Introduction: The treatment of squamous cell carcinoma of the tongue base has evolved with concomitant chemoradiation replacing surgery ± radiotherapy for advanced stages of disease. This study examines 10-year treatment outcomes in this patient population over the time-span of the changing treatment paradigm. Methods: A cohort of 290 patients was followed for 3442 months (median15 months). Survival analysis was done using Kaplan- Meier curves and log-rank test for comparing sub-groups. Cox’s proportional hazard models were used to determine the predictors of 10-year survival after treatment. Results: The mean age of the cohort was 62.2 years (SD=12.4 years), 79.7% were males, and 86.4% had Stage III or IV disease at presentation. The overall median survival time was 16 months (95% CI=9.5, 22.5 months) with 23% of patients surviving the 10-year period. The 10-year disease specific and disease free survival was similar at 30 and 31 months respectively. Survival varied significantly (P< 0.05) with stage of disease on presentation. Survival probability at 10 years was 37% for stage I disease and 26%, 28% and 23% for stages II -IV respectively. Patients younger than 65 years had better overall survival when compared to those 65 or more: 29% versus 14%, respectively (P<0.0001). Similarly, females had better 10-year survival as compared to males, 31% vs. 21%, respectively; however, this difference was not statistically significant (P>0.10). A lateral location of the tumor had a better survival outcome when compared to a midline location of (27% or 28% versus 9%, P<0.0025). With univariate analysis 10-year survival of 54% was observed with chemoradiation, and 45% for surgey+radiotherapy (P<0.0001). Multivariate models demonstrated an independent effect of stage, gender, age, and initial treatment modality on overall survival. Treatment with radiotherapy and chemotherapy reduced the risk of death over 10 years by 89% (HR=0.11; 95% CI=0.1, 0.2; P<0.0001) and surgery + radiotherapy reduced the risk of death over 10 years by 87% (HR=0.13; 95% CI=0.1, 0.2; P<0.0001). Conclusion: Independent of cancer stage, gender, and age, treatment modality predicts 10-year survival of patients with base of tongue cancer. Similar survival is observed following chemoradiation and surgey+radiotherapy.

Introduction: The treatment of squamous cell carcinoma of the tongue base has evolved with concomitant chemoradiation replacing surgery ± radiotherapy for advanced stages of disease. This study examines 10-year treatment outcomes in this patient population over the time-span of the changing treatment paradigm. Methods: A cohort of 290 patients was followed for 3442 months (median15 months). Survival analysis was done using Kaplan- Meier curves and log-rank test for comparing sub-groups. Cox’s proportional hazard models were used to determine the predictors of 10-year survival after treatment. Results: The mean age of the cohort was 62.2 years (SD=12.4 years), 79.7% were males, and 86.4% had Stage III or IV disease at presentation. The overall median survival time was 16 months (95% CI=9.5, 22.5 months) with 23% of patients surviving the 10-year period. The 10-year disease specific and disease free survival was similar at 30 and 31 months respectively. Survival varied significantly (P< 0.05) with stage of disease on presentation. Survival probability at 10 years was 37% for stage I disease and 26%, 28% and 23% for stages II -IV respectively. Patients younger than 65 years had better overall survival when compared to those 65 or more: 29% versus 14%, respectively (P<0.0001). Similarly, females had better 10-year survival as compared to males, 31% vs. 21%, respectively; however, this difference was not statistically significant (P>0.10). A lateral location of the tumor had a better survival outcome when compared to a midline location of (27% or 28% versus 9%, P<0.0025). With univariate analysis 10-year survival of 54% was observed with chemoradiation, and 45% for surgey+radiotherapy (P<0.0001). Multivariate models demonstrated an independent effect of stage, gender, age, and initial treatment modality on overall survival. Treatment with radiotherapy and chemotherapy reduced the risk of death over 10 years by 89% (HR=0.11; 95% CI=0.1, 0.2; P<0.0001) and surgery + radiotherapy reduced the risk of death over 10 years by 87% (HR=0.13; 95% CI=0.1, 0.2; P<0.0001). Conclusion: Independent of cancer stage, gender, and age, treatment modality predicts 10-year survival of patients with base of tongue cancer. Similar survival is observed following chemoradiation and surgey+radiotherapy.

Thesis (doctoral)--Karolinska Institutet, Stockholm, 1989. / Extra t.p. with thesis statement inserted. Includes bibliographical references.

Thesis (doctoral)--Karolinska Institutet, Stockholm, 1989. / Extra t.p. with thesis statement inserted. Includes bibliographical references.

