SEARCH

SEARCH BY CITATION

Keywords:

  • squamous cell carcinoma;
  • paranasal sinus;
  • incidence and survival of sinonasal cancers

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

BACKGROUND

Paranasal sinus squamous cell carcinomas (PNSSCC) account for 3% of all head and neck malignancies. There has been little information on the trends in incidence and survival, and no randomized trials have been conducted to guide therapy.

METHODS

Patients with PNSSCC reported to the Surveillance, Epidemiology, and End Results (SEER) Program from 1973 through 2009 were categorized by sex, age, year of diagnosis, primary site, stage, and treatment. The incidence and survival were then compared across different demographic and disease-related categories by calculating rate ratios (RRs) and mortality hazard ratios along with the corresponding 95% confidence intervals (CIs).

RESULTS

In total, 2553 patients with PNSSCC were identified. While incidence of PNSSCC showed a gradual decline, survival remained largely unchanged. The proportion of patients with advanced disease decreased from 14.7% during the period from 1983 to 1992 to 12.4% during 1993-2002 and to 9.5% during 2003-2009. Compared with whites, incidence was higher among African Americans (RR 1.63; 95% CI, 1.39, 1.90) and among all other racial groups (RR, 1.78; 95% CI: 1.53-2.07). After adjusting for age, sex, disease stage, tumor site, and treatment, mortality among African American patients also was increased (hazard ratio, 1.22; 95% CI, 1.04-1.43). Among patients with localized disease, the relation between race and mortality was no longer evident once the results were controlled for tumor classification.

CONCLUSIONS

The current findings point to racial disparities in the incidence of PNSSCC and, to a lesser extent, in the outcome of patients with PNSSCC. Although there has been a decline in the proportion of patients presenting with advanced PNSSCC, the overall survival remained stable over time. Cancer 2013;119:2602–2610. © 2013 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Malignant tumors originating from the nose or the paranasal sinuses are rare,[1-4] representing 3% of all malignancies of the head and neck region.[1-3, 5, 6] They arise in air-filled cavities, which usually become infiltrated by tumor before signs and symptoms develop; therefore, most patients at diagnosis present with advanced stage disease and have extensive involvement of adjacent sites.[7] The tumor sites are the maxillary, ethmoid, frontal, sphenoid, and accessory sinuses and the overlapping accessory sinuses. These tumor sites are at close proximity to the orbit, base of skull, and the central nervous system, which presents difficulty in the management[3, 7] and accounts for the poor outcomes and low 5-year survival.[8] Most patients with paranasal sinus squamous cell carcinoma (PNSSCC) undergo surgical resection and receive irradiation. Histologic types of the malignant tumors of the paranasal sinuses include squamous cell carcinomas, adenocarcinoma, adenoid cystic carcinoma, mucoepidermoid carcinoma, transitional cell carcinoma, undifferentiated carcinoma, and verrucous carcinoma. All published literature agrees that PNSSCC, the focus of our current analysis, is the most predominant type of solid paranasal sinus cancer.[1, 3, 5, 6, 8]

There is a scarcity of prospectively collected data addressing management options and outcomes because of the rarity of this disease.[9] Prognostic factors for PNSSCC include disease stage, site of origin, histopathology, and patient characteristics.[9]

Relatively high rates of PNSSCC are reported in Asia and Africa, especially among Japanese men[10, 11] and in individuals with environmental and occupational exposure to wood dust, nickel, industrial fumes, and textile dust.[5, 6, 11] High levels of textile dust and asbestos and formaldehyde have been associated with PNSSCC.[12, 13] Chronic sinusitis, allergies, nasal polyposis, and tobacco smoking also are suggested risk factors.[14, 15]

There has been little information describing the incidence and survival of PNSSCC; and, to our knowledge, there have been no randomized trials to guide therapy. In this study, we analyzed the incidence trend and survival outcomes for patients diagnosed with PNSSCC in the Surveillance, Epidemiology, and End Results (SEER) database.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

Data were obtained from the National Cancer Institute's SEER Program, which is the main source for cancer statistics in the United States and includes information on incidence, prevalence, and survival from specific geographic areas representing 28% of the US population. SEER also compiles reports on cancer mortality for the entire country.[16] SEER data collection began in the early 1970s and gradually expanded. Data for most analyses were obtained from the currently operating 18 SEER registries; however, incidence rate calculations were limited to the 9 original registries (San Francisco/Oakland, Connecticut, Detroit, Hawaii, New Mexico, Seattle, Utah. and Atlanta) to evaluate changes in incidence since 1973.

