Quality of Life Among Patients With Ductal Carcinoma In Situ

Victoria J Dunsmore; Bradley S Snyder; Ilana F Gareen; Constance D Lehman; Seema A Khan; Justin Romanoff; Constantine Gatsonis; Ralph L Corsetti; Habib Rahbar; Derrick W Spell; Linda K Han; John R Bumberry; Kathy D Miller; Joseph A Sparano; Christopher Comstock; Elyse Park; Lynne I Wagner; Ruth C Carlos

doi:10.1001/jamanetworkopen.2025.18887

. 2025 Jul 3;8(7):e2518887. doi: 10.1001/jamanetworkopen.2025.18887

Quality of Life Among Patients With Ductal Carcinoma In Situ

Victoria J Dunsmore ^1,^✉, Bradley S Snyder ², Ilana F Gareen ³, Constance D Lehman ⁴, Seema A Khan ⁵, Justin Romanoff ², Constantine Gatsonis ⁶, Ralph L Corsetti ⁷, Habib Rahbar ⁸, Derrick W Spell ⁹, Linda K Han ¹⁰, John R Bumberry ¹¹, Kathy D Miller ¹⁰, Joseph A Sparano ¹², Christopher Comstock ¹³, Elyse Park ⁴, Lynne I Wagner ¹, Ruth C Carlos ¹⁴

¹University of North Carolina at Chapel Hill

²Center for Biostatistics and Health Data Science, Brown University School of Public Health, Providence, Rhode Island

³Center for Biostatistics and Health Data Science, Department of Epidemiology, Brown University School of Public Health, Providence, Rhode Island

⁴Massachusetts General Hospital, Boston

⁵Northwestern University, Chicago, Illinois

⁶Center for Biostatistics and Health Data Science, Department of Biostatistics, Brown University School of Public Health, Providence, Rhode Island

⁷Tulane University School of Medicine, New Orleans, Louisiana

⁸University of Washington, Seattle

⁹Gulf South NCORP, New Orleans, Louisiana

¹⁰Indiana University, Bloomington

¹¹Mercy Hospital Springfield, Springfield, Missouri

¹²Montefiore Medical Center-Weiler Hospital, Bronx, New York

¹³Memorial Sloan Kettering Cancer Center, New York, New York

¹⁴Columbia University, New York, New York

Accepted for Publication: March 28, 2025.

Published: July 3, 2025. doi:10.1001/jamanetworkopen.2025.18887

^✉

Corresponding Author: Victoria J. Dunsmore, PhD, The University of North Carolina at Chapel Hill, 115 Winifred Dr, Morrisville, NC 27560 (vdunsmo@unc.edu).

Author Contributions: Mr Snyder had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Dunsmore, Gareen, Lehman, Gatsonis, Rahbar, Miller, Sparano, Comstock, Park, Carlos.

Acquisition, analysis, or interpretation of data: Snyder, Gareen, Lehman, Khan, Romanoff, Gatsonis, Corsetti, Rahbar, Spell, Han, Bumberry, Miller, Sparano, Park, Wagner, Carlos.

Drafting of the manuscript: Dunsmore, Snyder, Park, Carlos.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Snyder, Gatsonis, Park.

Obtained funding: Lehman, Gatsonis, Sparano, Park.

Administrative, technical, or material support: Gareen, Khan, Corsetti, Spell, Han, Miller, Sparano, Park.

Supervision: Dunsmore, Lehman, Corsetti, Spell, Miller, Sparano, Comstock, Park.

Conflict of Interest Disclosures: Dr Dunsmore reported grants from National Cancer Institute's (NCI) National Research Service Award (No. T32 CA116339) outside the submitted work. Mr Snyder reported grants from Brown University School of Public Health to aid statistical analysis performed under the aegis of Eastern Cooperative Oncology Group–American College of Radiology Imaging Network (ECOG-ACRIN) during the conduct of the study. Dr Lehman reported grants from GE Healthcare, consultant fees from Hologic, and a role as cofounder and consultant with Clairity outside the submitted work. Dr Khan reported advisory work with Hahah Therapeutics outside the submitted work. Dr Romanoff reported grants from NCI's National Clinical Trials Network (No. U10CA180794) during the conduct of the study. Dr Rahbar reported grants from National Institutes of Health (NIH)/NCI funds to institution and grants from Guerbet, LLC outside the submitted work; he reported consulting fees from Guerbet, LLC outside the submitted work; he reported support for role as co-investigator in a prior study not related to this work or topic supported by GE Healthcare; he reported travel support to attend the ECOG-ACRIN and European Society of Breast Imaging meetings; and he reported receiving an honoraria to teach at the American College of Radiology Education Center outside the submitted work. Dr Han reported honoraria from Hologic and Cardinal Health outside the submitted work. Dr Comstock reported consulting fees from Bayer Medical and Recentric Healthcare Technologies outside the submitted work. Dr Wagner reported consulting fees from Celgene/Bristol Myers Squibb outside the submitted work. Dr Carlos reported grants from ECOG-ACRIN during the conduct of the study; she reported salary support from American College of Radiology as editor in chief of Journal of the American College of Radiology; she reported receiving a speaker honorarium from Canon Medical; she reported receiving travel reimbursement from the Radiological Society of North America, Association of Academic Radiology, GE Medical, and The Academy of Radiology and Biomedical Imaging Research outside the submitted work. No other disclosures were reported.

