Introduction: Sinonasal inverted papilloma (SNIP) is an uncommon benign tumor with a high recurrence rate, significant malignant potential and unknown etiology. The population in Hong Kong is unique in its high population density and having a majority of Chinese people who are ethnically and geographically predisposed to certain cancers. Research on the etiology, pathogenesis, diagnosis and treatment of the neoplasm and comparison with reported findings from other parts of the world may contribute to management of the condition in terms of prevention, staging and treatment. / Aims: The aim of the thesis is to describe the common and unique clinico-pathological characteristics of SNIP in Hong Kong and compare these with reported characteristics in populations from other geographical areas and in other races with the expectation of attaining new insights into the diagnosis and management of SNIP in Hong Kong patients. / Methods: Four studies designed to evaluate the risk factors, viral associations, cell-cycle protein expression, radiological features, clinical features, treatment approaches and treatment outcomes were conducted. The findings of these studies were compared with those reported from different geographical areas of the world. Study 1: Evaluation of the risk factors associated with SNIP by a case-controlled epidemiological study of 50 patients with SNIP and 150 matched control patients. Study 2: Evaluation of the prevalence of human papillomavirus (HPV), Epstein-Barr virus (EBV), p21 and p53 expression in SNIP and comparison with reports from the literature. In a case-control study, 73 SNIP, 48 nasal polyps (NP) and 85 hypertrophied turbinates (HT) specimens were examined by polymerase chain reaction (PCR) for HPV. Seventy-three SNIP, 30 NP and 32 HT specimens were examined by in-situ hybridization (ISH) for EBV and by immunohistochemistry (IHC) for p21 and p53. SNIP results were compared with those of NP and HT (as controls). Study 3: Evaluation of the radiological signs, accuracy of prediction of tumor origin and extent, and accuracy of preoperative staging of SNIP of plain computed tomography by an observational study of plain CT scans and operative findings from 30 patients with SNIP. Study 4: Evaluation of the clinico-pathological features and treatment outcomes of SNIP in 56 patients seen between 1990 and 2008 with follow-up of more than 2 years and comparison with the results of the literature. / Results and conclusions: There are certain unique clinico-pathological features of sinonasal inverted papilloma (SNIP) in Hong Kong which are related to its predominantly Chinese population, high population density, heavy pollution and, accessible and efficient specialist services. Concordant with the results of another case-control study in the literature, the study described herein demonstrated that occupational chemical exposure, but not smoking, is a risk factor for SNIP. This is the first case-control study demonstrating that alcohol intake, allergic rhinitis, nasal polyposis, sinusitis, non-sinonasal papilloma and non-sinonasal malignancy are not risk factors for SNIP. The low prevalence of HPV in non-malignant sinonasal inverted papillomas (NMIPs) in Hong Kong suggests that it does not play a significant pathogenic role. The absence of EBV in SNIPs in Hong Kong concurs with most reports that EBV is not a causative agent. The high p21 and low p53 expression in SNIPs compared with the average values reported from other studies further support the presence of a non-p53-dependent p21 regulatory pathway. The higher prevalence of both HPV and p53 in malignant sinonasal inverted papilloma (MIP) than in NMIP agrees with other reports that both could be markers of malignant transformation. However, their inverse relation suggests they are independent factors. Although most plain CT signs are the same as those previously reported and not pathognomonic for SNIP, the high predictive value of the “pedunculation sign and absence of intra-tumor calcification are unique to Hong Kong patients. Concordant with other reports, “bony strut or focal hyperostosis is highly accurate in predicting the site of SNIP origin. This is the first report on the accuracy of preoperative CT staging, which is slightly lower than that of preoperative MRI staging (80% versus 86%). The estimated incidence of SNIP (2.4/1,000,000/year) is low but may be an underestimation as the number from the private sector is undetermined. The male:female ratio of SNIP patients in both Hong Kong and Asia is low, suggesting a geographic or racial influence on sex predilection. The absence of extrasinonasal extension, low rates of cellular atypia, dysplasia and synchronous malignancy in the Hong Kong SNIPs may reflect less aggressive tumor behavior as well as accessibility to efficient specialist services. The distribution of tumor origins, presenting symptoms and presenting stages of the Hong Kong SNIPs are similar to those reported elsewhere. The higher recurrence rate in the Hong Kong series is related to inadequate treatment of the tumor origin and inadequate conversion to combined external approaches in the early cases. Contrary to previously reported statistics, combined extranasal approaches were used more often in secondary cases than in primary cases. As in previously reported series, the recurrence rate in secondary cases tended to be higher than that in primary cases. Concordant with previous reports from the endoscopic era, most recurrences in Hong Kong occurred at the original tumor site and were discovered within the first 2 years after surgery. The average time of diagnosis of the first recurrence was much shorter than that of the pre-endoscopic era (1.2 years vs. 4.3 years). As reported elsewhere, about one-third of recurrences required combined external approaches for salvage. This is the first report comparing 2-, 5- and 10-year-follow-up results, and suggesting a minimum of 2 years’ follow-up before reporting results to avoid underestimation of recurrences. / Sham, Cheuk-lun. / Thesis (M.