oral Diagnosis, Piracicaba Dental School, University of Campinas,
Piracicaba, Brazil
2Department of Oral Surgery and Pathology, School of Dentistry, Federal University
of Minas Gerais, Belo Horizonte, Brazil
3Cancer and Neurobiology Laboratory, Experimental Research Center, Porto Alegre
Clinical Hospital, Federal University of Rio Grande do Sul, Porto Alegre, Brazil
4Children’s Cancer Institute, Porto Alegre, Brazil
5Experimental Pathology Unit, Clinics Hospital of Porto Alegre, Federal University of
Rio Grande do Sul, Porto Alegre, RS, Brazil.
6 Pediatric Oncology Service, Clinical Hospital, Federal University of Rio Grande do
Sul, Porto Alegre, Rio Grande do Sul, Brazil
7Department of Pharmacology, Institute for Basic Health Sciences, Federal
University of Rio Grande do Sul, Porto Alegre, Brazil
8Department of Oral Pathology, School of Dentistry, Federal University of Rio
Grande do Sul, Porto Alegre, RS, Brazil.
#Contributed equally to this article
39
Corresponding author:
Manoela Domingues Martins
Universidade Federal do Rio Grande do Sul
Faculdade de Odontologia
Rua Ramiro Barcelos, 2492, sala 503
CEP: 90035-003
Santana, Porto Alegre RS
Brazil
Phone: 55-51-33085011
Conflicts of interest
The authors declare no potential conflicts of interest.
Word count: 6.624
40
Abstract
To evaluate the expression of brain-derived neurotrophic factor (BDNF), its
tyrosine kinase receptor B (TrkB) and two downstream targets of this pathway, Akt
and ribosomal protein S6 (RPS6), in normal oral mucosa (NOM), leukoplakia (OL)
and oral squamous cell carcinoma (OSCC), and correlate this expression with
OSCC patients’ outcomes, cell senescence and “stemness” profile. Ten cases
of NOM, 33 OL and 72 primary OSCC were included. Cases of OSCC were
assembled in tissue micro arrays (TMAs). Immunohistochemical analysis for BDNF,
TrkB, p-TrkB, p-Akt and p-RPS6 were performed in all cases. Cell senescence and
stemness profile of OSCC were evaluated through p16 and BMI-1
immunohistochemical expression, respectively. The slides were scanned into high-
resolution images and immunohistochemical expression was quantified through
digital analysis. Cut-off values were determined to evaluate the association of each
marker with clinco-pathologic features. OSCC presented increased expression of
BDNF/TrkB/Akt pathway compared to NOM and OL. OSCC diagnosed in advanced
clinical stages presented an up-regulation of BDNF and p-TrkB. Moreover,
overexpression of BDNF and p-Akt were identified as predictors of poor disease-
specific survival, with hazard ratios of 2.83 and 2.19 respectively. The increase in
stemness profile was correlated with a decrease in p-TrkB and p-Akt expression.
Taken together these data suggest that BDNF/TrkB/Akt pathway is involved in the
acquisition of the malignant phenotype during oral carcinogenesis. Furthermore, the
increased availability of BDNF influences OSCC prognosis presumably by activating
downstream targets such as Akt. Patients with increased expression of these
proteins were more likely to experience shorter survival time.
Keywords: head and neck cancer, biomarker, predictive value, prognosis,
immunohistochemistry.
41
Introduction
Oral cancer is the most common malignancy of the head and neck region
and represents a global health problem with 354,860 cases estimated in 2018.1
Prospective new cases, for both sexes and all ages, increased from 363,626
worldwide in 2020 to 450,870 in 2030. 1 In Brazil, the National Cancer Institute José
Alencar Gomes da Silva (INCA) estimated 11,200 new cases of oral cancer in men
and 3,500 in women in Brazil in the year 2018 (biennium 2018-2019). In this way, it
occupies the 5th position between all types of cancers in men and 12º in women.
Approximately 90% of the oral malignancies are squamous cell carcinomas (SCC)
and are associated with alcohol and tobacco consumption.2 While the detection and
treatment of most malignancies has improved over the last several decades, high
mortality rate of oral squamous cell carcinoma (OSCC) has remained unchanged.3
As a result, 125,384 deaths attributed to OSCC occur per annum. 1 These data
demonstrate that prevention strategies are not effective; the diagnosis is still late
and current treatments have limited effectiveness.4,5
Oral carcinogenesis is a multi-step process that involves genetic and
epigenetic alterations resulting in oral mucosa modifications that culminate with
cancer manifestation. A significant number of OSCC can be preceded by
asymptomatic clinical lesions collectively referred as oral potentially malignant
disorders (OPMDs), with oral leukoplakia (OL) being the most prevalent disorder.6,7
These lesions are defined as “altered epithelium with an increased potential for
progression to OSCC.7 Histopathological examination revealed that OPMDs could
vary from non-dysplastic (epithelial hyperkeratosis, hyperplasia, acanthosis) to
different degrees of dysplasia. However, the evolution from dysplasia to invasive
carcinoma is not yet well understood.6-8 The understanding of mechanisms driving
oral carcinogenesis and OSCC behavior is a key step to improving oral cancer
outcomes and development of new candidates for targeted therapies.9
There is now an extensive literature reporting the role of deregulation of
growth factors and their receptors in the activation of intracellular signaling
pathways involved in the acquisition of the malignant phenotype and tumor
behavior.10-12 They are involved in several cell process such as differentiation,
42
survival, mobility and cell growth in different types of cancer. In this sense, the
brain-derived neurotrophic factor (BDNF) pathway and tropomyosin-related kinase
B receptor (TrkB) has been extensively studied in pancreatic, lung, colon, prostate,
breast, bladder, neuroblastoma, medulloblastoma, multiple myeloma, among
others.13-16
BDNF is a member of the neurotrophins family that are growth factors that
participate in the differentiation, proliferation and survival of neural cells and
neurogenesis in the central and peripheral nervous systems. The major ligand for
BDNF are the cell membrane receptors, especially TrkB.16,17 BDNF receptor binding
in TrkB elicits its dimerization through conformational changes and
autophosphorylation of tyrosine residues in the intracellular environment.18 In
cancer, stimulation of the BDNF/TrkB signaling pathway promotes the activation of
signaling molecules such as Akt, STAT3, Src, ERK and MAPK resulting in
increased cell proliferation, resistance to apoptosis, invasion, metastasis and
resistance to chemotherapy. Thus, increased BDNF/TrkB expression has been
associated with more aggressive behavior of different malignant tumors and worse
prognosis.14,19-25
In a recent literature review, our group evidenced that the activation of the
BDNF/TrkB pathway in squamous cell carcinoma of the head and neck and salivary
gland cancer leads to a more aggressive phenotype and induces the proliferation,
invasion of SCC cells. In addition, this signaling pathway promotes tumor related
angiogenesis and increased resistance of neoplastic cells to cisplatin.26 The
association of BDNF/TrkB signaling pathway activation in HNSCC with tumor
aggressiveness has been described, however important aspects never before
evaluated still need to be elucidated.26,27 The association of this pathway and the
population of cancer stem cells, recognized for their role in tumor resistance, have
never been analyzed. There are no studies evaluating the role of this pathway
throughout the oral carcinogenesis process. Lastly, the capacity of these markers to
predict OSCC prognosis still needs to be validated in different populations to
confirm its real prognostic value. Thus, the aim of the present study was to evaluate
the expression of BDNF/TrkB and two downstream targets of this pathway, Akt and
43
ribosomal protein S6 (RPS6) during oral carcinogenesis and correlate it with OSCC
cell senescence, stemness profile and patient’s outcomes.
Materials and methods
This study was approved by the Ethics Committee on Human Research
(approval No. 56334716.9.0000.5327).
