Meta-analysis of the association between emphysematous change on thoracic computerized tomography scan and recurrent pneumothorax

OBJECTIVES

At least a third of patients go on to suffer a recurrence following a first spontaneous pneumothorax. Surgical intervention reduces the risk of recurrence and has been advocated as a primary treatment for pneumothorax. But surgery exposes patients to the risks of anaesthesia and in some cases can cause chronic pain. Risk stratification of patients to identify those most at risk of recurrence would help direct the most appropriate patients to early intervention. Many studies have addressed the role of thoracic computerized tomography (CT) in identifying those individuals at increased risk of recurrence, but a consensus is lacking.

AIM

Our objective was to clarify whether CT provides valuable prognostic information for recurrent pneumothorax.

DESIGN

Meta-analysis.

METHODS

We conducted an exhaustive search of the literature for thoracic CT imaging and pneumothorax, and then performed a meta-analysis using a random effects model to estimate the common odds ratio and standard error.

RESULTS

Here, we show by meta-analysis of data from 2475 individuals that emphysematous change on CT scan is associated with a significant increased odds ratio for recurrent pneumothorax ipsilateral to the radiological abnormality (odds ratio 2.49, 95% confidence interval 1.51-4.13).

CONCLUSIONS

The association holds true for primary spontaneous pneumothorax when considering emphysematous changes including blebs and bullae. Features, such as bullae at the azygoesophageal recess or increased Goddard score similarly predicted recurrent secondary pneumothorax, as shown by subgroup analysis. Our meta-analysis suggests that CT scanning has value in risk stratifying patients considering surgery for pneumothorax.

Abstract P1-09-04: Association between a breast cancer polygenic risk score and contralateral breast cancer risk

Background

Breast cancer patients are at significant risk of a second contralateral, breast cancer (CBC). Identification of women at high or low CBC risk could improve patient management decisions. Previous research has shown that breast cancer-associated single nucleotide polymorphisms (SNPs) summarized in a polygenic risk score (PRS) predict the risk of a first breast cancer with an odds ratio (OR) per 1 SD of 1.55 (95% confidence interval (95%CI)=1.52-1.58) (77-SNP PRS). The aim of this study was to evaluate the association between a recently developed PRS and CBC risk.

Methods

We identified 19 studies from the Breast Cancer Association Consortium (BCAC) with follow-up information on participating patients and at least 10 patients diagnosed with CBC. This included 38,228 females of European ancestry diagnosed with first invasive breast cancer since 1990. Genotyping was done using the iCOGS array or OncoArray, with genotypes for SNPs not on the arrays estimated by imputation. We used a 313-SNP PRS, optimized for prediction of overall (first) breast cancer in the BCAC dataset. Metachronous CBC risk by PRS was quantified using univariable and multivariable Cox regression analyses stratified by country and adjusted for multiple patient, tumor, and treatment characteristics. We assessed PRS interaction with age, family history, adjuvant systemic therapy, and ER-status.

Results

Median time to develop a CBC (N=1,046) after a first breast cancer was 5.8 years (range 0.3-21.9). Higher PRS was associated with increased CBC risk: hazard ratio (HR) per 1 SD=1.31 (95%CI=1.23-1.39). Patients in the highest and lowest 5% of the PRS had 1.95 fold and 0.67 fold risks of CBC, respectively, compared with patients in the middle quintile. Adjustments for age, year of diagnosis, family history, tumor size, nodal status, ER-status, or treatment (chemotherapy, endocrine therapy, radiotherapy) did not substantially alter these results. We found an interaction with age at first breast cancer diagnosis (Pinteraction=.002); the PRS was associated with an increased CBC risk for patients aged ≥40 years (HR=1.37, 95%CI=1.28-1.47), but not for patients <40 years (HR=1.06, 95%CI=0.93-1.21).

Conclusion

The PRS is predictive for the development of CBC in patients ≥40 years at first breast cancer diagnosis. For this group, the PRS could be incorporated in CBC risk prediction models to help define high and low risk patients, and hence optimize screening and treatment strategies.

Citation Format: Kramer I, Hooning MJ, Breast Cancer Association Consortium (BCAC), Mavaddat N, Canisius S, Keeman R, van den Broek AJ, Steyerberg E, Hauptmann M, Pharoah PD, Easton DF, Hall P, Schmidt MK. Association between a breast cancer polygenic risk score and contralateral breast cancer risk [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P1-09-04.

Rare, protein-truncating variants in ATM, CHEK2 and PALB2, but not XRCC2, are associated with increased breast cancer risks

BACKGROUND

Breast cancer (BC) is the most common malignancy in women and has a major heritable component. The risks associated with most rare susceptibility variants are not well estimated. To better characterise the contribution of variants in ATM, CHEK2, PALB2 and XRCC2, we sequenced their coding regions in 13 087 BC cases and 5488 controls from East Anglia, UK.

METHODS

Gene coding regions were enriched via PCR, sequenced, variant called and filtered for quality. ORs for BC risk were estimated separately for carriers of truncating variants and of rare missense variants, which were further subdivided by functional domain and pathogenicity as predicted by four in silico algorithms.

RESULTS

Truncating variants in PALB2 (OR=4.69, 95% CI 2.27 to 9.68), ATM (OR=3.26; 95% CI 1.82 to 6.46) and CHEK2 (OR=3.11; 95% CI 2.15 to 4.69), but not XRCC2 (OR=0.94; 95% CI 0.26 to 4.19) were associated with increased BC risk. Truncating variants in ATM and CHEK2 were more strongly associated with risk of oestrogen receptor (ER)-positive than ER-negative disease, while those in PALB2 were associated with similar risks for both subtypes. There was also some evidence that missense variants in ATM, CHEK2 and PALB2 may contribute to BC risk, but larger studies are necessary to quantify the magnitude of this effect.

CONCLUSIONS

Truncating variants in PALB2 are associated with a higher risk of BC than those in ATM or CHEK2. A substantial risk of BC due to truncating XRCC2 variants can be excluded.

Weibull regression with Bayesian variable selection to identify prognostic tumour markers of breast cancer survival

As data-rich medical datasets are becoming routinely collected, there is a growing demand for regression methodology that facilitates variable selection over a large number of predictors. Bayesian variable selection algorithms offer an attractive solution, whereby a sparsity inducing prior allows inclusion of sets of predictors simultaneously, leading to adjusted effect estimates and inference of which covariates are most important. We present a new implementation of Bayesian variable selection, based on a Reversible Jump MCMC algorithm, for survival analysis under the Weibull regression model. A realistic simulation study is presented comparing against an alternative LASSO-based variable selection strategy in datasets of up to 20,000 covariates. Across half the scenarios, our new method achieved identical sensitivity and specificity to the LASSO strategy, and a marginal improvement otherwise. Runtimes were comparable for both approaches, taking approximately a day for 20,000 covariates. Subsequently, we present a real data application in which 119 protein-based markers are explored for association with breast cancer survival in a case cohort of 2287 patients with oestrogen receptor-positive disease. Evidence was found for three independent prognostic tumour markers of survival, one of which is novel. Our new approach demonstrated the best specificity.

PALB2, CHEK2 and ATM rare variants and cancer risk: data from COGS

BACKGROUND

The rarity of mutations in PALB2, CHEK2 and ATM make it difficult to estimate precisely associated cancer risks. Population-based family studies have provided evidence that at least some of these mutations are associated with breast cancer risk as high as those associated with rare BRCA2 mutations. We aimed to estimate the relative risks associated with specific rare variants in PALB2, CHEK2 and ATM via a multicentre case-control study.

METHODS

We genotyped 10 rare mutations using the custom iCOGS array: PALB2 c.1592delT, c.2816T>G and c.3113G>A, CHEK2 c.349A>G, c.538C>T, c.715G>A, c.1036C>T, c.1312G>T, and c.1343T>G and ATM c.7271T>G. We assessed associations with breast cancer risk (42 671 cases and 42 164 controls), as well as prostate (22 301 cases and 22 320 controls) and ovarian (14 542 cases and 23 491 controls) cancer risk, for each variant.

RESULTS

For European women, strong evidence of association with breast cancer risk was observed for PALB2 c.1592delT OR 3.44 (95% CI 1.39 to 8.52, p=7.1×10-5), PALB2 c.3113G>A OR 4.21 (95% CI 1.84 to 9.60, p=6.9×10-8) and ATM c.7271T>G OR 11.0 (95% CI 1.42 to 85.7, p=0.0012). We also found evidence of association with breast cancer risk for three variants in CHEK2, c.349A>G OR 2.26 (95% CI 1.29 to 3.95), c.1036C>T OR 5.06 (95% CI 1.09 to 23.5) and c.538C>T OR 1.33 (95% CI 1.05 to 1.67) (p≤0.017). Evidence for prostate cancer risk was observed for CHEK2 c.1343T>G OR 3.03 (95% CI 1.53 to 6.03, p=0.0006) for African men and CHEK2 c.1312G>T OR 2.21 (95% CI 1.06 to 4.63, p=0.030) for European men. No evidence of association with ovarian cancer was found for any of these variants.

