A Demographic Study of Intraoral Malignancies in Minnesota from 1993-2008

A Demographic Study of Intraoral Malignancies in Minnesota from 1993-2008

Matthew D. Reilland, B.S.,* and Ioannis G. Koutlas, D.D.S., M.S.**:

Oral cancer accounts for less than 3% of all cancers in the United States1,2,3 and represents the eighth most common malignancy among men and the thirteenth among women,4 with approximately 30,000 new cases diagnosed each year.3,5,6,7 The number of men with oral cancer is consistently higher by two- to threefold compared to  women.2,8,9,10,11 In the United States, more than 7,500 (2%) deaths per year1 are associated with oral cancer.3,5,6,7 For squamous cell carcinoma (SCCa), the most common form of oral cancer, tobacco, alcohol, the combination of the two, as well as high risk HPV viruses are important risk factors.12,13 For rarer forms of oral cancer, EBV and immunosuppresion increase the risk for malignancy. Although mortality rates of oropharyngeal cancer in the United States decreased significantly from 1988 to 2004,14,15 there  is still significant morbidity and mortality associated with this type of cancer.  However, these findings are not uniform; in fact, in some areas rates have increased.16,17 

In  Minnesota, oropharyngeal cancer, excluding hematopoietic malignancies, represents 2.3% of all cancers, with incidence rates and gender trends similar to those reported at the  national level.15 Approximately 530 cases of oropharyngeal cancer are diagnosed each year, and 115 people die annually of this disease.15 

Previously, few reports have dealt solely with intraoral malignancies in Minnesota.18The scope of this study was to review the clinicopathologic characteristics and geographic distribution in the state of  Minnesota of intraoral malignancies as accessioned in the University of Minnesota Oral and Maxillofacial Pathology Laboratories (UMOMPL) between 1993 and 2008. Additionally, trends comparing total number and types of malignancies with population, as well as number of biopsies, were investigated.

Materials and Methods

The following diagnostic terms were searched for in the UMOMPL database: carcinoma, sarcoma, melanoma, metastatic, malignant, myeloma, leukemia, and lymphoma. The information was entered into a spreadsheet and organized by county, type of malignancy, age, gender, site in mouth, and race. For location in the oral cavity, cases were grouped into 14 sites: buccal mucosa/ vestibule, tongue (lateral, ventral, and dorsal), palate (soft and hard), and maxillary tuberosity, gingiva/alveolar mucosa/alveolar ridge,  tonsillar area, retromolar area, mandible, maxilla, floor of mouth, lip, sinus, salivary gland, overlapping/contiguous, and undetermined. The overlapping/ contiguous location  included any diagnosis that involved more than one of the previously listed sites. 

Adjustments were made to exclude diagnostic terms such as carcinoma “in situ”, and to eliminate duplicate records of patients with recurrent disease. Additionally, cases diagnosed as “possible” malignancies as well as those considered extraoral malignancies were removed. Cases of myeloma, lymphoma, or leukemia were grouped into one  category: hematopoietic malignancies. Salivary gland, odontogenic, and unclassified malignancies were grouped separately. If a patient’s address was not provided, the  contributor’s was used, assuming the region of referral would be the same. Frequencies for each type of malignancy were calculated statewide as well as for each county in  Minnesota. To calculate the incidence per capita, population values statewide and for each county were used from the 2000 census, retrieved from the Minnesota State Demographic Center website.19The total number of malignancies was divided by the total number of people residing in the county. This value was multiplied by 100,000 to bring  the small percentages to a comprehensible range. The per capita data for total malignancies, as well as for each type of malignancy for each county, were calculated, summed,  and then divided by the number of counties, 87 giving a statewide average to be used as a comparison for individual county per capita values. In order to conceptualize this  data in a geographic sense, the software Mapviewer 7.0 was used to create a map of Minnesota displaying the per capita values for each county. Special interest was placed in the ten most populated counties; in descending order, Hennepin, Ramsey, Dakota, Anoka, Washington, St. Louis, Stearns, Olmsted, Wright, and Scott counties.  The five most common intraoral malignancies – SCCa, hematopoietic, salivary gland, metastatic, and sarcoma – were extensively studied (Table 1). 

