A national case-control study of Ewing's sarcoma family of tumours in Australia

Limited population-based epidemiologic information is available on Ewing's sarcoma family of tumours (ESFT), a rare group of neoplasms. Several associations have been noted on a few studies but results were not consistent, except for exposure to farming among cases and their parents. Here we present the non-farm findings of a nationwide case-control study of ESFT in children and young adults in Australia. The analysis included 106 persons with confirmed ESFT and 344 population-based controls selected randomly via telephone. Information was collected by interview (84% face to face). We found a strong and significant association of ESFT with hernias, in particular hernia repaired in hospital (OR = 5.6, 95% Cl 1.3-6.4). Among other factors, there was a near doubling of risk for males, and male cases had their pubertal signs earlier (started shaving earlier) than male controls. There was also an increased risk of ESFT at higher levels of self-assessed exercise, but no other factor really stood out. For pregnancy-related factors, there was a tripling of risk for glandular fever, a doubling of risk for urinary tract infection and a near doubling of risk for X-rays during or just before pregnancy, but these estimates were not significant. In addition, there was a large number of inverse associations with medical conditions (specifically bone disorders), case exposure to medications, vaccinations and X-rays, with ultrasound during the pregnancy having the most certain effects. We conclude that, although the aetiology of ESFT remains obscure, overall there is strong evidence of an association with inguinal hernia; this can now be added to the farm-associated risk reported by others and us. The other associations reported here await replication and refinement in future studies. (C) 2003 Wiley-Liss, Inc.

Double-strand break repair gene polymorphisms and risk of breast or ovarian cancer

Deficiencies in DNA repair have been hypothesized to increase cancer risk and excess cancer incidence is a feature of inherited diseases caused by defects in DNA damage recognition and repair. We investigated, using a case-control design, whether the double-strand break repair gene polymorphisms RAD51 5' untranslated region -135 G > C, XRCC2 R188H G > A, and XRCC3 T241M C > T were associated with risk of breast or ovarian cancer in Australian women. Sample sets included 1,456 breast cancer cases and 793 age-matched controls ages under 60 years of age, 549 incident ovarian cancer cases, and 335 controls of similar age distribution. For the total sample and the subsample of Caucasian women, there were no significant differences in genotype distribution between breast cancer cases and controls or between ovarian cancer cases and combined control groups. The crude odds ratios (OR) and 95% confidence intervals (95% CI) associated with the RAD51 GC/CC genotype frequency was OR, 1.10; 95% CI, 0.80-1.41 for breast cancer and OR, 1.22; 95% CI, 0.92-1.62 for ovarian cancer. Similarly, there were no increased risks associated with the XRCC2 GA/AA genotype (OR, 0.98; 95% CI, 0.76-1.26 for breast cancer and OR, 0.93; 95% CI, 0.69-1.25 for ovarian cancer) or the XRCC3 CT/TT genotype (OR, 0.92; 95% Cl, 0.77-1.10 for breast cancer and OR, 0.87; 95% CI, 0.71-1.08 for ovarian cancer). Results were little changed after adjustment for age and other measured risk factors. Although there was little statistical power to detect modest increases in risk for the homozygote variant genotypes, particularly for the rare RAD51 and XRCC2 variants, the data suggest that none of these variants play a major role in the etiology of breast or ovarian cancer.