STUDY POPULATION
A manual retrospective search was performed in the archives of the
Pathology Laboratory at the Clinics Hospital of Porto Alegre - Brazil to identify cases
of primary OSCC diagnosed between 1996 and 2010. A total of eighty-seven cases
of primary oral squamous cell carcinoma (OSCC) were retrospectively collected and
retrieved from the archives of the Head and Neck Department of the Clinics Hospital
of Porto Alegre (Brazil). Clinical data were collected from patient's medical files and
hospital records like age, sex, tumor location, tumor stage, tumor recurrence, follow-
up, survival time (time difference between treatment and either the date of death or
last follow-up) and disease-free survival (time between treatment and the date of
recurrence).
Twenty-four cases of oral leukoplakia (OL) were selected in the archives of
Oral Pathology of the Federal University of Rio Grande do Sul. Additionally, ten
cases of normal oral mucosa (NOM) obtained from mucocele specimens were
retrieved in the Federal University of Rio Grande do Sul for comparison purposes.
TISSUE MICROARRAY (TMA) CONSTRUCTION
OSCC specimens that were retrospectively collected were arranged into
tissue microarray (TMA) blocks for immunohistochemical analysis. TMA
construction was performed as previously.28,29 Briefly three representative areas of
the invasive front were elected in the H&E slides of each case using an objective
marker (Nikon Corp, Tokyo, Japan). A manual tissue arrayer (Sakura Co, Japan)
was used to cut the respective cylindrical cores (2.0mm in diameter each) after
matching the marked slides over the original paraffin block. Three cores of normal
44
mucosa inserted in the left upper corner of each recipient block for orientation. To
enable the interpretation of TMAs, a map indicating the exact position of each case
was performed.
IMMUNOHISTOCHEMISTRY
Histological sections of NOM, OL and OSCC TMAs slides were subjected to
immunohistochemical staining for BDNF, TrkB, phospho-TrkB (p-TrkB), phospho-
Akt (p-Akt), phospho-ribossomal protein S6 (p-RPS6). Furthermore, OSCC were
also subjected to moloney murine leukemia virus insertion site 1 (BMI1) and p16
immunohistochemical analysis. Briefly, blocks were sectioned (3 μm) and placed on
silanized slides. The slides were subsequently deparaffinized in xylene and
hydrated in descending grades of ethanol. Antigen retrieval was performed for 30
min in a citrate buffer solution heated to 90oC in a water bath. The slides were then
incubated with the primary antibodies: BDNF (1:750, EPR1292, Abcam), TrkB
(1:1000, Polyclonal, Abcam), p-TrkB Y706 + Y070 (1:100, Polyclonal, Abcam), p-
Akt s473 (1:200, EP2109Y, Abcam), p-RPS6 S235 + S236 (1:200, Polyclonal,
Abcam), BMI1 (1:200, Polyclonal, Abcam) and p16 (1:1500, 2DA12, Abcam). All
slides were then exposed to avidin–biotin complex and horseradish peroxidase
reagents (LSAB Kit; Dako Cytomation). The reactions were revealed with
diaminobenzidine tetrahydrochloride (DAB; Novocastra, Newcastle, UK) and
counterstained with Mayer’s hematoxylin. Negative controls were obtained through
incubation with nonimmune serum instead of primary antibodies. Positive controls
for BDNF, TrkB, p-TrkB, p-Akt, p-RPS6, BM1 and p16 were human brain tissue,
cervical carcinoma, human pancreas tissue, and human brain tissue respectively.
Only brown color regardless of the color intensity will be considered as positive
marking.
DIGITAL ANALYSIS
The immunohistochemical slides were scanned into high-resolution images
using the Aperio Scanscope CS Slide Scanner (Aperio Technologies Inc, Vista,
CA). The digital images obtained in .svs format were visualized using the
ImageScope software (Aperio Technologies Inc., Vista, CA). Digital images were
45
analyzed for cytoplasm immunomarking with PixelCount V9 (Aperio Technologies
Inc, Vista, CA, USA) by using specific input parameters (hue value, 0.1; hue width,
0.5; color saturation threshold, 4 e -002; and intensity threshold ranging from 0 to
240), the percentage of positivity was calculated and classified in 3 categories
according to their intensity range as weak (from 155-240), moderate (from 120 to
155), and strong staining (from 0 to 120). For membrane immunomarking, the
algorithm Membrane V9 (Aperio Technologies Inc, Vista, CA, USA) was used and
the percentage of positivity was calculated using (3+) Percent Cells, (2+) Percent
Cells, (1+) Percent Cells. The final score of each tumor was calculated as the sum
of the percentage of each category multiplied by their intensity scores using the
following formule: ([%weak × 1]+[%moderate×2]+[%strong×3]). The results always
ranged from 100 to 300. The digital images were analyzed for nuclear
immunomarking with Nuclear V9 (Aperio Technologies Inc, Vista, CA, USA), using
Total Nuclei number and Percent Positive Nuclei.
STATISTICAL ANALYSIS
The clinical and immunohistochemical data were analyzed using SPSS
software (IBM Corporation, Armonk, NY), version 18.0. Initially, a descriptive
analysis of clinic-pathologic features was performed for OSCC and OL. The
Kruskal-Wallis test was used to compare raw values between NOM, OL and OSCC
and significant differences were then analyzed using Dunn’s post-hoc test adjusted
for Bonferroni error correction. Receiver Operating Characteristic (ROC) curves was
constructed to establish the best cut-off point using “alive or deceased due to
OSCC” as main outcome. The cut-off point was determined as the value presenting
the highest sensitivity along with a good specificity. The proteins expression was
then dichotomized in high(high) and low(low) expression. Correlation between clinic-
pathological features and proteins expression was evaluated using the chi-square
test or fisher’s exact test. Binary logistic regression was used to determine the odds
ratio. The univariable Cox proportional hazard regression model was performed to
evaluate the prognostic value of BDNF/TrkB pathway expression in patients’
disease-specific survival (DSS). This value was also evaluated using a multivariable
model controlled for age, gender and clinical stage. Kaplan-Meier cumulative
46
disease-specific survival curves were generated and compared using the log-rank
test. Spearman correlation test was used to determine the correlation of BDNF/TrkB
pathway proteins expression with BMI1 (stemness profile) and p16 (senescence
profile). For all tests, p≤0.05 was considered indicative of statistical significance.
Results
STUDY POPULATION
Seventy-two cases of OSCC and 33 cases of OL were included in the
present study. The clinic-demographic profile of OSCC and OL patients is
presented in Tables 1 and 2, respectively.
BDNF/TRKB/Akt SIGNALING PATHWAY IS MORE EXPRESSED IN OSCC
THAN OL AND NOM
We first sought to understand the patterns of expression of proteins related
to BDNF/TrkB pathway in NOM, OL and OSCC. We observed that all evaluated
specimens were positive for BDNF, TrkB, p-TrkB, p-Akt and p-S6, however some
important differences were observed among NOM, OL and OSCC (Figure 1A).
Digital analysis was performed to determine the score of expression in each case
considering positive the brown cytoplasmic staining. The mean score of all markers
for NOM, OL and OSCC is presented in Figure 1B. NOM and OL showed similar
expression of BDNF, TrkB, p-Akt and p-S6. They differed only concerning p-TrkB
expression. OL had a higher score compared to NOM. OSCC presented increased
expression of BNDF, TrkB, p-TrkB and p-Akt compared to NOM. Moreover, the
malignant cells also had increased expression compared to OL for BDNF, TrkB and
p-Akt. Interestingly, p-RPS6 expression decreased in OSCC compared to NOM and
OL.
BDNF/TRKB/Akt PATHWAY IS MORE EXPRESSED IN ADVANCED OSCC
CASES AND IS ASSOCIATED WITH POOR OUTCOMES
To evaluate the association between clinical features and outcomes with
BDNF, TrkB, p-TrkB, p-Akt and p-RPS6 expression, the variables were
dichotomized as previously reported in the statistical analysis section. The cut-off
47
values used for BDNF, TrkB, p-TrkB, p-Akt and p-RPS6 were respectively 210, 195,
125, 142 and 134. Based on these values the patients were divided in two groups:
high and low expression.
