The epidemiology of salivary gland tumors can be analyzed in several ways. The most fundamental distinction that epidemiologists draw is between descriptive and analytic epidemiology. In its most basic form, descriptive epidemiology is the discipline of identifying and reporting both the pattern and frequency of disease occurrence within a population—that is, who gets it and how much of it is there. In comparison, analytic epidemiology focuses on the search for the causes of disease—who gets it and what puts them at risk of getting it. In this chapter, we examine the descriptive epidemiology of malignant tumors of the salivary glands. We will touch on some elements of the analytic epidemiology of salivary gland neoplasms, but a more detailed consideration of this, including a discussion of the molecular-genetic basis for salivary gland disease, is given in Chapter 6.
Descriptive epidemiology is the discipline of identifying and reporting both the pattern and frequency of disease occurrence in a population.
Analytic epidemiology is the search for the causes of disease.
Epidemiological studies of malignant salivary gland tumors are useful for the clinician for several important reasons. The presentation complex of a salivary gland mass can be similar for both malignant and benign, or neoplastic and nonneoplastic, processes. In the light of this, and in the absence of a one-stop definitive test for salivary gland malignancy, a detailed knowledge of epidemiology, including variation in incidence between age, sex, and populations at risk, is helpful in directing the work-up, treatment, and counseling of patients with salivary gland disease. The epidemiology of salivary gland cancer is also central to the wider consideration of why patients develop salivary gland cancer.
The concept of disease incidence is important when considering epidemiological data. The incidence of salivary gland cancer can be defined as the number of new cases within a defined population at risk over a specific period of time. Conventionally, this will be presented as the incidence rate, which is the ratio of the number of new cancers occurring in a population during a year to the number of individuals who were at risk for the given cancer, generally expressed as the number of cancers per 100 000 persons. Another measure of incidence commonly found in studies is the age-adjusted rate. This is a statistical method that facilitates comparisons of populations, taking into account age-distribution differences between the populations. Incidence rates can also be adjusted for the distribution of other characteristics, such as race/ethnicity. Studies that help identify significant variation in the incidence or prevalence of disease between populations provide helpful guidance for clinical and basic science research.
The incidence of salivary gland cancer is the number of new cases in a defined population at risk over a specific period of time.
The incidence rate is the ratio of the number of new cancers occurring in a population during a year to the number of individuals who were at risk for the given cancer, generally expressed as the number of cancers per 100 000 persons.
For rare conditions like salivary gland carcinoma, the distinction between recording cases of “new-onset” disease and “new recurrent disease” may become blurred, leading to overestimation of disease incidence. Other sources of bias in reporting epidemiological data relate to the populations that are studied. Much of the literature reporting the incidence and distribution of salivary gland cancer comes from the experience of institutions and clinicians with a subspecialty interest. The definition of the population at risk for these studies is important, as most often the data will relate to treated cases (case series/case loads), rather than the true incidence of a controlled population over time. That is not to say that the data from these studies is without use, only that the results should be considered within the context of the population being examined. Finally, historical series and reports are often incomplete owing to difficulties in the classification of tumors by site—i.e., major versus minor salivary gland and histological subtypes.1 Indeed, many early studies only refer to parotid tumors, others to major salivary glands, or combined malignant and benign, so that distinguishing between true differences in these populations is exigent.
The proliferation of cancer statistics published on the Internet allows access to epidemiological data previously available only to the most determined of investigators. Many regional and national cancer registries now publish detailed figures and analyses relating to cancer incidence and mortality. Unfortunately for the head and neck oncologist, much of the data available from these web sites contains little detailed information about head and neck cancer and its subsites. To standardize the interpretation of data on the incidence and mortality of head and neck cancer, many cancer registries combine both anatomic and histological subtypes, rendering any useful analysis somewhat limited. A policy document published by the United Kingdom Association of Cancer Registries (UKACR) summarizes the difficulties faced: “For epidemiological purposes, a standard grouping may be necessary or useful, particularly in reporting the incidence for planning purposes. C00 to C14, together with C30 to C32 should form the core set of tumors included in an overall category of head and neck cancers.”2 Variations in the coding systems used—i.e., diagnostic coding (ICD-10, ICD-10-CM) versus pathological coding (ICD–O–3)—further confound matters. Many databases accrue data on the basis of surgical interventions or pathological studies only, and institutional case series studies that give details of the incidence and prevalence of benign versus malignant disease, or data reporting anatomic distributions, may be subject to significant selection bias by the authors—i.e., they may only include patients who had surgery. When considering the results of these reports and the way in which they relate to clinical practice, therefore, the reader should always ask, “What is the population these data refer to and how accurate and complete are they?”
Salivary gland cancer is a rare condition, and most data collections are hospital-based and subject to bias. Little information about salivary gland cancer is given in most cancer registry–based collections and Internet platforms.
Before the development of Internet-based national cancer registries, most studies reporting epidemiological data for salivary gland malignancies were single institutions or author-based.
