PHG Foundation | www.phgfoundation.org

3 August 1998The Office of Genetics and Disease Prevention of the US Centers for Disease and the editors of the American Journal of Epidemiology and Epidemiological Reviews have announced that they are collaborating to launch the Human Genome Epidemiology Network (HuGE Net). The aim of HuGE Net, as set out in an editorial in the July issue of the American Journal of Epidemiology [Khoury, M. and Dorman, J.S. (1998) Am. J. Epidemiol. 148, 1-3] is to put the information emerging from the human genome project in an epidemiological context, by establishing an international network for collecting and disseminating peer-reviewed epidemiological information on human genes. A Web site has been set up to store the information, which will include such material as review articles, research papers, systematic reviews, meta-analyses, consensus conferences, editorials and workshop reports.

Comment: This appears to be a very interesting and useful development for public health genetics. Prototype HuGE Net gene reviews are already up on the Web site for the BRCA1 gene, the CFTR (cystic fibrosis) gene and the HbS (sick cell anaemia) allele of the beta globin gene. The PHGU will monitor the development of HuGE Net and its Web site, and highlight important new additions as they appear.

18 August 1998Suggestions that the normal function of BRCA1 is to help repair damaged DNA are strengthened by a report that cell lines lacking this gene are unable to repair DNA damaged by oxidising agents, and are readily killed by ionising radiation [Gowen, L.C. et al. (1998) Science 281, 1009-1012]. BRCA1-deficient cells are defective in the process of transcription-coupled repair, in which the cell preferentially repairs active genes.

Comment: Understanding the normal function of BRCA1 may eventually lead to more effective treatments for hereditary breast cancer. In the shorter term, the hypersensitivity of BRCA1-deficient cells to radiation may indicate that radiotherapy should be particularly effective in killing tumour cells in BRCA1 mutation carriers.

10 August 1998An editorial in the BMJ urges caution about the much-vaunted announcement that a common mutation in the alpha2-microglobulin (A2M) gene may be a risk factor for late-onset Alzheimer's disease (see recent Newsletter item) [Edwardson, J. and Morrison, C. (1998) BMJ 317, 361-362 (link to full text, from this Web site only, by permission of the BMJ)]. Writing after the 6th International Conference on Alzheimer's Disease and Related Disorders, held recently in Amsterdam, the authors point out that several groups at the conference said they had not been able to confirm the association, which was uncovered by an approach called 'family-based association', in which the genotype of individuals with Alzheimer's disease was compared with the genotype of their unaffected siblings. Population-based studies are needed before the A2M mutation can be confirmed as an Alzheimer's risk factor.

Comment: A salutary reminder that no genetic association or linkage can be considered firm - especially in the field of psychiatric and behavioural genetics - until it is independently confirmed. But the public health conclusions remain the same: whether the A2M association is confirmed or not, there is at present no justification for presymptomatic testing for genetic risk factors for late-onset Alzheimer's disease.  

18 August 1998Genetic variation in proteins important for immune responses may explain why some people develop persistent arthritis after infection with Lyme disease, while others do not [Gross, D.M. et al. (1998) Science 281, 703-706]. In people who have a particular variant of one of the HLA genes (which encode proteins that participate in recognising 'foreign' antigens), a protein on the surface of the Lyme disease bacterium appears to trigger an autoimmune response against a similar human protein, leading to immune attack on joint tissues, particularly in the knee.

Comment: There are still many unanswered questions about the role of genetic factors in disease responses, but it may be feasible (and cost-effective) in the future to test people for disease-response variants in order to determine the likely course of the disease and thus what treatment is appropriate. More immediately, this work has important implications for vaccine development: any vaccine based on the Lyme disease protein that triggers the autoimmune response might itself cause arthritis. 

3 August 1998A large UK study of two-year-old twins suggests that delay in learning to speak (known as specific language impairment or SLI) is a largely genetic trait [Dale, P.S. et al. (1998) Nature Neurosci. 1, 324-328; see also Editorial on p. 259 of the August issue of the journal]. Although for 95% of the population genetics makes a relatively small contribution (about 25%) to vocabulary acquisition at this age, a much higher heritability (73%) was found for the 5% of children who are slowest to begin speaking. In other words, these children do not represent the tail of a normal distribution, but have a distinct genetic condition.

Comment: The discovery of the genes underlying SLI is likely to be some way off, and it may well be a genetically heterogeneous condition. Even if the genes can be found, it seems very unlikely that there would ever be justification for genetic testing or screening for the variants associated with SLI. The value of uncovering the genetic basis of SLI may lie rather in what it could reveal about how the brain functions in language acquisition. In public health terms, if SLI is indeed largely genetic, what are the practical consequences? Firstly, do children with genetically-based SLI go on to have further language and/or behavioural difficulties, or are they just late starters who eventually catch up with their peers? If the former is the case, can anything be done about SLI? Rather than landing us in the realms of genetic determinism, the recognition that SLI is a genetic condition might on the contrary lead to the discovery that these children respond a quite different range of interventions from those that have been found to improve language acquisition in the 'normal' population.

18 August 1998Oral contraceptive use may help prevent ovarian cancer in women carrying mutations in the BRCA genes [Narod, S. et al. (1998) N. Engl. J Med. 339, 424-428 (Summary)]. It has been known for some time that the Pill has a protective effect against ovarian cancer in women in the general population, but this is the first evidence that it may also work in women with a strong hereditary risk of the disease (lifetime risk approximately 45% in BRCA1 carriers and 25% in BRCA2 carriers; see summary on Breast cancer for further information about the BRCA genes and cancer risk). In the study, 207 BRCA mutation carriers who had ovarian cancer were compared with 179 of their unaffected sisters. About half of these sisters themselves carried BRCA mutations. Oral contraceptive use for less than 3 years decreased the risk of ovarian cancer by 20%, and the protective effect increased to 60% for long-term use (more than 6 years).

Comment: There are several caveats associated with this study (see Editorial by Rubin in the same issue of NEJM). Ideally, the control group should have been all unaffected mutation carriers, for example, and the fact that one-third of the control women had undergone oophorectomy might have affected the results. Nevertheless, the well-documented protective effect of oral contraceptive use in unselected women does increase confidence that the results of this study will be borne out by future research. The possibility of some protection by oral contraceptive use provides a welcome additional option for women who are known to carry BRCA mutations, though the protective effect of oophorectomy is greater. Contraceptive use may be advisable for women whose mutation status is unknown but who are members of high-risk families. One important unresolved question is whether oral contraceptive use might increase the already high risk of breast cancer in women with BRCA mutations.