We used the International Classification of Diseases for Oncology, third edition (ICD-O-3) to select all cases reported to the SEER Program from 1973 through 2009 that originated from 1 of the following primary sites (and ICD-O-3 codes): maxillary sinus (C.31.0); ethmoid sinus (C.31.1); frontal sinus (C.31.2); sphenoid sinus (C31.3); overlapping lesion of accessory sinuses (C.31.8); and accessory sinus, not otherwise specified (C31.9). Only invasive malignancies classified as squamous cell carcinomas (ICD-O-3 codes 8050 through 8089) were included. Cases were excluded from the analyses if they were of unknown age, identified from death certificates only, or were not microscopically confirmed.

All patients with PNSSCC reported during the study period from 1973 to 2009 were characterized according to sex, race, age at diagnosis, and disease stage. Tumor-directed therapy was grouped into 4 categories: none, surgery alone, radiation alone, and surgery plus radiation. Because treatment information was available from 1983 onward, only patients who were diagnosed from 1983 through 2009 were examined in the analyses involving type of tumor-directed therapy.

Cancer staging has undergone multiple changes since the early 1970s; therefore, we used the “SEER historic stage A” variable, because it covered the entire study period. The SEER historic stage A classification scheme categorizes cancer cases as localized, regional, distant, or unstaged based on the following definitions[16]:

  • Localized cancer is cancer that is limited to the organ in which it began without evidence of spread;
  • Regional cancer is cancer that has spread beyond the original (primary) site to nearby lymph nodes or organs and tissues;
  • Distant cancer is cancer that has spread from the primary site to distant organs or distant lymph nodes; and
  • Unstaged cancer is cancer for which there is not enough information to indicate a stage.

Incidence rates of PNSSCC were calculated for the entire study period by sex and race and were expressed as the number of cases per 1,000,000 individuals per year accompanied by 95% confidence intervals (CIs). The differences in rates between men and women and among whites, African Americans, and individuals of all other racial groups were examined by calculating rate ratios (RRs) and the corresponding 95% CIs, as described by Tiwari et al.[17] Changes in incidence over time were expressed as the annual percentage change (APC) and were examined further using the Joinpoint Regression Program (Control and Population Sciences, National Cancer Institute, Bethesda, Md). The APC was calculated by fitting a least-squares regression line to the natural logarithm of the rates using the calendar year as a regressor variable. The incidence data were further analyzed using joinpoint models, which are aimed at evaluating longitudinal data for a change in trend. Changes in trend were tested for statistical significance using a Monte Carlo permutation method.

The survival of patients with PNSSCC was analyzed using several methods, and the total follow-up was extended to 10 years (120 months) postdiagnosis. First, we calculated 5-year and 10-year cumulative observed survival (OS), which was defined as the proportion of patients who survived beyond a given interval. We then calculated relative survival (RS), which was obtained by dividing OS among cancer patients by the expected survival in the general population with the same age, race, and sex characteristics. Unlike frequency analyses, which included all records, survival analyses excluded cases that represented second or later primary cancer. Patients known to be alive but without documented survival time also were excluded. The 5-year and 10-year OS and RS estimates were evaluated overall and by sex and race. In addition, 5-year RS estimates were examined over time using joinpoint regression, and the resulting trend was expressed as the APC and its 95% CI. These survival calculations were based on the actuarial life-table method. The follow-up data for survival analyses extended through the end of 2009.

