Risk Factors and Mortality Among Women With Interval Breast Cancer vs Screen-Detected Breast Cancer

Key Points Question What are the established risk factors for women with interval breast cancer (IBC) vs screen-detected breast cancer (SBC), and how might they be associated with mortality outcomes? Findings In this cohort study of 18 194 Korean women who underwent mammographic breast cancer screening, breast density, obesity, and hormone replacement therapy use were associated with IBC compared with SBC. Overall mortality of IBC was comparable with that of SBC. Meaning These findings suggest that breast density is associated with IBC; additional research is warranted to determine whether a shortened screening period and new screening strategies are needed for women with risk factors for IBC.


Introduction
According to GLOBOCAN estimates, female breast cancer was the most common incident cancer type in 2020, with approximately 2.3 million new cases, accounting for 24.5% of all cancer diagnoses. 1In Korea, breast cancer is the most common cancer type in women aged 35 to 64 years, and its incidence has been increasing since nationwide cancer statistics were first reported in 1999. 2 To reduce the burden of breast cancer through early detection and treatment, mammographic breast cancer screening is recommended for average-risk women in many countries. 3In Korea, mammography is provided biennially to women aged 40 years or older as a nationwide organized screening program. 4east cancer diagnosed after a negative screening but before the next scheduled mammogram is referred to as interval breast cancer (IBC). 5,6There are 2 types: (1) "true" IBC, in which cancer did not present on the previous mammogram and thus developed during the screening interval; and (2)   tumors missed on the previous mammogram (ie, false-negative results). 5Compared with screendetected breast cancer (SBC), IBC has more aggressive characteristics and a worse prognosis. 5evious studies have identified the following risk factors for IBC: high breast density, current hormone replacement therapy (HRT) use, young age, premenopausal status, and family history of breast cancer. 7Previous studies focused on IBC risk factors and characteristics have been conducted mainly in Western countries, where early mammographic breast cancer screening is recommended.However, little is known about the risk factors for IBC among Asian women.
Identification of epidemiologically associated risk factors for IBC and survival rates of Asian women with IBC compared with those with SBC would be useful for improving screening strategies for women in Asia.Therefore, this study aimed to compare established risk factors for IBC and SBC among Korean women using data from a nationwide mammographic screening program.Furthermore, mortality outcomes of IBC were compared with those of SBC.

Methods
The institutional review board of Hanyang University College of Medicine, Republic of Korea, approved the protocol for this cohort study.In addition, the National Health Insurance Sharing Service System (NHIS) approved the use of the NHIS database, which was constructed after individual identities were anonymized.The need for informed consent was thus waived due to secondary data analysis.The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Study Design and Population
This study used data from the National Health Insurance Service National Health Information Database (NHIS-NHID).As a single compulsory health insurance system in Korea, the NHIS-NHID includes information on demographics, health care use, vital statistics, and national health screening results for the entire Korean population. 8is study included data from all women who underwent national mammographic breast cancer screening between January 1, 2009, and December 31, 2012 (Figure).The mammographic screening film was read by a trained radiologist and the results were recorded according to the Breast Imaging Reporting and Data System, Fourth Edition (BI-RADS 4).A positive screening result was defined as a BI-RADS final assessment of 0, 4, or 5 (incomplete, suspicious abnormality, or highly suggestive of malignancy), and negative screening results were defined as 1, 2, or 3 (negative, benign finding, or probably benign) based on the American College of Radiology guideline. 9Participants were excluded if they met the following criteria: (1)

Definition of IBC and SBC
Breast cancer was defined as any breast cancer event according to International Classification of Diseases, Tenth Edition (ICD-10) codes for invasive breast cancer (C50.0-C50.9)and ductal carcinoma in situ (D05.0-D05.9)and as any cancer event registered with the rare incurable disease code for cancer.If a breast cancer diagnosis occurred within 6 months from the date of mammographic screening among women with positive screening results, the case was defined as SBC, considering the time of referral to a hospital for further examination and pathologic confirmation.The 6-month period to define SBC was applied in another claim-based study in Korea. 11Among women with negative screening results, cancers diagnosed within the first 6 months after negative screening were considered false-negative results and were excluded.Considering the definition of IBC and the screening cycle of other studies, [12][13][14] the case was defined as IBC if the diagnosis of breast cancer occurred during the period 6 to 24 months after negative breast cancer screening.Considering the recommendation of annual breast cancer screening in other guidelines, 3 IBC was further classified as breast cancer diagnosed either 6 to 12 months or 12 to 24 months after negative screening.Breast cancer was defined as a combination of the ICD-10 codes for invasive breast cancer and ductal carcinoma in situ and the special code for cancer. 15Based on this definition, we identified 8702 women with IBC (2359 were diagnosed between 6 and 12 months after negative screening, and 6343 were diagnosed between 12 and 24 months) from 5 628 035 with negative results (0.15%) and 9492 women with SBC from 1 144 776 with positive results (0.83%).

Established Risk Factors for Breast Cancer Considered in This Study
During cancer screening, participants were asked to complete a self-administered questionnaire about their health behaviors and risk factors for breast cancer.

