Effect of baseline oestradiol serum concentration on the efficacy of anastrozole for preventing breast cancer in postmenopausal women at high risk: a case-control study of the IBIS-II prevention trial.

BACKGROUND: An increased risk of breast cancer is associated with high serum concentrations of oestradiol and testosterone in postmenopausal women, but little is known about how these hormones affect response to endocrine therapy for breast cancer prevention or treatment. We aimed to assess the effects of serum oestradiol and testosterone concentrations on the efficacy of the aromatase inhibitor anastrozole for the prevention of breast cancer in postmenopausal women at high risk. METHODS: In this case-control study we used data from the IBIS-II prevention trial, a randomised, controlled, double-blind trial in postmenopausal women aged 40-70 years at high risk of breast cancer, conducted in 153 breast cancer treatment centres across 18 countries. In the trial, women were randomly assigned (1:1) to receive anastrozole (1 mg/day, orally) or placebo daily for 5 years. In this pre-planned case-control study, the primary analysis was the effect of the baseline oestradiol to sex hormone binding globulin (SHBG) ratio (oestradiol-SHBG ratio) on the development of all breast cancers, including ductal carcinoma in situ (the primary endpoint in the trial). Cases were participants in whom breast cancer was reported after trial entry and until the cutoff on Oct 22, 2019, and who had valid blood samples and no use of hormone replacement therapy within 3 months of trial entry or during the trial. For each case, two controls without breast cancer were selected at random, matched on treatment group, age (within 2 years), and follow-up time (at least that of the matching case). For each treatment group, we applied a multinominal logistic regression likelihood-ratio trend test to assess what change in the proportion of cases was associated with a one-quartile change in hormone ratio. Controls were used only to determine quartile cutoffs. Profile likelihood 95% CIs were used to indicate the precision of estimates. A secondary analysis also investigated the effect of the baseline testosterone-SHBG ratio on breast cancer development. We also assessed relative benefit of anastrozole versus placebo (calculated as 1 - the ratio of breast cancer cases in the anastrozole group to cases in the placebo group). The trial was registered with ISRCTN (number ISRCTN31488319) and completed recruitment on Jan 31, 2012, but long-term follow-up is ongoing. FINDINGS: 3864 women were recruited into the trial between Feb 2, 2003, and Jan 31, 2012, and randomly assigned to receive anastrozole (n=1920) or placebo (n=1944). Median follow-up time was 131 months (IQR 106-156), during which 85 (4·4%) cases of breast cancer in the anastrozole group and 165 (8·5%) in the placebo group were identified. No data on gender, race, or ethnicity were collected. After exclusions, the case-control study included 212 participants from the anastrozole group (72 cases, 140 controls) and 416 from the placebo group (142 cases, 274 controls). A trend of increasing breast cancer risk with increasing oestradiol-SHBG ratio was found in the placebo group (trend per quartile 1·25 [95% CI 1·08 to 1·45], p=0·0033), but not in the anastrozole group (1·06 [0·86 to 1·30], p=0·60). A weaker effect was seen for the testosterone-SHBG ratio in the placebo group (trend 1·21 [1·05 to 1·41], p=0·011), but again not in the anastrozole group (trend 1·18 [0·96 to 1·46], p=0·11). A relative benefit of anastrozole was seen in quartile 2 (0·55 [95% CI 0·13 to 0·78]), quartile 3 (0·54 [0·22 to 0·74], and quartile 4 (0·56 [0·23 to 0·76]) of oestradiol-SHBG ratio, but not in quartile 1 (0·18 [-0·60 to 0·59]). INTERPRETATION: These results suggest that serum hormones should be measured more routinely and integrated into risk management decisions. Measuring serum hormone concentrations is inexpensive and might help clinicians differentiate which women will benefit most from an aromatase inhibitor. FUNDING: Cancer Research UK, National Health and Medical Research Council (Australia), Breast Cancer Research Foundation, and DaCosta Fund.

Are better AI algorithms for breast cancer detection also better at predicting risk? A paired case-control study.