After OS and RS calculations, we constructed Kaplan-Meier curves with corresponding log-rank tests for statistical significance to examine patient survival according to PNSSCC stage, type of tumor-directed therapy, and decade of diagnosis. The follow-up for survival analyses by decade was limited to 5-years to allow comparisons between earlier and later intervals. In addition, we used multivariable Cox proportional hazards models to examine the association between survival and various patient-related, disease-related, and treatment-related characteristics with the primary focus on disease stage and treatment type. Additional covariates the model included age, sex, race, and primary site. Multivariable survival analyses were conducted for 2 sets of patients: those diagnosed with PNSSCC since 1973 and those with localized disease (N0/M0) diagnosed after 2003. The second set was used specifically to assess the effect of tumor (T)-classification on survival, because TNM stage became available in the SEER data only since 2004. The results of these multivariable models are expressed as adjusted hazard ratios (HRs) and are reported along with the corresponding 95% CIs. Proportional hazard assumptions were tested by examining log-minus-log plots for each variable.[18] In addition, all models were examined for interactions and colinearity among covariates. The analyses were performed using SPSS 17.0 for Windows (LEAD Technologies, Inc., Chicago Ill) and SEER*Stat version 6.4.4 (National Cancer Institute, Bethesda, Md) statistical software packages.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

In total, 2553 patients with PNSSCC were identified in the 18 SEER registries from 1973 through the end of 2009. Nearly half of all patients (49.4%) were between ages 60 years and 79 years, whereas patients aged <40 years and those aged ≥80 years contributed only 4.3% and 15.8% of cases, respectively, as indicated in Table 1. Men accounted for 63.6% of PNSSCC cases, approximately 75% of all patients were white, and 76% presented with a tumor originating in the maxillary sinuses. The distributions of patient demographic characteristics and cancer primary site did not differ appreciably over time. By contrast, there was evidence that the proportion of PNSSCC cases diagnosed with advanced disease (SEER historic stage A coded as “distant”) decreased over time from 14.7% during 1983 to 1992, up to 12.4% during 1993 to 2002, and to 9.5% during 2003 to 2009. The stage distribution in the earlier 10-year interval (1973-1982) could not be assessed, because all cases in the time period were coded as “unstaged” or “blank.”

Table 1. Characteristics of Patients With Paranasal Sinus Squamous Cell Carcinoma by Decade: 18 Surveillance, Epidemiology, and End Results Registries, 1973 to 2009
 No. of Patients (%)
Patient CharacteristicAll Patients, N = 25531973-1982, N = 4311983-1992, N = 4501993-2002, N = 8282003-2009, N = 844
  1. a

    Other races included Asian or Pacific Islander, American Indian, Alaska Native, or unspecified.

Sex     
Men1623 (63.6)282 (65.4)281 (62.4)529 (63.9)531 (62.9)
Women930 (36.4)149 (34.6)169 (37.6)299 (36.1)313 (37.1)
Race     
White1932 (75.7)337 (78.2)340 (75.6)620 (74.9)635 (75.2)
Black324 (12.7)57 (13.2)48 (10.7)116 (14)103 (12.2)
Othera297 (11.6)37 (8.6)62 (13.8)92 (11.1)106 (12.6)
Age, y     
<40109 (4.3)18 (4.2)12 (2.7)38 (4.6)41 (4.9)
40-49239 (9.4)31 (7.2)42 (9.3)82 (9.9)84 (10)
50-59539 (21.1)116 (26.9)89 (19.8)166 (20)168 (19.9)
60-69629 (24.6)126 (29.2)125 (27.8)190 (22.9)188 (22.3)
70-79634 (24.8)96 (22.3)119 (26.4)207 (25)212 (25.1)
≥80403 (15.8)44 (10.2)63 (14)145 (17.5)151 (17.9)
Disease stage     
Localized204 (8)0 (0)61 (13.6)72 (8.7)71 (8.4)
Regional1536 (60.2)0 (0)292 (64.9)605 (73.1)639 (75.7)
Distant249 (9.8)0 (0)66 (14.7)103 (12.4)80 (9.5)
Unstaged564 (22.1)431 (100)31 (6.9)48 (5.8)54 (6.4)
Disease site     
Maxillary sinus1947 (76.3)347 (80.5)343 (76.2)629 (76)628 (74.4)
Ethmoid sinus267 (10.5)31 (7.2)44 (9.8)94 (11.4)98 (11.6)
Frontal sinus48 (1.9)13 (3)4 (0.9)16 (1.9)15 (1.8)
Sphenoid sinus105 (4.1)16 (3.7)18 (4)22 (2.7)49 (5.8)
Overlapping, nonspecified186 (7.3)24 (5.6)41 (9.1)67 (8.1)54 (6.4)