Ergebnisse einer retrospektiven Studie an der Universität Würzburg: Patienten und Methoden: In einer retrospektiven Studie wurden Einflussfaktoren auf die Lokoregionäre Kontrolle, das Gesamtüberleben und das rezidivfreie Überleben von 106 Patienten, mit histologisch gesicherten Oropharynxkarzinomen (28 T1, 46 T2, 25 T3 und 7 T4 Tumore, mit lymphatischer Beteiligung in 78 Fällen), mit uni- und multivariaten Analysen untersucht. Das mediane Alter bei Primärdiagnose betrug 55 Jahre. Es wurde eine mediane Nachbeobachtungszeit 36 Monaten erreicht (zwischen 5 bis 126 Monate). In 18 Fällen (17%) konnte der Primärtumor in sano entfernt werden (Sicherheitsabstand > 3mm). In 34 Fällen (32%) bestand ein knapper Sicherheitsabstand (definitionsgemäß < als 3mm) und in 54 Fällen (51%) waren die Resektatränder nicht frei von Tumorzellen (R1 Resektion). Patienten, welche eine Chemotherapie aufgrund des erhöhten Rezidivrisikos erhielten, machten 24% (25 Patienten) des Patientenkollektivs aus. Behandlungskonzept Das Tumorbett des Primärtumors und die zervikalen lymphatischen Abflussgebiete erhielten mediane Bestrahlungsdosen von 56 Gy (2 Gy/ Behandlung, 5 Fraktionen pro Woche). Patienten mit R0-Resektion erhielten Bestrahlungsdosen von 56-60Gy. Bei Patienten mit knappen Resektatrand wurde das Tumorbett mit einer höheren Dosis von 60-66Gy bestrahlt, R1 Resektionen wurden mit einer Boost-Aufsättigung bis zu einer Gesamtdosis von 66-70 Gy behandelt. Patienten im UICC-Stadium 4, mit erhöhtem Rezidivrisiko, machten 24% (25 Patienten) des Patientenkollektivs aus. Diese Patienten erhielten je nach Nierenfunktion und Blutbild eine zusätzliche Chemotherapie mit Cisplatin (40mg/m² wöchentlich) in 1-4 Zyklen, sowie eine Boost-Aufsättigung des Tumorbettes bis zu einer Gesamtdosis von 66-70 Gy. Ergebnisse der univariaten Analysen mittels Kaplan-Maier Plot Verfahren: lokoregionäre Kontrolle Mit einer medianen Nachbeobachtungszeit von 36 Monaten wurde eine 5 Jahres Rezidivfreiheit bei 87% der Patienten erzielt. Davon wurden 80% der Rezidive innerhalb der ersten 24 Monate diagnostiziert. Bei Patienten mit R0 Status wurde in 16,7 % ein Rezidiv diagnostiziert, bei Patienten mit R1 Situation in 17% und bei Patienten mit knappen Resektatrand wurde nur in 6 % ein Rezidiv diagnostiziert. Als statistisch signifikanter Einflussfaktor des Rezidiv erwies sich nur das Gesamttumorvolumen. Gesamtüberlebensrate Es wurde eine 3- und 5- Jahresüberlebensrate von 75% und 66% erreicht. Die 5JÜR bezüglich der Radikalität der Resektion erreichte bei R0 Resektion 61%, 71% bei Patienten mit knappen Resektatrand und 65% bei R1 Situation. Bei Patienten mit einem T1 Tumorstadium ergab sich eine 5JÜR von 82%, bei T2 67%, bei T3 52% und für Patienten im T4 Stadium ergab sich eine 5JÜR von 43 %. Patienten mit N0-Status verzeichneten eine 5JÜR von 68%, mit N1-Status 82%, N2a,b-Status 68%, N2c-Status 36% und N3-Status ergab 43%. Patienten ohne adjuvante Chemotherapie erzielten eine 5JÜR von 69% und Patienten, die aufgrund des erhöhten Rezidivrisikos eine Chemotherapie erhielten, erreichten 56%. Die Einflussgröße der Rezidiventwicklung erbrachte eine 5JÜR von 13%, wogegen sie bei Patienten ohne Rezidiv 75% betrug. Patienten, welche einen 2. Tumor entwickelten, verzeichneten eine 5JÜR von 45% gegenüber 71% bei Patienten ohne 2.Tumor. Der Vergleich der Bestrahlungsdosen im Tumorbett ergab, dass Patienten mit einer Gesamtdosis unter/gleich 66Gy eine 5JÜR von 71% erreichten und 62% bei Gesamtdosen über 66Gy. Die 5JÜR bezüglich des Tumorvolumens des Primärtumors, inklusive der befallenen Lymphknoten, erbrachte in der ersten Gruppe von unter 10ml Tumorvolumen 77%, von 10 bis 20ml 83%, von 20 bis 50ml 52% und in der vierten Gruppe mit über 50ml Tumorvolumen 33%. Mit einem Grading von 2 wurde 69% und mit einem Grading von 3 wurde bei Patienten eine 5JÜR von 61% berechnet. In der univariaten Analyse mittels des Kaplan-Maier-Plot-Verfahrens, zeigte sich in der 5-Jahres Überlebenskurve eine Signifikanz der Einflussgrößen Tumorstadium (p-Wert 0,003), Rezidivereignis (p-Wert 0,000), 2.Tumor (p-Wert 0,001) und Tumorvolumen (p-Wert 0,000). Rezidivfreies Überleben Das rezidivfreie Überleben betrug nach 3 Jahren 68% und nach 5 Jahren 64%. Bezüglich der Radikalität der Resektion ermittelte man für Patienten mit R0 Resektion nach 5 