Figure 2 illustrates all the significant associations between proteins involved
in BDNF/TrkB pathway with clinical features and outcomes. At the time of diagnosis
patients with BDNFhigh expression presented a 2.62-fold higher chance of having
nodal metastasis (Figure 2A). Additionally, patients diagnosed in advanced clinical
stages (TNM III or IV) presented elevated expression of the growth factor
(BDNFhigh) and its phosphorylated receptor (p-TrkBhigh). The odds ratios for these
associations were 8.0 and 5.4 for BDNF and p-TrkB, respectively (Figures 2B and
2C).
During the follow-up period, patients with p-Akthigh expression presented
2.72-fold higher chance of having recurrences compared to patients with p-Aktlow
expression (Figure 2D). Concerning the survival of OSCC patients, we observed
that both BDNF and p-Akt were associated with poor outcomes. A higher chance to
decease during follow-up of 3.54 for BDNFhigh expression and 2.83 for p-Akthigh
expression were identified (Figures 2E and 2F).
BDNF AND p-AKT EXPRESSION ARE ASSOCIATED WITH OSCC DISEASE-
SPECIFIC SURVIVAL
The next step was to identify the capacity of proteins involved in BDNF/TrkB
pathway to predict the DSS. The hazard ratio and p values obtained from Cox
Regression hazard regression model are presented in Table 3. In our analysis,
BDNF and p-Akt were identified as biomarkers associated with survival in OSCC
patients. Univariate analysis demonstrated that the chance to die due to the disease
in each time point was 2.83-fold higher in patients with BDNFhigh expression
compared to BDNFlow and 2.19-fold higher in patients with p-Akthigh expression
compared to p-Aktlow. When the model was controlled for age, gender and clinical
stage, the hazard ratio values slightly changed and the p value increased to 0.09 for
BDNF analysis and 0.08 for p-Akt.
A visual inspection of the survival curves (Figure 3) reveals the evident
difference among patients with BDNFhigh and BDNFlow expression and among
48
patients with p-Akthigh and p-Aktlow. Moreover, despite no significant difference was
encountered, a tendency of different survival pattern can also be observed between
patients with p-TrkBhigh and p-TrkBlow expression. After 50 months of follow-up the
curves tend to distance themselves, demonstrating that patients with p-TrkBhigh
expression have higher chances of deceasing after this time point. The log-rank test
revealed significant difference between the curves for BDNF and p-Akt markers,
corroborating with the results from Cox regression.
BDNF/TRKB/Akt PATHWAY INCREASE EXPRESSION CORRELATES WITH
MODIFICATION IN THE STEMNESS PROFILE
Immunohistochemical analysis of p16 and BMI1 were used to evaluate cell
senescence and the stemness profile, respectively. No association was observed
between p16 expression and the BDNF/TrkB pathway proteins (Figure 4A),
indicating that this pathway appears to have no impact on cell senescence. On the
other hand, BMI1 expression was significantly correlated with p-TrkB and p-Akt. An
increase in both proteins resulted in the decrease of BMI1 evidenced by a negative
correlation coefficient (Figure 4A) and descending curves in the scatter plot graphs
(Figure 4B and 4C).
Discussion
BDNF/TrkB axis has been implicated in multiple biological processes
including tumorigenesis and tumor behavior.12-16 The binding of BDNF to TrkB
triggers promotes the activation of numerous signaling pathways including Akt,
STAT3, Src, ERK and MAPK, which play important roles in malignant
transformation, tumor progression, metastasis, angiogenesis, and therapy
resistance.14,19-26 Although this pathway has been previously studied in OSCC26,27
we identified some important gaps in the literature, which were addressed in the
present study. Our results significantly expand the knowledge concerning the role of
this pathway in oral cancer. We demonstrated that during oral carcinogenesis an
increase in TrkB activation occur at an early stage, while the increase of BDNF, by
the other side, is observed when epithelial cells already achieved the malignant
phenotype. In addition, we identified for the first time the association of this pathway
49
with the stemness capacity of OSCC cells. Interestingly, there is an inverse
correlation between the p-TrkB and p-Akt with the stemness profile of OSCC.
Finally, we validate in our population the prognostic value of BDNF and p-Akt. Both
proteins were capable of predicting a shorter survival time in OSCC.
Modifications in the epithelial tissue during oral carcinogenesis have been
studied to understand the mechanisms involved with cell signaling regulation, cell
proliferation and acquisition of malignant phenotypes. Nevertheless, no specific
molecular marker to predict malignant transformation of oral lesions is established
in the literature. Overall, our results reveled that BDNF/TrkB/Akt pathway was found
in the epithelial cells of the NOM, OL and OSCC samples. For the first time, we
identified that OSCC exhibit higher levels of BDNF, TrkB and p-Akt compared to
OL, demonstrating that this pathway is more active in the malignant neoplasm
already established compared to the most prevalent OPMD. It can be presumed
that BDNF/TrkB/Akt modifications are associated with the progression of oral
carcinogenesis and acquisition of a malignant phenotype. Indeed, in vitro studies
demonstrated that BDNF stimulus is capable of increasing the migratory and
invasive capacity of OSCC cells and inducing epithelial-mesenchymal
transition,20,27,30 which are key mechanisms involved with the capacity of epithelial
cells to invade the connective tissue through basement membrane rupture. In the
present study, the follow-up information on OL patients was not available, limiting
our analysis of associations of these markers with clinical outcomes. We believe it is
important to further evaluate if an increased expression of BDNF/TrkB/Akt is
capable of predicting malignant transformation of OL during follow-up. Interestingly,
p-RPS6 expression was down-regulated in OSCC compared to NOM and OL. P-
RPS6 and p-Akt proteins are downstream target of phosphatidylinositol 3-kinase
mammalian target of rapamycin mTOR (PI3K-mTOR) pathway that is implicated in
epithelial tumors development and progression. mTORC1 activation results in
RPS6 phosphorylation. The higher expression of RPS6 in NOM and OL suggest
that it might be involved in mechanisms related to cell differentiation. Results from
other studies by our group corroborate with this hypothesis. We observed the
importance of mTORC1 activation in keratoacanthoma, a well-differentiated
epithelial neoplasm, with reduced rates of proliferation and invasion31 and in
50
epithelial cells during the late stages of skin wound healing, when terminal
differentiation.32 By the other way, in OSCC we observed an overexpression of Akt
at serine 473 (p-AktSer473), indicating the activation of mTOR2 that is more
associated with malignant tumors based on activation of tumor cells migration and
invasion. Therefore, we believe that Akt overexpression in oral cancer triggers
mTORC2 activation rather then mTORC1. This specific activation has been
previously demonstrated in oral cancer cells and has been associated with clinically
advanced tumors33 and in vitro chemoresistance.34
Growth factors are recognized as major regulators of tumor progression,
controlling events such as cell proliferation, invasion, migration and drug-resistance.