CONCLUSIONS

This report adds to accumulating evidence that at least some variants in these genes are associated with an increased risk of breast cancer that is clinically important.

Association of vitamin D levels and risk of ovarian cancer: a Mendelian randomization study

BACKGROUND

In vitro and observational epidemiological studies suggest that vitamin D may play a role in cancer prevention. However, the relationship between vitamin D and ovarian cancer is uncertain, with observational studies generating conflicting findings. A potential limitation of observational studies is inadequate control of confounding. To overcome this problem, we used Mendelian randomization (MR) to evaluate the association between single nucleotide polymorphisms (SNPs) associated with circulating 25-hydroxyvitamin D [25(OH)D] concentration and risk of ovarian cancer.

METHODS

We employed SNPs with well-established associations with 25(OH)D concentration as instrumental variables for MR: rs7944926 (DHCR7), rs12794714 (CYP2R1) and rs2282679 (GC). We included 31 719 women of European ancestry (10 065 cases, 21 654 controls) from the Ovarian Cancer Association Consortium, who were genotyped using customized Illumina Infinium iSelect (iCOGS) arrays. A two-sample (summary data) MR approach was used and analyses were performed separately for all ovarian cancer (10 065 cases) and for high-grade serous ovarian cancer (4121 cases).

RESULTS

The odds ratio for epithelial ovarian cancer risk (10 065 cases) estimated by combining the individual SNP associations using inverse variance weighting was 1.27 (95% confidence interval: 1.06 to 1.51) per 20 nmol/L decrease in 25(OH)D concentration. The estimated odds ratio for high-grade serous epithelial ovarian cancer (4121 cases) was 1.54 (1.19, 2.01).

CONCLUSIONS

Genetically lowered 25-hydroxyvitamin D concentrations were associated with higher ovarian cancer susceptibility in Europeans. These findings suggest that increasing plasma vitamin D levels may reduce risk of ovarian cancer.

Adult body mass index and risk of ovarian cancer by subtype: a Mendelian randomization study

BACKGROUND

Observational studies have reported a positive association between body mass index (BMI) and ovarian cancer risk. However, questions remain as to whether this represents a causal effect, or holds for all histological subtypes. The lack of association observed for serous cancers may, for instance, be due to disease-associated weight loss. Mendelian randomization (MR) uses genetic markers as proxies for risk factors to overcome limitations of observational studies. We used MR to elucidate the relationship between BMI and ovarian cancer, hypothesizing that genetically predicted BMI would be associated with increased risk of non-high grade serous ovarian cancers (non-HGSC) but not HGSC.

METHODS

We pooled data from 39 studies (14 047 cases, 23 003 controls) in the Ovarian Cancer Association Consortium. We constructed a weighted genetic risk score (GRS, partial F-statistic = 172), summing alleles at 87 single nucleotide polymorphisms previously associated with BMI, weighting by their published strength of association with BMI. Applying two-stage predictor-substitution MR, we used logistic regression to estimate study-specific odds ratios (OR) and 95% confidence intervals (CI) for the association between genetically predicted BMI and risk, and pooled these using random-effects meta-analysis.

RESULTS

Higher genetically predicted BMI was associated with increased risk of non-HGSC (pooled OR = 1.29, 95% CI 1.03-1.61 per 5 units BMI) but not HGSC (pooled OR = 1.06, 95% CI 0.88-1.27). Secondary analyses stratified by behaviour/subtype suggested that, consistent with observational data, the association was strongest for low-grade/borderline serous cancers (OR = 1.93, 95% CI 1.33-2.81).

CONCLUSIONS

Our data suggest that higher BMI increases risk of non-HGSC, but not the more common and aggressive HGSC subtype, confirming the observational evidence.

Hypertension and mortality in the Golestan Cohort Study: A prospective study of 50 000 adults in Iran

High blood pressure has been the second most important determinant of disease burden in Iran since the 1990s. Despite well-recognized evidence on the association of high blood pressure and mortality in other countries, this relationship has not been fully investigated in the demographic setting of Iran. The current study is the first large-scale longitudinal study of this association in Iran. Briefly, 50 045 subjects between 40 and 75 years of age have been recruited and followed. Blood pressure measurements were carried out at baseline. Causes of death were reported and verified by verbal autopsy throughout the follow-up period. The outcomes of interest were all-cause deaths and deaths due to ischemic heart disease (IHD) or stroke. Cox proportional hazards regression models were used to estimate hazard ratios (HRs). A total of 46 674 subjects free from cardiovascular disease at baseline were analyzed. Absolute mortality rates increased along with increasing systolic or diastolic blood pressure above 120 and 80 mm Hg, respectively. Adjusted HRs (95% confidence intervals) for each 20 mm Hg increase in systolic blood pressure in all age groups were 1.18 (1.13-1.23) for all-cause mortality, 1.21 (1.13-1.31) for deaths due to IHD and 1.50 (1.39-1.63) for deaths due to stroke. Unadjusted and adjusted HRs were higher in younger subjects and decreased with increasing age of the participants. High blood pressure is a serious threat to the health of Iranians. The entire health-care system of Iran should be involved in a comprehensive action plan for controlling blood pressure.

Astronomical algorithms for automated analysis of tissue protein expression in breast cancer

BACKGROUND

High-throughput evaluation of tissue biomarkers in oncology has been greatly accelerated by the widespread use of tissue microarrays (TMAs) and immunohistochemistry. Although TMAs have the potential to facilitate protein expression profiling on a scale to rival experiments of tumour transcriptomes, the bottleneck and imprecision of manually scoring TMAs has impeded progress.

METHODS

We report image analysis algorithms adapted from astronomy for the precise automated analysis of IHC in all subcellular compartments. The power of this technique is demonstrated using over 2000 breast tumours and comparing quantitative automated scores against manual assessment by pathologists.

RESULTS

All continuous automated scores showed good correlation with their corresponding ordinal manual scores. For oestrogen receptor (ER), the correlation was 0.82, P<0.0001, for BCL2 0.72, P<0.0001 and for HER2 0.62, P<0.0001. Automated scores showed excellent concordance with manual scores for the unsupervised assignment of cases to 'positive' or 'negative' categories with agreement rates of up to 96%.

CONCLUSION

The adaptation of astronomical algorithms coupled with their application to large annotated study cohorts, constitutes a powerful tool for the realisation of the enormous potential of digital pathology.

Abstract P3-08-02: Common variants at 10p14 and 1p11.2 display heterogeneity in breast cancer associations by E-cadherin tumor tissue expression in two independent datasets

Background: E-cadherin is a tumor suppressor gene involved in cell-cell adhesion, epithelial-to-mesenchymal transitions (EMT) and invasion. Loss of E-cadherin expression is strongly associated with lobular breast cancers, which exhibit single cell patterns of infiltration and are often estrogen receptor positive. We sought to determine if relative risk estimates for 19 established breast cancer susceptibility loci were modified by E-cadherin breast tumor tissue expression.

Methods: Case-control analyses included up to 1885 invasive breast cancer cases and 2366 age and site matched controls aged 20–74 years from the Polish Breast Cancer Study (PBCS), a population based case-control study conducted in Poland from 2000–2003. Genotyping of the 19 single nucleotide polymorphisms (SNPs) was performed using TaqMan® assays. Tissue expression of E-cadherin was assessed using immunohistochemical (IHC) staining of tissue microarrays and IHC results were scored as the product of percent positive tumor cells × intensity. Tumors having a score of &lt;10 were classified as E-cadherin low and those with a score ≥10 as E-cadherin high. Polytomous logistic regression models adjusted for age and study site were used to estimate odds ratios (OR) and 95% confidence intervals (95% CI) for each breast cancer subtype, defined by E-cadherin expression levels, compared to controls. Case-only data from the PBCS (N = 797) and the Study of Epidemiology and Risk Factors in Cancer Heredity, SEARCH (N = 2155) was used in logistic regression models to test for heterogeneity of SNPs by E-cadherin expression.

Results: Three SNPs suggested significant heterogeneity by E-cadherin expression in the PBCS: rs2046210 at 6q25.1(ESR1) [per-allele ORs (95% CI); 1.53 (1.19–1.98) for E-cadherin low tumors and 0.99 (0.86–1.14) for E-cadherin high tumors, P-heterogeneity = 0.002]; rs1045485 at 10p14 (CASP8) [per-allele ORs (95% CI); 0.62 (0.41–0.93) for E-cadherin low tumors and 0.98 (0.81–1.18) for E-cadherin high tumors, P-heterogeneity = 0.04]; and rs11249433 at 1p11.2 (NOTCH2/FCGR1B) [per-allele ORs (95% CI); 1.29 (1.02–1.64) for E-cadherin low tumors and 1.01 (0.88–1.15) for E-cadherin high tumors, P-heterogeneity = 0.06]. Combined case-only analysis of PBCS and SEARCH for these three SNPs showed significant heterogeneity by E-cadherin expression for rs11249433 [Interaction OR (95% CI); 1.19 (1.05–1.36), P-heterogeneity = 0.007] and rs1045485 [Interaction OR (95% CI); 0.69 (0.53–0.90), P-heterogeneity = 0.007]. The association with rs2046210 [Interaction OR (95% CI); 1.12 (0.61–2.03), P-heterogeneity = 0.73] did not remain significant in combined analyses.