BACKGROUND: There is increasing evidence that artificial intelligence (AI) breast cancer risk evaluation tools using digital mammograms are highly informative for 1-6 years following a negative screening examination. We hypothesized that algorithms that have previously been shown to work well for cancer detection will also work well for risk assessment and that performance of algorithms for detection and risk assessment is correlated. METHODS: To evaluate our hypothesis, we designed a case-control study using paired mammograms at diagnosis and at the previous screening visit. The study included n = 3386 women from the OPTIMAM registry, that includes mammograms from women diagnosed with breast cancer in the English breast screening program 2010-2019. Cases were diagnosed with invasive breast cancer or ductal carcinoma in situ at screening and were selected if they had a mammogram available at the screening examination that led to detection, and a paired mammogram at their previous screening visit 3y prior to detection when no cancer was detected. Controls without cancer were matched 1:1 to cases based on age (year), screening site, and mammography machine type. Risk assessment was conducted using a deep-learning model designed for breast cancer risk assessment (Mirai), and three open-source deep-learning algorithms designed for breast cancer detection. Discrimination was assessed using a matched area under the curve (AUC) statistic. RESULTS: Overall performance using the paired mammograms followed the same order by algorithm for risk assessment (AUC range 0.59-0.67) and detection (AUC 0.81-0.89), with Mirai performing best for both. There was also a correlation in performance for risk and detection within algorithms by cancer size, with much greater accuracy for large cancers (30 mm+, detection AUC: 0.88-0.92; risk AUC: 0.64-0.74) than smaller cancers (0 to < 10 mm, detection AUC: 0.73-0.86, risk AUC: 0.54-0.64). Mirai was relatively strong for risk assessment of smaller cancers (0 to < 10 mm, risk, Mirai AUC: 0.64 (95% CI 0.57 to 0.70); other algorithms AUC 0.54-0.56). CONCLUSIONS: Improvements in risk assessment could stem from enhancing cancer detection capabilities of smaller cancers. Other state-of-the-art AI detection algorithms with high performance for smaller cancers might achieve relatively high performance for risk assessment.

Long-term prediction by DNA methylation of high-grade cervical intraepithelial neoplasia: Results of the ARTISTIC cohort.

Methylation markers have shown potential for triaging high-risk HPV-positive (hrHPV+) women to identify those at increased risk of invasive cervical cancer (ICC). Our aim was to assess the performance of the S5 DNA methylation classifier for predicting incident high-grade cervical intraepithelial neoplasia (CIN) and ICC among hrHPV+ women in the ARTISTIC screening trial cohort. The S5 classifier, comprising target regions of tumour suppressor gene EPB41L3 and L1 and L2 regions of HPV16, HPV18, HPV31, and HPV33, was assayed by pyrosequencing in archived hrHPV+ liquid-based samples from 343 women with high-grade disease (139 CIN2, 186 CIN3, and 18 ICC) compared to 800 hrHPV+ controls. S5 DNA methylation correlated directly with increasing severity of disease and inversely with lead time to diagnosis. S5 could discriminate between hrHPV+ women who developed CIN3 or ICC and hrHPV+ controls (p <.0001) using samples taken on average 5 years before diagnosis. This relationship was independent of cytology at baseline. The S5 test showed much higher sensitivity than HPV16/18 genotyping for identifying prevalent CIN3 (93% vs. 61%, p = .01) but lower specificity (50% vs. 66%, p <.0001). The S5 classifier identified most women at high risk of developing precancer and missed very few prevalent advanced lesions thus appearing to be an objective test for triage of hrHPV+ women. The combination of methylation of host and HPV genes enables S5 to combine the predictive power of methylation with HPV genotyping to identify hrHPV-positive women who are at highest risk of developing CIN3 and ICC in the future.

Predictable changes in the accuracy of human papillomavirus tests after vaccination: review with implications for performance monitoring in cervical screening.

Vaccination against human papillomavirus (HPV) is changing the performance of cytology as a cervical screening test, but its effect on HPV testing is unclear. We review the effect of HPV16/18 vaccination on the epidemiology and the detection of HPV infections and high-grade cervical lesions (CIN2+) to evaluate the likely direction of changes in HPV test accuracy. The reduction in HPV16/18 infections and cross-protection against certain non-16/18 high-risk genotypes, most notably 31, 33, and/or 45, will likely increase the test's specificity but decrease its positive predictive value (PPV) for CIN2+. Post-vaccination viral unmasking of non-16/18 genotypes due to fewer HPV16 co-infections might reduce the specificity and the PPV for CIN2+. Post-vaccination clinical unmasking exposing a higher frequency of CIN2+ related to non-16/18 high-risk genotypes is likely to increase the specificity and the PPV of HPV tests. The effect of HPV16/18 vaccination on HPV test sensitivity is difficult to predict based on these changes alone. Programmes relying on HPV detection for primary screening should monitor the frequency of false-positive and false-negative tests in vaccinated (younger) vs. unvaccinated (older) cohorts, to assess the outcomes and performance of their service.

Development and evaluation of a method to assess breast cancer risk using a longitudinal history of mammographic density: a cohort study.