To analyze trends, the following data were utilized: (a) Minnesota population, (b) number of biopsies, and (c) number of malignancies diagnosed. Analysis of proportions 
was also performed calculating and regressing the percentages of different types of malignancies (SCCa, hematopoietic, salivary, and other) against the years studied. Finally, analysis of compositions was performed by comparing the average compositions of the malignancies for the first (1993-2000) and second half (2001-20008) of the study using a special statistical package for compositional data analysis.20 
 
Results
According to information from the Minnesota Department of Health (MDH), UMOMPL reports approximately 15-25% of total oral cavity malignancies, excluding non-Hodgkin’s  lymphomas. Although they are considered the second most frequent type of oral malignancy, non-Hodgkin’s lymphomas are reported as a separate category among types of  cancer irrespective of location. From a total of approximately 75,000 UMOMPL records, 1,047 (1.4%) malignancies met our selection criteria. 
 
Incidence by Type of Malignancy
The five most common malignancies were SCCa (78.8%), hematopoietic (7.93%), salivary gland (7.55%), metastatic disease (2.67%), and sarcomas (1.53%). 
 
Incidence by County
The per capita values calculated for each county for all types of malignancies reported are shown in Table 1. Tables 2, 3, 4, 5,and 6 show the per capita values for the five  most common types of malignancies for each county. The statewide averages for each type of malignancy are shown in Table 7. The geographic distribution of total  malignancies based on the occurrence per 100,000 people is depicted in Figure 1. The raw numbers in the ten most populous counties are summarized in Table 8. 
 
Other than the ten most populous counties, a few counties presented interesting data worth mentioning. Clearwater County had the highest incidence rate for total malignancies 
(83.1/100,000), followed by Crow Wing County (70.8/100,000).  As would be expected, these two counties also had the highest per capita values for SCCa’s; 71.2 and 63.5 per  100,000 people, respectively. For hematopoietic malignancies, Pope County recorded the highest rate (17.8/100,000), and Swift County had the highest incidence of salivary  gland tumors (8.4/100,000).
 
Interestingly, 19 counties (Kittson, Marshall, Red Lake, Norman, Mahnomen, Clay, Wilkin, Otter Tail, Traverse, Yellow Medicine, Pipestone, Murray, Cottonwood, Rock, Nobles,  Jackson, Cook, Carlton, and Houston) did not have any intraoral malignancies recorded in our files during this 16-year period. 
 
Gender
Males presented a higher incidence rate (53.5%) of total malignancies when compared to females (46.5%). The distribution of gender for the three most common malignancies was as follows: For SCCa, 55% patients were male, 45% female; for hematopoietic and salivary gland malignancies, males represented 53 and 30 percent, respectively. 
 
Ethnicity
Caucasians showed the highest frequency of intraoral malignancies at 90.6%, with less than 2% occurring in other ethnic groups. One should note that in 7.64% of the reports,  the patients’ ethnicity was not reported. 
 
Age
The average age for total malignancies was 64, with that of SCCa and hematopoietic malignancies being slightly higher, 65 and 67, respectively. The average age for salivary  tumors and sarcomas were lower, 48 and 40, respectively. The average age for metastatic disease was 63. For all the reported malignancies, the average age for men was 62.2  and for women 65.9. 
 
Site Distribution According to Cancer Subtypes
Of the 14 sites, the tongue was the most common site, with 23% of all reported malignancies (Table 9). The five most common types of malignancies and their incidences in the respective 14 sites are shown in Table 10. SCCa was present most commonly on the tongue (28%), followed by the gingiva/alveolar mucosa/alveolar ridge (19.15%). Both hematopoietic and salivary gland malignancies showed the highest incidence on the palate (34.9% and 41.8%, respectively) and buccal mucosa/vestibule (15.7% and 20.3%, respectively). The mandible was the most common site for metastatic disease (57%), followed by maxillary/ mandibular gingiva (29%). Of the 16 sarcomas, the mandible and the 
maxilla had the highest occurrences, at 31.25%. 
 