Conclusion: Our findings provide evidence that associations for breast cancer susceptibility loci vary by E-cadherin tumor tissue expression, which has not been described previously. Specifically, our results suggest that the genetic markers rs11249433 and rs1045485 may preferentially modify risk for tumors with low or absent E-cadherin expression in two independent data sets.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-08-02.

Abstract P3-08-05: Title: Copy number variation and risk of chemotherapy-related infection

Background: Copy-number variations (CNVs) are DNA changes that result in regions of the genome being either duplicated (copy number gains) or deleted (copy number losses). CNVs have been identified in 12% of the human genome. Although CNVs do not appear to have a major role in disease susceptibility, it is known that they are involved in drug metabolism, toxicity and response. Infection is a common dose and/or treatment limiting chemotherapy-related toxicity. Infections can be a major cause of morbidity and mortality. Although a limited number of studies have investigated the association of single nucleotide polymorphisms (SNPs) with severity of infection, no studies so far have identified CNV associations with chemotherapy-related infection. In a non-oncological setting, CNVs have been noted to be associated with altered risk of certain types of infections and severe sepsis. This study aims to evaluate whether CNV are associated with the risk of chemotherapy-related infection in a genome wide association study (GWAS) of breast cancer patients recruited to clinical trials and treated with chemotherapy regimens including paclitaxel, epirubicin, cyclophosphamide, gemcitabine, 5fluorouracil and methotrexate.

Method: DNA from blood and saliva samples were collected as part of the pharmacogenetics GWAS, PGSNPS. Chemotherapy-related infection was assessed using the NCI Common Toxicity Criteria (CTC). Patients (n = 1921 samples) were classified as cases (NCI CTC grades 2–4) or as controls (NCI CTC grades 0–1). The Affymetrix SNP6.0 array, which contains more than 946,000 probes for the detection of CNVs and more than 906,600 SNPs, was used as the copy number estimation platform. DNA was segmented using DNAcopy (circular binary segmentation). CNVs were called based on their median and standard deviation. Log-ratios larger than the median plus 2 times the standard deviation were considered gains and log-ratios with values smaller than 2.5 times the standard deviation were considered as losses. Only regions with a copy number alteration present in at least a 10% of the samples were considered for the analysis. Fisher's exact test was performed on each of these regions and the variable ‘infection’. 2-side p-values were obtained and adjusted using Benjamin-Hochberg correction.

Results: One CNV region (frequency of gain 0.16) showed an association with increased risk of infection (Odds Ratio=1.79, 95% Confidence Interval [1.34–2.40], P = 6.37 × 10−5). Five other CNV regions demonstrated copy number gains and three CNV regions showed copy number losses that were associated with an increased risk of infection (P = 10−4). All regions had a frequency of gain or loss &gt; 0.10 and were therefore not rare variants.

Conclusion: These data suggest that CNVs may play a role in increased risk of chemotherapy-related toxicity. The most significant CNVs associated with infection are currently undergoing validation using an independent cohort of adjuvant breast cancer patients with similar exposure to chemotherapy and comparable documentation of chemotherapy-related infection.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-08-05.

Aurora kinase A outperforms Ki67 as a prognostic marker in ER-positive breast cancer

Background:

Proliferation has emerged as a major prognostic factor in luminal breast cancer. The immunohistochemical (IHC) proliferation marker Ki67 has been most extensively investigated but has not gained widespread clinical acceptance.

Methods:

We have conducted a head-to-head comparison of a panel of proliferation markers, including Ki67. Our aim was to establish the marker of the greatest prognostic utility. Tumour samples from 3093 women with breast cancer were constructed as tissue microarrays. We used IHC to detect expression of mini-chromosome maintenance protein 2, Ki67, aurora kinase A (AURKA), polo-like kinase 1, geminin and phospho-histone H3. We used a Cox proportional-hazards model to investigate the association with 10-year breast cancer-specific survival (BCSS). Missing values were resolved using multiple imputation.

Results:

The prognostic significance of proliferation was limited to oestrogen receptor (ER)-positive breast cancer. Aurora kinase A emerged as the marker of the greatest prognostic significance in a multivariate model adjusted for the standard clinical and molecular covariates (hazard ratio 1.3; 95% confidence interval 1.1–1.5; P=0.005), outperforming all other markers including Ki67.

Conclusion:

Aurora kinase A outperforms other proliferation markers as an independent predictor of BCSS in ER-positive breast cancer. It has the potential for use in routine clinical practice.

Comparison of methods for handling missing data on immunohistochemical markers in survival analysis of breast cancer

BACKGROUND

Tissue micro-arrays (TMAs) are increasingly used to generate data of the molecular phenotype of tumours in clinical epidemiology studies, such as studies of disease prognosis. However, TMA data are particularly prone to missingness. A variety of methods to deal with missing data are available. However, the validity of the various approaches is dependent on the structure of the missing data and there are few empirical studies dealing with missing data from molecular pathology. The purpose of this study was to investigate the results of four commonly used approaches to handling missing data from a large, multi-centre study of the molecular pathological determinants of prognosis in breast cancer.

PATIENTS AND METHODS

We pooled data from over 11,000 cases of invasive breast cancer from five studies that collected information on seven prognostic indicators together with survival time data. We compared the results of a multi-variate Cox regression using four approaches to handling missing data - complete case analysis (CCA), mean substitution (MS) and multiple imputation without inclusion of the outcome (MI-) and multiple imputation with inclusion of the outcome (MI+). We also performed an analysis in which missing data were simulated under different assumptions and the results of the four methods were compared.

RESULTS

Over half the cases had missing data on at least one of the seven variables and 11 percent had missing data on 4 or more. The multi-variate hazard ratio estimates based on multiple imputation models were very similar to those derived after using MS, with similar standard errors. Hazard ratio estimates based on the CCA were only slightly different, but the estimates were less precise as the standard errors were large. However, in data simulated to be missing completely at random (MCAR) or missing at random (MAR), estimates for MI+ were least biased and most accurate, whereas estimates for CCA were most biased and least accurate.

CONCLUSION

In this study, empirical results from analyses using CCA, MS, MI- and MI+ were similar, although results from CCA were less precise. The results from simulations suggest that in general MI+ is likely to be the best. Given the ease of implementing MI in standard statistical software, the results of MI+ and CCA should be compared in any multi-variate analysis where missing data are a problem.

A large-scale meta-analysis to refine colorectal cancer risk estimates associated with MUTYH variants

Background:

Defective DNA repair has a causal role in hereditary colorectal cancer (CRC). Defects in the base excision repair gene MUTYH are responsible for MUTYH-associated polyposis and CRC predisposition as an autosomal recessive trait. Numerous reports have suggested MUTYH mono-allelic variants to be low penetrance risk alleles. We report a large collaborative meta-analysis to assess and refine CRC risk estimates associated with bi-allelic and mono-allelic MUTYH variants and investigate age and sex influence on risk.

Methods:

MUTYH genotype data were included from 20 565 cases and 15 524 controls. Three logistic regression models were tested: a crude model; adjusted for age and sex; adjusted for age, sex and study.

Results:

All three models produced very similar results. MUTYH bi-allelic carriers demonstrated a 28-fold increase in risk (95% confidence interval (CI): 6.95–115). Significant bi-allelic effects were also observed for G396D and Y179C/G396D compound heterozygotes and a marginal mono-allelic effect for variant Y179C (odds ratio (OR)=1.34; 95% CI: 1.00–1.80). A pooled meta-analysis of all published and unpublished datasets submitted showed bi-allelic effects for MUTYH, G396D and Y179C (OR=10.8, 95% CI: 5.02–23.2; OR=6.47, 95% CI: 2.33–18.0; OR=3.35, 95% CI: 1.14–9.89) and marginal mono-allelic effect for variants MUTYH (OR=1.16, 95% CI: 1.00–1.34) and Y179C alone (OR=1.34, 95% CI: 1.01–1.77).

Conclusions:

Overall, this large study refines estimates of disease risk associated with mono-allelic and bi-allelic MUTYH carriers.

A genome-wide association scan on estrogen receptor-negative breast cancer

Introduction

Breast cancer is a heterogeneous disease and may be characterized on the basis of whether estrogen receptors (ER) are expressed in the tumour cells. ER status of breast cancer is important clinically, and is used both as a prognostic indicator and treatment predictor. In this study, we focused on identifying genetic markers associated with ER-negative breast cancer risk.