BACKGROUND: Women with dense breasts have an increased risk of breast cancer. However, breast density is measured with variability, which may reduce the reliability and accuracy of its association with breast cancer risk. This is particularly relevant when visually assessing breast density due to variation in inter- and intra-reader assessments. To address this issue, we developed a longitudinal breast density measure which uses an individual woman's entire history of mammographic density, and we evaluated its association with breast cancer risk as well as its predictive ability. METHODS: In total, 132,439 women, aged 40-73 yr, who were enrolled in Kaiser Permanente Washington and had multiple screening mammograms taken between 1996 and 2013 were followed up for invasive breast cancer through 2014. Breast Imaging Reporting and Data System (BI-RADS) density was assessed at each screen. Continuous and derived categorical longitudinal density measures were developed using a linear mixed model that allowed for longitudinal density to be updated at each screen. Predictive ability was assessed using (1) age and body mass index-adjusted hazard ratios (HR) for breast density (time-varying covariate), (2) likelihood-ratio statistics (ΔLR-χ2) and (3) concordance indices. RESULTS: In total, 2704 invasive breast cancers were diagnosed during follow-up (median = 5.2 yr; median mammograms per woman = 3). When compared with an age- and body mass index-only model, the gain in statistical information provided by the continuous longitudinal density measure was 23% greater than that provided by BI-RADS density (follow-up after baseline mammogram: ΔLR-χ2 = 379.6 (degrees of freedom (df) = 2) vs. 307.7 (df = 3)), which increased to 35% (ΔLR-χ2 = 251.2 vs. 186.7) for follow-up after three mammograms (n = 76,313, 2169 cancers). There was a sixfold difference in observed risk between densest and fattiest eight-category longitudinal density (HR = 6.3, 95% CI 4.7-8.7), versus a fourfold difference with BI-RADS density (HR = 4.3, 95% CI 3.4-5.5). Discriminatory accuracy was marginally greater for longitudinal versus BI-RADS density (c-index = 0.64 vs. 0.63, mean difference = 0.008, 95% CI 0.003-0.012). CONCLUSIONS: Estimating mammographic density using a woman's history of breast density is likely to be more reliable than using the most recent observation only, which may lead to more reliable and accurate estimates of individual breast cancer risk. Longitudinal breast density has the potential to improve personal breast cancer risk estimation in women attending mammography screening.

Quantifying the effects of risk-stratified breast cancer screening when delivered in real time as routine practice versus usual screening: the BC-Predict non-randomised controlled study (NCT04359420).

BACKGROUND: Risk stratification as a routine part of the NHS Breast Screening Programme (NHSBSP) could provide a better balance of benefits and harms. We developed BC-Predict, to offer women when invited to the NHSBSP, which collects standard risk factor information; mammographic density; and in a sub-sample, a Polygenic Risk Score (PRS). METHODS: Risk prediction was estimated primarily from self-reported questionnaires and mammographic density using the Tyrer-Cuzick risk model. Women eligible for NHSBSP were recruited. BC-Predict produced risk feedback letters, inviting women at high risk (≥8% 10-year) or moderate risk (≥5-<8% 10-year) to have appointments to discuss prevention and additional screening. RESULTS: Overall uptake of BC-Predict in screening attendees was 16.9% with 2472 consenting to the study; 76.8% of those received risk feedback within the 8-week timeframe. Recruitment was 63.2% with an onsite recruiter and paper questionnaire compared to <10% with BC-Predict only (P < 0.0001). Risk appointment attendance was highest for those at high risk (40.6%); 77.5% of those opted for preventive medication. DISCUSSION: We have shown that a real-time offer of breast cancer risk information (including both mammographic density and PRS) is feasible and can be delivered in reasonable time, although uptake requires personal contact. Preventive medication uptake in women newly identified at high risk is high and could improve the cost-effectiveness of risk stratification. TRIAL REGISTRATION: Retrospectively registered with clinicaltrials.gov (NCT04359420).

Evaluation of an AI Model to Assess Future Breast Cancer Risk.

Background Accurate breast cancer risk assessment after a negative screening result could enable better strategies for early detection. Purpose To evaluate a deep learning algorithm for risk assessment based on digital mammograms. Materials and Methods A retrospective observational matched case-control study was designed using the OPTIMAM Mammography Image Database from the National Health Service Breast Screening Programme in the United Kingdom from February 2010 to September 2019. Patients with breast cancer (cases) were diagnosed following a mammographic screening or between two triannual screening rounds. Controls were matched based on mammography device, screening site, and age. The artificial intelligence (AI) model only used mammograms at screening before diagnosis. The primary objective was to assess model performance, with a secondary objective to assess heterogeneity and calibration slope. The area under the receiver operating characteristic curve (AUC) was estimated for 3-year risk. Heterogeneity according to cancer subtype was assessed using a likelihood ratio interaction test. Statistical significance was set at P < .05. Results Analysis included patients with screen-detected (median age, 60 years [IQR, 55-65 years]; 2044 female, including 1528 with invasive cancer and 503 with ductal carcinoma in situ [DCIS]) or interval (median age, 59 years [IQR, 53-65 years]; 696 female, including 636 with invasive cancer and 54 with DCIS) breast cancer and 1:1 matched controls, each with a complete set of mammograms at the screening preceding diagnosis. The AI model had an overall AUC of 0.68 (95% CI: 0.66, 0.70), with no evidence of a significant difference between interval and screen-detected (AUC, 0.69 vs 0.67; P = .085) cancer. The calibration slope was 1.13 (95% CI: 1.01, 1.26). There was similar performance for the detection of invasive cancer versus DCIS (AUC, 0.68 vs 0.66; P = .057). The model had higher performance for advanced cancer risk (AUC, 0.72 ≥stage II vs 0.66

Breast cancer polygenic risk scores derived in White European populations are not calibrated for women of Ashkenazi Jewish descent.

