Mutations in the CHEK2 gene have been associated with increased risks of breast, prostate and colon cancer. In contrast, a previous report suggests that individuals with the I157T missense variant of the CHEK2 gene might be at decreased risk of lung cancer and upper aero-digestive cancers. To confirm this hypothesis, we genotyped 895 cases of lung cancer, 430 cases of laryngeal cancer and 6391 controls from Poland for four founder alleles in the CHEK2 gene, each of which has been associated with an increased risk of cancer at several sites. The presence of a CHEK2 mutation was protective against both lung cancer [odds ratio (OR) = 0.3; 95% confidence interval (CI) 0.2-0.5; P = 3 x 10(-8)] and laryngeal cancer (OR = 0.6; 95% CI 0.3-0.99; P = 0.05). The basis of the protective effect is unknown, but may relate to the reduced viability of lung cancer cells with a CHEK2 mutation. Lung cancers frequently possess other defects in genes in the DNA damage response pathway (e.g. p53 mutations) and have a high level of genotoxic DNA damage induced by tobacco smoke. We speculate that lung cancer cells with impaired CHEK2 function undergo increased rates of cell death.

For the past two decades, simple shunt devices inserted either primarily, at the time of laryngectomy, or later as a secondary procedure, have mainly supplanted the other methods of voice rehabilitation, with the exception of an occasional patient who has acquired good esophageal speech, or for whom external devices may be the only practical method of voice production.

ARLTS1--a member of ADP-ribosylation factor family, is a newly described candidate tumour suppressor gene. Recent studies show that a nonsense polymorphism, G446A (Trp149Stop), in ARLTS1 gene is significantly more frequent in familial cancer cases than in sporadic cancer cases. This study presents analysis of the germ-line G446A polymorphism in the ARLTS1 gene among 1686 consecutively collected patients with breast cancer, prostate cancer, malignant melanoma, thyroid papillary cancer or laryngeal cancer in Poland. The G446A allele was present in 1.81% (9/497) breast cancer patients, 1.46% (5/343) prostate cancer patients, 1.76% (7/397) melanoma patients, 1.65% (3/182) thyroid papillary carcinoma patients and 2.68% (8/299) of laryngeal cancer patients. The frequency of this polymorphism in the control group was 1.45% (8/552). Differences in the frequency of the G446A polymorphism between case and control groups were not statistically significant. In addition, there was no significant difference in the number of Cancer Familial Aggregations (CFA) among breast, prostate, thyroid or laryngeal cancer cases harbouring the G446A polymorphism, when compared to the G446A negative cases. Interestingly out of the CFA melanoma cases, 4/6 (66.6%) were found to harbour the change compared to only 20.2% (69/341) sporadic melanoma cases. This difference was statistically significant (p = 0.02, OR = 7.8). The results of this study suggest that the G446A in ARLTS1 gene is probably not associated with an increased risk of sporadic breast cancer, prostate cancer, melanoma, thyroid papillary cancer or laryngeal cancer. Moreover, the G446A polymorphism is not significantly more frequent in CFA cases except for families in which the proband had melanoma. To confirm this result more cases of melanoma should be analysed.

Recent studies suggest that there are groups of genes that predispose simultaneously to both early-onset breast and laryngeal cancer. Studies were performed on a large series of unselected patients with laryngeal cancer diagnosed in Szczecin, Poland. Pedigrees of 683 laryngeal cancer patients were analysed for the frequency of early-onset and late-onset breast cancer among first degree relatives. The observed frequencies of breast cancer in these families were compared to those expected. In addition, common mutations/variants in the 3 genes BRCA1, NOD2 and CYP1B1, known to be associated with early-onset breast cancer, were assessed to determine their frequency in 348 unselected laryngeal cancers. The average age at diagnosis of LC among patients, who had relatives affected by BC diagnosed under the age of 50 years was 57.62. In comparison LC patients reporting a first degree relative affected by BC diagnosed above 50 years of age, had an average age of diagnosis of 66.00 years, which was significantly different (p=0.0064). Similarly, the average age of diagnosis of BC among patients with LC diagnosed under age of 50 years was 46.7 years and whereas LC patients with tumors diagnosed above 50 years had relatives diagnosed with breast cancer at an average age of 53.37 years, which was significantly different (p=0.02). From the 348 consecutive ascertained laryngeal cancer patients who had molecular studies undertaken, breast cancers among first degree relatives were found in 18 families including 8 with breast cancers diagnosed less than 50 years of age. A molecular basis was identified (the CYP1B1 355T/T genotype) in only 2 of the 8 early cases suggestive of there being additional, as yet unknown genes that are associated with an early-onset laryngeal-breast cancer phenotype.

This study indicates that leaving hyoid bone is a major risk factor for developing pseudoepiglottis.