Methods

We conducted a genome-wide association analysis of 285,984 single nucleotide polymorphisms (SNPs) genotyped in 617 ER-negative breast cancer cases and 4,583 controls. We also conducted a genome-wide pathway analysis on the discovery dataset using permutation-based tests on pre-defined pathways. The extent of shared polygenic variation between ER-negative and ER-positive breast cancers was assessed by relating risk scores, derived using ER-positive breast cancer samples, to disease state in independent, ER-negative breast cancer cases.

Results

Association with ER-negative breast cancer was not validated for any of the five most strongly associated SNPs followed up in independent studies (1,011 ER-negative breast cancer cases, 7,604 controls). However, an excess of small P-values for SNPs with known regulatory functions in cancer-related pathways was found (global P = 0.052). We found no evidence to suggest that ER-negative breast cancer shares a polygenic basis to disease with ER-positive breast cancer.

Conclusions

ER-negative breast cancer is a distinct breast cancer subtype that merits independent analyses. Given the clinical importance of this phenotype and the likelihood that genetic effect sizes are small, greater sample sizes and further studies are required to understand the etiology of ER-negative breast cancers.

COGENT (COlorectal cancer GENeTics): an international consortium to study the role of polymorphic variation on the risk of colorectal cancer

It is now recognised that a part of the inherited risk of colorectal cancer (CRC) can be explained by the co-inheritance of low-penetrance genetic variants. The accumulated experience to date in identifying these variants has served to highlight difficulties in conducting statistically and methodologically rigorous studies and follow-up analyses. The COGENT (COlorectal cancer GENeTics) consortium includes 20 research groups in Europe, Australia, the Americas, China and Japan. The overarching goal of COGENT is to identify and characterise low-penetrance susceptibility variants for CRC through association-based analyses. In this study, we review the rationale for identifying low-penetrance variants for CRC and our proposed strategy for establishing COGENT.

HapMap-based study of the 17q21 ERBB2 amplicon in susceptibility to breast cancer

ERBB2 is frequently amplified in breast tumours as part of a wide region of amplification on chromosome 17q21. This amplicon contains many candidate genes for breast cancer susceptibility. We used a genetic association study design to determine if common genetic variation (frequency>or=5%) in a 400-kb region surrounding ERBB2 and containing the PPARBP, CRK7, NEUROD2, PPP1R1B, STARD3, TCAP, PNMT, CAB2, ERBB2, C17ORF37, GRB7 and ZNFN1A3 genes, was associated with breast cancer risk. Sixteen tagging single-nucleotide polymorphisms (tSNPs) selected within blocks of linkage disequilibrium from the HapMap database, one HapMap singleton SNP, and six additional SNPs randomly selected from dbSNP were genotyped using Taqman in a large study set of British women (2275 cases, 2280 controls). We observed no association between any of the genotypes or associated haplotypes and disease risk. In order to simulate unidentified SNPs, we performed the leave-one-out cross-validation procedure on the HapMap data; over 90% of the common genetic variation was well represented by tagging polymorphisms. We are therefore likely to have tagged any common variants present in our population. In summary, we found no association between common genetic variation in the 17q21 ERBB2 amplicon and breast cancer risk in British women.

Single-nucleotide polymorphisms in the RB1 gene and association with breast cancer in the British population

A substantial proportion of the familial risk of breast cancer may be attributable to genetic variants each contributing a small effect. pRb controls the cell cycle and polymorphisms within it are candidates for such low penetrance susceptibility alleles, since the gene has been implicated in several human tumours, particularly breast cancer. The purpose of this study was to determine whether common variants in the RB1 gene are associated with breast cancer risk. We assessed 15 tagging single-nucleotide polymorphisms (SNPs) using a case–control study design (n⩽4474 cases and n⩽4560 controls). A difference in genotype frequencies was found between cases and controls for rs2854344 in intron 17 (P-trend=0.007) and rs198580 in intron 19 (P-trend=0.018). Carrying the minor allele of these SNPs appears to confer a protective effect on breast cancer risk (odd ratio (OR)=0.86 (0.76–0.96) for rs2854344 and OR=0.80 (0.66–0.96) for rs198580). However, after adjusting for multiple testing these associations were borderline with an adjusted P-trend=0.068 for the most significant SNP (rs2854344). The RB1 gene is not known to contain any coding SNPs with allele frequencies ⩾5% but several intronic variants are in perfect linkage disequilibrium with the associated SNPs. Replication studies are needed to confirm the associations with breast cancer.

Development of a protocol for evaluation of mammographic surveillance services in women under 50 with a family history of breast cancer

BACKGROUND

Preliminary retrospective data suggest it is possible to identify impalpable breast cancer in women presenting with a family history of breast cancer under the age of 50, by using regular mammography. In consequence, this service is offered in a number of centres in the UK. The effectiveness of such a service, however, has not been fully evaluated.

METHODS

We propose to perform such an evaluation in a cohort of 20000 women under the age of 50 with a significant family history of breast cancer, given regular mammographic surveillance over 5 years. Comparison of surgical and pathological data with completed and ongoing population screening trials using analysis techniques of varying complexity will be performed to obtain an accurate prediction of future breast-cancer mortality reduction. The formal aims are: i) to estimate the difference in breast-cancer mortality in women under the age of 50 with a significant family history of breast cancer having regular mammography, compared with those not being screened; ii) to estimate the cost-effectiveness of regular mammography in this group of women, compared with no screening. The increase in health service resource use attributable to such a policy will be compared with no screening, and costed. Incremental cost-effectiveness ratios of implementing the standardised mammography strategy compared with no screening will be presented in terms of the additional cost per cancer detected, per life saved and per life-year saved.

Incidence of gastric cancer and breast cancer in CDH1 (E-cadherin) mutation carriers from hereditary diffuse gastric cancer families

BACKGROUND & AIMS

Germline mutations in CDH1 are known to cause hereditary diffuse gastric cancer (HDGC). Breast and colorectal cancer have also been reported in CDH1-associated HDGC. The purpose of this study was to estimate the cumulative risk of gastric and breast cancer in CDH1 mutation carriers.

METHODS

Family data were collected by member groups of the International Gastric Cancer Linkage Consortium. Eligible families had at least 3 cases of diffuse gastric cancer, and at least 1 affected member had tested positive for a mutation in CDH1. Eleven families met these criteria. We used the pedigree information to estimate penetrance using the MENDEL program. The conditional likelihood of the pedigree was maximized given the phenotype of the pedigree and genotype of the index case at ascertainment. We parameterized the model in terms of log relative risks for mutation carriers compared with risks in the general population of the United Kingdom. Noncarriers of the gene were assumed to develop the disease at population incidence rates.

RESULTS

The estimated cumulative risk of gastric cancer by age 80 years was 67% for men (95% confidence interval [95% CI], 39-99) and 83% for women (95% CI, 58-99). For women, the cumulative risk of breast cancer was 39% (95% CI, 12-84). The combined risk of gastric cancer and breast cancer in women was 90% by age 80 years.

CONCLUSIONS

These penetrance estimates should be useful for genetic counseling in multiple-case families. However, they may not apply to individuals with a minimal family history, in whom the risks may be lower.

Polymorphisms in CYP1A1 and smoking: no association with breast cancer risk

Several studies have investigated polymorphisms in CYP1A1 and breast cancer risk with inconsistent results. We have carried out a population based case-control study of the Thr461Asn and Ile462Val polymorphisms in CYP1A1 to clarify their importance in determining breast cancer susceptibility. A total of 1873 cases and 712 controls were genotyped for Thr461Asn and 1948 cases and 1355 controls were genotyped for Ile462Val. We have also investigated a putative interaction between smoking and CYP1A1 genotype and breast cancer risk using a case only study design. The genotype distribution of Thr461Asp in controls was close to that expected under Hardy-Weinberg equilibrium (HWE). We detected no significant differences in genotype frequencies between breast cancer cases and controls (P = 0.68). Compared with the Thr/Thr homozygotes there was no significant risk for either the Thr/Asp heterozygote [OR = 1.1 (95% CI 0.8-1.4)] or the Asp/Asp homozygote [OR = 0.4 (0.02-6.1)]. The genotype distribution of Ile462Val in controls was also close to that expected under HWE with no significant differences between breast cancer cases and the controls (P = 0.44). No significant risk was found for either the Ile/Val heterozygote [OR = 0.8 (0.6-1.1)] or the Val/Val homozygote [OR = 2.7 (0.3-24)] compared with the Ile/Ile homozygotes. Furthermore, subgroup analyses revealed no effect of age or menopausal status on genotypic risks, and we found no evidence for an interaction between genotype and smoking habit or alcohol consumption and susceptibility to breast cancer. We combined our data for the Ile462Val polymorphism with those from four other published studies, but even with >5000 subjects, none of the genotype-associated risks achieved statistical significance, and there was no consistent pattern to the risks associated with increasing Val allele dosage [Ile/Val OR = 0.9 (0.7-1.1), Val/Val OR = 2.3 (0.4-12), and Val carrier OR = 1.0 (0.9-1.1)].