PURPOSE: Polygenic risk scores (PRSs) are a major component of accurate breast cancer (BC) risk prediction but require ethnicity-specific calibration. Ashkenazi Jewish (AJ) population is assumed to be of White European (WE) origin in some commercially available PRSs despite differing effect allele frequencies (EAFs). We conducted a case-control study of WE and AJ women from the Predicting Risk of Cancer at Screening Study. The Breast Cancer in Northern Israel Study provided a separate AJ population-based case-control validation series. METHODS: All women underwent Illumina OncoArray single-nucleotide variation (SNV; formerly single-nucleotide polymorphism [SNP]) analysis. Two PRSs were assessed, SNV142 and SNV78. A total of 221 of 2243 WE women (discovery: cases = 111; controls = 110; validation: cases = 651; controls = 1772) and 221 AJ women (cases = 121; controls = 110) were included from the UK study; the Israeli series consisted of 2045 AJ women (cases = 1331; controls = 714). EAFs were obtained from the Genome Aggregation Database. RESULTS: In the UK study, the mean SNV142 PRS demonstrated good calibration and discrimination in WE population, with mean PRS of 1.33 (95% CI 1.18-1.48) in cases and 1.01 (95% CI 0.89-1.13) in controls. In AJ women from Manchester, the mean PRS of 1.54 (1.38-1.70) in cases and 1.20 (1.08-1.32) in controls demonstrated good discrimination but overestimation of BC relative risk. After adjusting for EAFs for the AJ population, mean risk was corrected (mean SNV142 PRS cases = 1.30 [95% CI 1.16-1.44] and controls = 1.02 [95% CI 0.92-1.12]). This was recapitulated in the larger Israeli data set with good discrimination (area under the curve = 0.632 [95% CI 0.607-0.657] for SNV142). CONCLUSION: AJ women should not be given BC relative risk predictions based on PRSs calibrated to EAFs from the WE population. PRSs need to be recalibrated using AJ-derived EAFs. A simple recalibration using the mean PRS adjustment ratio likely performs well.

Variability Among Breast Cancer Risk Classification Models When Applied at the Level of the Individual Woman.

BACKGROUND: Breast cancer risk models guide screening and chemoprevention decisions, but the extent and effect of variability among models, particularly at the individual level, is uncertain. OBJECTIVE: To quantify the accuracy and disagreement between commonly used risk models in categorizing individual women as average vs. high risk for developing invasive breast cancer. DESIGN: Comparison of three risk prediction models: Breast Cancer Risk Assessment Tool (BCRAT), Breast Cancer Surveillance Consortium (BCSC) model, and International Breast Intervention Study (IBIS) model. SUBJECTS: Women 40 to 74 years of age presenting for screening mammography at a multisite health system between 2011 and 2015, with 5-year follow-up for cancer outcome. MAIN MEASURES: Comparison of model discrimination and calibration at the population level and inter-model agreement for 5-year breast cancer risk at the individual level using two cutoffs (≥ 1.67% and ≥ 3.0%). KEY RESULTS: A total of 31,115 women were included. When using the ≥ 1.67% threshold, more than 21% of women were classified as high risk for developing breast cancer in the next 5 years by one model, but average risk by another model. When using the ≥ 3.0% threshold, more than 5% of women had disagreements in risk severity between models. Almost half of the women (46.6%) were classified as high risk by at least one of the three models (e.g., if all three models were applied) for the threshold of ≥ 1.67%, and 11.1% were classified as high risk for ≥ 3.0%. All three models had similar accuracy at the population level. CONCLUSIONS: Breast cancer risk estimates for individual women vary substantially, depending on which risk assessment model is used. The choice of cutoff used to define high risk can lead to adverse effects for screening, preventive care, and quality of life for misidentified individuals. Clinicians need to be aware of the high false-positive and false-negative rates and variation between models when talking with patients.

Dynamic data-enabled stratified sampling for trial invitations with application in NHS-Galleri.

BACKGROUND: Participants of health research studies such as cancer screening trials usually have better health than the target population. Data-enabled recruitment strategies might be used to help minimise healthy volunteer effects on study power and improve equity. METHODS: A computer algorithm was developed to help target trial invitations. It assumes participants are recruited from distinct sites (such as different physical locations or periods in time) that are served by clusters (such as general practitioners in England, or geographical areas), and the population may be split into defined groups (such as age and sex bands). The problem is to decide the number of people to invite from each group, such that all recruitment slots are filled, healthy volunteer effects are accounted for, and equity is achieved through representation in sufficient numbers of all major societal and ethnic groups. A linear programme was formulated for this problem. RESULTS: The optimisation problem was solved dynamically for invitations to the NHS-Galleri trial (ISRCTN91431511). This multi-cancer screening trial aimed to recruit 140,000 participants from areas in England over 10 months. Public data sources were used for objective function weights, and constraints. Invitations were sent by sampling according to lists generated by the algorithm. To help achieve equity the algorithm tilts the invitation sampling distribution towards groups that are less likely to join. To mitigate healthy volunteer effects, it requires a minimum expected event rate of the primary outcome in the trial. CONCLUSION: Our invitation algorithm is a novel data-enabled approach to recruitment that is designed to address healthy volunteer effects and inequity in health research studies. It could be adapted for use in other trials or research studies.