Jahren ein rezidivfreies Überleben von 61%, 71% bei knappen Resektatrand und 61% bei R1 Situation. Patienten mit einem T1 Stadium erreichten ein 5 jähriges rezidivfreies Überleben in 82%, mit T2 Stadium 67%, mit T3 Stadium 48% und Patienten im T4 Tumorstadium erzielten 43 %. Patienten mit N0-Status verzeichneten ein 5 jähriges rezidivfreies Überleben von 64%, mit N1-Status 82%, N2a,b-Status 68%, N2c-Status 27% und ein N3-Status ergab 43%. Patienten ohne adjuvante Chemotherapie erreichten in 68% und Patienten, welche eine Chemotherapie erhielten, erreichten ein 5 jähriges rezidivfreies Überleben in 52%. Patienten, welche einen 2. Tumor entwickelten, verzeichneten eine 5 jähriges rezidivfreies Überleben von 45%, gegenüber 68% bei Patienten ohne 2.Tumor. Die Gegenüberstellung der Bestrahlungsdosen im Tumorbett ergab, dass Patienten mit einer Gesamtdosis unter/gleich 66Gy ein rezidivfreies 5-jähriges Überleben von 69% erreichten, hingegen Patienten mit mehr als 66Gy Bestrahlungsdosis 60% erzielten. Das 5 jährige rezidivfreie Überleben in Bezug auf das Tumorvolumen des Primärtumors, inklusive der befallenen Lymphknoten, erbrachte in der ersten Gruppe von unter 10ml Tumorvolumen 77%, von 10 bis 20ml 79%, in der dritten Gruppe von 20 bis 50ml 48% und in der vierten Gruppe mit über 50ml Tumorvolumen wurde nach 5 Jahren ein rezidivfreies Überleben von 33% verzeichnet. Mit einem Grading von 2, wurde 66% und mit einem Grading von 3 ergaben sich für die Patienten ein 5 jähriges rezidivfreies Überleben von 61%. In der univariaten Analyse mittels des Kaplan-Maier-Plot-Verfahren, zeigte sich in der Kurve für das 5-jährige rezidivfreie Überleben, eine Signifikanz der Einflussgrößen Tumorstadium (p-Wert 0,003), Lymphknotenstatus (p-Wert 0,048), Chemotherapie (p-Wert 0,047), 2.Tumor (p-Wert 0,003) und Tumorvolumen (p-Wert 0,000). Ergebnisse der multivarianten Analysen In einer multivariaten Cox-Regressions Analyse erwiesen sich die Einflussgrößen des Tumorstadiums und die Entwicklung eines 2. Tumors, bezüglich des Gesamt- und des rezidivfreien Überlebens, als statistisch signifikant. Das Tumorstadium konnte, in Bezug auf das Gesamtüberleben, eine Signifikanz von 0,015 ermittelt werden. Im Hinblick auf das rezidivfreie Überleben konnte ihm eine Signifikanz von 0,03 zugeschreiben werden. Die Einflussgröße des 2.Tumors ergab für das Gesamtüberleben eine Signifikanz von ebenfalls 0,015 und eine Signifikanz von 0,025 bezüglich des rezidivfreien Überlebens. Schlussfolgerung: Mit dem Therapiekonzept konnte eine Verbesserung der 5JÜR und des 5-jährigen rezidivfreien Überlebens erzielt werden. Die Patienten mit knappen Resektatrand wiesen durchweg bessere Ergebnisse auf als Patienten mit R0-Resektion. Als Konsequenz dieser Ergebnisse müsste man eine Angleichung des bisherigen Therapiekonzeptes der R0 Patienten an das der knapp resezierten Patienten vornehmen. Bei Patienten mit einem primär erhöhtem Rezidivrisiko, welche eine simultane Radiochemotherapie erhielten, erzielte man mit diesem Therapiekonzept eine Angleichung der 5JÜR an Patienten ohne dieses. Es zeigte sich hierbei in der multivariaten Analyse, sowohl beim Gesamtüberleben, als auch beim rezidivfreien Überleben kein statistisch signifikanter Unterschied (Gesamtüberleben p-Wert 0,064, rezidivfreies Überleben p-Wert 0,085). / Results of a retrospective study at the University of Wuerzburg: Patients and methods: In a retrospective study factors of influence on the locoregional control, the overall survival and the disease free survival of 106 patients with hisological approved oropharyngeal cancer (28 T1, 46 T2, 25 T3 and 7 T4 tumors with lymphatic involment in 78 cases) were tested with uni and multivariant anyalysis. The mediane age at the date of the primar diagnose was 55 years. A mediane follow up surveillance of 36 months could be achived (between 5 to 126 months). In 18 cases (17%) the primary tumor could be removed in sano (safety margin >3mm). In 34 cases (32%) were detected close resection margins (<3mm) and in 54 cases the resection margins were not free of tumor cells (R1 resection). Patients, who were treated with a chemotherapy, because of the increased recurrency risk, were 24% of the patient database. The concept of medical treatment: The tumorbed of the primary tumor and the cervical lymphatic drain