Many cancer cells also acquire the capacity to synthesize and release growth
factors to which they are responsive, leading to constitutive pathway activation.35
Up-regulated growth factors might represent promising therapeutic targets and their
identification is therefore of paramount importance. In the present study, we
demonstrated that BDNF is overexpressed in OSCC cells, suggesting that
neoplastic cells have the capacity to produce this growth factor inducing an
autocrine loop with TrkB activation. Dudas et al., (2011) have also demonstrated
that OSCC-associated fibroblasts also secrete BDNF, indicating that paracrine
stimulus also occurs in the tumor microenvironment.30 The persistent activation of
BDNF/TrkB pathway may promote a more aggressive clinical behavior. In the
present cohort, patients with BDNF high expression had an increased risk of
presenting nodal metastasis at the diagnosis. Previous in vitro studies had already
established a correlation between BDNF stimulus and increased migratory/invasive
capacity of OSCC cells.20,27 Besides, TrkB expression has also been previously
correlated with lymphovascular invasion and lymph vessel density in OSCC
samples.23,27 Combined, these data demonstrate the importance of BDNF/TrkB axis
in the establishment of OSSC nodal metastasis, which is recognized as a major
indicator of poor prognosis.36-38
Nodal metastasis and clinical stage usually are effective in predicting the
overall survival of OSCC patients.36 However, patients diagnosed within the same
clinical stage commonly present distinct behavior during follow-up. Therefore,
biomarkers are continuously being investigated in order to distinguish molecular or
51
cellular signatures with a greater predictive value. Herein, we identified BDNF and
Akt as biomarkers associated with patients outcomes in a cohort of 72 patients with
primary OSCC. The higher expression of both proteins was capable of predicting a
shorter survival time of OSCC patients in univariate analysis. In multivariate
analysis the model was controlled for multiple important factors such as age, gender
and clinical stage. In this analysis, the p value of these associations increased to
0.09 and 0.08 for BDNF and p-Akt, respectively. Despite these values are not
normally considered significant, we believe that the high hazard ratios observed for
BDNF and p-Akt, 2.19 and 2.48 respectively, still have clinical relevance. They
indicate that BDNF and Akt have value in predicting DSS in OSCC patients. Despite
TrkB expression was not significantly associated with DSS in our analysis, survival
curves demonstrated a trend of poor prognosis in patients with p-TrkBhigh
expression. Our results corroborate with previous findings that demonstrated the
predictive value of BDNF/TrkB/Akt axis in OSCC patients. Recently, Moriwaki et al.
identified that up regulation of BDNF was capable to predict OSCC disease-free
survival in a cohort of 44 patients. Interestingly, these authors also observed that
higher TrkB expression was associated with an important but not significant trend of
poor prognosis.27 Sasahira et al. detected a significant association of TrkB positive
expression with disease-free survival in their study comprising 102 patients with
OSCC. BDNF was not evaluated in this study23 and Akt was not evaluated in neither
of these studies.23,27 Yet, other groups have also detected Akt as a prognostic
marker in OSCC.39,40 These findings reveal that over-activation of axis leads to a
more aggressive behavior and therefore represent a promising therapeutic target.
The ideal model of cancer treatment has evolved from conventional non-
specific agents to selective mechanism-based drugs.41 This new paradigm of target
therapy can achieve high levels of success specially if allied with precision
medicine. The molecular heterogeneity that can be observed within a similar clinical
cancer phenotype imposes this necessity. In precision medicine, genomic,
proteomic, transcriptomic and metabolomic markers can be used to identify patients
who are likely to benefit from a specific treatment.42 Our results demonstrated that
patients with overexpression of BDNF had significantly lower survival rates. This
subset of patients could benefit from drugs targeting the BDNF/TrkB axis by using
52
either TrkB antagonists or monoclonal antibodies against BDNF or TrkB. Currently
several Trk-inhibitors are in clinical development and being tested in clinical
trials,43,44 however most of drugs are multitarget and act on all Trk receptors as well
as other pathways. Cazorla et al., developed a low-molecular-weight TrkB ligand,
ANA-12, capable of inhibiting the downstream processes of TrkB without modifying
TrkA and TrkC functions.45 The effect of this compound was recently evaluated in
OSCC cell lines. Selective inhibition of TrkB blocked proliferation and migration of
poorly differentiated OSCC cells and also suppressed the tumor growth in an
orthotopic transplantation mouse model using these same cell lines.27 The effect of
ANA-12 in other mechanisms involved with OSCC progression, such as cell cycle
regulation, senescence and percentage of cancer stem cells, remain unexplored.
In the present study we evaluated in OSCC the correlation of BDNF/TrkB/Akt
axis activation with cell senescence and cancer stem cells (CSC), using p16 and
BMI-1 markers for this purposes. The p16INK4A gene encodes the p16 protein, that is
well known for its capacity to inhibit the kinase activities of the cyclin D-dependent
kinases CDK4 and CDK6 and therefore decelerate cell progression from G1 to S
phase, inducing cell senescence.46 No significant correlation was observed between
BDNF/TrkB/Akt and p16 positive cells. This result suggests that activation of this
pathway has no influence on triggering of cell senescence thought up-regulation of
cyclin-dependent kinase inhibitor p16. BMI1 is a component of the polycomb
repressive complex 1 recognized to have a significant role in the maintenance of
normal and CSCs.47 CSCs represent a subset of cells within the tumor bulk that
have stem cell-like proprieties, such as self-renewal and multipotency.48 They have
a unique potential for tumor initiation and are the main responsible for tumor relapse
after treatment.48 BMI1 acts as an epigenetic modifier protein enabling the
maintenance of self-renewal capacity of CSCs. Thus, BMI1 is highly enriched in
CSCs.47 Recently, Yin et al. (2015) demonstrated that BDNF stimulates the
maintenance of CSC in breast cancer and inhibition of TrkB leads to ablation of this
population of tumor cells preventing the recurrence of these tumors in animal
model.49 Our study was the first to evaluate the association between BDNF/TrkB
axis with a marker of stemness capacity in OSCC. Remarkbly, we observed an
inverse association of TrkB and Akt with BMI1 expression. Tumors with higher TrkB
53
and Akt that were associated with poor outcomes showed lower BMI1 labeling.
Nowadays, the CSC phenotype is considered as a transient state that depends of
microenvironment and therapeutic pressures. In melanoma, CSC oscillates
between quiescent and active states depending of PI3K/AKT signaling pathway.50
Therefore, we believe that some factors such as BDNF, TrkB and Akt could modify
CSC plasticity in oral cancer resulting in modification in BMI1 expression. In
agreement with this theory, some authors observed higher BMI1 expression in the
initially/well-differentiated HNSCCs and declined in poorly
differentiated/progressed tumors.51,52 Also, some authors found a statistically
significant correlation between lack of BMI-1 immunoexpression and poor prognosis
of OSCC patients.52,53 The relationship between the BDNF/TrkB/Akt pathways
with CSC plasticity need to be better evaluated in future studies.
In conclusion, our data demonstrate that BDNF/TrkB/Akt signaling pathway is
more expressed in OSCC compared to OL and NOM indicating that they are
involved in the acquisition of the malignant phenotype during oral carcinogenesis.
Furthermore, the increased availability of BDNF influences OSCC prognosis and
shorter survival time presumably by activating downstream targets such as Akt. In
addition, BDNF/TrkB/Akt pathway increased expression was correlates with
modification in the stemness profile by decreasing BMI1 expression. This study
supports that BDNF/TrkB is an attractive therapeutic target for OSCC.
Acknowledgments
The authors are grateful to Flavia Rejane Giusti for her technical support and
the Biostatistics Unit of the Postgraduate Research Group of the Porto Alegre
University Hospital for assisting the statistical analysis. The authors are also grateful
to Brazilian Federal Agency for Support and Evaluation of Graduate Education
(CAPES- finance code 001), the São Paulo State Research Foundation (FAPESP
2016/21785-4) for student scholarship and Children’s Cancer Institute-ICI. This
study was funded by the Postgraduate Research Group of the Hospital de Clínicas
de Porto Alegre (GPPG/FIPE: 16–0437) and Graduation Program in Oral Medicine
54
and Oral Pathology of Piracicaba Dental School, UNICAMP (FOPUNICAMP).
Manoela Domingues Martins, Rafael Roesler and Pablo Agustin Vargas are
research fellows funded by the Brazilian National Council for Scientific and
Technological Development (CNPq).