Evidence for further breast cancer susceptibility genes in addition to BRCA1 and BRCA2 in a population-based study

We used data from a population based series of breast cancer patients to investigate the genetic models that can best explain familial breast cancer not due to the BRCA1 and BRCA2 genes. The data set consisted of 1,484 women diagnosed with breast cancer under age 55 registered in the East Anglia Cancer registry between 1991-1996. Blood samples taken from the patients were analysed for mutations in BRCA1 and BRCA2. The genetic models were constructed using information on breast and ovarian cancer history in first-degree relatives and on the mutation status of the index patients. We estimated the simultaneous effects of BRCA1, BRCA2, a third hypothetical gene BRCA3, and a polygenic effect. The models were assessed by likelihood comparisons and by comparison of the observed numbers of mutations and affected relatives with the predicted numbers. BRCA1 and BRCA2 could not explain all the familial clustering of breast cancer. The best-fitting single gene model for BRCA3 was a recessive model with a disease allele frequency 24% and penetrance 42% by age 70. However, a polygenic model gave a similarly good fit. The estimated population frequencies for BRCA1 and BRCA2 mutations were similar under both recessive and polygenic models, 0.024 and 0.041%, respectively. A dominant model for BRCA3 gave a somewhat worse fit, although the difference was not significant. The mixed recessive model was identical to the recessive model and the mixed dominant very similar to the polygenic model. The BRCA3 genetic models were robust to the BRCA1 and BRCA2 penetrance assumptions. The overall fit of all models was improved when the known effects of parity on breast and ovarian cancer risks were included in the model-in this case a polygenic model fits best. These findings suggest that a number of common, low-penetrance genes with additive effects may account for the residual non-BRCA1/2 familial aggregation of breast cancer, but Mendelian inheritance of an autosomal recessive allele cannot be ruled out.

Ovarian cancer survival in Ashkenazi Jewish patients with BRCA1 and BRCA2 mutations

PURPOSE

To compare the clinical characteristics and survival of Ashkenazi Jewish ovarian cancer patients with and without BRCA1 and BRCA2 mutations.

METHODS

An unselected series of 118 Ashkenazi Jewish ovarian cancer patients were screened for the three common founder mutations in BRCA1 and BRCA2. Patient survival and other clinical characteristics of the tumours were compared in patients with BRCA1 or BRCA2 mutations and those without mutations.

RESULTS

Twenty-seven individuals with invasive carcinomas were found to have mutations (14 with 185delAG and one with 5382insC in BRCA1 and 12 with 6174delT in BRCA2). No mutations were identified in the 20 patients with borderline tumours. For the invasive carcinomas, there was a survival advantage for BRCA1 and BRCA2 patients compared to patients without mutations, though the differences were not statistically significant. There were no significant differences in the histopathological characteristics of the tumours between the patient groups.

CONCLUSION

These results are similar to those of other studies and suggest that ovarian cancer in BRCA1 and BRCA2 mutation carriers may have a distinct clinical behaviour.

Apparent human BRCA1 knockout caused by mispriming during polymerase chain reaction: implications for genetic testing

We report an apparent BRCA1 homozygous knockout that, on further analysis, was found to be an artefact of the polymerase chain reaction. This finding has two important implications. First, it challenges results of a previous study that reported a homozygous knockout associated with the same BRCA1 mutation. Second, our findings suggest that mispriming caused by mismatched primers at the site of single-nucleotide polymorphisms, leading to preferential amplification of one allele, may represent a significant proportion of instances of mutation-detection insensitivity. This may have major implications for the sensitivity of all polymerase chain reaction-based mutation-detection methods in clinical genetic testing laboratories.

Degenerate oligonucleotide primed-polymerase chain reaction-based array comparative genomic hybridization for extensive amplicon profiling of breast cancers : a new approach for the molecular analysis of paraffin-embedded cancer tissue

We have developed a protocol for degenerate oligonucleotide-primed-polymerase chain reaction-based array comparative genomic hybridization (array CGH) that, when combined with a laser microdissection technique, allows the analysis of cancer cell populations isolated from routine, formalin-fixed, paraffin-embedded tissue samples. Comparison of copy number changes detected by degenerate oligonucleotide-primed-polymerase chain reaction-based array CGH to those detected by conventional array CGH or fluorescence in situ hybridization, demonstrated that amplifications can be reliably detected. Using a genomic microarray containing 57 oncogenes, we screened a total of 28 breast cancer samples and obtained a detailed amplicon profile that is the most comprehensive to date in human breast cancer. The array CGH method described here will allow the genetic analysis of paraffin-embedded human cancer materials for example in the context of clinical trials.

A common variant in BRCA2 is associated with both breast cancer risk and prenatal viability

Inherited mutations in the gene BRCA2 predispose carriers to early onset breast cancer, but such mutations account for fewer than 2% of all cases in East Anglia. It is likely that low penetrance alleles explain the greater part of inherited susceptibility to breast cancer; polymorphic variants in strongly predisposing genes, such as BRCA2, are candidates for this role. BRCA2 is thought to be involved in DNA double strand break-repair. Few mice in which Brca2 is truncated survive to birth; of those that do, most are male, smaller than their normal littermates and have high cancer incidence. Here we show that a common human polymorphism (N372H) in exon 10 of BRCA2 confers an increased risk of breast cancer: the HH homozygotes have a 1.31-fold (95% CI, 1.07-1.61) greater risk than the NN group. Moreover, in normal female controls of all ages there is a significant deficiency of homozygotes compared with that expected from Hardy-Weinberg equilibrium, whereas in males there is an excess of homozygotes: the HH group has an estimated fitness of 0.82 in females and 1.38 in males. Therefore, this variant of BRCA2 appears also to affect fetal survival in a sex-dependent manner.

Frequent loss of BRCA1 mRNA and protein expression in sporadic ovarian cancers

Germline mutations in the BRCA1 gene cause inherited susceptibility to breast and ovarian cancers. However, somatic mutations of BRCA1 are rare in sporadic breast and ovarian tumours. To establish whether BRCA1 is altered during the development of sporadic ovarian cancer by mechanisms other than somatic mutation, we have analysed 57 sporadic epithelial ovarian tumours for BRCA1 protein and RNA expression. Reduced or absent protein expression was observed in 90% of tumours. Decreased protein expression was significantly associated with a reduction in the levels of RNA expression. Somatic mutations of BRCA1 and LOH at the BRCA1 locus were detected in 3.5% and 44% of informative tumours, respectively; there was no significant correlation between the levels of protein and RNA expression and the DNA mutation and/or LOH status. Together, these data suggest that expression of BRCA1 is down-regulated at the level of transcription during the development of sporadic ovarian cancers.

Ovarian and breast cancer risks to women in families with two or more cases of ovarian cancer

There are few published estimates of the risk of developing breast or ovarian cancer in women with a strong family history of ovarian cancer. As these women commonly present to family cancer clinics, accurate cancer risk estimates are needed. We have estimated these risks in women from families with 2 or more confirmed ovarian cancers in first-degree relatives using data from the UKCCCR Familial Ovarian Cancer Register. The number of cancers observed in more than 10,000 person years of follow-up was compared with the number expected based on national-, age-, sex- and period-specific incidence rates. The relative risk of ovarian cancer was found to be 7.18 (95% CI 3.82-12.3), declining from 16.0 (6.40-32.9) in women under 50 to 4.38 (1.60-9.52) in women 50 years of age and older. For breast cancer, the relative risk for women under 50 was 3.74 (2.04-6. 28) and 1.79 (1.02-2.90) for women 50 years of age and older (average RR 2.36, 1.59-3.37). These correspond to absolute risks by age 70 of 11% for ovarian cancer and 15% for breast cancer. When the analyses were restricted to families that had been negative for mutations in the breast/ovarian cancer susceptibility genes, BRCA1 and BRCA2, the ovarian cancer risk was 11.59 (3.12-29.7) and that of breast cancer 3.32 (1.52-6.31). As well as having clinical relevance, our finding may suggest that other breast/ovarian cancer genes are segregating in these families, though the possibility of undetected BRCA1/2 mutations must also be considered.