Funding/Support: This study was coordinated by the ECOG-ACRIN Cancer Research Group (Peter J. O’Dwyer, MD, and Mitchell D. Schnall, MD, PhD, group co-chairs) and supported by the National Cancer Institute of the National Institutes of Health under award Nos. UG1CA189828, UG1CA233180, UG1CA189859, UG1CA189854, UG1CA189822, UG1CA233328, UG1CA233320, UG1CA233290, and UG1CA233373. Dr Dunsmore was also supported by the NIH Cancer Care Quality Training Program No. T32 CA116339-16.

Role of the Funder/Sponsor: The funding agency (the National Cancer Institute of the National Institutes of Health) was involved in the design and conduct of the study through protocol (and addendum) approval and serving as a nonvoting member of the data safety monitoring board. However, the funding agency had no role in the collection, management, analysis, and interpretation of the data, no role in the preparation, review, or approval of the manuscript, and no role in the decision to submit the manuscript for publication.

Disclaimer: This content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data Sharing Statement: See Supplement 2.

Additional Contributions: The authors would like to acknowledge Jennifer L. Sabol, MD, for exceptional efforts in patient recruitment. Dr Sabol did not receive additional compensation beyond the terms of her employment for her contribution.

^✉

Corresponding author.

PMCID: PMC12232187 PMID: 40608342

Key Points

Question

Are clinical or social determinants of health factors associated with health-related quality of life in the 2-year period following surgery for ductal carcinoma in situ (DCIS) breast cancer?

Findings

In this cohort study including 296 patients with DCIS, minoritized race (including American Indian or Alaska Native, Asian, Black, multiracial, and unknown or not reported) and receiving more than 1 surgery were associated with worsening long-term mental health, and nonprivate insurance was associated with worsening of both mental and physical health. Non-White women who received more than 1 surgery experienced a sustained decline in mental health substantially larger than the defined minimal important change.

Meaning

Interventions aimed at improving mental health outcomes for at-risk DCIS patients are a crucial next step.

This cohort study and secondary analysis of a nonrandomized clinical trial examines health-related quality of life by race following surgery for ductal carcinoma in situ breast cancer.

Abstract

Importance

Limited longitudinal data exist regarding health-related quality of life (HRQL) following surgery for ductal carcinoma in situ (DCIS) breast cancer.

Objective

To assess individual- and neighborhood-level factors associated with longitudinal trajectories of mental and physical HRQL among individuals with DCIS eligible for breast conservation surgery.

Design, Setting, and Participants

This cohort study was an ancillary to a prospective, nonrandomized clinical trial of women with DCIS breast cancer between March 2015 and April 2016 at 75 US institutions, community practices, and academic centers coordinated by the Eastern Cooperative Oncology Group–American College of Radiology Imaging Network (ECOG-ACRIN) Cancer Research Group (E4112). Eligible participants were recently diagnosed with unilateral DCIS who were eligible for wide local excision (WLE) and had a diagnostic mammogram within 3 months of study registration. Patient-reported outcome (PRO) questionnaires were administered to participants at the time of registration, after surgeon consultation but presurgery, at the first postoperative visit, 12 months postsurgery, and 24 months postsurgery. Self-reported social determinants of health (SDOH) and clinical history and surgery received were also collected. Data were analyzed from June 2024 to November 2024.

Main Outcomes and Measures

PRO measures assessing patient knowledge of DCIS and perception of being informed; HRQL, measured using the 10-item Patient-Reported Outcomes Measurement Information System with mental and physical health subscales.

Results

Among the 296 women, the median age at enrollment was 60 years (range, 34-87 years) (11 Asian [4%], 41 Black [14%], 229 White [77%]); 147 participants (50%) reported at least 1 family member with breast cancer. The majority of participants had non-Hispanic ethnicity (280 [95%]), had private insurance (227 [77%]), resided in areas of low-to-moderate deprivation (ADI median, 44 [range, 1-99]), and received a single WLE (185 [63%]). Participants exhibited good knowledge of DCIS (median, 80 [range, 20-100]) and a high perception of being informed (median, 10 [range, 3-10]). Being American Indian or Alaska Native, Asian, Black, multiracial, or not reporting race (χ² = 6.8 [df, 2]; P = .03), having insurance other than private (χ² = 14.3 [df, 2]; P < .001), and having more than 1 surgery (χ² = 12.3 [df, 4]; P = .02) were associated with decreases in mental health at 24 months compared with baseline. Having private insurance and a low perception of being informed of one’s treatment were associated with increases in physical health at 2 years.

Conclusions and Relevance

In this cohort study of women with DCIS, in the 24 months following diagnosis, nonprivate insurance, minoritized race, and number of surgeries received were associated with reduced mental HRQL. Nonprivate insurance was also associated with reduced physical HRQL.