Author contributions
J Kern and VP Wagner contributed equally in the acquisition, analysis and
interpretation of data, writing the manuscript, revising it critically, approved the final
version to be published and agree to be accountable for all aspects of the work. FP
Fonseca, EFS Pilar and GK Amaral-Silva contributed in the analysis and
interpretation of data, revising the manuscript critically, approved the final version to
be published and agree to be accountable for all aspects of the work. CB de Farias,
L Gregianin, R Roesler and PA Vargas contributed in design of the study, revising
the manuscript critically, approved the final version to be published and agreed to
be accountable for all aspects of the work. MD Martins contributed in design of the
study, analysis and interpretation of data, writing the manuscript, revising it critically,
approved the final version to be published and agree to be accountable for all
aspects of the work.
Compliance with ethical standards
Conflict of interest - The authors declare that they have no conflict of interest.
55
References
1. Global Health Observatory (Globocan). Geneva: World Health Organization; 2018. who.int/gho/database/en/. Accessed Nov 16, 2018. Google Scholar.
2. Wyss A, Hashibe M, Chuang S, et al. Cigarette, cigar, and pipe smoking and the risk of head and neck cancers: pooled analysis in the international head and neck cancer epidemiology consortium. Am J Epidemiol 2013;178:679–690.
3. Warnakulasuriya S. Living with oral cancer: epidemiology with particular reference to prevalence and life-style changes that influence survival. Oral Oncol. 2010;46:407-410.
4. Winn DM, Lee YC, Hashibe M, Boffetta P, consortium I. The INHANCE consortium: toward a better understanding of the causes and mechanisms of head and neck cancer. Oral Dis. 2015;21:685-693
5. Jansen JFA, Parra C, Lu Y, et al. Evaluation of head and neck tumors with functional MRI. Magn Reson Imaging Clin N Am 2016;24:123–133.
6. Dionne KR, Warnakulasuriya S, Zain RB, et al. Potentially malignant disorders of the oral cavity: Current practice and future directions in the clinic and laboratory. Int J Cancer 2015;136:503–515.
7. Speight PM, Khurram SA, Kujan O. Oral potentially malignant disorders: risk of progression to malignancy. Oral Surg Oral Med Oral Pathol Oral Radiol 2018;125:612–627.
8. Chaves ALF, Silva AG, Maia FM, et al. Reduced CD8+ T cells infiltration can be associated to a malignant transformation in potentially malignant oral epithelial lesions. Clin Oral Investig 2018; 10.1007/s00784-018-2622-8.
9. Villa A, Sonis S. Oral leukoplakia remains a challenging condition. Oral Dis 2018;24:179–183.
10. Cohen Y, Goldenberg-Cohen N, Shalmon B, Shani T, Oren S, Amariglio N, Dratviman-Storobinsky O, Shnaiderman-Shapiro A, Yahalom R, Kaplan I, Hirshberg A. Mutational analysis of PTEN/PIK3CA/AKT pathway in oral squamous cell carcinoma. Oral Oncol. 2011;47:946-950.
11. Alyasiri NS, Mehdi SJ, Alam MS, Ali A, Mandal AK, Gupta S, Singh I, Rizvi MM. PTEN-mediated AKT activation contributes to the reduced apoptosis among Indian oral squamous cell carcinoma patients. J Cancer Res Clin Oncol. 2012;138:103-109
12. Thomaz A, Jaeger M, Buendia M, Bambini-Junior V, Gregianin LJ, Brunetto AL, Brunetto AT, de Farias CB, Roesler R BDNF/TrkB Signaling as a Potential Novel Target in Pediatric Brain Tumors: Anticancer Activity of Selective TrkB Inhibition in Medulloblastoma Cells. J Mol Neurosci. 2016; 59:326-333
13. Lai PC, Chiu TH, Huang YT. Overexpression of BDNF and TrkB in human bladder cancer specimens. Oncol Rep. 2010;24:1265-1270.
56
14. Roesler R, de Farias CB, Abujamra AL, Brunetto AL, Schwartsmann G. BDNF/TrkB signaling as an anti-tumor target. Expert Rev Anticancer Ther. 2011;11:1473-1475.
15. Brunetto de Farias CB, Heinen TE, Dos Santos RP, et al. BDNF/TrkB signaling protects HT-29 human colon cancer cells from EGFR inhibition. Biochem Biophys Res Commun 2012; 425: 328─332.
16. Akil H, Perraud A, Jauberteau MO, Mathonnet M.Tropomyosin-related kinase B/brain derived-neurotrophic factor signaling pathway as a potential therapeutic target for colorectal cancer. World J Gastroenterol. 2016.14;22:490-500.
17. Schecterson LC, Bothwell M. Neurotrophin receptors: Old friends with new partners. Dev Neurobiol. 2010;70:332-338.
18. Liu C, Chan CB, Ye K. 7,8-dihydroxyflavone, a small molecular TrkB agonist, is useful for treating various BDNF-implicated human disorders. Transl Neurodegener. 2016;5:2.
19. Fujikawa H, Tanaka K, Toiyama Y, Saigusa S, Inoue Y, Uchida K, Kusunoki M. High TrkB expression levels are associated with poor prognosis and EMT induction in colorectal cancer cells. J Gastroenterol. 2012;47:775-784.
20. Kupferman ME, Jiffar T, El-Naggar A, Yilmaz T, Zhou G, Xie T, Feng L, Wang J, Holsinger FC, Yu D, Myers JN. TrkB induces EMT and has a key role in invasion of head and neck squamous cell carcinoma. Oncogene. 2010;29:2047-2059
21. Guo D, Hou X, Zhang H, Sun W, Zhu L, Liang J, Jiang X. More expressions of BDNF and TrkB in multiple hepatocellular carcinoma and anti-BDNF or K252a induced apoptosis, supressed invasion of HepG2 and HCCLM3 cells. J Exp Clin Cancer Res. 30:972011.
22. Okamura K, Harada T, Wang S, Ijichi K, Furuyama K, Koga T, Okamoto T, Takayama K, Yano T, Nakanishi Y. Expression of TrkB and BDNF is associated with poor prognosis in non-small cell lung cancer. Lung Cancer. 2012;78:100-106.
23. Sasahira T, Ueda N, Yamamoto K, et al. Trks are novel oncogenes involved in the induction of neovascularization, tumor progression, and nodal metastasis in oral squamous cell carcinoma. Clin Exp Metastasis 2013;30:165–176.
24. Jia S, Wang W, Hu Z, Shan C, Wang L, Wu B, Yang Z, Yang X, Lei D. BDNF mediated TrkB activation contributes to the EMT progression and the poor prognosis in human salivary adenoid cystic carcinoma. Oral Oncol. 2015;51:64-70.
25. Kim MS, Lee WS, Jeong J, et al. Induction of metastatic potential by TrkB via activation of IL6/JAK2/STAT3 and PI3K/AKT signaling in breast cancer. Oncotarget 2015;6:40158–40171.
26. Moraes JK, Wagner VP, Fonseca FP, et al. Uncovering the role of brain-derived neurotrophic factor/tyrosine kinase receptor B signaling in head and neck malignancies. J Oral Pathol Med 2018;47:221–227.
57
27. Moriwaki K, Ayani Y, Kuwabara H, et al. TRKB tyrosine kinase receptor is a potential therapeutic target for poorly differentiated oral squamous cell carcinoma. Oncotarget 2018;9:25225-25243.
28. Vargas PA, Speight PM, Bingle CD, et al. Expression of PLUNC family members in benign and malignant salivary gland tumours. Oral Dis 2008; 14:613–619.
29. Fonseca FP, de Andrade BA, Rangel AL, et al. Tissue microarray is a reliable method for immunohistochemical analysis of pleomorphic adenoma. Oral Surg Oral Med Oral Pathol Oral Radiol 2014;117:81–88.
30. Dudás J, Bitsche M, Schartinger V, et al. Fibroblasts produce brain-derived neurotrophic factor and induce mesenchymal transition of oral tumor cells. Oral Oncol 2011;47:98─103.
