No abstract

will be interesting to see how the reformed NHS deals with this question. This is vet another reason for our erring on the side of financial caution. ConclusionWe have to ask ourselves whether or not we are really enhancing patient choice and using monev to the best effect. What is absolutely certain is that we are spending a lot of management and clinical time making decisions on a relatively small number of patients. This time may not be used to its best advantage.

In a small initial study of oncogene activation in differentiated thyroid cancer, using DNA transfection and tumorigenicity assays, we identified activating ras mutations in 80% (4/5) of follicular carcinomas, but in only 20% (2/10) of papillary carcinomas (Lemoine et al., 1988). However, Suarez et al. (1988), using the 'focus' transformation assay, reported two out of a total of three papillary carcinomas to have activated ras oncogenes, while Fusco et al. (1987), also using DNA transfection techniques, failed to find any activating ras mutations in 20 papillary cases analysed. These conflicting results indicated the need for further study of the prevalence of ras mutations in thyroid cancer.We now report a further series of papillary carcinomas, analysed by polymerase chain reaction (PCR) amplification and oligonucleotide probing of archival material, and summarise our data on the incidence and type of mutation in a total of 33 differentiated thyroid carcinomas.Wax-embedded tissue blocks of eight papillary thyroid carcinomas (up to 7 years old) were obtained from the archives of the Department of Pathology, University of Wales College of Medicine. Sections of 5.m were deparaffinised twice with xylene, washed twice with ethanol and vacuum desiccated (Shibata et al., 1988). Dried pellets were mixed with sterile water, denatured for 10 min at 95°C and rapidly cooled to 4'C. PCR incubations were prepared with a Perkin Elmer/Cetus Gene Amp kit according to the manufacturers' protocol. Final concentrations were 10 mM Tris pH 8.3, 50 mM KCI, 1.5 mM MgCl2, 0.2 mM of each dNTP, 1 tIM of each of a pair of amplimers and 2.5 units of Taq DNA polymerase (Saiki et al., 1986). Amplimers were 20 bases long, each pair enclosing an amplified region of 100 bp. Samples received 50 cycles of amplification in a Cetus thermal cycler, with 1 min denaturation at 94°C, 2.5 min annealing at 55'C and 3 min extension at 72°C. A 10 pl aliquot of each incubation was run on a 2.5% agarose gel to check for specific amplification, and reactions were repeated until sharp single bands were obtained.After an initial quantitation, 5-12gdl of each PCR-amplified mixture was diluted to 200 t l I in 1O mM Tris pH 8.0/ 1 mM EDTA, denatured at 95°C for 5 min and cooled rapidly to 4'C. Replicate slotblots, prepared using Hybond-N filters (Amersham) and a Millipore vacuum slotblot apparatus were prehybridised at 56°C for 30 min in 3 M tetramethylammonium chloride (TEMAC)/50 mM Tris pH 7.5/2 mM EDTA/0.3% SDS/5 x Denhardts solution, 100 flg ml-' denatured herring sperm DNA (Wood et al., 1985). Filters were hybridised for 1-2 h at 56°C with 32P-endlabelled 20mer oligonucleotide probes specific for mutations altering aminoacids at the 12, 13, and 61 positions of each of the three human ras genes. As a check on quantitation, filters were first autoradiographed after non-stringent washes in 2 x SSPE/0.1% SDS at room temperature. They were then Correspondence:

We have collated the results of cystic fibrosis (CF) mutation analysis conducted in 22 laboratories in the United Kingdom. A total of 9,807 CF chromosomes have been analysed, demonstrating 56 different mutations so far observed and accounting for 86% of CF genes in the native Caucasian population of the United Kingdom. delta F508 is the most common at 75.3% of CF mutations (range 56.5-83.7%), followed by G551D (3.08%; range 0.71-7.60%), G542X (1.68%; range 0.85-3.66%), 621 + 1 (G > T) (0.93%; range 0.41-3.16%), 1717-1(G > A) (0.57%; range 0.17-1.14%), 1898 + 1)(G > A) (0.46%), R117H (0.46%), N1303K (0.46%), and R553X (0.46%). The data show a clear geographical variation in the distribution of some of the mutations, most notably a marked regional variation in the distribution of 621 + 1 (G > T) and 1989 + 1(G > A), which are both apparently more frequent in Wales. R560T and R117H appear to be more frequent in Ireland and Scotland, and G551D more frequent in Scotland. In summary, these data illustrate that the mutations present within a particular population need to be defined in order to provide meaningful carrier screening and testing for rare mutations in affected individuals. Furthermore, it is apparent that the ethnic origin of a patient, even within a small country such as the United Kingdom, should be taken into account.

Our results suggest that ultrasonographic detection of fetal echogenic bowel is not associated with an increased prevalence of cystic fibrosis mutations in pregnancies at low risk for this disease.