Randomised trial of population-based BRCA testing in Ashkenazi Jews: long-term secondary lifestyle behavioural outcomes.

OBJECTIVE: Ashkenazi-Jewish (AJ) population-based BRCA testing is acceptable, cost-effective and amplifies primary prevention for breast & ovarian cancer. However, data describing lifestyle impact are lacking. We report long-term results of population-based BRCA testing on lifestyle behaviour and cancer risk perception. DESIGN: Two-arm randomised controlled trials (ISRCTN73338115, GCaPPS): (a) population-screening (PS); (b) family history (FH)/clinical criteria testing. SETTING: North London AJ-population. POPULATION/SAMPLE: AJ women/men >18 years. EXCLUSIONS: prior BRCA testing or first-degree relatives of BRCA-carriers. METHODS: Participants were recruited through self-referral. All participants received informed pre-test genetic counselling. The intervention included genetic testing for three AJ BRCA-mutations: 185delAG(c.68_69delAG), 5382insC(c.5266dupC) and 6174delT(c.5946delT). This was undertaken for all participants in the PS arm and participants fulfilling FH/clinical criteria in the FH arm. Patients filled out customised/validated questionnaires at baseline/1-year/2-year/3-year follow-ups. Generalised linear-mixed models adjusted for covariates and appropriate contrast tests were used for between-group/within-group analysis of lifestyle and behavioural outcomes along with evaluating factors associated with these outcomes. Outcomes are adjusted for multiple testing (Bonferroni method), with P < 0.0039 considered significant. OUTCOME MEASURES: Lifestyle/behavioural outcomes at baseline/1-year/2-year/3-year follow-ups. RESULTS: 1034 participants were randomised to PS (n = 530) or FH (n = 504) arms. No significant difference was identified between PS- and FH-based BRCA testing approaches in terms of dietary fruit/vegetable/meat consumption, vitamin intake, alcohol quantity/ frequency, smoking behaviour (frequency/cessation), physical activity/exercise or routine breast mammogram screening behaviour, with outcomes not affected by BRCA test result. Cancer risk perception decreased with time following BRCA testing, with no difference between FH/PS approaches, and the perception of risk was lowest in BRCA-negative participants. Men consumed fewer fruits/vegetables/vitamins and more meat/alcohol than women (P < 0.001). CONCLUSION: Population-based and FH-based AJ BRCA testing have similar long-term lifestyle impacts on smoking, alcohol, dietary fruit/vegetable/meat/vitamin, exercise, breast screening participation and reduced cancer risk perception.

Postoperative radiation rates in stage IIA1 cervical cancer: Is surgical treatment justified? An Israeli Gynecologic Oncology Group Study.

OBJECTIVES: Data on the outcome of stage IIA1 cervical cancer is limited, as these tumors comprise a small percentage of early tumors. NCCN guidelines suggest consideration of surgical management for small tumors with vaginal involvement. Our objective was to evaluate the risk of adjuvant radiotherapy in stage IIA1 cervical cancer and its associated features, in order to improve selection of patients for surgical management. METHODS: A retrospective cohort study comparing surgically treated cervical cancer patients with stage IB1 and stage IIA1 disease. Women treated between 2000 and 2015 in ten Israeli medical centers were included. Patient and disease features were compared between stages. The relative risk (Fisher's exact test) of receiving post-operative radiation was calculated and compared for each risk factor. A general linear model (GLM) was used for multivariable analysis. RESULTS: 199 patients were included, of whom 21 had stage IIA1 disease. Most features were comparable for stage IB1 and stage IIA1 disease, although patients with vaginal involvement were more likely to have close surgical margins (23.8% vs 8.5%, p = 0.03). Patients with stage IIA1 disease were more likely to receive radiation after surgery (76% vs. 46%, RR = 1.65 (1.24-2.2), p = 0.011). Vaginal involvement as well as depth of stromal invasion, LVSI and lymph node metastases were independent predictors of radiation on multivariable general linear modeling. CONCLUSIONS: Cervical cancer patients with vaginal involvement are highly more likely to require postoperative radiation. We recommend careful evaluation of these patients before surgical management is offered.