received a radiation dose of 56 Gy (2 Gy/ treatment, 5 fractions a week). Patients with R0 resections received doses from 56 to 60 Gy. Patients with close resections margins received a higher dose of 60 to 66 Gy at the tumorbed and patients with R1 resections were treated with a boost up to 66 to 70Gy. Patients with UICC 4 status with an increased recurrency risk received an additional chemotherapy with cisplatin (40mg/m² a week) in 1 to 4 cycles as well a boost of the tumorbed up to 66 to 70 Gy. Results of the univariant analyses with Kaplan-Maier Plot method: Locoregional control With a median follow up surveillance of 36 months a 5 year disease free survival of 87% was achieved. 80% of the recurrencies were detected within the first 24 months. Patients with R0 status had in 16,7%, patients with R1 status in 17% and patients with close resection margins had in only 6% the diagnose of a recurrence. The only factor of influence with statistical significance was the overall tumor volume. Overall survival A 3 and 5 year overall survival rate of 75% and 66% was achieved. The 5 year survival rate considering the radicality of the resection was 61% with R0 resection, 71% with close resection margins and 65% with R1 situation. Patients with T1 tumor status had a 5 year overall survival rate of 82%, 67% with T2, 52% with T3 and patients with T4 status hat a 5 year survival rate of 43%. Patients with N0 status achieved a 5 year overall survival of 68%, with N1 status 82%, N2a/b status 68%, N2c status and patients with N3 status achieved 43%. Patients without an adjuvative chemotherapy achieved a 5 year survival rate of 69% and patients, who received a chemotherapy because of the increase risk of recurrence achieved a 56% 5 year survival rate. Patients who developed a recurrence hat a 5 year survival rate of 13%, whereas the survival rate of patients without a recurrence was 75%. Patients who developed a second primary tumor had a 5 year survival rate of 45%, versus 71% without a second primary tumor. The comparison of the radiation dose in the tumorbed recorded, that patients with an overall radiation dose of 66Gy or less achieved a 5 year survival rate of 71% and patients with more than 66 Gy radiation dose achieved 62%. The 5 year overall survival considering the overall tumor volume recorded in the first group with less than 10ml 77%, in the second group with 10 to 20ml 83%, in the third group with 20 to 50ml 52% and in the forth group with more than 50ml tumor volume 33% 5 year survival rate. Patients with a grading of 2 had a 69% 5 years survival rate and patients with a grading of 3 had a 61% 5 years survival rate. The univariate analysis with the Kaplan-Maier Plot method detected following factors of influence of the 5 year overall survival: tumor status (p-value 0,003), tumor recurrence (p-value 0,000), a second tumor (p-value 0,001) and the overall tumor volume (p-value 0,000). Disease free survival The disease free survival was 68% after 3 years and 64% after 5 years. Regarding the radicality of the resection for patients with R0 resection was detected a 5 years disease free survival of 61%, 71% for patients with close resection margins and 61% with R1 resection. Patients with a T1 tumor status achieved a 5 years disease free survival rate of 82%, with a T2 status 67%, with a T3 status 48% and patients with a T4 status achieved 43%. Patients with N0 status had a 5 years disease free survival of 64%, with N1 status 82%, with N2a/b status 68%, with N2c status 27% and patients with N3 status 43%. Patients without a chemotherapy achieved 68% and patients who received a chemotherapy had a 5 years disease free survival of 52%. With the development of a second tumor, patients had a 5 years disease free survival of 45%, versus 68% survival rate without a second tumor. The comparison of the radiation dose in the tumorbed resulted, that patients with an overall radiation dose of 66 Gy or less had a 5 years survival rate of 69%, whereas patients with more than 66Gy radiation dose had 60%. The 5 years disease free survival regarding the overall tumor volume of the primar tumor achieved in the first group with less than 10ml tumor volume 77%, from 10 to 20ml 79%, in the third group from 20 to 50ml 48% and in the fourth group with more than 50ml a 5 years disease