Breast cancer risk stratification in women of screening age: Incremental effects of adding mammographic density, polygenic risk, and a gene panel.

PURPOSE: There is great promise in breast cancer risk stratification to target screening and prevention. It is unclear whether adding gene panels to other risk tools improves breast cancer risk stratification and adds discriminatory benefit on a population basis. METHODS: In total, 10,025 of 57,902 women aged 46 to 73 years in the Predicting Risk of Cancer at Screening study provided DNA samples. A case-control study was used to evaluate breast cancer risk assessment using polygenic risk scores (PRSs), cancer gene panel (n = 33), mammographic density (density residual [DR]), and risk factors collected using a self-completed 2-page questionnaire (Tyrer-Cuzick [TC] model version 8). In total, 525 cases and 1410 controls underwent gene panel testing and PRS calculation (18, 143, and/or 313 single-nucleotide polymorphisms [SNPs]). RESULTS: Actionable pathogenic variants (PGVs) in BRCA1/2 were found in 1.7% of cases and 0.55% of controls, and overall PGVs were found in 6.1% of cases and 1.3% of controls. A combined assessment of TC8-DR-SNP313 and gene panel provided the best risk stratification with 26.1% of controls and 9.7% of cases identified at <1.4% 10-year risk and 9.01% of controls and 23.3% of cases at ≥8% 10-year risk. Because actionable PGVs were uncommon, discrimination was identical with/without gene panel (with/without: area under the curve = 0.67, 95% CI = 0.64-0.70). Only 7 of 17 PGVs in cases resulted in actionable risk category change. Extended case (n = 644)-control (n = 1779) series with TC8-DR-SNP143 identified 18.9% of controls and only 6.4% of stage 2+ cases at <1.4% 10-year risk and 20.7% of controls and 47.9% of stage 2+ cases at ≥5% 10-year risk. CONCLUSION: Further studies and economic analysis will determine whether adding panels to PRS is a cost-effective strategy for risk stratification.

Supporting the implementation of new healthcare technologies by investigating generalisability of pilot studies using area-level statistics.

BACKGROUND: Implementation of new technologies into national health care systems requires careful capacity planning. This is sometimes informed by data from pilot studies that implement the technology on a small scale in selected areas. A critical consideration when using implementation pilot studies for capacity planning in the wider system is generalisability. We studied the feasibility of using publicly available national statistics to determine the degree to which results from a pilot might generalise for non-pilot areas, using the English human papillomavirus (HPV) cervical screening pilot as an exemplar. METHODS: From a publicly available source on population indicators in England ("Public Health Profiles"), we selected seven area-level indicators associated with cervical cancer incidence, to produce a framework for post-hoc pilot generalisability analysis. We supplemented these data by those from publicly available English Office for National Statistics modules. We compared pilot to non-pilot areas, and pilot regimens (pilot areas using the previous standard of care (cytology) vs. the new screening test (HPV)). For typical process indicators that inform real-world capacity planning in cancer screening, we used standardisation to re-weight the values directly observed in the pilot, to better reflect the wider population. A non-parametric quantile bootstrap was used to calculate 95% confidence intervals (CI) for differences in area-weighted means for indicators. RESULTS: The range of area-level statistics in pilot areas covered most of the spectrum observed in the wider population. Pilot areas were on average more deprived than non-pilot areas (average index of multiple deprivation 24.8 vs. 21.3; difference: 3.4, 95% CI: 0.2-6.6). Participants in HPV pilot areas were less deprived than those in cytology pilot areas, matching area-level statistics. Differences in average values of the other six indicators were less pronounced. The observed screening process indicators showed minimal change after standardisation for deprivation. CONCLUSIONS: National statistical sources can be helpful in establishing the degree to which the types of areas outside pilot studies are represented, and the extent to which they match selected characteristics of the rest of the health care system ex-post. Our analysis lends support to extrapolation of process indicators from the HPV screening pilot across England.

Effective methylation triage of HPV positive women with abnormal cytology in a middle-income country.

The S5-methylation test, an alternative to cytology and HPV16/18 genotyping to triage high-risk HPV-positive (hrHPV+) women, has not been widely validated in low-middle-income countries (LMICs). We compared S5 to HPV16/18 and cytology to detect cervical intraepithelial neoplasia Grade 2 or worse (CIN2+) and CIN3+ in hrHPV+ women selected from a randomized pragmatic trial of 2661 Colombian women with an earlier-borderline abnormal cytology. We included all hrHPV+ CIN2 and CIN3+ cases (n = 183) age matched to 183

Mammographic density, endocrine therapy and breast cancer risk: a prognostic and predictive biomarker review.