Trends Over Time
Over the 16-year course, Minnesota’s population increased in size (Figure 2), with the number of biopsies reported to UMOMPL increasing as well (Figure 3). During this time, although the number of malignancies diagnosed increased in accordance with the increase of the number of biopsies (Figure 4), the percentage of malignancies over biopsies decreased (Figure 5). The percentages of different kinds of malignancies (SCCa, hematopoietic, salivary, and other) were calculated and regressed against the years; no  statistically significant effects were found (Figure 6). The average compositions of the malignancies for the first (1993-2000) and second half (2001-20008) of the study are  shown in Figure 7 (A and B, respectively). These compositions differed significantly (P < 0.044). This was mainly due to a decrease of hematopoietic malignancies and an increase in the salivary malignancies between the two time periods.
 
Discussion
The scope of this study was to present the state distribution and summarize the clinicopathologic characteristics of intraoral malignancies recorded during a 16-year period at  the University of Minnesota Oral and Maxillofacial Pathology Laboratory. This laboratory is dedicated almost solely to intraoral biopsies, and is the only laboratory in the state  devoted to and specialized in the diagnosis of such specimens. According to the Minnesota Department of Health (MDH), it accounts for the reporting of about 15-25% of  intraoral malignancies occurring statewide per year. In Minnesota, 1.3% of all deaths can be attributed to oropharyngeal cancer, which is slightly lower than the nation as a  whole.14,15 Although mortality rates have decreased significantly from 1988 to 2004,15 oral cancer remains among the most debilitating and disfiguring cancers.21 The decrease  in the percentage of malignancies over the number of biopsies is encouraging, although both raw numbers increased during the studied period. The increase in cases of intraoral  cancer may be related to the enhanced awareness of dental practitioners.
 
In the 16-year period covered in this study, 1.4% of all diagnoses were malignancies, with SCCa being the most common form, accounting for 78.8% of reported malignancies. This percentage is lower than the expected number (more than 90-95%) of all oral malignancies. There is no obvious reason why the number of SCCa in our series is smaller than expected, although with the recent austerity measures against smoking paired with increased tobacco taxes ($1.57/pack, higher than the national average),22 one would expect  the numbers of those smoking to decrease over time, resulting in decrease the incidence of SCCa. In the United States, the overall incidence of SCCa has decreased over a 30- year span; in 1973-1976 the rate was 13.2/100,000 persons, whereas in 2000-2003 it had decreased to 10.2/100,000.23 Also, one must note that the UMOMPL deals almost  exclusively with oral mucosa biopsies; thus pharyngeal cases of squamous cell carcinoma are hardly ever encountered. 
 
A rising incidence of hematopoietic malignancies, specifically non-Hodgkin’s lymphomas,24 may explain the higher incidence of hematopoietic malignancies in our study, representing nearly eight percent of all intraoral malignancies. Unfortunately, an accurate incidence of lymphomas initially presenting in or affecting only the oral cavity cannot  be well established because of systemic involvement by the disease. Of the 83 hematopoietic malignancies reported, 14 of these cases had a history of systemic non-Hodgkin’s lymphoma. Also, we encountered a higher incidence of malignant minor salivary gland tumors, representing 7.5% of all cancers in our study, or 0.85 cases/100,000 when compared to 0.1-1.5 cases/100,000 nationwide (data extrapolated from a comprehensive study of tumors of the salivary glands).25 
 
In this study we focused primarily on data from the ten most populous counties. However, certain counties that were less populated showed a higher incidence of malignancies.  For example, Clearwater and Crow Wing counties reported the highest rates for total malignancies, as well as for SCCa. Ramsey County consistently reported a higher incidence  rate for all types of malignancies compared to the other metro area counties. Further investigation is necessary to determine the cause. It is possible that there is a higher rate  of tobacco or alcohol use, or a decreased awareness in the oral screening process. In the United States, SEER reports that just 20% of Americans over the age of 40 have had a formal oral cancer screening done.14
 
When it comes to age, the average age for men and women in our study was in accordance with those reported by the MDH and nation (men: 62; women: 66). Also, in  Minnesota, less than 20% of oral cancers are diagnosed before the age of 50, which is also in accordance with our study (19.5%).15  
 
Males are consistently affected with oral cancer more frequently than females, with a 2:1 to 3:1 ratio.2,8,9,10,11 Specifically in Minnesota, according to MDH, the average annual incidence of oral cancer from 2000- 2004 was 15.6 and 6.1 per 100,000, for males and females, respectively.15 However, in our study, the ratio of males to females was favoring  males only slightly. This is partially attributed to the surprisingly high incidence of malignant minor salivary gland tumors in women (70%). 
 