free survival of 33% could be detected. With a grading of 2 the 5 years disease free survival rate was 66% and with a grading of 3 patients achieved a 61% rate. In the univariate analysis with the Kaplan-Maier Plat method following factors of influence were detected: tumor status (p-value 0,003), lymph node status (p-value 0,048), chemotherapy (p-value 0,047), second tumor (p-value 0,003) and the overall tumor volume (p-value 0,000). Results of the multivariate analysis In the cox regression analysis the tumor status and the development of a second tumor were detected as factors of influence of the 5 year overall survival rate and the 5 year disease free survival survival rate. Regarding to the overall survival rate for the tumor status was detected a statistical significance of 0,015 and regarding the disease free survival was detected a significance of 0,025. The factor of influence of the second tumor had a significance for the overall survival of 0,015 and a significance regarding the disease free survival of 0,025. Conclusion: With the concept of medical treatment there could be achieved an improvement of the 5 year overall survival and the 5 year disease free survival. Patients with close resection margins had throughout better results as patients with a R0 resection. The consequence of those results should be an adjustment of the present concept of medical treatment of the R0 patients to the concept of patients with close resection margins. Patients who received a simultaneous radiochemotherapy, because of the increased risk of recurrence, the present concept of medical treatment effected an adjustment of the 5 year overall survival rate to patients without an increased risk of recurrence. There was no statistical significance in the multivariate analysis at the overall survival as well as at the disease free survival (overall survival p-value 0,064, disease free survival p-value 0,085).

Squamous cell carcinoma (SCC) of the oropharynx affects nearly 50,000 individuals in the United States each year, and, with the rising incidence of the human papillomavirus (HPV), the number of patients diagnosed with SCC is expected to continue to grow (American Cancer Society 2018; Coughlan and Frick 2012). Oropharyngeal squamous cell carcinoma (OPSCC) has traditionally been treated with wide surgical extirpation often involving removal of portions of the oral cavity, pharynx, and jaw; this kind of surgery can be disfiguring and has been associated with significant post-operative complications (Brickman and Gross 2014). In the late 20th century, clinicians began favoring the use of chemoradiation therapy instead of surgery in an effort to spare patients the morbidity associated with surgical techniques at the time (Mercante et al. 2015). While chemoradiation offers excellent survival for patients with SCC, this therapeutic strategy has been observed to have its own debilitating post-treatment side effects (Hamilton and Paleri 2017). An important advancement in the management of OPSCC occurred about 20 years ago with the advent of transoral robotic surgery (TORS), a surgical technique that uses a robotic system to operate through the natural opening of the mouth. Proponents of TORS suggest that the technology improves on conventional surgery and may provide patients with functional outcomes superior to those seen with chemoradiation with no sacrifice in survival (Yeh et al. 2015; Hay et al. 2017). This review investigates the validity of the concept that TORS has significant advantages in the modern-day treatment of OPSCC. This report includes three components. First, the TORS technology, its advantages, and its drawbacks are explained. Second, relevant medical literature is reviewed to provide an understanding of the rationale for utilizing TORS in the treatment of OPSCC. Review and analysis of published reports show that TORS can provide patients with excellent post-operative function, good quality of life, and acceptable survival rates. Notable exceptions include patients with advanced disease. Third, this review discusses future studies that will better inform caregivers about the utility of TORS in the treatment of OPSCC. TORS is a relatively new technology that seems to offer the possibility of helping to improve the lives of patients with OPSCC.