BACKGROUND: Endocrine therapy is effective at preventing or treating breast cancer. Some forms of endocrine therapy have been shown to reduce mammographic density. Reduced mammographic density for women receiving endocrine therapy could be used to estimate the chance of breast cancer returning or developing breast cancer in the first instance (a prognostic biomarker). In addition, changes in mammographic density might be able to predict how well a woman responds to endocrine therapy (a predictive biomarker). The role of breast density as a prognostic or predictive biomarker could help improve the management of breast cancer. OBJECTIVES: To assess the evidence that a reduction in mammographic density following endocrine therapy for breast cancer prevention in women without previous breast cancer, or for treatment in women with early-stage hormone receptor-positive breast cancer, is a prognostic or predictive biomarker. SEARCH METHODS: We searched the Cochrane Breast Cancer Group Specialised Register, CENTRAL, MEDLINE, Embase, and two trials registers on 3 August 2020 along with reference checking, bibliographic searching, and contact with study authors to obtain further data. SELECTION CRITERIA: We included randomised, cohort and case-control studies of adult women with or without breast cancer receiving endocrine therapy. Endocrine therapy agents included were selective oestrogen receptor modulators and aromatase inhibitors. We required breast density before start of endocrine therapy and at follow-up. We included studies published in English. DATA COLLECTION AND ANALYSIS: We used standard methodological procedures expected by Cochrane. Two review authors independently extracted data and assessed risk of bias using adapted Quality in Prognostic Studies (QUIPS) and Risk Of Bias In Non-randomised Studies - of Interventions (ROBINS-I) tools. We used the GRADE approach to evaluate the certainty of the evidence. We did not perform a quantitative meta-analysis due to substantial heterogeneity across studies. MAIN RESULTS: Eight studies met our inclusion criteria, of which seven provided data on outcomes listed in the protocol (5786 women). There was substantial heterogeneity across studies in design, sample size (349 to 1066 women), participant characteristics, follow-up (5 to 14 years), and endocrine therapy agent. There were five breast density measures and six density change definitions. All studies had at least one domain as at moderate or high risk of bias. Common concerns were whether the study sample reflected the review target population, and likely post hoc definitions of breast density change. Most studies on prognosis for women receiving endocrine therapy reported a reduced risk associated with breast density reduction. Across endpoints, settings, and agents, risk ratio point estimates (most likely value) were between 0.1 and 1.5, but with substantial uncertainty. There was greatest consistency in the direction and magnitude of the effect for tamoxifen (across endpoints and settings, risk ratio point estimates were between 0.3 and 0.7). The findings are summarised as follows. Prognostic biomarker findings: Treatment Breast cancer mortality Two studies of 823 women on tamoxifen (172 breast cancer deaths) reported risk ratio point estimates of ~0.4 and ~0.5 associated with a density reduction. The certainty of the evidence was low. Recurrence Two studies of 1956 women on tamoxifen reported risk ratio point estimates of ~0.4 and ~0.7 associated with a density reduction. There was risk of bias in methodology for design and analysis of the studies and considerable uncertainty over the size of the effect. One study of 175 women receiving an aromatase inhibitor reported a risk ratio point estimate of ~0.1 associated with a density reduction. There was considerable uncertainty about the effect size and a moderate or high risk of bias in all domains. One study of 284 women receiving exemestane or tamoxifen as part of a randomised controlled trial reported risk ratio point estimates of ~1.5 (loco-regional recurrence) and ~1.3 (distance recurrence) associated with a density reduction. There was risk of bias in reporting and study confounding, and uncertainty over the size of the effects. The certainty of the evidence for all recurrence endpoints was very low. Incidence of a secondary primary breast cancer Two studies of 451 women on exemestane, tamoxifen, or unknown endocrine therapy reported risk ratio point estimates of ~0.5 and ~0.6 associated with a density reduction. There was risk of bias in reporting and study confounding, and uncertainty over the effect size. The certainty of the evidence was very low. We were unable to find data regarding the remaining nine outcomes prespecified in the review protocol. Prevention Incidence of invasive breast cancer and ductal carcinoma in situ (DCIS) One study of 507 women without breast cancer who were receiving preventive tamoxifen as part of a randomised controlled trial (51 subsequent breast cancers) reported a risk ratio point estimate of ~0.3 associated with a density reduction. The certainty of the evidence was low. Predictive biomarker findings: One study of a subset of 1065 women from a randomised controlled trial assessed how much the effect of endocrine therapy could be explained by breast density declines in those receiving endocrine therapy. This study evaluated the prevention of invasive breast cancer and DCIS. We found some evidence to support the hypothesis, with a risk ratio interaction point estimate ~0.5. However, the 95% confidence interval included unity, and data were based on 51 women with subsequent breast cancer in the tamoxifen group. The certainty of the evidence was low. AUTHORS' CONCLUSIONS: There is low-/very low-certainty evidence to support the hypothesis that breast density change following endocrine therapy is a prognostic biomarker for treatment or prevention. Studies suggested a potentially large effect size with tamoxifen, but the evidence was limited. There was less evidence that breast density change following tamoxifen preventive therapy is a predictive biomarker than prognostic biomarker. Evidence for breast density change as a prognostic treatment biomarker was stronger for tamoxifen than aromatase inhibitors. There were no studies reporting mammographic density change following endocrine therapy as a predictive biomarker in the treatment setting, nor aromatase inhibitor therapy as a prognostic or predictive biomarker in the preventive setting. Further research is warranted to assess mammographic density as a biomarker for all classes of endocrine therapy and review endpoints.