Trial Registration

ClinicalTrials.gov Identifier: NCT02352883

Introduction

Ductal carcinoma in situ (DCIS) is a nonobligate precursor breast cancer accounting for approximately 20% to 25% of newly diagnosed lesions that are commonly classified as malignant.^1,2 Rates of DCIS diagnosis have risen significantly with widespread screening mammography.^1,3 DCIS, often referred to as preinvasive or stage 0 cancer, is considered a key precursor to invasive breast cancer (IBC),⁴ with high-grade subtypes more likely to progress to IBC if left untreated.⁴ The goal of treatment in DCIS is to reduce the risk of this progression. Surgical treatment options for DCIS vary and include wide local excision (WLE), with or without radiation, as well as mastectomy.^5,6 Survival following these treatments is high, with 10-year disease-specific survival exceeding 98%,⁵ but patient choice in receiving these treatments varies. We previously demonstrated that the process of treatment decision-making in DCIS is dynamic and influenced by surgeon recommendation, self-reported importance of keeping one’s breast, and cancer worry.⁷

What is less known is how treatment impacts long-term health-related quality of life (HRQL) in the survivorship period for women with newly diagnosed DCIS. A 2020 review showed that HRQL among women with DCIS is comparable with women with IBC, and differences in HRQL have been related to surgery type.⁸ When examining social determinants of health (SDOH) among DCIS patients, authors found no differences in HRQL by age, race, or neighborhood deprivation index.⁹ When examining patient-clinician dynamics, 1 study found that shared decision-making was positively related to HRQL in cancer.¹⁰ However, previous work across SDOH and shared decision-making has predominantly focused on cross-sectional assessments of individuals across the survivorship spectrum. Furthermore, the limited longitudinal information on HRQL is often available only for women with IBC.¹¹ Uniquely missing are assessments of longitudinal trajectories of HRQL and factors that influence these trajectories in newly diagnosed DCIS. In addition, most work in HRQL using the Patient-Reported Outcomes Measurement Information System (PROMIS) measure has focused only on mental health outcomes rather than separate physical and mental health outcomes among people with cancer.¹² As a result, previous research in HRQL outcomes has often lacked the nuanced understanding of factors that may differentially impact physical and mental health. To fill this gap, we assessed the longitudinal trajectories of mental and physical HRQL and individual- and neighborhood-level factors associated with these trajectories among a prospective cohort of women with newly diagnosed DCIS participating in the Eastern Cooperative Oncology Group–American College of Radiology Imaging Network (ECOG-ACRIN) Cancer Research Group (E4112) nonrandomized clinical trial.¹³

Methods

The study was approved by the National Cancer Institute Division of Cancer Prevention, the local institutional review board at each participating site, and the Brown University institutional review board. Written informed consent was obtained from all participants (NCT02352883). Reporting of this study follows the Strengthening of the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

Data and Sample

This was an ancillary study to a prospective, nonrandomized clinical trial coordinated by the ECOG-ACRIN Cancer Research Group (E4112) that enrolled women with unilateral DCIS. Primary results, recruitment methods, data collection processes, and participating institutions have been previously described.^7,13 Patient-reported outcome (PRO) questionnaires were scheduled to be provided to participants at the time of registration, after surgeon consultation but presurgery, at the first postoperative visit, 12 months postsurgery, and 24 months postsurgery (Figure 1). See eMethods in Supplement 1 for details on PRO data collection.

Figure 1. — HRQL indicates health-related quality of life; MRI, magnetic resonance imaging; PRO, patient-reported outcome; PROMIS-10, 10-Item Patient Reported Outcomes Measurement Information System.

Covariates

Clinical Factors

Surgery received was categorized as initial WLE as the sole surgery, initial mastectomy as the sole surgery, or more than 1 surgery. The latter category included women who underwent multiple WLEs to ensure clear margins (surgical margin of less than 2 mm or evidence of microinvasive or invasive carcinoma) and women who converted to mastectomy after WLE. Patients self-reported whether they had a family history of breast cancer.

SDOH Factors

SDOH collected at baseline included age, race, and ethnicity. Race and ethnicity were self-reported. Race categories included American Indian or Alaska Native, Asian, Black, White, multiple races, and not reported or unknown. Given the limited number of American Indian or Alaska Native, Asian, Black, and multiracial individuals, we grouped race into 2 categories (White and non-White). Additional covariates included insurance status and area deprivation index (ADI). The ADI is a composite measure that ranks neighborhoods by socioeconomic disadvantage and includes covariates such as income, educational level, and employment at the census block level.¹⁴ ADI scores range from 1 as the lowest to 100 as the highest level of disadvantage.¹⁴

PRO Questionnaires

We adapted items from the Breast Cancer Surgery Decision Quality Instrument developed by Sepucha et al¹⁵ to examine patient knowledge of DCIS and perception of being informed.¹⁶ Patient knowledge regarding DCIS was measured at the first postoperative visit using 5 multiple-choice knowledge items. Knowledge score represented the proportion of questions answered correctly (range, 0% to 100%). Perception of being informed was collected at presurgery consultation using 1 item, “On a scale from 0 to 10, where 10 means extremely well informed and 0 means not informed at all, how informed did you feel about surgical options for breast cancer?” (range, 0-10). All relevant PROs are contained in eMethods in Supplement 1.

Outcomes

Mental and physical HRQL was assessed at baseline, first postoperative visit, 12 months postsurgery, and 24 months postsurgery using the Patient Reported Outcomes Measurement Information System 10-item questionnaire (PROMIS-10) for global physical and mental health.¹⁷ All PROMIS-10 survey items were assessed using a 5-point Likert scale, except for the average pain rating, which used a 10-point Likert scale. Summed raw scores were converted into physical and mental T scores per PROMIS-10 published criteria (physical T score range, 16.2-67.7; mental T score range, 21.2-67.6),¹⁸ which served as the HRQL outcomes of interest.