31. Dillenburg CS, Martins MD, Meurer L, et al. Keratoacanthoma of the lip - Activation of the mTOR pathway, tumor suppressor proteins, and tumor senescence. Medicine (Baltimore) 2015;94:e1552.
32. Neto NFO, Batista AC, Weissheimer C, et al. Mucoadhesive Formulation with Curcumina Longa L. Extract accelerates wound healing in skin and oral mucosa ulcers. Oral Surg Oral Med Oral Pathol Oral Radiol 2018:126:e174
33. Naruse T, Yanamoto S, Okuyama K, et al. Therapeutic implication of mTORC2 in oral squamous cell carcinoma. Oral Oncology 2017;65:23–32. 34. Zheng G, Jia X, Peng C, et al. The miR-491-3p/mTORC2/FOXO1 regulatory loop modulates chemo-sensitivity in human tongue cancer. Oncotarget 2016;6:6931–6943. 35. Witsch E, Sela M, Yarden Y. Roles for growth factors in cancer progression. Physiology 2010;25:85–101.
36. Dissanayaka W, Pitiyage G, Kumarasiri PVR, et al. Clinical and histopathologic parameters in survival of oral squamous cell carcinoma. Oral Surg Oral Med Oral Pathol Oral Radiol 2012;113:518–525.
37. Li S, Sun Y, Gao D. Role of the nervous system in tumor metastasis. Oncology Letters 2013;5:1101–1111.
38. Jiffar T, Yilmaz T, Lee J, et al. Brain derived neutrophic factor (BDNF) coordinates lymphovascular metastasis through a fibroblast-governed paracrine axis in the tumor microenvironment. Cancer Cell Microenviron. 2017;4:e1566.
39. Lim J, Kim JH, Paeng JY, et al. Prognostic value of activated Akt expression in oral squamous cell carcinoma. J Clin Pathol 2005;58:1199–1205.
40.Oliveira L.R.; Ribeiro-Silva A. Prognostic significance of immunohistochemical biomarkers in oral squamous cell carcinoma. Int J Oral Maxillofac Surg. 2011;40:298-307
58
41. Vanneman M, Dranoff G. Combining immunotherapy and targeted therapies in cancer treatment. Nat Rev Cancer 2012;12:237–251.
42. Polverini PJ, D’Silva NJ, Lei YL. Precision therapy of head and neck squamous cell carcinoma. J Dent Res 2018;97:614–621.
43. Khotskaya YB, Holla VR, Farago AF, et al. Targeting TRK family proteins in cancer. Pharmacol Ther 2017;173:58–66.
44. Bailey JJ, Schirrmacher R, Farrell K, et al. Tropomyosin receptor kinase inhibitors: an updated patent review for 2010-2016 – Part I. Expert Opin on Ther Pat 2017;27:733–751.
45. Cazorla M, Prémont J, Mann A, Girard N, Kellendonk C, Rognan D. Identification of a low-molecular weight TrkB antagonist with anxiolytic and antidepressant activity in mice. J Clin Invest. 2011;121:1846-1857.
46. Mirzayans R, Andrais B, Hansen G, et al. Role of p16 ink4a in replicative senescence and DNA damage-induced premature senescence in p53-deficient human cells. Biochem Res Int 2012:951574.
47. Siddique HR, Saleem M. Role of BMI1, a stem cell factor, in cancer recurrence and chemoresistance: preclinical and clinical evidences. Stem Cells 2012;30:372–378.
48. Krishnamurthy S, Nör JE. Head and neck cancer stem cells. J Dent Res 2012;91:334–340.
49. Yin B, Ma ZY, Zhou ZW, Gao WC, Du ZG, Zhao ZH, Li QQ. The TrkB+ cancer stem cells contribute to post-chemotherapy recurrence of triple-negative breast cancers in an orthotopic mouse model. Oncogene. 2015;34:761-770.
50. Touil Y, Zuliani T, Wolowczuk I, et al. The PI3K/AKT signaling pathway controls the quiescence of the low-Rhodamine123-retention cell compartment enriched for melanoma stem cell activity. Stem Cells 2013;31:641–651.
51. Häyry V, Mäkinen LK, Atula T, et al. Bmi-1 expression predicts prognosis in squamous cell carcinoma of the tongue. Br J Cancer 2010;102:892–897.
52. Yamazaki H, Mori T, Yazawa M, Maeshima AM, Matsumoto F, Yoshimoto S, Ota Y, Kaneko A, Tsuda H, Kanai Y. Stem cell self-renewal factors Bmi1 and HMGA2 in head and neck squamous cell carcinoma: clues for diagnosis.Lab Invest. 2013;93:1331-1338
53. Tamatani T, Takamaru N, Ohe G, et al. Expression of CD44, CD44v9, ABCG2, CD24, Bmi-1 and ALDH1 in Stage I and II oral squamous cell carcinoma and their association with clinicopathological factors. Oncol Lett 2018;16:1133–1140.
59
Figures and Table Legends:
Table 1. Clinico-demographic features of OSCC patients included in the
present study (n=72).
Table 2. Clinico-demographic features of OL patients included in the present
study (n=33).
Table 3. Association between proteins involved in BDNF/TrkB pathway and
OSCC disease-specific survival estimated by univariate Cox regression hazard
regression model and multivariate Cox regression hazard regression model
controlled for age, gender and clinical stage.
Figure 1. BDNF/TrkB pathway in NOM, OL and OSCC. (A) Representative
images of BDNF, TrkB, p-TrkB, p-Akt and p-S6 in NOM, OL and OSCC. (B) Mean
expression of BDNF, TrkB, p-TrkB, p-Akt and p-S6 in NOM, OL and OSCC.
Different uppercase letters in columns denote significant difference (p<0,05,
Kruskal-Wallis test followed by post-hoc Tukey test adjusted for Bonferroni error
correction).
Figure 2. Stacked bar graphs illustrating significant associations of proteins
involved in BDNF/TrkB pathway with OSCC clinical features and outcomes. (A)
Association between BDNF status and nodal status at diagnosis. (B) Association
between p-TrkB status and TNM stage at diagnosis. (C) Association between BDNF
status and TNM stage at diagnosis. (D) Association between p-Akt status and
recurrence during follow-up. (D) Association between BDNF status and patients’
outcome during follow-up. (E) Association between p-Akt status and patients’
outcome during follow-up. Odds ratio (OR) and 95% confidence interval (CI)
determined by binary logistic regression and p value determined by Fisher’s exact
test.
60
Figure 3. Kaplan-Meier disease specific survival curves for OSCC patients.
(A) Overall disease specific survival. (A) Disease specific survival according to
BDNF status. (B) Disease specific survival according to TrkB status. (C) Disease
specific survival according to p-TrkB status. (D) Disease specific survival according
to p-Akt status. (D) Disease specific survival according to p-Akt status. Survival
curves were compared using the log-rank test.
Figure 4. BDNF/TrkB pathway correlation with cell senescence and
“stemness” profile. (A) Spearman correlation coefficients’ between BDNF/TrkB
pathway proteins and BMI1 (stem cell marker) and p16 (senescence marker). (B)
Scatter-plot between p-TkrB and BMI-1. Note that an increase in p-TrkB is
correlated with a decrease in BMI-1. (C) Scatter-plot between p-Akt and BMI1. Note
that an increase in p-Akt is correlated with a decrease in BMI-1.
61
Table 1. Clinico-demographic features of OSCC patients included in the
present study (n=72).