When we analyzed the stage data by tumor site, the proportions of patients with advanced disease were the highest for carcinomas of the sphenoid sinuses (11%) and the ethmoid sinuses (12%). The lowest proportion of advanced tumors was observed for the frontal sinuses (5%). The percentage of patients who received neither surgery nor radiation was about the same across all tumor sites (range, 14%-17%). The frequency of radiation treatment (with or without surgery) ranged from 69% for ethmoid tumors to 83% for frontal tumors. Only 31% patients with sphenoid tumors underwent surgery (either alone or in combination with radiation) compared with 52% to 54% for the other PNSSCC sites.

There was a statistically significant time trend (APC −1.5; 95% CI: −2.0, −1.0) showing a modest decrease in the overall incidence of PNSSCC between 1973 and 2009, but with no discernible inflection points (Fig. 1). Compared to whites, the incidence was higher among African Americans (RR 1.63; 95% CI, 1.39, 1.90) and among individuals of all other racial groups (RR, 1.78; 95% CI: 1.53-2.07). The rates of PNSSCC in women were significantly lower than the corresponding rates among men (RR 0.47; 95% CI: 0.42, 0.52).

image

Figure 1. Age adjusted rates of paranasal sinus squamous cell carcinoma over time and by sex and race (9 Surveillance, Epidemiology, and End Results registries, 1973-2009). APC indicates annual percentage change; CI, confidence interval. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com]

Download figure to PowerPoint

Figure 2 indicates that the 5-year RS remained relatively stable over the years with a statistically nonsignificant APC of 0.4 (95% CI, −0.4, 1.1). Overall the 5-year and 10-year OS estimates were 30.2% (95% CI, 28.1%-32.2%) and 21% (95% CI, 19%-23%), respectively. The corresponding 5-year and 10-year RS estimates were 34.8% (95% CI, 32.4%-37.1%) and 28.3% (95% CI, 25.7%-31%). The sex-specific and race-specific OS and RS estimates generally were higher in men than in women and were lower among African Americans than among whites (Fig. 2). The Kaplan-Meier curves (Figs. 3, 4) demonstrated that the survival of patients with PNSSCC differed significantly by stage and by treatment type (P < .001 for both). Survival was lower for patients who were diagnosed in the earliest decade (1973-1982) than for those who were diagnosed after 1982; however, subsequent time periods (1983-1992, 1993-2002, and 2003-2009) demonstrated no discernible difference (Fig. 5). Patients with early stage disease had better survival than those with advanced stage disease. Also, patients who underwent surgical therapy as part of the treatment plan had better survival outcomes. Notably, the survival of patients who either underwent surgery (HR, 0.41; 95% CI, 0.33-0.50) or received radiation (HR, 0.58; 95% CI, 0.49-0.68) was significantly better than that of patients who did not receive any tumor-directed therapy. The benefit of surgery and radiation was similar to that of surgery alone (HR, 0.39; 95% CI, 0.33-0.46) (Figs. 3, 4).

image

Figure 2. The survival of patients with paranasal sinus squamous cell carcinoma is illustrated over time and by sex and race (18 Surveillance, Epidemiology, and End Results registries, 1973-2009). APC indicates annual percentage change; CI, confidence interval; OS, observed survival; RS, relative survival. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com]

Download figure to PowerPoint

image

Figure 3. The survival of patients with paranasal sinus squamous cell carcinoma is illustrated by Surveillance, Epidemiology, and End Results (SEER) historic stage (18 SEER registries, 1983-2009).