Familial predisposition to breast cancer in a British population: implications for prevention

The aims of this study were to estimate the familial relative risk of breast cancer according to the age of the at risk individual and the age at which the relative was affected, and to estimate the proportion of the general population in several breast cancer risk categories because of a family history and, thus estimate the potential to reduce the overall breast cancer burden using interventions targeted at women at increased risk because of family history. Familial relative risks were computed by comparing breast cancer incidence in relatives of 2809 breast cancer cases from a population based case series with that expected from general population incidence rates. The proportion of the general population in different categories of risk according to family history was estimated from the relative risk of breast cancer for that category and the proportion of cases in that category. 389 (13.8%) cases had at least 1 first degree relative with breast cancer. The relative risk of breast cancer in sisters of index cases was 1.90 (95% confidence interval (CI) 1.63-2.36) and that in mothers 1.73 (1.52-1. 97). The risk to mothers of cases diagnosed under 50 years of age tended to decrease in older mothers, but no age effect was seen for mothers of cases diagnosed >/=50 years of age. There was no evidence that relative risk to sisters declined with age. For women with 2 affected first degree relatives the relative risk was 2.85 (2.12-3. 76). From these data, we estimate that in the general population 6. 8% of women under the age of 50 years and 9.7% of women aged 50-65 years have at least 1 first degree relative affected with breast cancer. Two per cent of women under 50 years have a family history which confers an increased risk of at least 2.5-fold. An intervention targeted at this group that reduced breast cancer morbidity by 20% would reduce the total burden of breast cancer in this age group by 1.1% at most. A family history of breast cancer is quite common in the general population, but preventive interventions targeted at women at high risk of breast cancer because of family history will have limited impact on breast cancer morbidity as a whole.

Polymorphisms in the human aromatase cytochrome P450 gene (CYP19) and breast cancer risk

The aromatase enzyme catalyses the conversion of androgens to oestrogens in the oestrogen biosynthesis pathway. Because increased exposure to oestrogens is considered to be a risk factor for breast cancer, the human aromatase gene (CYP19) is a plausible candidate for low penetrance breast cancer susceptibility. Preliminary reports have suggested that specific alleles of a TTTA repeat may be associated with differences in breast cancer risk. We have identified two new polymorphisms in the CYP19 gene: a TCT insertion/deletion in intron 4 and a G-->T substitution in intron 6, which have rare allele frequencies of 0.35 and 0.45, respectively, in the British population. Comparison was made between the frequencies of these alleles and those of the TTTA repeat in up to 599 breast cancer cases and 433 normal controls from the East Anglian, British population. We found strong linkage disequilibrium between the alleles of these three loci, but no significant association of any alleles with breast cancer risk. The maximum odds ratios observed were: 1.03 (95% CI 0.68-1.55) for the intron 4 TCT insertion/deletion polymorphism [del/del versus ins/ins]; 1.56 (95% CI 0.63-3.83) for the intron 4 [TTTA](10) allele; 1.29 (95% CI 0. 75-2.21) for the intron 6 G-->T polymorphism [TT versus GG]. We conclude that the CYP19 gene has no major role in common breast cancer incidence in the British population.

Familial gastric cancer: overview and guidelines for management

Families with autosomal dominant inherited predisposition to gastric cancer have been described. More recently, germline E-cadherin/CDH1 mutations have been identified in hereditary diffuse gastric cancer kindred. The need to have protocols to manage and counsel these families in the clinic led a group of geneticists, gastroenterologists, surgeons, oncologists, pathologists, and molecular biologists to convene a workshop to produce consensus statements and guidelines for familial gastric cancer. Review of the available cancer pathology from people belonging to families with documented germline E-cadherin/CDH1 mutations confirmed that the gastric cancers were all of the diffuse type. Criteria to define the different types of familial gastric cancer syndromes were agreed. Foremost among these criteria was that review of histopathology should be part of the evaluation of any family with aggregation of gastric cancer cases. Guidelines for genetic testing and counselling in hereditary diffuse gastric cancer were produced. Finally, a proposed strategy for clinical management in families with high penetrance autosomal dominant predisposition to gastric cancer was defined.

No association between androgen or vitamin D receptor gene polymorphisms and risk of breast cancer

Endogenous hormone exposure is known to alter breast cancer susceptibility and genes responsive to such hormones are plausible candidates for predisposition genes. We have examined polymorphisms in genes for two members of the nuclear receptor superfamily which are expressed in breast tissue and known to moderate rates of cell proliferation in a case-control association study: the androgen receptor (AR) and the vitamin D receptor (VDR). We have used two series of Caucasian female breast cancer cases, one incident and one prevalent, and compared both with two sets of matched controls from the East Anglian region of Britain. Since the results are similar in the two series we have combined them. The AR poly[Gly](n) and poly[Gln](n) tracts were genotyped in a total of 508 female breast cancer cases and 426 controls. The VDR TaqI polymorphism was analysed in 951 cases and 627 controls drawn from the same population series. There were no significant differences between cases and controls for either the AR or VDR polymorphisms. Compared with individuals with two short alleles (<22 repeats) of the AR poly[Gln](n) tract, the odds ratios and 95% confidence intervals (95% CI) for individuals with one or two long alleles were 0.82 (95% CI 0.62-1.09) and 1.31 (95% CI 0.87-1.97), respectively. Heterozygotes and homozygotes for the VDR TaqI cutting site had odds ratios of 1.01 (95% CI 0.81-1.27) and 0.97 (95% CI 0.71-1.32), respectively. None of the AR or VDR polymorphisms investigated has a major effect on risk of breast cancer in the British population.

A systematic review of genetic polymorphisms and breast cancer risk

Studies investigating the relationship between common genetic variants and cancer risk are being reported with rapidly increasing frequency. We have identified 46 published case-control studies that have examined the effect of common alleles of 18 different genes on breast cancer risk. Of these, 12 report statistically significant associations, none of which were reported by more than one study. However, many of the studies were small: 10 of the 46 had 80% power or greater to detect a rare allele homozygote relative risk <2.5. We therefore combined the results of individual studies to obtain more precise estimates of risk. Statistically significant differences in genotype frequencies were found in three case-control comparisons of unselected cases. These were for CYP19 (TTTA)n polymorphism [(TTTA)10 carrier odds ratio (OR) = 2.33; P = 0.002], the GSTP1 Ile105Val polymorphism (Val carrier OR = 1.60; P = 0.02), and the TP53 Arg72Pro polymorphism (Pro carrier OR = 1.27; P = 0.03). In addition, the GSTM1 gene deletion was found to be significantly associated with postmenopausal breast cancer (null homozygote OR = 1.33; P = 0.04). There was also some evidence that homozygotes for the PR PROGINS allele are protected against breast cancer, although this result was of borderline statistical significance. For polymorphisms in BRCA1, COMT, CYP17, CYP1A1, NAT1, and NAT2, the best estimate of risk either from the individual studies or the meta-analyses was sufficiently precise to exclude a relative risk of 1.5 or greater. For the polymorphisms in EDH17B2, ER, CYP2D6, CYP2E1, GSTT1, HSP70, and TNFalpha, the risk estimates, although nonsignificant, were insufficiently precise to exclude a moderate risk (>1.5). Precise estimation of the risks associated with these and other as yet untested genes, as well as investigation of more complex risks arising from gene-gene and gene-environment interactions, will require much larger studies.

Somatic mutations in the p53 gene and prognosis in breast cancer: a meta-analysis

Many studies have investigated the association between alterations in the p53 gene and clinical outcome of breast cancer, and most investigators have reported poorer overall and disease-free survival (as indicated by a relative hazard (RH) greater than one) in breast cancer cases with somatic mutations in p53. However, different studies have produced widely differing RH estimates, ranging from no risk (RH = 1) to a relative hazard of 23, and not all of these results have been statistically significant. We have therefore reviewed all the published studies that have investigated the association between somatic mutations in the p53 gene and breast cancer prognosis and used standard techniques of meta-analysis to combine the results of these studies to produce a more precise estimate of the prognostic significance of p53 mutations. Eleven studies investigated overall survival in a total of 2319 unselected cases. The RH estimates from these ranged from 1 to 23.4 with a combined RH estimate of 2.0 (confidence interval 1.7-2.5). Three studies investigated the role of p53 in node-negative patients and in these, the combined estimate of RH was 1.7 (1.2-2.3). For three studies of node-positive breast cancer the combined risk estimate was 2.6 (1.7-3.9). The inclusion of p53 mutation screening in large breast cancer clinical trials seems warranted in the light of these results. Analysis of large numbers of cases matched for stage and therapy will allow definitive clarification of the value of p53 mutational status in prognostication, and possibly choice of therapy.

Increased frequency of TP53 mutations in BRCA1 and BRCA2 ovarian tumours

We screened 81 ovarian tumours (30 BRCA1 associated, 18 BRCA2 associated, and 33 sporadic) for somatic TP53 mutations using both DNA analysis and immunostaining. TP53 mutations were significantly more frequent in tumours with mutations in BRCA1 (70% by immunostaining and 60% by DNA analysis) and BRCA2 (67% and 50%) compared to sporadic controls (39% and 30%) (P = 0.009). A higher proportion of tumours with BRCA1 and BRCA2 mutations were poorly differentiated, and TP53 mutant tumours in all categories were also more likely to be poorly differentiated. The poor differentiation of tumours with BRCA1 and BRCA2 mutations may be directly related to the role of these genes in DNA repair, and the need to overcome cell cycle checkpoints, often through loss of TP53. These results are consistent with the model of BRCA-induced tumorigenesis in which loss of checkpoint control is necessary for tumour development.