In 1971, Eneroth reported experience of 2513 salivary gland tumors treated at the Karolinska Hospital in Stockholm between 1919 and 1969.3 Five hundred and twenty-eight tumors (21%) were malignant. The most common histological type was mucoepidermoid carcinoma (23%), followed by adenoid cystic carcinoma (22%). The most common anatomic site was the parotid gland (n = 378, 72%), followed by the submandibular gland (n = 63, 12%).
Eveson and Cawson reported 2410 cases of primary salivary gland tumor reported to the British Salivary Gland Tumor Panel between 1975 and 1984.4 The site and histology were known for 2356 of the 2410 cases. Seventy-five percent (n = 1756) of the tumors were reported in the parotid, 14.7% of which (n = 258) were malignant, while 96 of 257 submandibular gland tumors (37%) were malignant. A total of 336 of the 2356 tumors (14.2%) were reported in the minor salivary gland, and 46.4% of them (n = 156) were classified as malignant. The most common malignancy in the minor salivary glands was adenoid cystic carcinoma, occurring in 44 of the 156 patients (13.1%). The peak incidence of malignant salivary gland tumors was in the seventh decade of life.
Spiro reported experience at the Memorial Sloan-Kettering Cancer Center in treating 2807 patients with salivary neoplasms between 1939 and 1973,5 1966 of whom had had no previous therapy. Malignant tumors were reported in 740 of the 1966 patients (37.6%). Mucoepidermoid carcinoma was most commonly encountered, followed by adenoid cystic carcinoma and adenocarcinoma. The most common site for malignant neoplasms was the minor salivary glands (318 of 390, 82%), followed by the submandibular gland (68 of 160, 43%) and the parotid gland (354 of 1416, 25%).
These series represent some of the largest experiences of their type, but interpretation of them must be guarded in the context of the numerous revisions that have taken place in the classification of salivary gland tumors. The reported frequency of many tumors is fairly diverse and is probably the result of variable classifications of the lesions6 over the years.
In a study in Quito, Ecuador, an Andean city with approximately 2 million inhabitants, the rates were reported as between 0.5 and 0.3 per 100 000. From 1982 to 1998, the authors carried out surgery for 308 salivary gland tumors. Malignant lesions were found in 58 cases (19%): 37 of 194 parotid gland tumors (19%), seven of 86 submandibular tumors (8%) and 14 of 28 minor salivary gland tumors (50%). Adenoid cystic carcinoma and mucoepidermoid carcinoma were the most common histological types.7
Jones et al. reviewed experience of treating 741 salivary gland tumors between 1974 and 2005; 260 of the lesions (35.1%) were malignant. Mucoepidermoid carcinoma was the most common malignant diagnosis, with 85 cases (32.7%), followed by adenoid cystic carcinoma with 62 cases (23.8%). In terms of location, malignant tumors occurred in the parotid gland in 62 cases, in the submandibular gland in 15, and in the sublingual gland in nine cases.6
The Netherlands Cancer Registry was established in 1989 to allow high-quality data to be collected from eight regional comprehensive cancer centers. National legislation ensures the registry contains data on virtually all cases of cancer treated in the Netherlands. For the period 1987 to 2007, the age-standardized incidence of salivary gland cancer was 0.6–0.9 per 100 000 (0.6–1.0 per 100 000 in males and 0.5–0.8 per 100 000 in females).8
Several authors have studied epidemiological data from Scandinavian national cancer registries. The Finnish Cancer Institute report of 2006/2007 reported age-adjusted incidence rates of 0.5–0.8 per 100 000 for salivary gland cancer.9 Luukkaa et al. examined all salivary gland cancers (n = 286) reported to the Finnish Cancer Registry between 1991 and 1996. The commonest tumor location was the parotid gland (n = 152, 64%), followed by the minor salivary glands (n = 46, 19%), the submandibular gland (n = 38, 16%), and the sublingual gland (n = 1, 0.4%). The most frequent histological types were adenoid cystic carcinoma (n = 65, 27%), mucoepidermoid carcinoma (n = 45, 19%), and acinic cell carcinoma (n = 41, 17%).10
Data from the Swedish Cancer Registry on malignant major and minor salivary gland tumors between 1960 and 1989 have been reported by Ostman et al.1 The total number of new malignant major salivary gland tumors was 2557, representing 0.3% of all newly diagnosed cancers. The incidence rate of malignant salivary gland tumors per 100 000 was 1.3 cases (1.4 for men and 1.2 for women).
A review of registrations for cancers diagnosed in England in 2006 (data from the Office for National Statistics, extracted by Cancer UK) shows that the age-adjusted incidence of salivary gland cancers was 0.8 per 100 000 persons (males 1.0, females 0.7).11 Bradley reported epidemiology statistics for salivary gland cancers in a defined population of 750 000 in Nottingham, England. Between 1988 and 2007, 147 salivary gland tumors were recorded, yielding an annual incidence per 100 000 population of 0.83–1.38.12
The age-standardized incidence of salivary gland cancer in
Europe ranges from 0.6 to 1.4 per 100 000. There is no significant gender difference.
< div class='tao-gold-member'>