Les infections à papillomavirus humains à haut-risque (HPV-HR) sont responsables de 100% des cancers cervico-utérins et de 50% des carcinomes de l'oropharynx. Les infections du col utérin sont généralement transitoires et bénignes. Cependant, en cas d'infection persistante, elles peuvent s'accompagner d'une progression vers des lésions (pré)cancéreuses du col utérin. Les facteurs viraux qui favorisent la persistance ou la clairance virale sont encore mal connus. Dans une première partie, nous avons étudié la méthylation des régions 3'L1 et LCR dans des frottis cervico-utérins HPV16+ / cytologie normale, prélevés chez 37 femmes qui présenteront soit une infection transitoire soit une infection persistante avec progression vers une lésion précancéreuse. Nous avons montré que certains îlots CpG présentaient des taux de méthylation différents en fonction de l'évolution ultérieure de l'infection HPV et pourraient être utilisés comme marqueurs prédictifs.Les cancers de l'oropharynx associés à une infection à HPV-HR présentent un pronostic plus favorable que ceux associés à une intoxication alcoolo-tabagique. Les mécanismes responsables de ce meilleur pronostic sont encore mal compris. Dans une seconde partie, nous avons caractérisé le statut HPV dans 202 cas de cancers de l'oropharynx. Au sein des 32 cancers présentant une infection HPV16 active (avec expression des ARNm E6/E7), nous avons étudié la méthylation des régions 3'L1 et LCR et l'intégration du génome viral. Nous avons montré que les niveaux de méthylation de certains sites CpG, notamment des sites de liaison à la protéine virale E2 (E2BS3 et E2BS4) sont fortement méthylés dans les formes épisomales et mixtes mais sont non méthylés dans les formes intégrées pures.Enfin dans une troisième partie, nous avons étudié des marqueurs de transition épithélio-mésenchymateuse (TEM) dans des modèles cellulaires et des cancers oropharyngés, en fonction du statut HPV. Une moindre TEM, connue pour être impliquée de façon importante dans les phénomènes d'invasion tumorale, pourrait expliquer le meilleur pronostic associé aux cancers HPV-induits. Nous avons montré que la vimentine, marqueur le plus représentatif de l'acquisition de capacités migratoires et invasives, est plus exprimée dans les cellules tumorales HPV positives que dans les cellules HPV négatives. Le pronostic différentiel observé entre les 2 étiologies des cancers oropharyngés ne serait donc pas corrélé à la TEM.Mots-clés : Cancer, HPV, col utérin, oropharynx, biomarqueurs, méthylation, intégration, TEM. / High-risk human papillomavirus (HR-HPV) infections are responsible for 100% of cervical cancers and 50% of oropharyngeal cancers. Cervical infections are usually transient and benign. Persistent infections may however progress and lead to cervical (pre)cancerous lesions. Viral factors contributing to persistence or clearance are poorly understood. In a fist part, we studied methylation of 3'L1 and LCR regions in HPV16+ / normal cytology cervical smears, taken in 37 women who will present either a transient infection, or a persistent infection with progression to a precancerous lesion. We demonstrated that some CpG sites harbored differential methylation rates in relation with later outcome of HPV infection and may be used as predictive biomarkers.HR-HPV related oropharyngeal cancers present a better prognostic that their alcohol-and-tobacco-induced counterparts. Mechanisms responsible for this better prognostic are mainly unknown. In a second part, we characterized the HPV status in 202 oropharyngeal cancers. In the 32 cases presenting with HPV16 active infections (expressing E6/E7 mRNA), we studied 3'L1 and LCR methylation and integration of viral genome. We showed that methylation rates of some CpG sites, especially in E2 binding sites (E2BS3 & 4), were strongly methylated in episomal and mixed forms but were unmethylated in purely integrated forms.In a third part, we studied epithelial-to-mesenchymal-transition (EMT) markers in cellular models and oropharyngeal cancers, according to HPV status. A lesser EMT, known as an important phenomenon implicated in tumor invasion, could explain the better prognostic associated with HPV-induced cancers. We showed that vimentin, the more representative marker for migration and invasion ability acquisition is more expressed in HPV+ tumor cells than in HPV-negative cells. So the differential prognosis observed between the 2 oropharyngeal cancer etiologies may not be linked to EMT.Key-words: Cancer, HPV, cervical cancer, oropharynx, biomarkers, methylation, integration, EMT.