Molecular genetics and the assessment of human cancers

The past 20 years have seen rapid advances in the understanding of the biology of human cancers, and a large body of evidence now supports the idea that accumulated genetic changes underlie the development of neoplasia. In this article, we have reviewed the current research into the genetic bases of cancer and discussed the potential clinical applications of recent advances, with particular reference to the possibility of using molecular genetic tests for pre-symptomatic screening, clinical diagnosis and clinical staging. The possibility that the genetic profile of a tumour, its 'molecular fingerprint', will improve the ability of oncologists to predict tumour behaviour and thus help to determine optimum treatment has also been considered. Although the potential for the application of molecular genetic technologies is enormous, these technologies have yet to be subjected to rigorous evaluation in a clinical setting, and much work needs to be done before they are adopted for use in routine clinical care.

Survival in familial, BRCA1-associated, and BRCA2-associated epithelial ovarian cancer. United Kingdom Coordinating Committee for Cancer Research (UKCCCR) Familial Ovarian Cancer Study Group

The natural history of hereditary and BRCA1- and BRCA2-associated epithelial ovarian cancer may differ from that of sporadic disease. The purpose of this study was to compare the clinical characteristics of BRCA1- and BRCA2-associated hereditary ovarian cancer, hereditary ovarian cancer with no identified BRCA1/2 mutation, and ovarian cancer in population-based controls. BRCA1 and BRCA2 mutation testing was carried out on index cases from 119 families with site-specific epithelial ovarian cancer or breast-ovarian cancer. We estimated overall survival in 151 patients from 57 BRCA1 and BRCA2 mutation families and compared it with that in 119 patients from 62 families in which a BRCA1/2 mutation was not identified. We compared clinical outcome and data on tumor histopathology, grade, and stage. We also compared survival in familial epithelial ovarian cancer, whether or not a mutation was identified, with that of an age-matched set of population control cases. Overall survival at 5 years was 21% (95% confidence interval, 14-28) in cases from BRCA1 mutation families, 25% (8-42) in BRCA2 mutation families, and 19% (12-26) in families with no identified mutation (P = 0.91). Survival in familial ovarian cancer cases as a whole was significantly worse than for population controls (P = 0.005). In the familial cases, we found no differences in histopathological type, grade, or stage according to mutation status. Compared to population control cases, mucinous tumors occurred less frequently in the familial cases (2 versus 12%, P<0.001), and a greater proportion of the familial cases presented with advanced disease (83% stage III/IV versus 56%; P = 0.001). We have shown that survival in familial ovarian cancer cases is worse than that in sporadic cases, whether or not a BRCA1/2 mutation was identified, perhaps reflecting a difference in biology analogous to that observed in breast cancer.

The genetics of inherited breast cancer

The isolation of BRCA1 and BRCA2 has greatly increased our understanding of the genetics of inherited breast and ovarian cancer although the functions of these genes are not yet fully understood. We will discuss the current state of knowledge about the function of BRCA1 and BRCA2 and summarize the cancer risks in women carrying a BRCA1 or BRCA2 mutation. We review the evidence for gene-gene and gene-environment interaction in modifying that risk, and discuss the contribution of BRCA1 and BRCA2 and other high penetrance genes to both inherited and sporadic breast cancer.

No association between a polymorphism in the steroid metabolism gene CYP17 and risk of breast cancer

A recent study showed an association between a single base substitution, T-->C, in the promotor region of the CYP17 gene, the risk of breast cancer and age at menarche in Asian, African-American and Latino women from California and Hawaii. The C allele was associated with increased risk of breast cancer, significantly so for patients presenting with advanced disease, whereas the TT genotype was associated with later age at menarche in control subjects. We attempted to confirm these findings in a large case-control study in East Anglia, England (835 cases and 591 control subjects). We found no evidence of an increased risk of breast cancer [odds ratio (OR) 1.10, confidence interval (CI) 0.89-1.37] or advanced breast cancer (OR 0.88, CI 0.38-2.01) in C allele carriers, nor any association between age at menarche and genotype. We conclude that these alleles do not significantly alter breast cancer risk in the English population.

Screening for breast and ovarian cancer: the relevance of family history

The recent identification of two breast and ovarian cancer susceptibility genes--BRCA1 and BRCA2--has received a lot of publicity. Public and professional expectations of the availability and utility of genetic testing have been raised and the importance of a family history of breast cancer overemphasized. In this chapter, we examine the significance of a family history of breast or ovarian cancer in determining individual risk. A strategy for management is proposed, based on stratifying women with such a history into three different categories of risk for breast cancer: high, moderate and low. Some of the more controversial aspects of screening for breast and ovarian cancer are reviewed, including the issue of management of women who are at increased risk of these cancers by virtue of a family history, genetic predisposition, or both. There is a need for further research to clarify the most appropriate management of those at moderate risk of developing these cancers. A management strategy for women at high risk is proposed. We believe that adoption of this strategy will strengthen consistent information giving from primary to tertiary care.

Family history and the risk of breast cancer: a systematic review and meta-analysis

An increased risk of breast cancer in women with a family history of breast cancer has been demonstrated by many studies using a variety of study designs. However, the extent of this risk varies according to the nature of the family history (type of relative affected, age at which relative developed breast cancer and number of relatives affected) and may also vary according to age of the individual. The aim of our study was to identify all the published studies which have quantified the risk of breast cancer associated with a family history of the disease, and to summarise the evidence from these studies, with particular emphasis on age-specific risks according to subject and relative age. Seventy-four published studies were identified. The pooled estimate of relative risk (RR) associated with various family histories was as follows: any relative, RR = 1.9 (95% CI, 1.7-2.0); a first-degree relative, RR = 2.1 (CI = 2.0, 2.2); mother, RR = 2.0 (CI = 1.8, 2.1); sister, RR = 2.3 (CI = 2.1, 2.4); daughter, RR = 1.8 (CI = 1.6, 2.0); mother and sister, RR = 3.6 (CI = 2.5, 5.0); and a second-degree relative, RR = 1.5 (CI = 1.4, 1.6). Risks were increased in subjects under age 50 and when the relative had been diagnosed before age 50.

Cost effectiveness of lowering cholesterol concentration with statins in patients with and without pre-existing coronary heart disease: life table method applied to health authority population

OBJECTIVES

To estimate the cost effectiveness of statins in lowering serum cholesterol concentration in people at varying risk of fatal cardiovascular disease and to explore the implications of changing the criteria for intervention on cost and cost effectiveness for a purchasing authority.

DESIGN

A life table method was used to model the effect of treatment with a statin on survival over 10 years in men and women aged 45-64. The costs of intervention were estimated from the direct costs of treatment, offset by savings associated with a reduction in coronary angiographies, non-fatal myocardial infarctions, and revascularisation procedures. The robustness of the model to various assumptions was tested in a sensitivity analysis.

SETTING

Population of a typical district health authority.

MAIN OUTCOME MEASURE

Cost per life year saved.

RESULTS

The average cost effectiveness of treating men aged 45-64 with no history of coronary heart disease and a cholesterol concentration > 6.5 mmol/l for 10 years with a statin was 136,000 pounds per life year saved. The average cost effectiveness for patients with pre-existing coronary heart disease and a cholesterol concentration > 5.4 mmol/l was 32,000 pounds. These averages hide enormous differences in cost effectiveness between groups at different risk, ranging from 6000 pounds per life year in men aged 55-64 who have had a myocardial infarction and whose cholesterol concentration is above 7.2 mmol/l to 361,000 pounds per life year saved in women aged 45-54 with angina and a cholesterol concentration of 5.5-6.0 mmol/l.

CONCLUSIONS

Lowering serum cholesterol concentration in patients with and without preexisting coronary heart disease is effective and safe, but treatment for all those in whom treatment is likely to be effective is not sustainable within current NHS resources. Data on cost effectiveness data should be taken into account when assessing who should be eligible for treatment.

Selective or universal neonatal BCG immunization: what policy for a district with a high incidence of tuberculosis?