OSCC patients Frequency (%) N
Gender
Male
Female
90.3%
9.7%
65 07
Age (mean ± SD) 59.83 (±10.51) Tobacco user
Yes/former user
Never
96.7%
3.3%
69 03
Packyears - only for smokers (mean ± SD) 44.92 (±24.73) Alcohol user
Yes/former user
Never
76.3%
23.7%
55 17
Pain
Yes
No
93.2%
6.8%
67 05
Clinical aspect
Ulcer
Others*
91.9
8.1
66 06
Site
Tongue
Floor of the mouth
Palate
Gingiva
75.0%
11.1%
8.3%
5.6%
54 08 06 04
TNM
I/II
III/IV
13.8%
86.2%
10 62
Nodal metastasis
Yes
No
59.7%
40.3%
43 29
Treatment
Surgery
Surgery + Radiotherapy
Others**
26.8%
53.5%
19.7%
19 39 14
Recurrence
Yes
No
39.7%
60.3%
29 43
Outcome
Alive
Deceased
63.4%
36.6%
46 26
* Macule, patch or nodule.
** Radiotherapy, Chemotherapy, Surgery (14.1%); Chemotherapy + Surgery (1.4%); Radiotherapy +
Chemotherapy (1.4%); Radiotherapy (2.8%)
62
Table 2. Clinico-demographic features of OL patients included in the present study
(n=33).
OL patients Frequency (%) N
Gender
Male
Female
51.5%
48.5%
17 16
Age (mean ± SD) 59.83 (±10.51) Tobacco user
Yes/former user
Never
72.7%
27.3%
24 09
Alcohol user
Yes/former user
Never
27.3%
72.7%
09 24
Site
Tongue
Floor of the mouth
Palate
Gingiva
Buccal mucosa
Labial mucosa
7.1%
17.9%
17.8%
25.0%
28.6%
3.6%
02 06 06 08 10 01
Outcome
No lesion
Lesion Stable
Malignant transformation
Unknown
6.1%
18.1%
3.1%
72.7%
02 06 01 24
63
Table 3. Association between proteins involved in BDNF/TrkB pathway and OSCC
disease-specific survival estimated by univariate Cox regression hazard regression
model and multivariate Cox regression hazard regression model controlled for age,
gender and clinical stage.
HR – hazard ratio; CI – confidence interval.
64
Figure 1. BDNF/TrkB pathway in NOM, OL and OSCC. (A) Representative
images of BDNF, TrkB, p-TrkB, p-Akt and p-S6 in NOM, OL and OSCC. (B) Mean
65
expression of BDNF, TrkB, p-TrkB, p-Akt and p-S6 in NOM, OL and OSCC.
Different uppercase letters in columns denote significant difference (p<0,05,
Kruskal-Wallis test followed by post-hoc Tukey test adjusted for Bonferroni error
correction).
Figure 2. Stacked bar graphs illustrating significant associations of proteins
involved in BDNF/TrkB pathway with OSCC clinical features and outcomes. (A)
Association between BDNF status and nodal status at diagnosis. (B) Association
between p-TrkB status and TNM stage at diagnosis. (C) Association between BDNF
66
status and TNM stage at diagnosis. (D) Association between p-Akt status and
recurrence during follow-up. (D) Association between BDNF status and patients’
outcome during follow-up. (E) Association between p-Akt status and patients’
outcome during follow-up. Odds ratio (OR) and 95% confidence interval (CI)
determined by binary logistic regression and p value determined by Fisher’s exact
test.
67
Figure 3. Kaplan-Meier disease specific survival curves for OSCC patients (months)
(A) Overall disease specific survival. (A) Disease specific survival according to
BDNF status. (B) Disease specific survival according to TrkB status. (C) Disease
68
specific survival according to p-TrkB status. (D) Disease specific survival according
to p-Akt status. (D) Disease specific survival according to p-Akt status. Survival
curves were compared using the log-rank test.
Figure 4. BDNF/TrkB pathway correlation with cell senescence and “stemness”
profile. (A) Spearman correlation coefficients’ between BDNF/TrkB pathway
69
proteins and BMI1 (stem cell marker) and p16 (senescence marker). (B) Scatter-
plot between p-TkrB and BMI-1. Note that an increase in p-TrkB is correlated with a
decrease in BMI-1. (C) Scatter-plot between p-Akt and BMI1. Note that an increase
in p-Akt is correlated with a decrease in BMI-1.
70
4 CONCLUSÃO
A revisão da literatura encontrada, demonstrou que a via BDNF / TrkB induz
a proliferação e invasão de células tumorais do carcinoma espinocelular oral e de
cabeça e pescoço, estimula a transição epitélio-mesênquima (EMT) e a
angiogênese e está associada ao aumento da resistência à cisplatina. Em relação
aos tumores de glândulas salivares, os estudos demonstraram que a via
BDNF/TrkB está significativamente associada à invasão perineural, migração,
indução de transição epitélio-mesênquima e metástase. Além disso, o valor
prognóstico da expressão de TrkB foi observado tanto no CEC de cabeça e
pescoço como em alguns subtipos de neoplasias de glândulas salivares.
Outros achados importantes obtidos nesta tese mostraram o papel da via de
sinalização BDNF / TrkB / Akt na carcinogênese oral tendo em vista o aumento da
expressão desta via de sinalização no CCEO em comparação com leucoplasia e
mucosa bucal normal indicando que eles estão envolvidos na aquisição do fenótipo
maligno. Além disso, o aumento da disponibilidade de BDNF influenciou o
prognóstico do CCEO e está associado à redução do tempo de sobrevida dos
pacientes. Além disso, a expressão aumentada da via BDNF / TrkB / Akt foi
correlacionada com a modificação no perfil de células tronco tumorais diminuindo a
expressão de BMI 1 e correlacionada com mTORC2 observada pela baixa
expressão de p-RPS6 em CCEO. Com estes resultados, pode-se considerar que a
via BDNF / TrkB /Akt seja um importante alvo terapêutico de forma isolada ou
potencializando a quimioterapia disponível, na busca de tratamento mais efetivo
para CCEO. Assim sendo, novos estudos avaliando o efeito da inibição desta via
em modelos de carcinogênese oral em animais e em CCEO se fazem necessários.
71
REFERÊNCIAS1
Alyasiri NS, Mehdi SJ, Alam MS, Ali A, Mandal AK, Gupta S, et al. PTEN-mediated
AKT activation contributes to the reduced apoptosis among Indian oral squamous
cell carcinoma patients. J Cancer Res Clin Oncol 2012;138(1):103─109
Barde Y, Edgar D, Thoenen H. Purification of a new neurotrophic factor from
mammalian brain. EMBO J 1982; 1(5): 549─553.
Cabarcos YT, Sori BS, López AM, Castillo CAL. Caracterización clínico e
histopatológica de la leucoplasia bucal. AMC 2018; 22( 4 ): 432─451.
Camoratto AM, Jani JP, Angeles TS, Maroney AC, Sanders CY, Murakata C, Neff
NT, et al. CEP-751 inhibits TRK receptor tyrosine kinase activity in vitro exhibits
anti-tumor activity. Int J Cancer 1997;72(4):673─679.
Cohen Y, Goldenberg-Cohen N, Shalmon B, Shani T, Oren S, Amariglio N, et al.
Mutational analysis of PTEN/PIK3CA/AKT pathway in oral squamous cell
carcinoma. Oral Oncol 2011;47(10):946─950.
Curado MP, Hashibe M. Recent changes in the epidemiology of head and neck
cancer. Curr Opin Oncol 2009;21(3):194─200.
de Farias CB, Heinen TE, dos Santos RP, Abujamra AL, Schwartsmann G, Roesler
R. BDNF/TrkB signaling protects HT-29 human colon cancer cells from EGFR
inhibition. Biochem Biophys Res Commun 2012; 425(2): 328─332.
Dionne KR, Warnakulasuriya S, Zain RB, Cheong SC. Potentially malignant
disorders of the oral cavity: current practice and future directions in the clinic and
laboratory. Int J Cancer 2015;136(3):503─515.
__________________________________________________________________
1De acordo com as normas da UNICAMP/FOP, baseadas na padronização do International Committee of
Medical Journal Editors – Vancouver Group. Abreviatura dos periódicos em conformidade com o PubMed.