Statistical Analysis

This ancillary study conducted longitudinal data analyses to assess the association between prespecified covariates of interest and mental and physical HRQL for women enrolled in E4112. Based on prior research, we defined the minimal important change (MIC) in HRQL as a 2-point difference on the PROMIS-10 T score scale.^18,19 Given the longitudinal nature of the data, associations with covariates were assessed using linear mixed models. Both univariable and multivariable models were fit. Covariate categorization and details on the statistical models are provided in eMethods in Supplement 1.

Data were analyzed from June to November 2024 using SAS version 9.4 software (SAS Institute) and R version 4.3.1 software (R Project for Statistical Computing). All reported P values are 2-sided, with the significance threshold set to .05. Because this ancillary study’s aim was secondary to the main trial, analyses were considered exploratory and hypothesis-generating, so adjustment for multiplicity of inference was not performed.

Results

Participant Characteristics

A total of 296 women were analyzed longitudinally (Figure 1). The median age at enrollment was 60 years (range, 34-87 years), and 147 (50%) reported at least 1 family member with breast cancer (Table 1). Most participants were White (229 [77%]), with the non-White category totaling 67 respondents (23%) (2 American Indian or Alaska Native [3%], 11 Asian [16%], 41 Black [61%], 1 multiple race [1%], 12 not reported or unknown [18%]). The majority of participants were of non-Hispanic ethnicity (280 [95%]), had private insurance (227 [77%]), resided in areas of low-to-moderate deprivation (ADI median, 44 [range, 1-99]), and received a single WLE (185 [63%]). Regarding patient-reported outcomes, women preferred shared decision-making (211 [71%]) and exhibited good patient knowledge of DCIS (median, 80 [range, 20-100]) and a high perception of being informed (median, 10 [range, 3-10]). The mean (SD) for baseline PROMIS-10 mental and physical T scores for the entire cohort were 51.5 (6.7) and 52.0 (8.1), respectively (Table 1).

Table 1. Distributional Summaries for PROMIS-10 Mental and Physical T Scores and Prespecified Covariates.

Covariates	Women, No. (%)
Covariates	Eligible with MRI performed and known final surgery status (n = 339)	HRQL longitudinal analysis set (n = 296)
Social determinants of health covariates
Age, median (range), y	60 (34-87)	60 (34-87)
Race
Non-White^a	77 (23)	67 (23)
White	262 (77)	229 (77)
Ethnicity
Hispanic	21 (6)	16 (5)
Non-Hispanic	318 (94)	280 (95)
Insurance status
Private	261 (77)	227 (77)
Other^b	78 (23)	69 (23)
ADI, median (range)	44 (1-100)	44 (1-99)
Clinical covariates
Family history of breast cancer	163 (48)	147 (50)
Type of surgery
1 WLE	215 (63)	185 (63)
Mastectomy	54 (16)	47 (16)
>1 Surgery	70 (21)	64 (22)
PRO covariates
Patient knowledge, median (range)	NA	80 (20-100)
Perception of being informed, median (range)	NA	10 (3-10)
Cancer worry, median (range)	NA	2.3 (1-4)
Decision autonomy preference
Surgeon	NA	24 (8)
Shared	NA	211 (71)
Patient	NA	61 (21)
Importance of keeping the breast, median (range)	NA	7 (0-10)
Importance of removing the breast, median (range)	NA	5 (0-10)
Importance of avoiding radiation, median (range)	NA	6 (0-10)
Importance of sex life, median (range)	NA	5 (0-10)
HRQL outcomes
PROMIS-10 mental T score, mean (SD)
Baseline	NA	51.5 (6.7)
Postoperative follow-up	NA	51.3 (7.9)
12 months	NA	51.4 (7.5)
24 months	NA	52.0 (7.9)
PROMIS-10 physical T score, mean (SD)
Baseline	NA	52.0 (8.1)
Postoperative follow-up	NA	50.3 (8.1)
12 months	NA	51.4 (8.1)
24 months	NA	51.9 (8.7)

Open in a new tab

Abbreviations: ADI, Area Deprivation Index; HRQL, health-related quality of life; MRI, magnetic resonance imaging; NA, not applicable; PRO, patient-reported outcome; PROMIS-10, 10-Item Patient-Reported Outcomes Measurement Information System; WLE, wide local excision.

^{^a}

Breakdown of non-White race: 2 American Indian/Alaskan Native, 11 Asian, 41 Black, 1 multiple race, 7 not reported, and 5 unknown.

^{^b}

Breakdown of other insurance: 52 Medicare, 7 Medicaid, 2 veterans sponsored, 1 Medicaid and Medicare, 1 military sponsored, 1 self-pay, 1 no means of payment (no insurance), and 4 unknown.

Of the 229 White women in the cohort, 143 (62%) received 1 WLE, 35 (15%) received a mastectomy, and 51 (22%) received more than 1 surgery; of the 67 non-White women in the cohort, 42 (63%) received 1 WLE, 12 (18%) received a mastectomy, and 13 (19%) received more than 1 surgery. Additional information regarding breast reconstruction rates after surgery can be found in eTable 1 in Supplement 1.