In the United Kingdom, BCG immunization of neonates provides good protection against military and meningeal disease, and probably against other forms of tuberculosis, in all ethnic groups. Serious adverse reactions to BCG immunization are rare. Cost-effectiveness studies of BCG immunization in neonates have not been reported and "universal' BCG immunization is not recommended in the United Kingdom. The Department of Health does recommend immunization for children and infants of immigrants from countries with a high prevalence of tuberculosis. There are problems associated with such a "selective' policy, as determining what is "high' prevalence and thus defining "at risk' groups is difficult, and there may be political and practical difficulties in its implementation. This may result in low coverage in eligible groups. A "universal' policy of BCG immunization for all neonates may be politically more acceptable and easier to implement in districts with high tuberculosis notification rates. Although there is no cost-effectiveness data to determine at what tuberculosis notification rate a universal policy should be adopted, an universal policy has been suggested for districts with overall notification rate of greater than 40 per 100,000. Within districts there may be large variations in tuberculosis notification rates between different areas. This is becoming more common with the amalgamation and merger of smaller districts into new larger purchasing organizations. New River Health Authority is such a district formed by the amalgamation of the former districts of Haringey with a high tuberculosis notification rate, and Enfield with a lower TB notification rate. In order to maximize coverage in the "at risk' neonates, a different neonatal BCG policy has been adopted in the two areas. This has been possible because of the flexibility of the mechanisms for contracting with different provider units. Although the overall notification rate was not thought to be sufficiently high to justify a "universal' neonatal policy throughout the district, a "universal' policy has been instituted in the main provider unit in the former district Haringey. A "selective' policy, subject to ongoing evaluation to ensure high coverage, continues to be operated by the main provider unit in the former district of Enfield.

Variation in prescribing of hypnotics, anxiolytics and antidepressants between 61 general practices

BACKGROUND

Although the desirability of reducing prescribing of hypnotics and anxiolytics has long been recognized, variation between practices in patterns of psychotropic drug prescribing has received little attention; factors underlying such variation are poorly understood.

AIM

The study aimed to describe the extent of variation between general practices in the prescribing of hypnotics, anxiolytics and antidepressants; it also aimed to analyse the influence of measures of practice population and general practice and general practitioner characteristics on any variation in prescribing volumes.

METHOD

Routinely collected prescribing data and practice population data, from April 1992 to March 1993, from all 61 practices in the Cambridge and Huntingdon Health Commission were analysed. Prescribing was measured as annual defined daily doses per 1000 practice population for each drug class. Data on variables relating to practice structure and general practitioner characteristics were obtained and analysed. Potentially influencing variables were investigated by multiple regression.

RESULTS

Between the highest and lowest prescribing practices there was an 11-fold difference, a 13-fold difference and an eightfold difference in the annual defined daily doses per 1000 practice population prescribed for hypnotics, for anxiolytics and for antidepressants, respectively. Strong positive correlations existed between volumes of prescribing of each drug class. The drugs prescribed in the greatest volumes were hypnotics. Practice population structure had some influence on psychotropic drug prescribing with high prescribing being associated with the proportions of temporary residents and women aged 65 years and over in the practice for all three classes of drug. Other factors, including presence of a practice counsellor, were not found to have a significant influence on psychotropic drug prescribing.

CONCLUSION

The degree of variation in prescribing volumes of psychotropic drugs between practices raises serious concerns. Further study is needed, but progress will be hampered until there is more clarity over the effectiveness and appropriateness of using these substances in the various illness, symptom and life-stress presentations seen in primary care.

Health promotion in primary care: modelling the impact of intervention on coronary heart disease and stroke

BACKGROUND

Since the health promotion programme for general practice was introduced in 1993, there has been some controversy about the potential effectiveness and likely cost-effectiveness of health promotion in general practice. The aim of this study was to generate a model to estimate the potential impact of the new general practice health promotion programme on coronary heart disease (CHD) and stroke, in terms of absolute reductions in mortality, in Cambridge Health Authority, a district with relatively low mortality rates for these two diseases.

METHODS

A model to estimate the impact of the programme was derived from local data for risk factor prevalence in men and women aged 45-74; local death rates for CHD and stroke; and published data on the effectiveness of general practice health promotion.

RESULTS

The expected number of deaths in the ten years following screening was estimated at 3203 deaths for CHD and 1075 for stroke. The model estimates a 12% reduction in CHD deaths and 17% fewer stroke deaths following the implementation of the programme.

CONCLUSIONS

Even in areas where risk factor prevalence is low and death rates are low, the health promotion programme has the potential to have a moderate impact on CHD and stroke mortality. The cost-effectiveness of the programme, however, has yet to be established.

Endemic goitre and cretinism in the Simbai and Tep-Tep areas of Madang Province, Papua New Guinea

In the late 1960s and early 1970s a mass campaign of iodized oil injections was carried out in Papua New Guinea as an interim measure to improve iodine status in the population and to prevent endemic goitre and endemic cretinism. Following informal reports of children with neurological abnormalities resembling endemic cretinism in two areas of Madang Province, the Simbai and Tep-Tep Subdistricts, surveys were conducted in these areas in 1985 to establish prevalence rates of goitre and cretinism in order to determine whether they had returned as public health problems. 42% of the Simbai population and 38% of the Tep-Tep population attended for examination. Amongst those surveyed, the visible goitre rate was low: 0.1% in the Simbai and 2.5% in the Tep-Tep area. Although the iodized oil patrols were carried out a decade previously, goitre does not seem to have re-emerged in the area to the levels reported previously. In the Simbai villages surveyed, there were nine individuals diagnosed as suffering from cretinism, three of whom were thought to have been born after the last iodized oil patrol in the area. In the Tep-Tep villages, eight subjects were diagnosed as suffering from cretinism, all of whom were thought to have been born before the iodized oil patrols. A number of younger children with neurological abnormalities which did not fulfil criteria for endemic cretinism were also seen in both areas. Because of the difficulties in diagnosing endemic cretinism in young children, it is not possible to conclude that endemic cretinism is no longer a problem in these areas. More work is needed to confirm the results of these studies. This will enable a rational control program to be designed and implemented.

A randomized controlled study of post-injection rest following intra-articular steroid therapy for knee synovitis

In order to assess the effect of 24 h observed bed rest following intra-articular steroid injection of the knee joint in patients with an inflammatory arthritis such as RA, AS or colitic arthropathy, 91 patients with inflammatory arthritis of one knee joint were randomized to receive 24 h bed rest in hospital following intra-articular steroid injection or were injected in outpatients. The clinical and laboratory assessments such as pain and stiffness on a 10-cm visual analogue scale, knee circumference (cm), 50 ft walking time (s), CRP and ESR were measured before receiving the steroid injection and at 3, 6, 12 and 24 weeks. Both groups of patients improved clinically and serologically at 3 weeks. By 12 weeks the degree of improvement in the pain score, stiffness score, knee circumference, 50 ft walking time and CRP was better in the rest group and these differences persisted to 24 weeks. For each outcome variable the summary measure of response was significantly better in the rest group compared to the no rest group. Intra-articular steroid injection of the knee joint followed by strict i inpatient bed rest for 24 h results in a greater degree of clinical and serological improvement, compared to routine outpatient injections for up to 6 months. It is therefore possible that 24-h post-injection rest will result in a prolonged duration of clinical response and reduce the need for frequent steroid injections and the risk of complications.

Two cases of cutaneous leishmaniasis in Malawi

Leishmaniasis is rarely encountered in southern Africa. Two cases were diagnosed by histopathology in northern Malawi in one year. One was an 18 months old child with negative human immunodeficiency virus 1 (HIV-1) serology and widespread skin lesions; the other an HIV-1 positive adult with 2 skin lesions. The Leishmania species responsible could not be identified, but the infection may be more prevalent in this region than previously thought.

Erythromelalgia--the role of hypnotherapy

Erythromelalgia is an unusual condition characterized by attacks of burning pain in the hands and feet with local congestion and increased skin temperature. We report a case of erythromelalgia, with transient hypertension and elevated urinary catecholamines successfully treated by hypnotherapy. Such an association has not to our knowledge been previously reported in English language publications.

Role of gonadal and adrenal steroids in the impairment of the male rat's sexual behaviour by hyperprolactinaemia

The sexual behaviour of male rats, castrated and testosterone-implanted, declined following induction of hyperprolactinaemia by domperidone. Treatment with oestradiol benzoate did not reverse this effect, and may have accentuated it. Oestradiol also amplified domperidone-induced hyperprolactinaemia. Testosterone or dihydrotestosterone (DHT) apparently delayed, but did not prevent, the gradual deterioration in sexual behaviour (prolonged ejaculation latencies) induced by domperidone, but this effect was not confirmed statistically. Adrenalectomy, followed by cortisol replacement, failed to prevent the behavioural effects of hyperprolactinaemia. No consistent changes in serum progesterone or corticosterone could be found in hyperprolactinaemic rats in which the adrenals had not been removed. In vitro formation of DHT from precursor testosterone was reduced in the amygdalae of hyperprolactinaemic rats, but not in the hypothalamus or caudal spinal cord. Oestradiol cytosol binding was unchanged in all brain areas, except for a small but significant increase in the anterior hypothalamus. These results do not support a role for altered adrenal activity in determining the effects of high levels of prolactin on sexual behaviour. There is evidence for an impaired formation of DHT in the brain, but this may account for only part of the behavioural changes observed. It is possible that the major effect of prolactin lies in neural systems directly responsive to it, rather than in altered steroid secretion or metabolism.