Download figure to PowerPoint

image

Figure 4. The survival of patients with paranasal sinus squamous cell carcinoma is illustrated according to tumor-directed treatment (18 Surveillance, Epidemiology, and End Results registries, 1983-2009).

Download figure to PowerPoint

image

Figure 5. The survival of patients with paranasal sinus squamous cell carcinoma is illustrated by decade (18 Surveillance, Epidemiology, and End Results registries, 1973-2009).

Download figure to PowerPoint

The independent effects of disease stage and treatment, controlling for patient demographic characteristics, are presented in Table 2. Compared with patients who had localized PNSSCC, those with more advanced stage disease had significantly elevated mortality (lower survival), with an HR of 2.06 (95% CI, 1.68-2.52) for regional disease and 3.41 (95% CI, 2.68-4.33) for distant disease. Other factors associated with higher mortality included older age and black race (Table 2). After adjusting for age, sex, disease stage, tumor site, and treatment, mortality among black patients with PNSSCC relative to their white counterparts was increased by 22% (HR, 1.22; 95% CI, 1.04-1.43). By contrast, sex and primary tumor site did not appear to be related to survival.

Table 2. Multivariable Analyses Looking at the Association Between Mortality-, Patient-, Disease-, and Treatment-Related Characteristics Among Patients With Paranasal Sinus Squamous Cell Carcinoma: 18 Surveillance, Epidemiology, and End Results Registries, 1983 to 2009
Patient CharacteristicHR95% CLP
  1. Abbreviations: CL, confidence limits; HR hazard ratio; Ref, referent category.

Age, y   
<501.00Ref 
50-691.421.19-1.70<.001
≥702.271.90-2.72<.001
Sex   
Women1.00Ref 
Men1.010.90-1.13.857
Race   
White1.00Ref 
Black1.221.04-1.43.015
Other0.970.83-1.14.732
Disease site   
Maxillary1.00Ref 
Ethmoid1.080.91-1.28.360
Frontal0.780.52-1.18.245
Sphenoid0.920.70-1.20.523
Not specified0.920.75-1.14.449
Disease stage   
Localized1.00Ref<.001
Regional2.061.68-2.52<.001
Distant3.412.68-4.33<.001
Tumor-directed treatment   
None1.00Ref 
Surgery only0.410.33-0.50<.001
Radiation only0.580.49-0.68<.001
Surgery and radiation0.390.33-0.46<.001

When multivariable survival analyses were limited to patients who had localized (N0/M0) disease, the effects of age and treatment remained unchanged, but the relation between black race and survival was no longer evident. In addition, there was a clear dose-response relation between T-classification and mortality (Table 3). Once the data were restricted to patients with localized disease diagnosed since 2004, all PNSSCC cases were limited to 2 sites: the maxillary and ethmoid sinuses.

Table 3. Multivariable Analyses Looking at the Association Between Mortality-, Patient-, Disease-, and Treatment-Related Characteristics Among Patients With N0/M0 Paranasal Sinus Squamous Cell Carcinoma: 18 Surveillance, Epidemiology, and End Results Registries, 2004 to 2009
Patient CharacteristicHR95% CLP
  1. Abbreviations: CL, confidence limits; HR hazard ratio; Ref, referent category.