Women were scheduled to complete the HRQL assessment at the baseline, first postoperative visit, 12 months postsurgery, and 24 months postsurgery time points. Approximately two-thirds of women had 3 or more assessments available for both the mental (197 of 296 [67%]) and physical T scores (191 of 296 [65%]) (maximum 4). A smaller number of women contributed T score data for only 1 time point (45 of 296 [15%] mental and 49 of 296 [17%] physical) or 2 time points (54 of 296 [18%] mental and 56 of 296 [19%] physical). The median and IQR for the timing of PRO questionnaire completion by time point were: baseline, 21 (13-35) days before surgery; first postoperative visit, 30 (16-61) days after surgery; 12 months postsurgery, 405 (385-535) days; and 24 months postsurgery, 755 (749-780) days. However, there was variability among women as to the timing of the PRO questionnaires. PROMIS-10 physical and mental T scores at all available time points are shown in eFigures 1 and 2 in Supplement 1, respectively. Additional information on data missingness by covariate can be found in eTable 2 in Supplement 1.

Mental Health

The longitudinal trajectory of mental health remained stable for the entire cohort (eFigure 3 in Supplement 1). However, differences in mental health trajectories were detected by SDOH and clinical covariates (Figure 2; Table 2). Being insured with plans other than private insurance was associated with worsening mental health over time compared with private insurance. Women with other types of insurance demonstrated a mean (SE) estimated decline through 24 months of 2.8 points (0.8) on the T score scale (P < .001). Race was also significantly associated with mental health trajectory, with non-White women reporting worsening mental health over time compared with White women, with a decline through 24 months of 2.2 (0.9) points (P = .03). Clinically, women who received more than 1 surgery reported poorer mental health over time compared with women who received 1 WLE, with a decline through 24 months of 1.9 (0.8) points (P = .01). The observed difference in baseline mental health for women who received different types of surgery is due to the nonrandomized nature of E4112; however, the nonparallel longitudinal trajectories for the surgery groups demonstrate this association (Figure 2). When examining the interaction between race and type of surgery, receipt of more than 1 surgery was associated with a greater decline in mental health for non-White women, who exhibited a sustained decline through 24 months of 5.7 (1.7) points. This decline was larger than that of White women with more than 1 surgery, who exhibited a more modest decrement through 24 months of 1.0 (0.9) points, although the P value for this comparison from the univariable model was above the significance threshold (P = .051). However, after adjustment for potential confounders, the longitudinal association was consistent for other insurance (comparison with private insurance: χ² = 8.3 [df, 2]; P = .02) and the comparison between White women and non-White women with more than 1 surgery achieved statistical significance (χ² = 8.4 [df, 2]; P = .02) (eTable 3 in Supplement 1).

Figure 2. — D, White with 1 WLE vs non-White (American Indian or Alaska Native, Asian, Black, multiple races, and not reported or unknown) with >1 surgery, P < .001; White with mastectomy vs non-White with >1 surgery, P = .03; White with >1 surgery vs non-White with >1 surgery, P = .051; non-White with 1 WLE vs non-White with >1 surgery, P = .04. PROMIS-10 indicates 10-item Patient-Reported Outcomes Measurement Information System.

Table 2. Differences in Longitudinal Trends in PROMIS-10 Mental and Physical T Scores by Covariate Level as Estimated by Univariable Linear Mixed Models.

Covariate	Level	Differences in longitudinal trend in the PROMIS-10 mental T score		Differences in longitudinal trend in the PROMIS-10 physical T score
Covariate	Level	Test statistic, χ² (df)^a	P value	Test statistic, χ² (df)^a	P value
Age	Continuous (per 5-y increase)	3.4 (2)	.18	3.5 (2)	.17
Race	White (vs non-White^b)	6.8 (2)	.03	2.4 (2)	.31
Ethnicity	Hispanic (vs non-Hispanic)	0.3 (2)	.88	0.5 (2)	.77
Insurance status	Private (vs other)	14.3 (2)	<.001	14.1 (2)	<.001
ADI (theoretical range 0-100)	Continuous (per 10-percentile increase)	5.3 (2)	.07	3.0 (2)	.22
Family history of breast cancer	Yes (vs no)	0.3 (2)	.85	0.5 (2)	.77
Type of surgery^c	1 WLE vs 1 mastectomy vs >1 surgery	12.3 (4)	.02	2.9 (4)	.58
Patient knowledge (theoretical range 0-100)	Continuous (per 10-unit increase)	2.4 (2)	.31	3.8 (2)	.15
Perception of being informed (theoretical range 0-10)	Continuous (per 1-unit increase)	2.2 (2)	.34	7.4 (2)	.03
Cancer worry (theoretical range 1-4)	Continuous (per 1-unit increase)	0.7 (2)	.70	2.2 (2)	.33
Decision autonomy preference	Shared vs surgeon vs patient	1.9 (4)	.76	2.5 (4)	.64
Importance of keeping the breast (range 0-10)	Continuous (per 1-unit increase)	0.2 (2)	.91	0.9 (2)	.64
Importance of removing breast (theoretical range 0-10)	Continuous (per 1-unit increase)	1.3 (2)	.54	0.1 (2)	.95
Importance of avoiding radiation (theoretical range 0-10)	Continuous (per 1-unit increase)	2.1 (2)	.34	2.1 (2)	.35
Importance of sex life (theoretical range 0-10)	Continuous (per 1-unit increase)	2.8 (2)	.25	5.5 (2)	.07

Open in a new tab

Abbreviations: ADI, Area Deprivation Index; df, degrees of freedom; PROMIS-10, 10-Item Patient-Reported Outcomes Measurement Information System; WLE, wide local excision.

