Is hepatitis C virus infection associated with thyroid cancer? A case‐control study

Some have reported an increased risk of auto-immune thyroid disease in hepatitis C virus (HCV) and HBV and in interferon therapy (Fernandez-Soto et al., 1998). A new link between HCV and thyroid cancer has been reported (Antonelli et al.,1999). An oncogenetic role of hepatitis C in the pathogenesis of different kinds of tumor has been suggested (Ferri t al.,1997; Heintges and Wands, 1997). In southern Italy, there is a high prevalence of HCV infection (Guadagninoet al., 1997) with high mortality and incidence rates of liver cancer (Montella et al., 1998) and high incidence rates of thyroid cancer (Franceschi et al.,1998). For these reasons, we designed a case-control study on HCV infection and thyroid cancer in southern Italy. Both cases and controls are patients who received surgery in our institute, the main institute of health and care in oncology of southern Italy. Cases are patients who received surgery for histologically confirmed papillary thyroid cancer (ICD-O 8340/3) between 1997 and 1999. Patients previously exposed to low-dose therapeutic external radiation of the head and neck area were excluded. There are 130 subjects with papillary thyroid cancer (24 men and 106 women). Patients with neither history of cancer nor auto-immune disease but who were hospitalized at the same time as the cases were used as controls. They included 186 subjects (70 men and 116 women) with the following diagnoses: breast dysplasia (51), fibroma or myoma of the uterus (40), prostatic adenoma (32), hernia (15), lipoma (11), prostatic hypertrophy (20), fistulas (11), moles (6). Cases and controls were selected from the hospital database (hospital tumor registry), which collects information on all patients, including who has benign tumors (fibroma, myoma, etc.) or neoplastic pathology without confirmation (prostatic hyperplasia, fistulas, etc.). The following data were collected at recruitment: date of admission, date of birth, histologically confirmed diagnosis, age at diagnosis ( #39, 40–49, 50–59,$60), and marital status. The serological diagnosis of HCV infection was tested by ELISA, and samples yielding positive results were retested by PCR. The relationship between thyroid cancer and HCV infection was assessed by means of odds ratios (ORs) and corresponding 95% confidence intervals (CIs). Unconditional logistic regression equations included terms for age, sex, and marital status (Breslow and Day, 1980). Cases were not individually matched to controls but were comparable according to age and area of residence. The distribution according to age and sex for cases and controls is reported in Table I. Among thyroid cancer cases (mean age 46 6 14), 12% were HCV-positive compared to 6% of control subjects (mean age 50 6 13). In the age group#39, there were no cases or controls positive for HCV. In the analysis, we used only 2 age groups:#49 and$50 . The prevalence of HCV for thyroid cancer in the #49 age group was 6% compared to 2% in controls and in the $50 age group, it was 23% compared to 9% in control subjects. Thyroid cancer displayed a statistically significant association with HCV infection, adjusted for age and gender (OR 5 3.1, 95% CI 1.3–7.4). The possible confounding effects caused by age and sex were controlled throughout the stratification for age ($50 year age group OR5 3.1, 95% CI 1.1–8.2;,50 year age group OR5 2.8, 95% CI 0.5–16.0); males OR 5 2.9, 95% CI 0.6–12.5; females OR 5 4.0, 95% CI 1.1–8.8) (Table II). The presence of risk in the$50 year age group was probably caused by latency characteristics of HCV, which show its pathogenic effects after many years; the most evident risk for women was attributed to the high incidence of thyroid pathology (thyroiditis and thyroid cancer) in women. Thyroid cancer is a relatively rare pathology, but it is becoming more frequent (e.g., in Italy, in 1983–1987, the standardized incidence rates per 100,000 inhabitants were 2.0 for males and 5.5 for females, while in 1988–1992, they were 2.5 for males and 7.0 for females) (Zanetti t al.,1992, 1997). Unfortunately, information that could explain the high prevalence of HCV among the Mediterranean population is not available. Moreover, as HCV infection was discovered only 10 years ago, its prevalence and importance were under-estimated during the initial stage. Thyroiditis is frequently found in patients with type C chronic hepatitis during interferon therapy (Fernandez-Soto et al., 1998). Some results suggest a weak association between HCV and thyroid autoimmunity in females (Huang et al., 1999). Thyroiditis has also been regarded as a pre-neoplastic condition (Okayasuet al.,1995). It is, therefore, plausible that the oncogenetic role of HCV is relevant not only for non-Hodgkin’s lymphoma and liver cancer but also for thyroid cancer (Antonelli et al.,1999; Heintges and Wands, 1997). It is possible that in such cancers the association is more detectable in countries with a high preval nce of HCV and particularly where life expectancy, unlike in non-developed nations, is higher than 50 years ( i.e., MediterTABLE I – DISTRIBUTION ACCORDING AGE AND SEX