Mammographic density change in a cohort of premenopausal women receiving tamoxifen for breast cancer prevention over 5 years.

BACKGROUND: A decrease in breast density due to tamoxifen preventive therapy might indicate greater benefit from the drug. It is not known whether mammographic density continues to decline after 1 year of therapy, or whether measures of breast density change are sufficiently stable for personalised recommendations. METHODS: Mammographic density was measured annually over up to 5 years in premenopausal women with no previous diagnosis of breast cancer but at increased risk of breast cancer attending a family-history clinic in Manchester, UK (baseline 2010-2013). Tamoxifen (20 mg/day) for prevention was prescribed for up to 5 years in one group; the other group did not receive tamoxifen and were matched by age. Fully automatic methods were used on mammograms over the 5-year follow-up: three area-based measures (NN-VAS, Stratus, Densitas) and one volumetric (Volpara). Additionally, percentage breast density at baseline and first follow-up mammograms was measured visually. The size of density declines at the first follow-up mammogram and thereafter was estimated using a linear mixed model adjusted for age and body mass index. The stability of density change at 1 year was assessed by evaluating mean squared error loss from predictions based on individual or mean density change at 1 year. RESULTS: Analysis used mammograms from 126 healthy premenopausal women before and as they received tamoxifen for prevention (median age 42 years) and 172 matched controls (median age 41 years), with median 3 years follow-up. There was a strong correlation between percentage density measures used on the same mammogram in both the tamoxifen and no tamoxifen groups (all correlation coeficients > 0.8). Tamoxifen reduced mean breast density in year 1 by approximately 17-25% of the inter-quartile range of four automated percentage density measures at baseline, and from year 2, it decreased further by approximately 2-7% per year. Predicting change at 2 years using individual change at 1 year was approximately 60-300% worse than using mean change at 1year. CONCLUSIONS: All measures showed a consistent and large average tamoxifen-induced change in density over the first year, and a continued decline thereafter. However, these measures of density change at 1 year were not stable on an individual basis.

A case-control evaluation of 143 single nucleotide polymorphisms for breast cancer risk stratification with classical factors and mammographic density.

Panels of single nucleotide polymorphisms (SNPs) stratify risk for breast cancer in women from the general population, but studies are needed assess their use in a fully comprehensive model including classical risk factors, mammographic density and more than 100 SNPs associated with breast cancer. A case-control study was designed (1,668 controls, 405 cases) in women aged 47-73 years attending routine screening in Manchester UK, and enrolled in a wider study to assess methods for risk assessment. Risk from classical questionnaire risk factors was assessed using the Tyrer-Cuzick model; mean percentage visual mammographic density was scored by two independent readers. DNA extracted from saliva was genotyped at selected SNPs using the OncoArray. A predefined polygenic risk score based on 143 SNPs was calculated (SNP143). The odds ratio (OR, and 95% confidence interval, CI) per interquartile range (IQ-OR) of SNP143 was estimated unadjusted and adjusted for Tyrer-Cuzick and breast density. Secondary analysis assessed risk by oestrogen receptor (ER) status. The primary polygenic risk score was well calibrated (O/E OR 1.10, 95% CI 0.86-1.34) and accuracy was retained after adjustment for Tyrer-Cuzick risk and mammographic density (IQ-OR unadjusted 2.12, 95% CI% 1.75-2.42; adjusted 2.06, 95% CI 1.75-2.42). SNP143 was a risk factor for ER+ and ER- breast cancer (adjusted IQ-OR, ER+ 2.11, 95% CI 1.78-2.51; ER- 1.81, 95% CI 1.16-2.84). In conclusion, polygenic risk scores based on a large number of SNPs improve risk stratification in combination with classical risk factors and mammographic density, and SNP143 was similarly predictive for ER-positive and ER-negative disease.

Errors in determination of net survival: cause-specific and relative survival settings.

BACKGROUND: Cause-specific and relative survival estimates differ. We aimed to examine these differences in common cancers where by possible identifying the most plausible sources of error in each estimate. METHODS: Ten-year cause-specific and relative survival were estimated for lung, breast, prostate, ovary, oesophagus and colorectal cancers. The cause-specific survival was corrected for misclassification of cause of death. The Pohar-Perme relative survival estimator was modified by (1) correcting for differences in deaths from ischaemic heart disease (IHD) between cancers and general population; or (2) correcting the population hazard for smoking (lung cancer only). RESULTS: For all cancers except breast and prostate, relative survival was lower than cause-specific. Correction for published error rates in cause of death gave implausible results. Correction for rates of IHD death gave slightly different relative survival estimates for lung, oesophagus and colorectal cancers. For lung cancer, when the population hazard was inflated for smoking, survival estimates were increased. CONCLUSION: Results agreed with the consensus that relative survival is usually preferable. However, for some cancers, relative survival might be inaccurate (e.g. lung and prostate). Likely solutions include enhancing life tables to include other demographic variables than age and sex, and to stratify relative survival calculation by cause of death.