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 215-237). / CHAPTER 1 / Introduction / Chapter 1.1 --- Sinonasal inverted papilloma --- p.1 / Chapter 1.1.1 --- Nomenclature and classification --- p.1 / Chapter 1.1.2 --- Etiology and pathogenesis --- p.4 / Chapter 1.1.3 --- Gross and histological pathology --- p.9 / Chapter 1.1.4 --- Site of tumor attachment and extension --- p.10 / Chapter 1.1.5 --- Association with malignancy --- p.11 / Chapter 1.1.6 --- Incidence and demographics --- p.11 / Chapter 1.1.7 --- Clinical features --- p.12 / Chapter 1.1.8 --- Radiological features --- p.13 / Chapter 1.1.9 --- Staging --- p.14 / Chapter 1.1.10 --- Treatment modalities --- p.16 / Chapter 1.1.11 --- Treatment outcomes, recurrence and residual disease --- p.17 / Chapter 1.2 --- Unique characteristics of the Hong Kong population --- p.19 / Chapter 1.3 --- Research aims, areas and hypotheses --- p.22 / Chapter 1.3.1 --- Research aims --- p.22 / Chapter 1.3.2 --- Research areas and hypotheses --- p.23 / Chapter 1.4 --- Research plan and methodology --- p.25 / Chapter 1.5 --- Summary of Chapter 1 --- p.28 / CHAPTER 2 / Risk factors associated with SNIP / Chapter 2.1 --- Background --- p.29 / Chapter 2.2 --- Hypothesis --- p.29 / Chapter 2.3 --- Patients and methods --- p.30 / Chapter 2.4 --- Results --- p.35 / Chapter 2.5 --- Discussion --- p.39 / Chapter 2.6 --- Summary of Chapter 2 --- p.45 / CHAPTER 3 / Evaluation of the prevalence of HPV, EBV, p21 and p53 expression in SNIP in Hong Kong and comparison with results reported in the literature / Chapter 3.1 --- Background --- p.46 / Chapter 3.2 --- Hypothesis --- p.48 / Chapter 3.3 --- Patients and methods --- p.48 / Chapter 3.4 --- Results --- p.55 / Chapter 3.4.1 --- Overall results --- p.55 / Chapter 3.4.2 --- Comparison of the results of HPV studies --- p.57 / Chapter 3.4.3 --- Comparison of the results of EBV studies --- p.68 / Chapter 3.4.4 --- Comparison of the results of p21 studies --- p.71 / Chapter 3.4.5 --- Comparison of the results of p53 studies --- p.74 / Chapter 3.5 --- Discussion --- p.79 / Chapter 3.5.1 --- HPV and SNIP --- p.79 / Chapter 3.5.2 --- EBV and SNIP --- p.95 / Chapter 3.5.3 --- p21 and SNIP --- p.98 / Chapter 3.5.4 --- p53 and SNIP --- p.103 / Chapter 3.6 --- Summary of Chapter 3 --- p.115 / CHAPTER 4 / Evaluation of the radiological signs, accuracy of prediction of tumor origin and extent, and accuracy of preoperative staging of SNIP by plain computed tomography / Chapter 4.1 --- Background --- p.117 / Chapter 4.2 --- Hypothesis --- p.118 / Chapter 4.3 --- Patients and methods --- p.118 / Chapter 4.4 --- Results --- p.120 / Chapter 4.5 --- Discussion --- p.135 / Chapter 4.6 --- Summary of Chapter 4 --- p.142 / CHAPTER 5 / Evaluation of the clinico-pathological features and treatment outcomes of SNIP and comparison with results reported in the literature / Chapter 5.1 --- Background --- p.143 / Chapter 5.2 --- Hypothesis --- p.144 / Chapter 5.3 --- Patients and methods --- p.144 / Chapter 5.4 --- Results --- p.148 / Chapter 5.4.1 --- Incidence --- p.148 / Chapter 5.4.2 --- Demographics --- p.150 / Chapter 5.4.3 --- Presenting symptoms --- p.154 / Chapter 5.4.4 --- Site of tumor origin --- p.156 / Chapter 5.4.5 --- Rate of association with malignancy --- p.158 / Chapter 5.4.6 --- Staging of disease (Krouse system) and recurrence rate --- p.163 / Chapter 5.4.7 --- Treatment approaches and recurrence rates --- p.165 / Chapter 5.4.8 --- Comparison between patients with and without previous surgery --- p.167 / Chapter 5.4.9 --- Time and site of recurrence --- p.170 / Chapter 5.4.10 --- Surgical approaches used in salvage surgery --- p.175 / Chapter 5.4.11 --- Complication rate --- p.176 / Chapter 5.5 --- Discussion --- p.178 / Chapter 5.5.1 --- Incidence --- p.178 / Chapter 5.5.2 --- Demographics --- p.179 / Chapter 5.5.3 --- Present symptoms and duration --- p.180 / Chapter 5.5.4 --- Sites of tumor origin --- p.181 / Chapter 5.5.5 --- Association with malignancy --- p.182 / Chapter 5.5.6 --- Disease stages (Krouse system) and recurrence rate --- p.184 / Chapter 5.5.7 --- Treatment approaches and recurrence rates --- p.186 / Chapter 5.5.8 --- Comparison between patients with and without previous surgery --- p.187 / Chapter 5.5.9 --- Time and site of recurrence --- p.189 / Chapter 5.5.10 --- Surgical approaches used in salvage surgery --- p.191 / Chapter 5.5.11 --- Complication rate --- p.192 / Chapter 5.5.12 --- Management principles based on clinico-pathological features --- p.193 / Chapter 5.6 --- Summary of Chapter 5 --- p.197 / CHAPTER 6 / Summary of thesis and future perspective --- p.201 / REFERENCES --- p.215 / APPENDIX / Questionnaire for study of risk factors of SNIP --- p.238