L’objectif de ces travaux est de connaître et évaluer les caractéristiques épidémiologiques des cancers de la cavité orale et de l’oropharynx. Ces deux localisations partagent des facteurs de risque en commun, et sont de fait souvent regroupées dans les études épidémiologiques. Cependant, la découverte de facteurs de risque spécifiques, telle l’infection par le virus du papillome humain pour les cancers de l’oropharynx, nous conduit à fournir des taux d’incidence spécifiques avec la classification anatomique de ces cancers. En réorganisant les données disponibles dans la base des données du Centre International de Recherche sur le Cancer, nous avons recherché les cas incidents au niveau mondial et recalculé les taux d’incidence dans les registres de 60 pays, pendant la période 1998-2002. La classification TNM n’est pas disponible dans les bases de données du CIRC. Nous avons identifié et contacté les registres du cancer qui ont déclaré son recueil. Cela nous a permis de créer et structurer une base des données innovante et inédite, dont les informations ont été analysées par rapport à la qualité. Finalement nous avons comparé la distribution de stades précoces et avancés dans 8 pays. Les résultats montrent que l’incidence des cancers de la cavité buccale et de l’oropharynx est très hétérogène au niveau mondial par rapport à la sous localisation des tumeurs, à l’âge d’incidence, au ratio homme/femme et au stade clinique. / Oral cavity and oropharynx cancers : International incidence and TNM classification in population-based cancer registries The aim of this work was to know and to evaluate the epidemiological patterns of oral cavity and ororpharynx cancers. These topographies share some common risk factors and they are often grouped in epidemiological studies. However, the implication of the human papilloma virus in oropharyngeal tumors lead us to provide incidence rates according to the anatomical classification of these tumors. We reorganized the incidence data available at the International Agency for Research on Cancer, for the period 1998-2002. Incidence rates were calculated for oral cavity and oropharynx cancers separately for 60 countries. As the TNM classification is not available on the IARC database we contacted the cancer registries that declared to abstract and collect it. Based on their data we created and structure a new, innovative and quality controlled. Finally, we compared the TNM stage distribution among 8 countries. The results show that the oral cavity and oropharynx cancers have a very heterogeneous distribution in the studied registries concerning tumor sub-sites, age of incidence, male to female ratio and clinical stage.

Chaque année 610 000 cancers sont diagnostiqués dans le monde induits par une infection à papillomavirus humains à haut-risque (HPV-HR). Bien que les carcinomes des voies aéro-digestives supérieures (VADS) soient principalement associés à une forte consommation de tabac et d’alcool, 20 à 25% sont causés par une infection à HPV, particulièrement l’HPV de type 16. Les patients HPV positifs présentent un meilleur survi global, pourtant ils sont diagnostiqués avec plus de métastases à distance que les patients HPV négatifs. Au travers d’une étude sur des lignées cellulaires dérivées des VADS, nous avons montré que toutes les lignées cellulaires HPV+ présentaient une intégration du génome d’HPV au sein du génome cellulaire, avec des profils d’intégration différents. Les lignées pouvant être utilisées comme modèles caractéristiques des tumeurs HPV+ et HPV- sont respectivement les lignées UPCI:SCC090 et FaDu. La première par ses capacités migratoire et proliférative et la seconde par sa faible agressivité et une mutation du gène cellulaire p53. Dans une étude portant sur une série rétrospective de cancers de l’oropharynx éligible à une résection chirurgicale, 6 cancers sur 40 soit 15% présentaient une infection à HPV16 active (expression de l’ARNm E6*I). Nous avons étudié les marqueurs de TEM dans ces cancers oropharyngés en fonction du statut HPV. Nous avons retrouvé une perte plus importante du marqueur épithélial cadhérine-E au sein du groupe HPV+, associée à une moins bonne survie globale.Au total, nous montrons que le statut HPV et les marqueurs de TEM semblent être deux facteurs indépendants, qui peuvent se combiner pour définir des niveaux pronostiques différents. / Each year, 610,000 cancers are diagnosed worldwide attributed to high risk human papillomavirus (HR-HPV) infection. Although head and neck squamous cell carcinoma (HNSCC) is mainly associated with tobacco and/or alcohol consumption, 20 to 25% are caused by HPV infection, particularly HPV type 16. Although patients with HPV+ tumors present a better overall survival, they are diagnosed with more lymph node metastasis than HPV-negative patients.Through a study of HNSCC derived cell lines, we showed that all HPV-positives cell lines harbored HPV genome integration through host genome, with different integration profiles. Cell lines identified as good HPV+ and HPV- tumors models are UPCI:SCC090 and FaDu respectively. The first one by its migratory and proliferative properties, the second through its poor aggressiveness and mutation of p53 cellular gene.In a study on a retrospective series of oropharyngeal carcinomas with surgical resection, 6 out of 40 cancers shown HPV16 active infection (expressing E6*I mRNA). We studied epithelial-to-mesenchymal transition (EMT) markers on this oropharyngeal cancers, according to HPV status. We found a larger loss of epithelial marker E-cadherin in HPV+ group and loss of this marker is associated with a worse overall survival.We showed that HPV and EMT status seem to be two independent factors that could combine differently to define different prognostic levels.