^{^a}

The difference in longitudinal trend was based on a joint significance test for the interaction terms between the covariate and the linear and quadratic terms for time, using the corresponding estimable contrast with appropriate degrees of freedom. A significant P value indicates that the estimated longitudinal curve for the PROMIS-10 mental (or physical) T score differs by covariate level and, thus, that patients at varying levels of the covariate experienced differing longitudinal trends in mental (or physical) health over time.

^{^b}

American Indian or Alaska Native, Asian, Black, multiple races, and not reported or unknown.

^{^c}

The omnibus contrast testing for equality in longitudinal trend across the 3 surgery types was statistically significant. Pairwise contrasts are as follows: 1 WLE vs >1 surgery, P = .01; 1 mastectomy vs >1 surgery, P = .37; 1 WLE vs 1 mastectomy, P = .09.

As an exploratory analysis, the group corresponding to more than 1 surgery (64 participants) was further subdivided into women who received multiple WLEs (53 [83%]) vs women who received a mastectomy after attempted WLE (11 [17%]), with longitudinal trajectories of mental health estimated for each subgroup (eFigure 4 in Supplement 1). Women with multiple WLEs saw a mean (SE) decline of 1.2 (0.8) points through 24 months, and women with a mastectomy following an attempted WLE saw a mean decline of 6.5 (2.1) points through 24 months.

Physical Health

The longitudinal trajectory of physical health remained stable for the entire cohort (eFigure 3 in Supplement 1). However, as with mental health, differences in physical health trajectories were detected by insurance status (Figure 3; Table 2). Women without private insurance reported worsening physical health over time, demonstrating a decline through 24 months of 3.2 (0.9) points on the T score scale (P < .001). In addition, women who reported a lower perception of being informed exhibited a larger improvement in physical health over time. There was no difference in the change in physical health over time by type of surgery. After adjustment for potential confounders, the association between other insurance and reduced physical health over 24 months remained evident (comparison with private insurance: χ² = 10.0 [df, 2]; P = .007) (eTable 4 in Supplement 1).

Discussion

In this longitudinal assessment of mental and physical health domains of HRQL, we found that non-White women with DCIS who received more than 1 surgery experienced a sustained decline in mental health. The magnitude of this decline in the mental health T score of almost 6 points 2 years after surgery was well beyond the MIC of 2. This decline was also notably worse than the corresponding decline of 1 point over 2 years experienced by White women who received more than 1 surgery. Furthermore, women with nonprivate insurance also experienced a minimal important decrement in both mental and physical health. Our study also found that while both physical and mental HRQL were associated with insurance status, physical health was also associated with decision-making factors, such as perception of being informed, while mental health was more associated with race and clinical factors including the number of surgeries.

Previous studies have shown that treatment for DCIS using mastectomy or WLE has comparable impact on physical and mental health 2 years after treatment.²⁰ Our work expands on this to include the potential association of receiving more than 1 surgery, as a 2015 meta-analysis found that repeat breast surgery has been shown to occur in about 22% of patients.²¹ We found that an increased number of surgeries was related to decreased mental health 2 years after treatment among women with DCIS, with an average decline in the PROMIS-10 mental T score (1.9) very close to the MIC of 2. However, although the number of patients is small, our data suggest that the subgroup of patients who received a mastectomy after attempted WLE experienced particularly poor mental health outcomes, with an average decline of more than 6 points. The use of diagnostic technologies such as breast MRI may aid in identifying tumor extent to assist surgical planning and potentially reduce the need for repeat surgeries.²²

In addition, we found that after controlling for age, SDOH (specifically identifying with a minoritized racial group), and having Medicare or Medicaid insurance correlated with worse mental health 2 years after treatment. Our results contrast with 2023 findings that suggest HRQL among women with DCIS did not significantly differ by race, neighborhood deprivation index, or treatment, although these findings were cross-sectional.⁹ Other studies supporting our findings show that private insurance was related to better HRQL,²³ and that racial disparities in HRQL persist across time when examining SEER-Medicare data.²⁴

Our work also assessed factors previously considered important to HRQL, namely decision-making factors, including knowledge and perception of being informed. Of these, only a patient’s perception of being informed was significantly associated with physical health in the early postoperative period. Women who rated their perception of being informed as low before surgery were more likely to experience an improvement in physical health scores. However, we would note that individuals with a low perception of being informed started the study with lower levels of physical health. Given that women with a higher perception of being informed began the study with better physical health, there may be a ceiling effect as to how much improvement these women could experience, especially given that these women had DCIS and not another type of cancer requiring more invasive or extensive treatment. Physical health scores converged 2 years after treatment across all levels of perception of being informed. Previous work has shown that physical health returns to baseline years after treatment.⁸ Our work adds nuance to this by accounting for SDOH, including lack of private insurance, which was associated with a minimal important decrease in physical health. None of the evaluated decision-making factors correlated with mental health.

