Norman, Aaron

Abstract Background Artificial intelligence (AI)–based, mammography breast cancer risk prediction models show improved discriminatory accuracy relative to clinical risk models. However, data on their calibration are limited. This study compared model performance of 3 clinical breast cancer risk models: Gail, Tyrer–Cuzick v8, and Breast Cancer Surveillance Consortium v3 with the MIRAI AI-risk model. Methods Digital mammograms were ascertained from a screening mammography cohort of 12 308 women within the Mayo Clinic Biobank with 250 incident breast cancers (176 invasive) within 5 years. We predicted 5-year breast cancer risk, estimated discriminatory accuracy (C-index), and calibration (observed-to-expected ratio) of both overall and invasive breast cancer and compared estimates using bootstrapping approaches. Results MIRAI demonstrated similar or improved discriminatory accuracy of overall breast cancer (C-index = 0.71, 95% confidence interval [CI] = 0.68 to 0.74) and invasive breast cancer (C-index = 0.71, 95% CI = 0.67 to 0.75) compared with clinical models (overall breast cancer: C-index = 0.59-0.68; invasive breast cancer: C-index = 0.60-0.68). MIRAI’s calibration for risk of overall breast cancer (observed-to-expected ratio = 0.96, 95% CI = 0.85 to 1.08) was improved compared with Gail (observed-to-expected ratio = 1.22, 95% CI = 1.07 to 1.38) and Breast Cancer Surveillance Consortium (observed-to-expected ratio = 1.38, 95% CI = 1.22 to 1.56) but similar to TC with volumetric percent density and polygenic risk score (observed-to-expected ratio = 0.99, 95% CI = 0.87 to 1.13). However, for low-risk women (approximately 50%), MIRAI overestimated risk of overall breast cancer. MIRAI also overestimated risk of invasive breast cancer across the risk spectrum (observed-to-expected ratio = 0.68, 95% CI = 0.58 to 0.78), whereas clinical models had good calibration (observed-to-expected ratio = 0.86–0.99). Conclusion MIRAI demonstrated stronger discriminatory accuracy than clinical models for 5-year overall and invasive breast cancer risk prediction but overestimated risk for both breast cancer endpoints. AI-based risk models should consider discriminatory accuracy and calibration for invasive cancer before implementation.

Summary Evidence for an association between insulin‐like growth factors (IGF) and multiple myeloma (MM) is inconsistent. We examined total IGF‐I concentrations and risk of MM by combining baseline serological data among UK Biobank participants ( n = 444 187; 732 incident MM) with a two‐sample Mendelian randomisation (MR) analysis using identified genetic variants associated with circulating total IGF‐I and IGF‐binding protein 3 (IGFBP‐3) in the InterLymph consortium (2434 MM and their 2567 controls). Finally, additional lymphoid neoplasm (LN) subtypes were included for comparison with the main hypothesis. Circulating IGF‐I level was positively associated with MM risk Hazard ratio‐HR‐per one standard deviation‐SD‐increase (HR 1‐SD = 1.11, 95% confidence interval [CI]: 1.01–1.22; p ‐value = 0.03), especially closer to diagnosis. Genetically inferred IGF‐I levels were associated with increased MM risk (odds ratio [OR] = 1.27, 95% CI: 1.05–1.54) but not with any other LNs. Genetically inferred IGFBP‐3 levels showed no associations with any LN evaluated. Corroborating previous findings, in a secondary analysis, IGF‐I levels were associated with the risk of chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL) in males with higher body mass index (HR 1‐SD in obese male = 1.36, 95% CI: 1.14–1.61). Our serological and MR analyses suggest a contributing role of IGF‐I in the susceptibility of MM; the CLL/SLL findings warrant further investigation considering sex‐specific adiposity.

Abstract Studies have identified genetic and epidemiologic factors associated with mammographic density (MD) phenotypes. However, MD-associated genetic variants only account for a small proportion of the total estimated heritability. Interrogating interactions between genetic and epidemiologic factors could potentially identify additional MD-associated loci, expand our understanding of the genetic basis of MD phenotypes, and clarify how epidemiologic factors modulate relationships between genetic variants and MD. We conducted six separate genome-wide, gene–environment (GxE) interaction analyses, applying 2 degrees of freedom (df) and 1df interaction tests, for each of three MD phenotypes (percent density, dense area (DA), and nondense area (NDA)). The six epidemiologic factors considered were height, ever parous, parity, ever menopausal hormone therapy, ever breastfeeding, and months of breastfeeding. We included European ancestry participants from multiple studies within the Markers of Density consortium and the Breast Cancer Association Consortium (n = 4895-16 218 depending on specific analyses). We identified 11 loci with genome-wide significant (P < 5 × 10−8) interaction tests including two novel common genetic signals interacting with parity (8p21.2) and ever breastfeeding (19p13.2) for NDA. Our results suggest that epidemiologic risk factors might influence relationships between common genetic variants and MD phenotypes at particular genomic loci.