In Minnesota, Caucasians comprise 86% of population, with African Americans following at 4%, and all other minorities accounting for the rest.13 Therefore, there is no surprising difference when 91% of patients with intraoral malignancies were Caucasian. Only 1.15% of the patients in our study were African Americans. Since almost all the contributors  are private practitioners, this discrepancy may indicate difficulties regarding barriers to care among minorities. 
 
In the present study, the tongue was the most frequent site, with 23% of all malignancies, followed by the gingiva/alveolar mucosa, and the floor of mouth. However, it is our opinion to address the most frequent subtypes of oral cancer individually rather than all intraoral malignancies together because sites of predilection differ among individual  subtypes. For example, SCCa affected the tongue and the gingiva/mucosa more frequently, while salivary gland and hematopoietic malignancies both occurred more frequently  on the palate and buccal mucosa/vestibule. Overall, site distribution of intraoral malignancy subtypes followed the known trends.2,3,8,26,27 
 
Conclusions
In summary, the clinicopathologic characteristics and geographic distribution of intraoral malignancies recorded in the UMOMPL in a 16-year period were reviewed. Depending  upon the site of involvement, this laboratory reports up to 25% of the total cases of oral cancer cases occurring per year in Minnesota. This value must be taken into  consideration when evaluating this data. With 75% or more of intraoral malignancies being reported apparently by other laboratories, the following must be considered: The 19  counties with no reported malignancies are not geographically close to UMOMPL, so there is potential for their biopsies to be sent elsewhere; counties with other nearby  laboratories may have underrepresentation in this data; and the counties nearest to UMOMPL may have an over-representation. Geographic distribution of specific types of  intraoral malignancies has not been addressed by the MDH. According to our data, certain counties (Ramsey, Clearwater, and Crow Wing) show higher incidences for intraoral  malignancies. Also, there was an unexpectedly high incidence of salivary gland malignancies among women. Additionally, risk factor distributions were not considered. The rates  of smoking and alcohol use in different locations throughout the state would influence these numbers. Barriers-to-care issues and the degree to which oral cancer screenings are  
performed to evaluate patients for pre-malignant/malignant lesions can influence the data as well. This is the first comprehensive study of the clinicopathologic characteristics of  solely intraoral malignancies over a 16-year period in the state of Minnesota. The data can be shared with the MDH to more specifically categorize the incidence of oral cancer in the state of Minnesota. 
 
Acknowledgments
Research Supported by UMNSOD Summer Research Program. The authors thank Apostolos P. Gerogopoulos, M.D., Ph.D., Regents’ Professor, for his invaluable help with the  statistical analysis; David Born, Ph.D., for the use of the software Mapviewer 7.0; and Joel Rudney, Ph.D., for his comments and suggestions. 
 