Young adulthood body mass index, adult weight gain and breast cancer risk: the PROCAS Study (United Kingdom).

BACKGROUND: We tested the hypothesis that body mass index (BMI) aged 20 years modifies the association of adult weight gain and breast cancer risk. METHODS: We recruited women (aged 47-73 years) into the PROCAS (Predicting Risk Of Cancer At Screening; Manchester, UK: 2009-2013) Study. In 47,042 women, we determined BMI at baseline and (by recall) at age 20 years, and derived weight changes. We estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for new breast cancer using Cox models and explored relationships between BMI aged 20 years, subsequent weight changes and breast cancer risk. RESULTS: With median follow-up of 5.6 years, 1142 breast cancers (post-menopausal at entry: 829) occurred. Among post-menopausal women at entry, BMI aged 20 years was inversely associated [HR per SD: 0.87 (95% CI: 0.79-0.95)], while absolute weight gain was associated with breast cancer [HR per SD:1.23 (95% CI: 1.14-1.32)]. For post-menopausal women who had a recall BMI aged 20 years <23.4 kg/m2 (75th percentile), absolute weight gain was associated with breast cancer [HR per SD: 1.31 (95% CIs: 1.21-1.42)], but there were no associations for women with a recall BMI aged 20 years of >23.4 kg/m2 (Pinteraction values <0.05). CONCLUSIONS: Adult weight gain increased post-menopausal breast cancer risk only among women who were <23.4 kg/m2 aged 20 years.

What are the benefits and harms of risk stratified screening as part of the NHS breast screening Programme? Study protocol for a multi-site non-randomised comparison of BC-predict versus usual screening (NCT04359420).

BACKGROUND: In principle, risk-stratification as a routine part of the NHS Breast Screening Programme (NHSBSP) should produce a better balance of benefits and harms. The main benefit is the offer of NICE-approved more frequent screening and/ or chemoprevention for women who are at increased risk, but are unaware of this. We have developed BC-Predict, to be offered to women when invited to NHSBSP which collects information on risk factors (self-reported information on family history and hormone-related factors via questionnaire; mammographic density; and in a sub-sample, Single Nucleotide Polymorphisms). BC-Predict produces risk feedback letters, inviting women at high risk (≥8% 10-year) or moderate risk (≥5 to < 8% 10-year) to have discussion of prevention and early detection options at Family History, Risk and Prevention Clinics. Despite the promise of systems such as BC-Predict, there are still too many uncertainties for a fully-powered definitive trial to be appropriate or ethical. The present research aims to identify these key uncertainties regarding the feasibility of integrating BC-Predict into the NHSBSP. Key objectives of the present research are to quantify important potential benefits and harms, and identify key drivers of the relative cost-effectiveness of embedding BC-Predict into NHSBSP. METHODS: A non-randomised fully counterbalanced study design will be used, to include approximately equal numbers of women offered NHSBSP (n = 18,700) and BC-Predict (n = 18,700) from selected screening sites (n = 7). In the initial 8-month time period, women eligible for NHSBSP will be offered BC-Predict in four screening sites. Three screening sites will offer women usual NHSBSP. In the following 8-months the study sites offering usual NHSBSP switch to BC-Predict and vice versa. Key potential benefits including uptake of risk consultations, chemoprevention and additional screening will be obtained for both groups. Key potential harms such as increased anxiety will be obtained via self-report questionnaires, with embedded qualitative process analysis. A decision-analytic model-based cost-effectiveness analysis will identify the key uncertainties underpinning the relative cost-effectiveness of embedding BC-Predict into NHSBSP. DISCUSSION: We will assess the feasibility of integrating BC-Predict into the NHSBSP, and identify the main uncertainties for a definitive evaluation of the clinical and cost-effectiveness of BC-Predict. TRIAL REGISTRATION: Retrospectively registered with clinicaltrials.gov (NCT04359420).

Risk Models for Breast Cancer and Their Validation.

Strategies to prevent cancer and diagnose it early when it is most treatable are needed to reduce the public health burden from rising disease incidence. Risk assessment is playing an increasingly important role in targeting individuals in need of such interventions. For breast cancer many individual risk factors have been well understood for a long time, but the development of a fully comprehensive risk model has not been straightforward, in part because there have been limited data where joint effects of an extensive set of risk factors may be estimated with precision. In this article we first review the approach taken to develop the IBIS (Tyrer-Cuzick) model, and describe recent updates. We then review and develop methods to assess calibration of models such as this one, where the risk of disease allowing for competing mortality over a long follow-up time or lifetime is estimated. The breast cancer risk model model and calibration assessment methods are demonstrated using a cohort of 132,139 women attending mammography screening in the State of Washington, USA.