Age, y   
<501.00Ref 
50-691.350.88-2.09.170
≥701.821.18-2.81.007
Sex   
Women1.00Ref 
Men1.030.78-1.36.823
Race   
White1.00Ref 
Black0.980.63-1.51.924
Other1.340.91-1.98.143
Disease site   
Maxillary1.00Ref 
Ethmoid0.910.62-1.32.610
Tumor classification   
T11.00Ref 
T21.770.88-3.56.108
T32.491.42-4.37.001
T43.642.12-6.28< .001
Tumor-directed treatment   
None1.00Ref 
Surgery only0.280.17-0.47< .001
Radiation only0.430.29-0.63< .001
Surgery and radiation0.260.18-0.38< .001

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES

The current results demonstrate that the maxillary sinus was the most common primary site of PNSSCC followed by the ethmoid and sphenoid sinuses. We observed a modest but statistically significant decline in the overall incidence of PNSSCC with no identifiable inflection points. The majority of patients with PNSSCC are men, and the age-adjusted incidence of disease is higher in males than in females. The incidence data also indicate that, although most cases occur among whites, PNSSCC is more likely to affect African Americans and other nonwhite populations.

Our results are consistent with those from earlier studies that demonstrated a preponderance of sinonasal cancers in men, among whites, and with predominantly maxillary sinus involvement.[1, 19] Unlike other reports, however, ours focused exclusively on squamous cell histology and encompassed more extensive and contemporary data from SEER. Other SEER-based studies have reviewed paranasal sinus cancers but included other histologies along with squamous cell carcinomas. To our knowledge, this study is the first SEER-based analysis focusing on PNSSCC in the adult population.[20]

It is important to note the modest and statistically nonsignificant increase in the 5-year survival for PNSSCC, which is in accordance with other reports.[2] The combination of radiation therapy and surgery in our data demonstrated superiority compared with radiation alone, which was the primary method of management before the mid-1980s,[1] although our analysis of survival trends demonstrated no discernible inflection points. Also, currently, relatively fewer patients present with distant metastasis, which may indicate improvements in diagnosis resulting in earlier detection. The finding that the survival of patients who received tumor-directed therapy was superior compared with those who did not may be a reflection of other factors, such as the true extent of disease at presentation or comorbidities that were not captured in the SEER data but precluded aggressive therapy.

It is noteworthy that racial differences in survival persisted after controlling for disease stage and treatment received. The reason behind this racial disparity in prognosis remains poorly understood and needs to be explored further. Possible explanatory hypotheses might include differences in post-treatment care, inadequate control for confounding factors because of limitations in SEER data (see below), or differences in tumor aggressiveness. On the other hand, the racial disparity in the analysis limited to localized disease was no longer evident after controlling for T stage.

The interpretation of our findings requires understanding of the strengths and limitations of the SEER data. The large sample size, as we previously noted elsewhere,[21, 22] enables SEER-based studies to have sufficient power for detecting relatively moderate associations and permits a variety of multivariable analyses. The population-based, as opposed to institution-based, identification of patients increases the generalizability of findings, and the active follow-up of patients improves the accuracy of survival analyses. Although institutional studies often have more detailed information about each patient, those studies usually are confined to major referral centers and may not be representative of the patients with PNSSCC who are treated in community hospitals and clinics.

The main limitations of this study pertain to the lack of data on certain important clinical and demographic variables. Although SEER data on surgery and radiation are reasonably complete, the information pertaining to systemic treatment like chemotherapy is usually missing and is not included in the public use files. In addition, SEER data do not contain information on such important predictors of survival as health insurance and socioeconomic status, all of which may determine access to and use of health care. Another important data item that may need to be considered is the effect of provider-related and facility-related characteristics, which cannot be addressed in the context of SEER-based research. For all of the above reasons, both cancer registry-based and institution-based studies provide useful, nonoverlapping information that contributes to the evidence despite their strengths and limitations.[23, 24]

In conclusion, we observed that patients with PNSSCC include a greater proportion of white men, and the age-adjusted disease incidence appears to differ by sex and is higher in nonwhites. Despite improvement in the timeliness of diagnosis and reported increases in surgical treatment, which is associated with a better prognosis, a secular trend in PNSSCC survival, if any, is modest and is not statistically significant. Overall survival appeared to be lower in African Americans compared with their white counterparts; however, among patients with localized disease, the relation between race and mortality was no longer evident once the results were controlled for T-classification.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SUPPORT
  8. CONFLICT OF INTEREST DISCLOSURES
  9. REFERENCES