Abstract Monoclonal gammopathy of undetermined significance (MGUS) is an asymptomatic premalignant disorder. The current standard of care is not to screen for MGUS, so it is often incidentally diagnosed in the clinic. It is unknown whether the outcomes of screened vs clinically detected MGUS differ. We compared the progression risk between screened vs clinical MGUS cohorts and assessed whether the MGUS detection method affected risk prediction of established clinical factors (score). We included 379 screened MGUS cases from the Olmsted County population-based study and 1384 patients with MGUS diagnosed during routine clinical evaluation at Mayo Clinic. Median follow-up time for the screened vs clinical cohort was 26.6 and 40.1 years, respectively. Accounting for death as a competing risk, the cumulative incidence of progression at 25 years was similar in the screened (11.1% [95% confidence interval [CI], 8.3-14.8]) vs clinical (10.1% [95% CI, 8.6-11.8]) MGUS cohorts, even when stratified by sex, age, or the baseline MGUS risk score. Overall, 0.9 (95% CI, 0.6-1.2) of patients with screened MGUS vs 1.0 (95% CI, 0.9-1.2) of those with clinically detected MGUS experienced disease progression for every 100 person-years of follow-up. MGUS detection method did not modify the association between MGUS risk score and progression risk (pinteraction = 0.217) and did not add to known risk factors for progression (likelihood ratio test; P = .839). Here, we show that progression risk among patients with screened vs clinically detected heavy-chain MGUS was similar. Future studies are needed to assess whether tailored follow-up of patients with screened MGUS affects clinical outcomes.

AbstractNon‐Hodgkin lymphoma (NHL) is composed of a heterogeneous collection of subtypes with considerable differences in genetics, biology and aetiology. Studies to date on physical activity and NHL risk have not had sufficient sample size to evaluate whether associations differ by subtype. We pooled data from nine case‐control studies to examine the association between moderate‐to‐vigorous intensity physical activity (MVPA) and risk of NHL overall and by subtype (diffuse large B‐cell lymphoma, follicular lymphoma, chronic lymphocytic leukaemia/small lymphocytic lymphoma, marginal zone lymphoma and mature T‐cell lymphoma). A total of 5653 cases and 9115 controls were included in the pooled analysis. Physical activity was harmonised across nine studies and modelled as study‐specific tertiles. Multinomial logistic regression was used to estimate the association between physical activity and NHL, adjusting for confounders. The overall odds of NHL was 13% lower among participants in the most active tertile of MVPA compared to the least active tertile (adjusted odds ratio = 0.87, 95% CI = 0.80, 0.95). Similar decreases were observed across NHL subtypes. In summary, in this pooled analysis of case‐control studies, physical activity was associated with a modest risk reduction for each NHL subtype examined and with overall NHL.

BackgroundThe objective of this randomized trial was to evaluate the short‐term effect of bilingual written and interpersonal education regarding mammographic breast density (MBD).MethodsLatinas aged 40 to 74 years who were presenting for screening mammography were recruited and randomized 1:1:1 to receive a letter with their mammogram and MBD results (usual care [UC]), a letter plus a brochure (enhanced care [ENH]), or a letter plus a brochure and telephonic promotora education (interpersonal care [INT]). Surveys were administered at enrollment (T0) and 2 weeks to 6 months after intervention delivery (T1). Differences were assessed with χ2, Kruskal‐Wallis, and McNemar tests and pairwise comparisons as appropriate. INT metrics and audio recordings were analyzed with descriptive statistics and qualitative content analysis.ResultsBetween October 2016 and October 2019, 943 of 1108 Latina participants (85%) completed both surveys. At T1, INT participants were more likely (P < .001) to report seeing their MBD results in the letter (70.2%) than UC (53.1%) or ENH participants (55.1%). The percentage of INT women who reported speaking with a provider about MBD (29.0%) was significantly greater (P < .001) than the percentage of UC (14.7%) or ENH participants (15.6%). All groups saw significant (P < .001) but nondifferential improvements in their knowledge of MBD as a masking and risk factor. In the INT group, the promotora delivered education to 77.1% of the 446 participants randomized to INT and answered questions at 28.3% of the encounters for an average of $4.70 per participant.ConclusionsAmong Latinas in a low‐resource setting, MBD knowledge may increase with written or interpersonal education, but with modest investment, interpersonal education may better improve MBD awareness and prompt patient‐provider discussions.

Abstract Antimullerian hormone (AMH) is a promising biomarker for ovarian reserve. In this study, we assessed AMH before and 1 year after initiation of adjuvant chemotherapy on National Surgical Adjuvant Breast and Bowel Project (NSABP)/NRG Oncology B-47 in female participants aged 42 years and younger (median age = 39 years). At baseline, median AMH was 1.2 ng/mL; 13 (4.7%) values were less than 0.1 ng/mL (the threshold for detectable levels, in the perimenopause and menopause range), and 57 values (20.6%) were less than 0.5 ng/mL. At 1 year, 215 (77.6%) were less than 0.1 ng/mL, and 264 (95.3%) were less than 0.5 ng/mL. Postchemotherapy menses were reported by 46.2% of participants. Multivariable logistic regression found that the odds of having postchemotherapy menses increased with younger age, higher body mass index, and higher prechemotherapy AMH, but not by trastuzumab administration or by the choice of chemotherapy (doxorubicin-cyclophosphamide followed by paclitaxel vs docetaxel-cyclophosphamide). We conclude that higher prechemotherapy AMH predicts a lower risk of chemotherapy-induced amenorrhea and that AMH 1 year after chemotherapy initiation is not informative in this setting because it is likely to be very low.

Abstract We evaluated the joint associations between a new 313-variant PRS (PRS313) and questionnaire-based breast cancer risk factors for women of European ancestry, using 72 284 cases and 80 354 controls from the Breast Cancer Association Consortium. Interactions were evaluated using standard logistic regression and a newly developed case-only method for breast cancer risk overall and by estrogen receptor status. After accounting for multiple testing, we did not find evidence that per-standard deviation PRS313 odds ratio differed across strata defined by individual risk factors. Goodness-of-fit tests did not reject the assumption of a multiplicative model between PRS313 and each risk factor. Variation in projected absolute lifetime risk of breast cancer associated with classical risk factors was greater for women with higher genetic risk (PRS313 and family history) and, on average, 17.5% higher in the highest vs lowest deciles of genetic risk. These findings have implications for risk prevention for women at increased risk of breast cancer.