72
Fujikawa H, Tanaka K, Toiyama Y, Saigusa S, Inoue Y, Uchida K, et al. High TrkB
expression levels are associated with poor prognosis and EMT induction in
colorectal cancer cells. J Gastroenterol 2012;47(7):775─784.
Global Health Observatory (Globocan). Geneva: World Health Organization; 2018.
who.int/gho/database/en/. Accessed Nov 16, 2018.
Guo D, Hou X, Zhang H, Sun W, Zhu L, Liang J, et al. More expressions of BDNF
and TrkB in multiple hepatocellular carcinoma and anti-BDNF or K252a induced
apoptosis, supressed invasion of HepG2 and HCCLM3 cells. J Exp Clin Cancer Res
2011; 30:97.
Haddad, M.D., and Dong M. Shin, M.D. Recent Advances in Head and Neck
Cancer Robert I. N Engl J Med 2008;359(11):1143─1154
Instituto Nacional de Câncer José Alencar Gomes da Silva (INCA); 2018.
www.inca.gov.br. Accessed Nov 16, 2018.
Internacional Agency for Research on Cancer (IARC): World Health Organization;
2018. www.iarc.fr. Accessed Nov 16, 2018.
Jansen JFA, Parra C, Lu Y, Shukla-Dave A. Evaluation of head and neck tumors
with functional MRI. Magn Reson Imaging Clin N Am 2016;24(1):123–133.
Jia S, Wang W, Hu Z, Shan C, Wang L, Wu B, Yang Z, Yang X, Lei D. BDNF
mediated TrkB activation contributes to the EMT progression and the poor
prognosis in human salivary adenoid cystic carcinoma. Oral Oncol
2015;51(1):64─70.
Kim MS, Lee WS, Jeong J, Kim SJ, Jim W. Induction of metastatic potential by TrkB
via activation of IL6/JAK2/STAT3 and PI3K/AKT signaling in breast cancer.
Oncotarget 2015;6(37):40158–40171.
Kupferman ME, Jiffar T, El-Naggar A, Yilmaz T, Zhou G, Xie T, et al. TrkB induces
EMT and has a key role in invasion of head and neck squamous cell carcinoma.
Oncogene 2010; 29(14): 2047─2059.
Lai PC, Chiu TH, Huang YT. Overexpression of BDNF and TrkB in human bladder
73
cancer specimens. Oncol Rep 2010;24(5):1265─1270.
Le Campion ACOV, Santos KCB, Carmo ES, Silva Jr FF, Peixoto FB, Ribeiro CMB,
et al. Characterization of diagnostic delay in oral and oropharyngeal cancer at two
referral centers. Cad Saúde Colet 2016; 24(2):178─184.
Levi-Montalcini R. The nerve growth factor 35 years later. Science 1987
237(4819):1154─1162.
Liu C, Chan CB, Ye K. 7,8-dihydroxyflavone, a small molecular TrkB agonist, is
useful for treating various BDNF-implicated human disorders. Transl Neurodegener
2016;5:2.
Marur S, Forastiere A. Head and neck squamous cell carcinoma: Update on
epidemiology, diagnosis, and treatment. Mayo Clin Proc 2016; 91(3):386─396.
Minturn JE, Evans AE, Villablanca JG, Yanik GA, Park JR, Shusterman S, et al.
Phase I trial of lestaurtinib for children with refractory neuroblastoma: a new
approaches to neuroblastoma therapy consortium study. Cancer Chemother
Pharmacol 2011;68(4):1057─1065.
Norris RE, Minturn JE, Brodeur GM, Maris JM, Adamson PC. Preclinical evaluation
of lestaurtinib (CEP-701) in combination with retinoids for neuroblastoma.Cancer
Chemother Pharmacol 2011;68(6):1469─1475.
Odate S, Onishi H, Nakamura K, Kojima M, Uchiyama A, Kato M, et al.
Tropomyosin-related kinase B inhibitor has potential for tumor regression and
relapse prevention in pulmonary large cell neuroendocrine carcinoma. Anticancer
Res 2013;33(9):3699─3703.
Okamura K, Harada T, Wang S, Ijichi K, Furuyama K, Koga T, et al. Expression of
TrkB and BDNF is associated with poor prognosis in non-small cell lung cancer.
Lung Cancer 2012;78(1):100─106.
Roesler R, de Farias CB, Abujamra AL, Brunetto AL, Schwartsmann G. BDNF/TrkB
signaling as an anti-tumor target. Expert Rev Anticancer Ther
2011;11(10):1473─1475.
74
Sasahira T, Ueda N, Yamamoto K, Bhawai UK, Kurihara M, Kirita T, et al. Trks are
novel oncogenes involved in the induction of neovascularization, tumor progression,
and nodal metastasis in oral squamous cell carcinoma. Clin Exp Metastasis
2013;30(2):165–176.
Schecterson LC, Bothwell M. Neurotrophin receptors: Old friends with new partners.
Dev Neurobiol 2010;70(5):332─338.
Scully C, Bagan JV. Recent advances in oral oncology 2008; squamous cell
carcinoma imaging, treatment, prognostication and treatment Outcomes. Oral Oncol
2009;45(6):25─30.
Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin 2017;
67(1):07─30.
Thiele CJ, Li Z, McKee AE. On Trk--the TrkB signal transduction pathway is an
increasingly important target in cancer biology. Clin Cancer Res 2009 Oct
1;15(19):5962─5967.
Thomaz A, Jaeger M, Buendia M, Bambini-Junior V, Gregianin LJ, Brunetto AL, et
al. BDNF/TrkB Signaling as a potential novel target in pediatric brain tumors:
anticancer activity of selective TrkB inhibition in medulloblastoma cells. J Mol
Neurosci 2016; 59(3): 326─333.
Vidal L, Aharon IB, Limon D, Cohen E, Popovtzer A. Role of Induction
Chemotherapy Prior to Chemoradiation in Head and Neck Squamous Cell Cancer—
Systematic Review and Meta-analysis. Cancer J 2017; 23(2):79─83.
Waal I. Oral potentially malignant disorders: is malignant transformation predictable
and preventable? Med Oral Patol Oral Cir Bucal 2014;19(4):386─390.
Warnakulasuriya S, Johnson NW, van der Waal I. Nomenclature and classification
of potentially malignant disorders of the oral mucosa. J Oral Pathol Med
2007;36(10):575–80.
Warnakulasuriya S. Living with oral cancer: epidemiology with particular reference
to prevalence and life-style changes that influence survival. Oral Oncol.
75
2010;46(6):407-410.
Winn DM, Lee YC, Hashibe M, Boffetta P, Consortium I. The Inhance consortium:
toward a better understanding of the causes and mechanisms of head and neck
cancer. Oral Dis 2015;21(6):685─693.
Yamamoto M, Sobue G, Yamamoto K, Terao S, Mitsuma T. Expression of mRNAs
for neurotrophic factors (NGF, BDNF, NT-3, and GDNF) and their receptors (p75
NGvR, TrkA, TrkB, and TrkC) in the adult human peripheral nervous system and
nonneural tissues. Neurochem Res 1996;21(8):929─938.
Yilmaz T, Jiffar T, de la Garza G, Lin H, Milas Z, Takahashi, et al. Theraputic
targeting of trk supresses tumor proliferation and enhances cisplatin activity in
HNSCC. Cancer Biol Ther 2010;10(6): 644─653.
Zhu L, Werner JA, Mandic R. Implications of tropomyosin-related kinase B (TrkB) in
head and neck cancer. Anticancer Res 2007; 27(5A): 3121─3126.
76
ANEXOS
ANEXO 1 APROVAÇÃO COMITÊ DE ÉTICA EM PESQUISA
77
78
79
80
81
82
ANEXO 2 RELATÓRIO ORIGINALIDADE
83