Characteristics of Women With IBC vs SBC
A total of 18 194 women were included in the study; 9492 had SBC and 8702 had IBC (

Mortality Risk Among Women With IBC vs SBC
The mean (SD) follow-up period after breast cancer diagnosis was 8.4 (1.9) years for women with IBC and 9.5 (2.1) years for those with SBC.
Age at menarche, y <15 Menopausal status Premenopause 7][18] Studies have reported that rates of IBC increase with an increasing duration of HRT use, and the association between HRT use and IBC is greater than that between HRT and SBC. 19,201][22] Previous studies have reported a higher proportion of premenopausal women with IBC than women with SBC. 23,24These associations are consistent with the results of this study.The mechanisms of the previous studies underlying the associations among HRT use, lower BMI, premenopause, and IBC were mainly explained by the effect of these factors on dense breasts and the masking effect or independent increased risk of IBC due to higher breast density. 22,25However,

6 772 811 Participants included in the analysis 674 304 Excluded 347 845 Aged <40 or >74 y at screening 179 128
Mortality and causes of death of individuals between 2009 and 2020 were confirmed by linking the NHIS-NHID data with mortality data from the Korea National Statistical Office (KNSO) using a unique 13-digit resident registration number.The KNSO death certificate includes the cause of death according to the ICD-10 code and the date of death.First, all-cause mortality, defined as death from any cause, was assessed.Causes of death were classified as breast cancer-related death (recorded cause of death as breast cancer) or deaths other than breast cancer (all other causes).Moreover, Died or diagnosed with any type of cancer before screening 147 259 Missing information on mammographic breast density 45 Missing information on breast cancer screening results 27 History of breast cancer before screening 6343 With interval breast cancer detected between 12 and 24 mo 2359 With interval breast cancer detected between 6 and 12 mo or 4, >75% glandular), body mass index (BMI [calculated as weight in kilograms divided by height in meters squared]; underweight, <18.5; normal, 18.5 to <23.0; overweight, 23.0 to <25.0; or obesity, Ն25.0), age at menarche (<15, 15-16, or Ն17 years), menopausal status (premenopausal or postmenopausal), oral contraceptive use (never or ever), HRT use after menopause (never or ever), parity (nulliparous, primiparous, or multiparous), breastfeeding experience (never or ever), family history of breast cancer in first-degree relatives (no or yes), current alcohol consumption (no or yes [<1 or Ն1 per week during the last year]), and physical activity (no or yes [<1 or Ն1 time per week at time of screening]).causes of death were also classified into cancer-related deaths (recorded causes of death as any cancer) and deaths other than cancer (all other causes).Statistical Analysis Descriptive statistics on breast cancer risk factors among women with SBC or IBC were compared using the t test or χ 2 test.Continuous variables are presented as means with SDs, and categorical variables are presented as numbers with percentages.Logistic regression with the aforementioned established risk factors for breast cancer was used to identify epidemiologically associated factors for IBC compared with SBC.Odds ratios (ORs) and 95% CIs for each factor with and without adjustments for other factors are presented.Analysis was performed for all participants with IBC and stratified by IBC diagnosed between 6 and 12 months and between 12 and 24 months.Given that mammographic density could affect IBC itself and its association with other factors, 16 a multiple logistic regression analysis of the association between established risk factors for breast cancer and IBC compared with SBC was performed, stratified by breast density (nondense [BI-RADS density categories 1-2] and dense [BI-RADS density categories 3-4]).Mortality outcomes for IBC compared with SBC were analyzed using Cox proportional hazards regression analysis.Follow-up for the mortality outcome was calculated from the date of breast cancer diagnosis to the date of death or December 31, 2020.Hazard ratios (HRs) for mortality risk are presented as unadjusted, adjusted for age at diagnosis, and adjusted for all of the aforementioned known risk factors for breast cancer for all-cause mortality, breast cancer-related death, cancerrelated death, deaths other than breast cancer, and deaths other than cancer.A 2-sided P < .05 was considered statistically significant.Statistical analysis was performed with SAS, version 9.4 (SAS Institute).Data were analyzed from March 1 to June 30, 2023.

Table 1 .
Characteristics of the IBC and SBC Groups a (continued) χ 2 test for categorical variables and t test for continuous variables.
a Unless indicated otherwise, values are presented as No. (%) of patients.bSubdivided by diagnosis time: within 6 and 12 months and 12 and 24 months.c

Table 2 .
Risk Factors for IBC vs SBC Multivariate logistic regression model adjusted for age at diagnosis, BI-RADS breast density, BMI, age at menarche, menopausal status, oral contraceptive use, HRT use among postmenopausal women, parity, breastfeeding experience, family history of breast cancer in first-degree relatives, alcohol consumption, and physical activity.Risk Factors and Mortality Among Women With Interval vs Screen-Detected Breast Cancer deaths compared with SBC.Otherwise, increased mortality or cause-specific mortality was not observed for IBC at 12 to 24 months compared with SBC.This study investigated differences in risk factors associated with IBC compared with those associated with SBC and mortality outcomes between IBC and SBC among Korean women who participated in a nationwide breast cancer screening program.In this study, increased breast density was associated with a lower likelihood of IBC; however, decreased BMI, HRT use after menopause, and alcohol consumption were associated with an increased likelihood of IBC compared with SBC.Age and family history of breast cancer had different associations according to the IBC diagnostic period after negative screening results.Younger age at the time of breast cancer diagnosis and family history of breast cancer were associated with an increased likelihood of IBC diagnosis 6 to 12 months after negative screening results but with a decreased likelihood of IBC diagnosis at 12 to 24 months.

Table 3 .
Mortality Among Women With IBC vs SBC Subdivided into 2 groups according to diagnosis time: within 6 and 12 months and 12 and 24 months.
a b Model 1: IBC as the outcome without adjustment.c Model 2: IBC as the outcome, adjusted for age at diagnosis.d Model 3: IBC as the outcome, adjusted for age at diagnosis; Breast Imaging Reporting and Data System, Fourth Edition breast density; body mass index; age at menarche; menopausal status; oral contraceptive use; hormone replacement therapy experience; parity; breastfeeding experience; family history of breast cancer in first-degree relatives; alcohol consumption; and physical activity.