A avaliação da presença de destruição óssea provocada por neoplasias malignas de boca e orofaringe é um fator de fundamental importância no estabelecimento da terapêutica adequada para o caso, como também, para a determinação do prognóstico do paciente. O presente estudo teve por objetivo determinar os aspectos clínicos (localização, forma de apresentação e estadiamento) que podem estar associadas com o potencial de infiltração do osso subjacente a lesão, como também determinar a sensibilidade e especificidade do exame físico. A população de estudo consistio de vinte e cinco pacientes (17 homens e 8 mulheres, média de idade de 57,88 anos) portadores de neoplasias malignas de boca e orofaringe atendidos no Ambulatório de Semiologia da Faculdade de Odontologia da Universidade de São Paulo – campus São Paulo, no período de agosto de 2003 a agosto de 2004, os quais foram submetidos ao exame clínico e a tomografia computadorizada (TC). A TC foi considerada o padrão ouro para a avaliação da presença de destruição óssea. Foi observada a presença de infiltração neoplásica para o tecido ósseo adjacente em 68% dos casos (17 pacientes). O exame físico dos pacientes revelou uma sensibilidade de 80% e especificidade de 87,50% na análise de comprometimento do osso, além de uma acurácia de 84%. As lesões que se apresentavam clinicamente como uma úlcera do tipo infiltrativa e lesões do tipo nodulares, não ulceradas, foram as que apresentaram maior potencial de infiltrar-se para o osso, 68,75% e 100% respectivamente. A localização do tumor em determinados sítios, também influenciou diretamente na presença de invasão óssea,principalmente lesões localizadas em região de gengiva, trígono retromolar, palato duro e orofaringe. O estadiamento das lesões revelou relação existente entre o tamanho do tumor e a presença de metástases à distância com a presença de infiltração da neoplasia para o tecido ósseo. Concluindo, observou-se que a identificação de determinados parâmetros clínicos como localização, forma de apresentação clinica, tamanho da lesão e a presença de metástases à distância, associado a um criterioso exame físico regional podem servir como valiosas ferramentas para a análise de envolvimento ósseo por neoplasias malignas de boca e orofaringe. / The assessment of bone destruction by oral and oropharynx malignant neoplasms is a critical factor in the therapeutic planning and to determine the patient prognostic. The aim of this study was to determine the clinical aspects (localization, clinical manifestation and stage) that can be associated with the potential of bone infiltration, and also determine the physical exam sensibility and specificity. The study population consisted of twenty five patients (17 men and 8 women, mean age 57.88 years-old), with malignant neoplasms of the mouth and oropharynx, of the Stomatology Clinic of the College of Dentistry at the Sao Paulo University - campus Sao Paulo, in the period of august 2003 to august 2004, who were submitted to a clinical and computed tomography (CT) examinations. CT was considered the gold standard to evaluate the presence of bone involvement. The presence of bone destruction by the tumor was observed in 68% of the cases (17 patients). The physical examination of the patients revealed 82% of sensibility, 87.50% of specificity, and 84% of accuracy in the assessment of bone invasion by these diseases. The lesions that were clinical considered to be infiltrative ulcer and nodular lesions, non-ulcerated, presented the highest potential to cause bone destruction, 68.75% and 100% respectively. The tumor localization in specific sites also influenced the presence of bone invasion, meanly with the lesions localized in the gingival, retromolar trigone, hard palate and oropharynx. The stage of the lesions revealed a relation between the size and the presence of distant metastasis, with the presence of invasion by the neoplasm. In conclusion, it was determined that the identification of some clinical parameters such localization, clinical presentation, lesion size and the presence of distant metastasis, associated with a perceptive regional physical exam must be use as a value tool is the assessment of bone destruction by oral and oropharynx malignant neoplasms.