References
1. Weinberg M, Estefan D. Assessing oral malignancies. Am Fam Physician 2002;65(7):1,379-84.
2. Silverman S. Demographics and occurrence of oral and pharyngeal cancers: The outcomes, the trends, the challenge. J Am Dent Assoc 2001; 132:7S-11S.
3. Horowitz A, Canto M, Child W. Maryland adults’ perspectives on oral cancer prevention and early detection. J Am Dent Assoc 2002;133:1,058-63.
4. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, Thun M. Cancer Statistics, 2008. CA Cancer J Clin 2008;58(2):71-96. 
5. Kademani D. Oral Cancer. Mayo Clin Proc 2007;82(7):878-87. 
6. Peacock Z, Pogrel M, Schmidt B. Exploring the reasons for delay in treatment of oral cancer. J Am Dent Assoc 2008;139:1346-52.
7. Shiboski C, Shibosk S, Silverman S. Trends in oral cancer rates in the United states, 1973-1996. Community Dent Oral Epidemiol 1,000;28:249-56.
8. Marocchio L, Lima J, Sperandio F, Correa L, Sousa S. Oral squamous cell carcinoma: an analysis of 1,564 cases showing advances in early detection. Eur J Oral Sci 010;52(2):267-73.
9. Scully C, Bagan J. Oral squamous cell carcinoma overview. Oral Oncology 2009;45:301-8. 
10. Suba Z. Gender-related hormonal risk factors for oral cancer. Path Oncol Res 2007;13(3):195-202. 
11. Cook M, Dawsey S, Freedman N, Inskip P et al. Sex disparities in cancer incidence by period and age. Cancer Epidemiol Biomarkers Prev 2009;18(4):1,174-82.
12. Horowitz AM, Drury, TF, Goodman, HS, Yellowitz, JA. Oral pharyngeal cancer Prevention and early detection: dentists’ opinions and practices. J Am Dent Assoc  2000;131;453-462.
13. Applebaum, E, Ruhlen, TN, Dronenberg FR, Hayes C et al. Oral cancer knowledge, attitudes and practices: a survey of dentists and primary care physicians in Massachusetts.  J Am Dent Assoc 2009;140;461-467.
14. National Cancer Institute, Surveillance, Epidemiology and End Results Database (NCI-SEER). http://seer.cancer.gov/about 
15. Brown M, Perkins C, Soler J, Bushhouse S. Cancer in Minnesota, 1988-2004. St. Paul, Minnesota: Minnesota Cancer Surveillance System, Minnesota Department of Health, 
September 2008.
16. Bunnell A, Pettit N, Reddout N, Sharma K, O’Malley S et al. Analysis of primary risk factors for oral cancer from select US states with increasing rates. Tobacco Induced  Diseases 2010;8:5. 
17. Kingsley K, O’Malley S, Ditmyer M, Chino M. Analysis of oral cancer epidemiology in the US reveals state specific trends: implications for oral cancer prevention. BMC Public  Health 2008;8:87-99. 
18. Bougquot J, Durland L, Weiland L. Layrngeal keratosis and carcinoma in the Rochester, MN, population 1935-1984. Cancer Detect Prev 1991;15(2):83-91.
19. Minnesota State Demographic Center [WWW document]. URL http://www.demography.state. mn.us/ [accessed 13 August 2010].
20. Aitchison J. The Statistical Analysis of Compositional Data. London UK: Chapman & Hall, 1986. 
21. Fedele S. Diagnostic aids in the screening of oral cancer. Head and Neck Oncology 2009;1:5. 
22. State cigarette excise tax rates and rankings [WWW document]. URL http://www.tobaccofreekids.org/research/factsheets/pdf/0097 pdf [accessed on 13 August 2010].
23. Goldenberg D, Brooksby C, Hollenbeak C. Age as a determinant of outcomes for patients with oral cancer. Oral Oncology 2009;45:e57-e61. 
24. Neville B, Damm D, Allen C, Bouquot J. Oral and Maxillofacial Pathology, Third Edition. St. Louis: Saunders Elsevier, 2009, pages 409-21, 595-8. 
25. Ellis G, Auclair P. Tumors of the Salivary Glands, Fourth Series, Fascicle 9. Washington, DC: American Registry of Pathology, 2008:26-28.
26. Bagan J, Sarrion G, Jimenez Y. Oral cancer: clinical features. Oral Oncology 2010;46:414-17. 
27. Gillison M. Current topics in the epidemiology of oral cavity and oropharyngeal cancers. Head and Neck 2007;29(8):779-92. 
28. Minnesota County Map [WWW document]. URL http://www.mapwatch.com/county-map/minnesota.shtml [accessed 10 August 2010]. 
 
 









*Mr. Reiland is a dental student at the University of Minnesota School of Dentistry, Minneapolis, MN 55455. Email is reila023@umn.edu.
**Dr. Koutlas is Associate Professor, Division of Oral and Maxillofacial Pathology, University of Minnesota School of Dentistry, Minneapolis, MN 55455.