© 2002 Nature Publishing Group

BRCA1 or BRCA2. This is consistent with pre-vious studies showing that BRCA1 andCHEK2 participate in the same biochemicalpathway, and also indicates that CHEK2might share a common biochemical pathwaywith BRCA2.

The CHEK2 protein is activated by theDNA-damage-activated kinase ATM. CHEK2blocks entry into mitosis at the G2 phase byactivating p53 and inhibiting the pro-mitotictyrosine phosphatase CDC25C. Consequently,cells with null mutations in CHEK2 cannotstop and repair damaged DNA; this leads togenomic instability and, potentially, cancer.

So far, the authors have screened only forthe CHEK2*1100delC mutation; other muta-tions in the same gene might also be expectedto contribute to breast cancer risk, with vary-ing penetrance depending on the severity ofthe mutation. So, the finding of this low-pene-trance susceptibility gene for breast cancermight allow us to remove several pieces fromthe chess board.

Michael StebbinsAssistant Editor, Nature Genetics

References and linksORIGINAL RESEARCH PAPER Meijers-Heijboer, H. et al.Low-penetrance susceptibility to breast cancer due toCHEK2*1100delC in noncarriers of BRCA1 and BRCA2mutations. Nature Genet. 31, 55–59 (2002)

328 | MAY 2002 | VOLUME 2 www.nature.com/reviews/cancer

If breast cancer genetics were a game of chess,we wouldn’t be close to winning it yet. We’vetaken two pieces off the board — BRCA1 andBRCA2 — but mutations in these genes accountfor only a small fraction of breast cancers.Hanne Meijers-Heijboers et al. have nowclaimed another pawn: they report in NatureGenetics that a truncating mutation in a cell-cycle checkpoint kinase, called CHEK2, mightbe responsible for 1% of all female breast cancercases and 9% of all male breast cancer cases.

The authors screened families that did nothave mutations in either BRCA1 or BRCA2, butdid have a history of breast cancer, for theCHEK2*1100delC mutation. This mutant geneyields a protein that is devoid of kinase activityin vitro. Of the 1,071 individuals screened, 55(5.1%) carried the CHEK2*1100delC mutation,compared with only 18 of 1,620 (1.1%) controlindividuals. When these numbers were brokendown further, a striking 13.5% of individualsfrom families that included cases of male breastcancer carried the variant. The authors concludethat in breast cancer families without BRCA1 orBRCA2 mutations, CHEK2*1100delC confers atwofold increased risk of breast cancer in womenand a tenfold increased risk in men.

The mutation does not increase the risk ofbreast cancer in people carrying mutations in

A simple urine test to detect cancer at anearly stage is every oncologist’s dream, butfor childhood neuroblastoma, for which justsuch a test is available, two groups report thatit doesn’t save lives.

These common childhood tumoursproduce catecholamines, the metabolites ofwhich can be measured in urine. This formsthe basis of a diagnostic test that ismandatory in Japan. But the validity of thestudy that first showed the test to improvesurvival has been questioned because datafrom the control group were collected beforedata from the screened group. Furthermore,childhood neuroblastoma can regressspontaneously, so detecting these tumoursmight actually increase mortality fromunnecessary surgery.

Two groups have now compared largepopulations of screened and unscreenedbabies contemporaneously. William G.Woods and colleagues screened over 425,000babies in the Quebec province, Canada, atthree weeks and six months of age and

followed them until the age of 8. Comparedwith four different populations ofunscreened children, the cumulativemortality rate from neuroblastoma was nolower in the screened group. Moreover, thetest failed to detect almost all the tumourswith poor prognostic indicators, such asMYCN amplifications.

Freimut Schilling and colleagues screenednearly 1.5 million babies at one year of age insix states in Germany, and compared theincidence of late-stage neuroblastoma (stage 4)and mortality with that in babies born at thesame time in the other ten states. Theincidence of stage-4 neuroblastoma andmortality were the same in the screened andunscreened groups, despite the fact that moreneuroblastomas were detected in thescreened population.

Both studies therefore argue againstpopulation screening for neuroblastoma ininfancy. They also highlight the need formore research into the two forms of thedisease — a highly aggressive one that

seems to evade the urine test, and a moreinsidious form that is readily detected butfrequently regresses of its own accord.

Cath Brooksbank

References and linksORIGINAL RESEARCH PAPER Woods, W. G. et al.Screening of infants and mortality due to neuroblastoma. N. Engl. J. Med. 346, 1041–1046 (2002) | Schilling, F. H. et al. Neuroblastoma screening at one year of age. N. Engl. J. Med. 346, 1041–1046 (2002)

Don’t take the p…

D I A G N O S T I C S

Checkmate

B R E A S T C A N C E R

H I G H L I G H T S