Strengths and Limitations

Study strengths include longitudinal evaluation of HRQL in a fairly large prospective cohort of women with DCIS and separate assessment of physical and mental health domains. However, there are several limitations to our work. First, our study was not randomized, so findings related to the type of surgery should be interpreted cautiously. Second, the number of non-White women who received more than 1 surgery and the number of women with mastectomy following attempted WLE were both small, so the reported findings of declines in mental health through 24 months well above the MIC for these groups, while suggestive, will require validation in a larger cohort. Rates of certain treatments tend to be related to SDOH and patient-related factors,^25,26 which could impact mental and physical health over and above treatment received. Third, while we controlled for age, we may not have sufficiently disentangled the effect of age, comorbidity, and Medicare or Medicaid insurance status. Fourth, there were fewer survey responses from later time points compared with the number of baseline surveys completed. This is due to long-term attrition, a common barrier in longitudinal work, potentially limiting generalizability. Fifth, our data were collected from 2015 to 2018 and may not capture more recent ecological changes that may currently contribute to HQRL. Nevertheless, we believe that our data continue to reflect the general trajectory, particularly for differences between White and non-White individuals. Furthermore, in the post–COVID-19 era, we believe that there may be wider gaps in HRQL between White and non-White individuals. Lastly, due to limited numbers, we were unable to assess more granular groupings of race and ethnicity beyond the non-White grouping. Future work should recruit a larger, more diverse pool of patients, as these differences may impact HRQL.

Conclusions

The 5-year survival rate for DCIS is exceptionally high; therefore, optimizing HRQL in this population is of paramount importance. Worryingly, mental health trajectories show sustained decrement at 2 years after surgery, particularly for non-White women, women without private insurance, and women who received more than 1 surgery. Future work may focus on a broader set of interventions tailored to at-risk groups who could be more likely to report worse mental health over time following surgery for DCIS.

Supplement 1.

eMethods.

eTable 1. Breast Reconstruction Rates Among Women in the HRQL Longitudinal Analysis Set Who Received a Mastectomy

eTable 2. Amount of Missing Data for Each Prespecified Covariate Among Women in the HRQL Longitudinal Analysis Set

eTable 3. Adjusted Joint Significance Tests for Longitudinal Trends From a Multivariable Linear Mixed Model for the PROMIS-10 Mental T Score

eTable 4. Adjusted Joint Significance Tests for Longitudinal Trends From a Multivariable Linear Mixed Model for the PROMIS-10 Physical T Score

eFigure 1. Scatterplot (With Jittering) of PROMIS-10 Mental T Scores, Including All Available Time Points (T0, T2, T3, T4)

eFigure 2. Scatterplot (With Jittering) of PROMIS-10 Physical T Scores, Including All Available Time Points (T0, T2, T3, T4)

eFigure 3. Model-Based Longitudinal Trajectories of the PROMIS-10 Mental and Physical T Score for the Entire Study Cohort Based on Estimated Marginal Mean Values

eFigure 4. Model-Based Longitudinal Trajectories of the PROMIS-10 Mental T Score by Type of Surgery Based on Estimated Marginal Mean Values

eReferences.

jamanetwopen-e2518887-s001.pdf^{(1.1MB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2518887-s002.pdf^{(15.4KB, pdf)}

References

1.Tomlinson-Hansen S, Khan M, Cassaro S. Breast Ductal Carcinoma In Situ. StatPearls Publishing; 2023. Updated January 2025. Accessed December 27, 2024. https://www.ncbi.nlm.nih.gov/books/NBK567766/ [PubMed] [Google Scholar]
2.American Cancer Society . Breast Cancer Facts & Figures 2022-2024. Published online 2022. Accessed December 27, 2024. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/breast-cancer-facts-and-figures/2022-2024-breast-cancer-fact-figures-acs.pdf
3.National Cancer Institute . SEER*Explorer: An interactive website for SEER cancer statistics. SEER Incidence Data, November 2022 Submission (1975-2020), SEER 22 registries (excluding Illinois and Massachusetts). Updated November 16, 2023. Accessed April 10, 2024. https://seer.cancer.gov/statistics-network/explorer/
4.Lesurf R, Aure MR, Haberg Mørk H, et al. Molecular features of subtype-specific progression from ductal carcinoma in situ to invasive breast cancer. Cell Reports no. 16. Published online July 26, 2016. Accessed December 27, 2024. https://www.cell.com/cell-reports/pdf/S2211-1247(16)30805-1.pdf [DOI] [PubMed]
5.Worni M, Akushevich I, Greenup R, et al. Trends in treatment patterns and outcomes for ductal carcinoma in situ. J Natl Cancer Inst. 2015;107(12):djv263. doi: 10.1093/jnci/djv263 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.National Comprehensive Cancer Network . NCCN guidelines for patients: ductal carcinoma in situ. Updated January 31, 2025. Accessed December 27, 2024. https://www.nccn.org/patientresources/patient-resources/guidelines-for-patients/guidelines-for-patients-details?patientGuidelineId=30
7.Fazeli S, Snyder BS, Gareen IF, et al. Association between surgery preference and receipt in ductal carcinoma in situ after breast magnetic resonance imaging: an ancillary study of the ECOG-ACRIN Cancer Research Group (E4112). JAMA Netw Open. 2022;5(5):e2210331. doi: 10.1001/jamanetworkopen.2022.10331 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Dominici LS, Rosenberg SM. Ductal carcinoma in situ (DCIS): the importance of patient-reported outcomes (PRO). Curr Breast Cancer Rep. 2020;12(2):90-97. doi: 10.1007/s12609-020-00363-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Kim H, Wang H, Demanelis K, et al. Factors associated with ductal carcinoma in situ (DCIS) treatment patterns and patient-reported outcomes across a large integrated health network. Breast Cancer Res Treat. 2023;197(3):683-692. doi: 10.1007/s10549-022-06831-w [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Kashaf MS, McGill E. Does shared decision making in cancer treatment improve quality of life? A systematic literature review. Med Decis Making. 2015;35(8):1037-1048. doi: 10.1177/0272989X15598529 [DOI] [PubMed] [Google Scholar]
11.Wagner LI, Gray RJ, Sparano JA, et al. Patient-reported cognitive impairment among women with early breast cancer randomly assigned to endocrine therapy alone versus chemoendocrine therapy: results from TAILORx. J Clin Oncol. 2020;38(17):1875-1886. doi: 10.1200/JCO.19.01866 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Tran TXM, Park J, Lee J, Jung YS, Chang Y, Cho H. Utility of the Patient-Reported Outcomes Measurement Information System (PROMIS) to measure primary health outcomes in cancer patients: a systematic review. Support Care Cancer. 2021;29(4):1723-1739. doi: 10.1007/s00520-020-05801-6 [DOI] [PubMed] [Google Scholar]
13.Lehman CD, Gatsonis C, Romanoff J, et al. Association of magnetic resonance imaging and a 12-gene expression assay with breast ductal carcinoma in situ treatment. JAMA Oncol. 2019;5(7):1036-1042. doi: 10.1001/jamaoncol.2018.6269 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Kind AJH, Buckingham WR. Making neighborhood-disadvantage metrics accessible—the neighborhood atlas. N Engl J Med. 2018;378(26):2456-2458. doi: 10.1056/NEJMp1802313 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Sepucha KR, Belkora JK, Chang Y, et al. Measuring decision quality: psychometric evaluation of a new instrument for breast cancer surgery. BMC Med Inform Decis Mak. 2012;12:51. doi: 10.1186/1472-6947-12-51 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.MGH Health Decision Sciences Center . Decision quality instruments. Accessed May 13, 2020. https://mghdecisionsciences.org/tools-training/decision-quality-instruments/
17.Hays RD, Bjorner JB, Revicki DA, Spritzer KL, Cella D. Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items. Qual Life Res. 2009;18(7):873-880. doi: 10.1007/s11136-009-9496-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.HealthMeasures . Meaningful Change. Accessed December 27, 2024. healthmeasures.net
19.Terwee CB, Peipert JD, Chapman R, et al. Minimal important change (MIC): a conceptual clarification and systematic review of MIC estimates of PROMIS measures. Qual Life Res. 2021;30(10):2729-2754. doi: 10.1007/s11136-021-02925-y [DOI] [PMC free article] [PubMed] [Google Scholar]
20.King MT, Winters ZE, Olivotto IA, et al. Patient-reported outcomes in ductal carcinoma in situ: a systematic review. Eur J Cancer. 2017;71:95-108. doi: 10.1016/j.ejca.2016.09.035 [DOI] [PubMed] [Google Scholar]
21.Fancellu A, Turner RM, Dixon JM, Pinna A, Cottu P, Houssami N. Meta-analysis of the effect of preoperative breast MRI on the surgical management of ductal carcinoma in situ. Br J Surg. 2015;102(8):883-893. doi: 10.1002/bjs.9797 [DOI] [PubMed] [Google Scholar]
22.Yabroff KR, Reeder-Hayes K, Zhao J, et al. Health insurance coverage disruptions and cancer care and outcomes: systematic review of published research. J Natl Cancer Inst. 2020;112(7):671-687. doi: 10.1093/jnci/djaa048 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Rincon MA, Smith AW, Yu M, Kent EE. Trends in racial/ethnic disparity of health-related quality of life in older adults with and without cancer (1998-2012). Cancer Epidemiol Biomarkers Prev. 2020;29(6):1188-1195. doi: 10.1158/1055-9965.EPI-19-0819 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Mendes, IS, da Silveira Souza, GA, das Neves MF, et al. Impact of mastectomy and breast-conserving surgery on quality of life of women after breast cancer. OMundo Saúde São Paulo. 2017;41(4):703-710. doi: 10.15343/0104-7809.20174104703710 [DOI] [Google Scholar]
25.Gumus M, Ustaalioglu BO, Garip M, et al. Factors that affect patients’ decision-making about mastectomy or breast conserving surgery, and the psychological effect of this choice on breast cancer patients. Breast Care (Basel). 2010;5(3):164-168. doi: 10.1159/000314266 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Byng D, Retèl VP, Engelhardt EG, et al. ; On Behalf Of The Grand Challenge Precision Consortium . Preferences of treatment strategies among women with low-risk DCIS and oncologists. Cancers (Basel). 2021;13(16):3962. doi: 10.3390/cancers13163962 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials