Intended for healthcare professionals

  1. Research
  2. Association of DCIS...
  3. Association of DCIS size and margin status with risk of developing breast cancer post-treatment: multinational, pooled cohort study
CCBYNC Open access
Research

Association of DCIS size and margin status with risk of developing breast cancer post-treatment: multinational, pooled cohort study

BMJ 2023; 383 doi: https://doi.org/10.1136/bmj-2023-076022 (Published 30 October 2023) Cite this as: BMJ 2023;383:e076022
  1. Renée S J M Schmitz, doctoral student1,
  2. Alexandra W van den Belt-Dusebout, postdoctoral researcher1,
  3. Karen Clements, project manager2,
  4. Yi Ren, biostatistician3,
  5. Chiara Cresta, master student1,
  6. Jasmine Timbres, doctoral student4,
  7. Yat-Hee Liu, data manager1,
  8. Danalyn Byng, doctoral student5,
  9. Thomas Lynch, senior researcher6,
  10. Brian A Menegaz, data manager7,
  11. Deborah Collyar, patient author8,
  12. Terry Hyslop, adjunct professor3,
  13. Samantha Thomas, biostatistician3,
  14. Jason K Love, data manager9,
  15. Michael Schaapveld, associate staff scientist10,
  16. Proteeti Bhattacharjee, project manager1,
  17. Marc D Ryser, assistant professor511,
  18. Elinor Sawyer, professor4,
  19. E Shelley Hwang, professor6,
  20. Alastair Thompson, professor7,
  21. Jelle Wesseling, professor11213,
  22. Esther H Lips, associate staff scientist1,
  23. Marjanka K Schmidt, professor114
  24. on behalf of the Grand Challenge PRECISION consortium
    1. 1Division of Molecular Pathology, Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, 1066 Amsterdam, Netherlands
    2. 2National Health Service England, Birmingham, UK
    3. 3Department of Biostatistics and Bioinformatics, Biostatistics Shared Resource Duke Cancer Institute, Durham, NC, USA
    4. 4School of Cancer and Pharmaceutical Science, King’s College London, London, UK
    5. 5Department of Population Health Sciences, Duke University Medical Center, Durham, NC, USA
    6. 6Department of Surgery, Duke Cancer Institute, Durham, NC, USA
    7. 7Department of Surgical Oncology, Baylor College of Medicine, Houston, TX, USA
    8. 8Patient Advocates in Research, Danville, CA, USA
    9. 9Department of Breast Surgical Oncology, MD Anderson Cancer Center, Houston, TX, USA
    10. 10Division of Psycho-oncology and Epidemiology, Netherlands Cancer Institute– Antoni van Leeuwenhoek Hospital, Amsterdam, Netherlands
    11. 11Department of Mathematics, Duke University, Durham, NC, USA
    12. 12Division of Diagnostic Oncology, Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Amsterdam, Netherlands
    13. 13Department of Pathology, Leiden University Medical Centre, Leiden, Netherlands
    14. 14Department of Clinical Genetics, Leiden University Medical Centre, Leiden, Netherlands
    1. Correspondence to: M K Schmidt mk.schmidt{at}nki.nl
    • Accepted 27 September 2023

    Abstract

    Objective To examine the association between size and margin status of ductal carcinoma in situ (DCIS) and risk of developing ipsilateral invasive breast cancer and ipsilateral DCIS after treatment, and stage and subtype of ipsilateral invasive breast cancer.

    Design Multinational, pooled cohort study.

    Setting Four large international cohorts.

    Participants Patient level data on 47 695 women with a diagnosis of pure, primary DCIS between 1999 and 2017 in the Netherlands, UK, and US who underwent surgery, either breast conserving or mastectomy, often followed by radiotherapy or endocrine treatment, or both.

    Main outcome measures The main outcomes were 10 year cumulative incidence of ipsilateral invasive breast cancer and ipsilateral DCIS estimated in relation to DCIS size and margin status, and adjusted hazard ratios and 95% confidence intervals, estimated using multivariable Cox proportional hazards analyses with multiple imputed data

    Results The 10 year cumulative incidence of ipsilateral invasive breast cancer was 3.2%. In women who underwent breast conserving surgery with or without radiotherapy, only adjusted risks for ipsilateral DCIS were significantly increased for larger DCIS (20-49 mm) compared with DCIS <20 mm (hazard ratio 1.38, 95% confidence interval 1.11 to 1.72). Risks for both ipsilateral invasive breast cancer and ipsilateral DCIS were significantly higher with involved compared with clear margins (invasive breast cancer 1.40, 1.07 to 1.83; DCIS 1.39, 1.04 to 1.87). Use of adjuvant endocrine treatment was not significantly associated with a lower risk of ipsilateral invasive breast cancer compared to treatment with breast conserving surgery only (0.86, 0.62 to 1.21). In women who received breast conserving treatment with or without radiotherapy, higher DCIS grade was not significantly associated with ipsilateral invasive breast cancer, only with a higher risk of ipsilateral DCIS (grade 1: 1.42, 1.08 to 1.87; grade 3: 2.17, 1.66 to 2.83). Higher age at diagnosis was associated with lower risk (per year) of ipsilateral DCIS (0.98, 0.97 to 0.99) but not ipsilateral invasive breast cancer (1.00, 0.99 to 1.00). Women with large DCIS (≥50 mm) more often developed stage III and IV ipsilateral invasive breast cancer compared to women with DCIS <20 mm. No such association was found between involved margins and higher stage of ipsilateral invasive breast cancer. Associations between larger DCIS and hormone receptor negative and human epidermal growth factor receptor 2 positive ipsilateral invasive breast cancer and involved margins and hormone receptor negative ipsilateral invasive breast cancer were found.

    Conclusions The association of DCIS size and margin status with ipsilateral invasive breast cancer and ipsilateral DCIS was small. When these two factors were added to other known risk factors in multivariable models, clinicopathological risk factors alone were found to be limited in discriminating between low and high risk DCIS.

    Introduction

    With the introduction of population based breast cancer screening1 and the more recent introduction of digital mammography,23 the incidence of ductal carcinoma in situ (DCIS) has increased more than fivefold in western countries.456 Currently, DCIS accounts for 20-25% of all newly detected breast cancers.7 Although DCIS is a potential precursor of invasive breast cancer, it is estimated that as much as 80% of DCIS lesions will never progress to invasive disease.891011 As potentially progressive, high risk DCIS is difficult to distinguish from potentially non-progressive, low risk DCIS, almost all women with DCIS are treated by mastectomy or breast conserving surgery followed by radiotherapy or endocrine treatment, or both.12 Yet no clear reduction of breast cancer mortality has been attributed to DCIS treatment.13 Furthermore, the incidence of invasive breast cancer has not decreased,4 suggesting a considerable reservoir of DCIS lesions exists that would never have progressed to invasive breast cancer. Consequently, concern is growing about the possible overtreatment of DCIS.48141516 The PRECISION (PREvent ductal Carcinoma In Situ Invasive Overtreatment Now) Cancer Grand Challenge Consortium aims to distinguish high risk from low risk DCIS to reduce the overtreatment of low risk DCIS.1417

    Some clinicopathological features such as high grade lesions and young age are associated with increased risk of subsequent ipsilateral DCIS and ipsilateral invasive breast cancer.1819 For other features, such as DCIS size and margin status, the association with risk of these two cancer types is less clear, and several relatively small studies (up to 2995 women with DCIS treated with breast conserving surgery only) have reported size to be associated with a higher risk.20212223 In contrast, a UK study of 24 779 women treated with breast conserving surgery and mastectomy for DCIS reported no significant association between increasing DCIS size and ipsilateral invasive breast cancer.19 Similarly, the literature has not reached consensus on the effect of DCIS margin status on risk of ipsilateral invasive breast cancer or ipsilateral DCIS. The UK study reported involved margins as an important risk factor for ipsilateral invasive breast cancer19 as did other studies.2425 Several smaller studies, however, did not find any associations between margin status and local recurrence,222627 or only did so in relation to ipsilateral DCIS.28 Consequently, substantially larger studies with longer term follow-up are needed to validate the association between DCIS size and involved margins with risk of subsequent ipsilateral DCIS or ipsilateral invasive breast cancer. A better understanding of these and other related risk factors could provide insight as to which women may be candidates for de-escalation of treatment, allowing for more personalised management strategies and avoiding overtreatment of women with low risk DCIS.

    Using a large, multinational, pooled cohort we investigated whether DCIS size and margin status are associated with risk of developing ipsilateral invasive breast cancer and ipsilateral DCIS. A secondary aim was to investigate whether DCIS size and margin status are associated with the subtype and stage of subsequent ipsilateral invasive breast cancer.

    Methods

    Study population

    We analysed four cohorts of women with DCIS from the Netherlands, UK, and US as part of the PRECISION DCIS Consortium. Table 1 and the supplementary methods describe the cohorts.

    Table 1

    Patient and cohort characteristics. Values are number (percentage) unless stated otherwise

    View this table:

    Data on participants’ characteristics, DCIS status, treatment, and follow-up had been collected for each of the four cohorts. For this study, we recoded and harmonised original data following a set codebook. Inclusion criteria were women older than 18 years with pure, primary, unilateral DCIS who had undergone surgery (with or without adjuvant treatment) between 1999 and 2017. We excluded women with DCIS accompanying micro-invasive or invasive disease or Paget’s disease, those who received chemotherapy or targeted therapy, those with missing information on treatment or follow-up, and those with follow-up shorter than six months or diagnosis with subsequent invasive breast cancer within six months after the primary DCIS diagnosis. As endocrine treatment is rarely prescribed for DCIS in the Netherlands, we excluded 148 women in the Dutch cohort who received endocrine treatment (supplementary figure S1).

    Definitions

    Contralateral and ipsilateral invasive breast cancer and ipsilateral DCIS were defined as an invasive breast cancer or DCIS diagnosed at least six months after the primary diagnosis of DCIS. Primary metastatic disease (n=32) without occurrence of contralateral breast cancer, and solitary axillary lesions (n=4) without an in-breast recurrence, were classified as ipsilateral invasive breast cancer. For 53 subsequent events, the laterality or invasive status of the event was unknown. For these women, follow-up was censored at time of the occurrence of the subsequent event. Use of endocrine treatment was defined as any such treatment initiated within six months after the primary diagnosis and, if known, a minimum duration of one year. Margin status was defined as the closest surgical margin width according to current treatment guidelines29; clear margins ≥2 mm and involved margins <2 mm margin width after final surgery (including re-resection). DCIS was divided into three size categories: <20 mm, 20-49 mm, and ≥50 mm. For multifocal disease, DCIS size was categorised on the basis of the largest diameter reported.

    Stage of subsequent event was defined as 0 (DCIS) to IV according to the eighth edition of the anatomic TNM staging criteria.30 For most subsequent ipsilateral invasive breast cancers, the number and location of positive lymph nodes was not available. As stages IIa-b and IIIa therefore could not be accurately distinguished, T0-2 N+ was defined as stage II. We categorised ipsilateral invasive breast cancers into four subtypes based on hormone receptor status (positive if oestrogen receptor positive or progesterone receptor positive; negative otherwise) and human epidermal growth factor receptor 2 (HER2) status: hormone receptor positive and HER2 negative, hormone receptor positive and HER2 positive, hormone receptor negative and HER2 positive, and hormone receptor negative and HER2 negative.

    Statistical analyses

    We harmonised and pooled the patient level data of the four cohorts. Time at risk of developing ipsilateral DCIS or ipsilateral invasive breast cancer started six months after the date of the primary DCIS diagnosis and ended at date of diagnosis of the subsequent event (ipsilateral DCIS or ipsilateral invasive breast cancer), death, diagnosis of a competing risk (death by any cause, diagnosis of contralateral invasive breast cancer and of ipsilateral DCIS when studying ipsilateral invasive breast cancer, or of ipsilateral invasive breast cancer when studying ipsilateral DCIS), or end of follow-up. Follow-up duration was right censored at 10 years.

    We used multiple imputation by chained equations to account for missing values for DCIS size (28%), margin status (14%), and grade (6%).3132 The imputation model employed an ordered logistic regression for DCIS size and grade and a nominal logistic regression for margin status.33 The multiple imputation model included the variables age at diagnosis, year of diagnosis, cohort, and treatment type. The imputation process was repeated 40 times. We combined results using Rubin’s rule.34 Imputed results are shown as the primary analyses31 (also see the supplementary material for analyses of the dataset including categories of missing information).

    Cumulative incidences of ipsilateral invasive breast cancer and ipsilateral DCIS were estimated considering the competing risks and grouped by tumour size, margin status after final surgery, and treatment.

    We used multivariable Cox proportional hazards models per treatment group in all pooled data and by cohort to estimate the association between DCIS size or margin status and the risk of subsequent ipsilateral invasive breast cancer or ipsilateral DCIS. Schoenfeld residuals and graphical methods were used to test the proportional hazards assumptions. Effect modification of the association between DCIS size or margin status and risk of ipsilateral invasive breast cancer or ipsilateral DCIS by other established risk factors of ipsilateral invasive breast cancer, such as age at diagnosis and DCIS grade, was examined. Data are reported as hazard ratios and 95% confidence intervals.

    Selection of variables to be evaluated for potential confounding was based on existing literature (ie, age at diagnosis, DCIS grade, treatment type, and year of diagnosis). These potentially relevant covariates were first evaluated individually then added to the model using forward selection. A covariate was considered a confounder when the difference between hazard ratios from the model with the covariate and the model without the covariate exceeded 10%. By this criterion, the only confounder identified was treatment type in the association between DCIS size and ipsilateral invasive breast cancer or ipsilateral DCIS. To ensure the results were comparable to those in the literature, however, variables included in the multivariable Cox regression analyses were DCIS size, margin status, age at diagnosis, DCIS grade, and treatment type. Furthermore, analyses were stratified by cohort. To assess whether the Dutch cohort, in which endocrine treatment was not prescribed, or hormone receptor negative DCIS influenced hazard ratios of ipsilateral invasive breast cancer and ipsilateral DCIS by treatment type, we performed two subgroup analyses: a cohort restricted to the UK and US based cohorts, and a cohort restricted to the UK and US based cohorts in women with oestrogen receptor positive DCIS.

    For the full model we calculated Harrell’s C index to assess the potential clinical usefulness of all clinical variables, despite the model being over-fitted and the study not specifically designed to develop a prediction model.

    We estimated cumulative relative incidences to compare the incidence of stage and subtype of ipsilateral invasive breast cancer after DCIS compared with age and calendar year specific expected stages and subtypes of primary invasive breast cancer in the Dutch female population. Joint Cox proportional hazards models adjusted for treatment were used to estimate the association between DCIS size or margin status and risk of subtype specific or stage specific ipsilateral invasive breast cancer.35 These analyses were carried out only in women with available information on stage or subtype of the ipsilateral invasive breast cancer.

    All tests of statistical significance were two sided. A P value of <0.05 was considered statistically significant. All analyses were performed using STATA/SE 15.0 (StataCorp, College Station, TX) or open-source software R version 4.04.

    Patient and public involvement

    The PRECISION patient advocates were involved in setting the research questions, interpreting the data, and writing up the results. All PRECISION patient advocates received the final draft of the manuscript and will be actively involved in the dissemination of the results to relevant patient communities.

    Results

    In total, 47 695 women were included in the analyses (supplementary figure S1). Table 1 presents the characteristics of the participants and lesions. Age at diagnosis ranged from 20 to 97 years (mean 59 years). Mean follow-up duration was 6.7 years. Fifteen per cent of participants received breast conserving surgery only, 36% breast conserving surgery with radiotherapy, 3% breast conserving surgery with endocrine treatment, and 14% breast conserving surgery with radiotherapy and endocrine treatment, and 32% underwent mastectomy. DCIS size was <20 mm in 46% of women, 20-49 mm in 19%, ≥50 mm in 7%, and missing in 28%. Margin status was clear (≥2 mm) in 81% of women, involved (<2 mm) in 5%, and missing in 14%. The 10 year cumulative incidence of ipsilateral invasive breast cancer and ipsilateral DCIS was 3.2% and 1.8%, respectively. The characteristics of participants with missing data were largely the same as those without missing data (supplementary table S1), although those with missing data more frequently received a diagnosis at a younger age and those with missing data on margin status more often had a mastectomy.

    DCIS size

    The 10 year cumulative incidence of ipsilateral invasive breast cancer according to DCIS size was 3.5% for <20 mm, 3% for 20-49 mm, and 2.7% for ≥50 mm. The corresponding values for ipsilateral DCIS were 1.9%, 2.0%, and 0.9% (fig 1). Cumulative incidences differed by treatment group (see supplementary figure 2 for cumulative incidences by DCIS size category) and were highest in women receiving breast conserving surgery only, slightly reduced in women receiving breast conserving surgery with radiotherapy, and lowest in women undergoing mastectomy (supplementary figure S2).

    Fig 1
    Fig 1

    Ten year cumulative incidence of subsequent iIBC and iDCIS. Competing risks were death by any cause, contralateral breast cancer, and, depending on analyses, iDCIS or iIBC. Cumulative incidences in relation to DCIS size were calculated in the full, pooled cohort (n=47 695). Cumulative incidences by DCIS margin status were calculated in the full, pooled cohort, after exclusion of women who underwent mastectomy (n=37 721). DCIS=ductal carcinoma in situ; iIBC=ipsilateral invasive breast cancer; iDCIS=ipsilateral ductal carcinoma in situ

    In multivariable Cox regression analyses adjusted for age, DCIS grade and margin status, and treatment, the risk of ipsilateral invasive breast cancer was not significantly increased in women receiving breast conserving surgery with or without radiotherapy for larger DCIS compared with DCIS <20 mm: the hazard ratio for 20-49 mm was 1.13 (95% confidence interval 0.94 to 1.37) and for ≥50 mm was 0.91 (0.54 to 1.54) (table 2). In women who received breast conserving surgery and radiotherapy, DCIS 20-49 mm was associated with an increased risk for ipsilateral invasive breast cancer compared with <20 mm, although this was not statistically significant (1.24, 0.99 to 1.56). In women who received breast conserving surgery with or without radiotherapy, the risk of ipsilateral DCIS was higher in those with DCIS 20-49 mm (1.38, 1.11 to 1.72) compared with <20 mm (table 2). In women who underwent mastectomy, the risks for ipsilateral invasive breast cancer and ipsilateral DCIS were increased with larger DCIS size, although this was not statistically significant (table 3). Supplementary tables S2 and S3 show the results of analyses within each cohort. Risks for ipsilateral invasive breast cancer and ipsilateral DCIS by size category were largely similar in sensitivity analyses on the non-imputed data, including missing information as a separate category (supplementary tables S4-S11).

    Table 2

    Risk of ipsilateral ductal carcinoma in situ or ipsilateral invasive breast cancer in women who received breast conserving surgery, using multivariable Cox proportional hazards analyses with multiple imputed data.* Data are hazard ratio (95% confidence interval); P value unless stated otherwise

    View this table:
    Table 3

    Risk of subsequent ipsilateral ductal carcinoma in situ or ipsilateral invasive breast cancer in all participants and in women who underwent mastectomy, using multivariable Cox proportional hazards analyses with multiple imputed data.* Data are hazard ratio (95% confidence interval); P value unless stated otherwise

    View this table:

    DCIS margin status

    The 10 year cumulative incidence of ipsilateral invasive breast cancer was higher in participants with DCIS with involved margins (5.8%) compared with DCIS with clear margins (3.9%). The corresponding values for ipsilateral DCIS were 4.5% and 2.5% (fig 1).

    In multivariable Cox regression analyses, adjusted risks for ipsilateral invasive breast cancer and ipsilateral DCIS were significantly higher for DCIS with involved margins compared with DCIS with clear margins in women who received breast conserving surgery with or without radiotherapy (ipsilateral invasive breast cancer 1.40, 1.07 to 1.83, and ipsilateral DCIS 1.39, 1.04 to 1.87) (table 2). In analyses that included women who underwent mastectomy, the risks for ipsilateral invasive breast cancer and ipsilateral DCIS did not increase significantly in women with involved margins (table 3). Supplementary tables S2 and S3 show the results of the analyses within each cohort. Risks for ipsilateral invasive breast cancer and ipsilateral DCIS by margin status were largely similar in sensitivity analyses on the non-imputed data, including missing information as a separate category (supplementary tables S4-S11).

    Other risk factors

    In stratified multivariable Cox regression analyses with multiple imputation data, higher grade of DCIS was not significantly associated with ipsilateral invasive breast cancer in any treatment group but was associated with a higher risk of ipsilateral DCIS in women who received breast conserving surgery with or without radiotherapy (grade 2: 1.42, 1.08 to 1.87; grade 3: 2.17, 1.66 to 2.83) (table 2). Use of endocrine treatment in addition to breast conserving surgery was not significantly associated with a lower risk of ipsilateral invasive breast cancer compared to treatment with breast conserving surgery only (0.86, 0.62 to 1.21). Use of endocrine treatment in addition to breast conserving surgery, however, was associated with a lower risk of ipsilateral DCIS compared with breast conserving surgery only (0.50, 0.30 to 0.82) (table 2). In subgroup analyses excluding the Dutch cohort, the hazard ratio was 1.07 (0.76 to 1.53) for ipsilateral invasive breast cancer in women who received breast conserving surgery with endocrine treatment compared with women who received breast conserving surgery only (supplementary table S12). Subgroup analyses excluding both Dutch women and women with oestrogen receptor negative DCIS showed a risk for ipsilateral invasive breast cancer of 1.03 (0.69 to 1.52) (supplementary table S13). Harrell’s C index of the multiple imputed, multivariable Cox regression models including all participants were 0.64 for ipsilateral invasive breast cancer and 0.74 for ipsilateral DCIS.

    Stage of ipsilateral invasive breast cancer

    Cumulative relative incidences, estimating observed stage of ipsilateral invasive breast cancer in the joint PRECISION cohort versus expected stages of invasive breast cancer in the Dutch population ranged between 0.99 and 1.01 for all stages (supplementary table S14), showing no difference in the stage of subsequent ipsilateral invasive breast cancer in women treated for DCIS compared with primary breast cancer in the Dutch female population. In multivariable joint Cox regression analyses, risks for stage III and IV ipsilateral invasive breast cancer were increased in women with DCIS ≥50 mm compared with <20 mm, (stage III: 3.93, 1.80 to 8.57; stage IV: 3.56, 1.84 to 6.91) (table 4). In contrast, no association was found between margin status and stage of ipsilateral invasive breast cancer.

    Table 4

    Risk of stage of subsequent event in multivariable joint Cox proportional hazards analyses* using multiple imputed data.† Data are hazard ratio (95% confidence interval); P value unless stated otherwise

    View this table:

    Subtype of ipsilateral invasive breast cancer

    Cumulative relative incidences for subtype of ipsilateral invasive breast cancer in the joint PRECISION cohort compared with subtypes of invasive breast cancer in the Dutch population ranged between 0.99 and 1.01 for all tumour subtypes (supplementary table S14), showing no difference in the subtype of subsequent ipsilateral invasive breast cancer in women treated for DCIS compared with primary breast cancer in the Dutch female population. In multivariable joint Cox regression analyses, the risk of hormone receptor negative and HER2 positive ipsilateral invasive breast cancer associated with DCIS >50 mm was significantly increased (2.64, 1.33 to 5.21) compared with <20 mm (table 5). Similarly, the risks were increased for hormone receptor negative and HER2 positive ipsilateral invasive breast cancer (2.21, 1.06 to 4.62) and for hormone receptor negative and HER2 negative ipsilateral invasive breast cancer (1.91, 1.11 to 3.26) after DCIS with involved margins compared to DCIS with clear margins.

    Table 5

    Risk of subtype of subsequent invasive breast cancer in multivariate joint Cox proportional hazards analyses* using multiple imputed data.† Data are hazard ratio (95% confidence interval); P value unless stated otherwise

    View this table:

    Discussion

    In this large, multinational, pooled analysis we studied the association between DCIS size and margin status with risk of developing ipsilateral invasive breast cancer and ipsilateral DCIS. We identified involved margins as a risk factor for both subsequent ipsilateral invasive breast cancer and subsequent ipsilateral DCIS in women receiving breast conserving surgery with or without radiotherapy for DCIS. Larger DCIS was associated with higher risk of ipsilateral DCIS. Similarly, a slightly increased risk of ipsilateral invasive breast cancer was found in women with larger DCIS, although this finding was not statistically significant. The potential clinical value of these clinical variables was shown to be moderate at best (Harrell’s C indices 0.64 for ipsilateral invasive breast cancer and 0.74 for ipsilateral DCIS). However, women with a large DCIS (≥50 mm) were more likely to develop stage III and IV ipsilateral invasive breast cancer compared to women with the smallest DCIS (<20 mm). No such association was found between involved margin status and higher stage of ipsilateral invasive breast cancer. Additional analyses on DCIS size or margin status and subtype of ipsilateral invasive breast cancer showed an association between larger DCIS and hormone receptor negative and HER2 positive ipsilateral invasive breast cancer and involved margin status, and hormone receptor negative ipsilateral invasive breast cancer.

    Strengths and limitations of this study

    Our study used large previously assembled cohorts to assess risk factors for the development of subsequent ipsilateral invasive breast cancer and ipsilateral DCIS after a primary diagnosis of DCIS. The use of multinational, patient level, registry data implies that results are widely applicable. We reported on associations between DCIS margin status and stage and subtype of subsequent ipsilateral invasive breast cancer, whereas to date, only one case series comprising 32 patients with ipsilateral invasive breast cancer or ipsilateral DCIS after treatment for DCIS has reported on stage of ipsilateral invasive breast cancer, in which only stage I and II ipsilateral invasive breast cancers were observed.36 Even in the current large cohort, the numbers of events in these subcategories were small; interpretation of these results should therefore be undertaken with care.

    This study also has some limitations. Firstly, the Netherlands Cancer Registry only recently registered data on margin status and DCIS size in the Dutch cohort. For earlier years, however, these data were manually extracted from 12 000 pathology reports provided by the Dutch Nationwide Pathology Databank: Palga,37 by two trained researchers (RSJMS and CC) independently, in a standardised manner specifically for this study. Any disagreements were resolved through consensus or by consulting a pathologist (JW). Secondly, as this study mainly used registry data that differed across cohorts, we could only use broad categories of DCIS size and margin status. We found heterogeneity in the categorisation of DCIS size in existing literature, making comparison between studies difficult. To introduce an objective categorisation, we determined size categories in line with the TNM criteria for invasive breast cancer.30 Thirdly, several values were missing in the data, with as much as 28% missing information on DCIS size. It was not possible to retrospectively collect this information. The characteristics of patients with missing data, compared to patients without missing data, showed that these patients were more often from the Dutch cohort and received a diagnosis in earlier years (supplementary table S1). Much of the missing information on margin status in the UK (Sloane) cohort was in those treated with mastectomy as it was not required to report margin status in women who had undergone mastectomy. Including those with missing data in an unknown category (although an imperfect approach compared with multiple imputation) did not alter the outcomes of this study using multiple imputation (supplementary tables S3-S10), suggesting that the missing data did not have an important influence on the study outcomes. Fourthly, baseline hazards might differ between the cohorts and countries. The results were not, however, altered when applying stratification by cohort (applied in the Cox regression models), again suggesting this did not influence the outcomes of this study. Results were largely similar among cohorts (supplementary tables S2 and S3), although the hazard ratio for margin status was not available for the US (MD Anderson Cancer Center) cohort owing to a lack of power, with only 47 occurrences of ipsilateral invasive breast cancer and 34 of ipsilateral DCIS. Furthermore, as these were registry data, some variables that potentially could be relevant were not available, such as detailed information on nodal status of the ipsilateral invasive breast cancer, comorbidities, menopausal status, hormone replacement therapy, and screening status.38 In future, such data will become available through new prospective DCIS registry cohorts or ongoing DCIS trials: Comparison of Operative versus Monitoring and Endocrine Therapy,39 Low Risk DCIS Trial (LORIS40 and LORD)41 trials. Lastly, although this cohort was large, the 10 year incidence of ipsilateral invasive breast cancer (n=981) and ipsilateral DCIS (n=658) were relatively low. Although this finding emphasises the benign nature of DCIS, it did lead to small number events in subgroup analyses, especially in patients who underwent mastectomy for DCIS. Therefore, these results should be interpreted with caution.

    Comparison with other studies

    Our results on DCIS size are consistent with a multi‐institutional nested case-control study including 939 patients with DCIS that found an association between larger DCIS and an increased risk of ipsilateral DCIS but not ipsilateral invasive breast cancer.23 However, a 15 year update of the National Surgical Adjuvant Breast and Bowel Project (NSABP) B‐17 (breast conserving surgery with or without radiotherapy) and B‐24 (breast conserving surgery with or without endocrine treatment) randomised controlled trials among 2622 women with DCIS reported no such association.26 These trials also studied the effect of margin status and concluded there was an association between involved margins and higher risk of ipsilateral invasive breast cancer and ipsilateral DCIS. Similarly, a retrospective study of 2996 women with DCIS and treated with breast conserving surgery with or without radiotherapy or endocrine treatment reported a trend (P for trend=0.087) in the association between increasing margin width and lower risk of recurrence (defined as any ipsilateral or metastatic breast event), although only in women who did not receive radiotherapy.42 In the current study, the increased risks of ipsilateral invasive breast cancer and ipsilateral DCIS in patients with involved margins who underwent breast conserving surgery was independent of adjuvant treatment, which was also observed in the Swedish Ductal Carcinoma in Situ (SweDCIS) randomised controlled trial cohort, comprising 1046 patients.43 Comparison between studies is difficult, however, owing to heterogeneity in the categorisation of DCIS size and in the definition of clear versus involved margins.

    The results of our study highlight the low risk nature of DCIS. Overall, 10 year cumulative incidence for ipsilateral invasive breast cancer was 3.2%. For reference, the 10 year risk of breast cancer in women aged 60 in the US general population is 3.5%.4445 Additionally, it has been reported that about 20% of all ipsilateral invasive breast cancers diagnosed after treatment for DCIS are not related to the primary DCIS,4647 implying that these breast cancer diagnoses are not a recurrent lesion but rather a second primary lesion. This would also explain why clinicopathological factors of DCIS are less useful in predicting subsequent ipsilateral invasive breast cancer but more useful in predicting ipsilateral DCIS.

    In our analyses, the use of endocrine treatment in addition to breast conserving surgery with or without radiotherapy was not significantly associated with a lower risk of ipsilateral invasive breast cancer compared to treatment with breast conserving surgery only. The benefits of endocrine treatment were also not observed in subgroup analyses that included only women with oestrogen receptor positive DCIS (supplementary table S13). Previous randomised trials of DCIS showed varying results on the potential benefit of endocrine treatment. A significantly lower risk of ipsilateral invasive breast cancer in patients who received breast conserving surgery with radiotherapy and endocrine treatment or placebo was reported in the NSABP B-24 trial.2648 In the UK, Australia, and New Zealand DCIS randomised controlled trial, however, only a lower risk of ipsilateral DCIS, and not of ipsilateral invasive breast cancer, was reported in patients who received endocrine treatment compared with those who did not49—this was also seen in our study.

    Conclusions and implications

    This pooled, multinational study provided an opportunity to review current clinical practice using real world data and enabled comprehensive, exploratory analyses of stage and type of ipsilateral invasive breast cancer after treatment for DCIS. The higher risk of stage III and IV breast cancer in women with a large DCIS (≥50 mm) found in our analyses, provides new insights into the characteristics of potentially harmful DCIS that may be expanded on and validated in future studies. Notably, even in this large cohort, numbers of events were limited, highlighting the low risk of subsequent breast cancer events after treatment for DCIS. Additionally, although the risks of ipsilateral invasive breast cancer and ipsilateral DCIS associated with DCIS size and margin status were increased, they were low. When these two factors were added to other perceived risk factors, the prognostic value of the combined factors was modest at best, highlighting the need for a shift in focus towards novel prognostic markers.

    What is already known on this topic

    • An understanding of which women with ductal carcinoma in situ (DCIS) are at high or low risk of subsequent events is urgently needed

    • It is still unclear to what extent current clinical variables are useful in this prediction—some clinicopathological features, such as high grade lesions and young age, are associated with increased risk of subsequent ipsilateral invasive breast cancer and ipsilateral DCIS

    • For other features, such as DCIS size and margin status, the association with risk of ipsilateral invasive breast cancer risk or ipsilateral DCIS is less clear

    What this study adds

    • DCIS size and margin status were associated with an increased risk of ipsilateral invasive breast cancer and ipsilateral DCIS, but the risks were low

    • When these two factors were added to other perceived risk factors, the prognostic value of the combined factors was modest at best, highlighting the need to focus on novel prognostic markers

    • The higher risk of stage III and IV breast cancer found in women with large DCIS (≥50 mm), offers new insights into the characteristics of potentially harmful DCIS that may be expanded on in future studies

    Ethics statements

    Ethical approval

    The data collection for the Dutch cohort was reviewed and approved by the Netherlands Cancer Registry and Palga review boards (data study Nos NCR: K12.281, K17.321; data study Nos Palga: lzv990, lzv2017-173). The dataset generated and analysed during the current study as the “MDACC cohort” was reviewed and collated as part of PA16-0612. A separate protocol, PA17-1020, governed the movement of data and specimens between PRECISION institutions. NCDB special study data were determined exempt by the Duke University Medical Centre institutional review board (Pro00081281).The Sloane study has been permitted to process personally identifiable data without consent under Regulation 5 of Statutory Instrument 2002 No 1438: The Health Service (Control of Patient Information) (15/CAG/0207) in line with the following clause: “quality assuring screening services to ensure they are effective and safe, and that any incidents are investigated and managed appropriately.” This statutory exemption to common law permits the processing of personally identifiable data, as part of the core remit of population screening. The Sloane Project was able to send clinical data to the Netherlands Cancer Institute under the ethics committee approval research ethics committee reference: 18/WM/0400, protocol No C38317/A24043, IRAS Project ID: 247823, and data sharing agreement between Public Health England/NHS England Reference ODR1819_019. Data protection and curation was done in compliance with current ethical and data protection regulations.

    Data availability statement

    The datasets generated and analysed during the current study as the “Dutch cohort” are not publicly available, as the study used external data from the Netherlands Cancer Registry. The datasets will be made available from the Netherlands Cancer Registry upon reasonable request (data request study Nos K12.2871 and K17.321). To apply for data access, please visit https://www.iknl.nl/en/ncr/apply-for-data. Sloane Project data that forms part of the PRECISION Project cannot be accessed directly from the PRECISION team. However, the Sloane Project group welcomes applications from the UK, European Economic Area, and international organisations to collaborate and release data. Any data are subject to a common governance framework, which will ensure correct adherence to confidentiality provisions, legal permission, and ethical approvals. The dataset generated and analysed during the current study as the “MDACC cohort” are not publicly available. Any requests for access of this data should be directed to Afutreal{at}mdanderson.org. The National Cancer Database (NCDB), under the guidance of the Commission on Cancer (CoC) leadership and committees, disseminates aggregate and individual data for CoC accredited programmes. NCDB special study data are primarily intended for internal use within CoC accredited cancer programmes. These data have permissible and non-permissible use and are therefore only able to be shared outside of the CoC accredited programme once an appropriate data sharing agreement has been established. For more information, please contact shelley.hwang{at}duke.edu. Data from the MDCC (MD Anderson Cancer Center) cohort can be made available upon request.

    Acknowledgments

    We thank the Grand Challenge PRECISION Consortium Steering Group: Jelle Wesseling (Netherlands Cancer Institute, Amsterdam, Netherlands), Alastair Thompson (Baylor College of Medicine, Houston, TX, USA), Serena Nik-Zainal (University of Cambridge, Cambridge, UK), Elinor J Sawyer (King’s College London, London, UK), Helen Davies (University of Cambridge), Andrew Futreal (MD Anderson Cancer Center), Nicholas Navin (MD Anderson Cancer Center), E Shelley Hwang (Duke University School of Medicine, Cancers 2022, 14, 3259 10 of 13 Durham, NC, USA), Jos Jonkers (Netherlands Cancer Institute), Jacco van Rheenen (Netherlands Cancer Institute), Fariba Behbod (Kansas University Medical Center, KS, USA), Esther H Lips (Netherlands Cancer Institute), Marjanka Schmidt (Netherlands Cancer Institute), Lodewyk F A Wessels (Netherlands Cancer Institute), Daniel Rea (University of Birmingham, Birmingham, UK), Proteeti Bhattacharjee (Netherlands Cancer Institute), Hilary Stobart (Independent Cancer Patients’ Voice, UK), Deborah Collyar (Patient Advocates in Research, USA), Donna Pinto (dcis411, USA), Ellen Verschuur (Borstkanker Vereniging Nederland, Netherlands), and Marja van Oirsouw (Borstkanker Vereniging Nederland, the Netherlands).

    We also thank the registration team of the Netherlands Comprehensive Cancer Organisation (IKNL) for the collection of data for the Netherlands Cancer Registry, as well as IKNL staff for scientific advice; the nationwide network and registry of histopathology and cytopathology in the Netherlands Palga: Dutch Nationwide Pathology Databank, for the collection of data as well as scientific advice; M van Seijen, E Groen, and S Alaeikhanehshir (Netherlands Cancer Institute, Amsterdam, Netherlands), for preparatory work on this project; and the Sloane Project Steering Group for its ongoing contributions to the Sloane Project which forms the UK data in the analyses.

    This work uses patient data collected by the UK National Health Service (NHS) as part of its care and support. The data are collated, maintained, and quality assured by the Screening Quality Assurance Service and the National Disease Registration Service, which are part of NHS England.

    Footnotes

    • Contributors: JW, EHL, and MKS contributed equally to this paper and are the guarantors of this manuscript. RSJMS, JW, EHL, MKS, AWBD, TL, DC, MDR, ES, ESH, and AT contributed to the study conceptualisation. RSJMS, JW, EHL, MKS, AWBD, MS, MDR, ES, ESH, and AT contributed to the study design and methodology. PB, KC, YR, CC, JT, YL, DB, BM, TH, ST, and JKL contributed to data collection and curation. PB contributed to project management and administration. RSJMS, AWBD, MS, and CC contributed to data analysis. RSJMS, AWBD, JW, EHL, and MKS participated in data interpretation and manuscript preparation. AWBD, JW, EHL, and MKS participated in supervision. All authors read and approved the final version of the manuscript. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

    • Funding: This work was supported through a Patient-Centered Outcomes Research Institute (PCORI) award (PCS-1505-30497). All statements in this manuscript are solely those of the authors and do not necessarily represent the views of PCORI, its board of governors, or methodology committee. The PRECISION project is supported by Cancer Research UK and by Dutch Cancer Society (KWF Kankerbestrijding) (ref C38317/A24043). Research at the Netherlands Cancer Institute is supported by institutional grants of the Dutch Cancer Society and of the Dutch Ministry of Health, Welfare, and Sport. The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication.

    • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/disclosure-of-interest/ and declare: support from Cancer Research UK and by KWF Kankerbestrijding and Patient-Centered Outcomes Research Institute; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years, no other relationships or activities that could appear to have influenced the submitted work.

    • The corresponding author (MKS) affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

    • Dissemination to participants and related patient and public communities: A lay summary will be drafted and sent to patient organisations, Palga, and Netherlands Cancer Registry. Additionally, the lay summary will be shared on the PRECISION website and with the organisations that funded this work.

    • Provenance and peer review: Not commissioned; externally peer reviewed.

    http://creativecommons.org/licenses/by-nc/4.0/

    This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

    References

    1. ↵
      1. Shapiro S,
      2. Coleman EA,
      3. Broeders M,
      4. et al.,
      5. International Breast Cancer Screening Network (IBSN) and the European Network of Pilot Projects for Breast Cancer Screening
      . Breast cancer screening programmes in 22 countries: current policies, administration and guidelines. Int J Epidemiol1998;27:735-42. doi:10.1093/ije/27.5.735 pmid:9839727
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. van Luijt PA,
      2. Fracheboud J,
      3. Heijnsdijk EA,
      4. den Heeten GJ,
      5. de Koning HJ,
      6. National Evaluation Team for Breast Cancer Screening in Netherlands Study Group (NETB)
      . Nation-wide data on screening performance during the transition to digital mammography: observations in 6 million screens. Eur J Cancer2013;49:3517-25. doi:10.1016/j.ejca.2013.06.020 pmid:23871248
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Vinnicombe S,
      2. Pinto Pereira SM,
      3. McCormack VA,
      4. Shiel S,
      5. Perry N,
      6. Dos Santos Silva IM
      . Full-field digital versus screen-film mammography: comparison within the UK breast screening program and systematic review of published data. Radiology2009;251:347-58. doi:10.1148/radiol.2512081235 pmid:19401569
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Bleyer A,
      2. Welch HG
      . Effect of three decades of screening mammography on breast-cancer incidence. N Engl J Med2012;367:1998-2005. doi:10.1056/NEJMoa1206809 pmid:23171096
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Punglia RS,
      2. Bifolck K,
      3. Golshan M,
      4. et al
      . Epidemiology, Biology, Treatment, and Prevention of Ductal Carcinoma In Situ (DCIS). JNCI Cancer Spectr2018;2:pky063. doi:10.1093/jncics/pky063 pmid:30627695
      OpenUrlCrossRefPubMed
    6. ↵
      1. Bluekens AM,
      2. Holland R,
      3. Karssemeijer N,
      4. Broeders MJ,
      5. den Heeten GJ
      . Comparison of digital screening mammography and screen-film mammography in the early detection of clinically relevant cancers: a multicenter study. Radiology2012;265:707-14. doi:10.1148/radiol.12111461 pmid:23033499
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Siegel RL,
      2. Miller KD,
      3. Jemal A
      . Cancer statistics, 2019. CA Cancer J Clin2019;69:7-34. doi:10.3322/caac.21551 pmid:30620402
      OpenUrlCrossRefPubMed
    8. ↵
      1. Ryser MD,
      2. Weaver DL,
      3. Zhao F,
      4. et al
      . Cancer Outcomes in DCIS Patients Without Locoregional Treatment. J Natl Cancer Inst2019;111:952-60. doi:10.1093/jnci/djy220 pmid:30759222
      OpenUrlCrossRefPubMed
    9. ↵
      1. Ryser MD,
      2. Worni M,
      3. Turner EL,
      4. Marks JR,
      5. Durrett R,
      6. Hwang ES
      . Outcomes of Active Surveillance for Ductal Carcinoma in Situ: A Computational Risk Analysis. J Natl Cancer Inst2015;108:djv372. doi:10.1093/jnci/djv372 pmid:26683405
      OpenUrlCrossRefPubMed
    10. ↵
      1. Erbas B,
      2. Provenzano E,
      3. Armes J,
      4. Gertig D
      . The natural history of ductal carcinoma in situ of the breast: a review. Breast Cancer Res Treat2006;97:135-44. doi:10.1007/s10549-005-9101-z pmid:16319971
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Sanders ME,
      2. Schuyler PA,
      3. Dupont WD,
      4. Page DL
      . The natural history of low-grade ductal carcinoma in situ of the breast in women treated by biopsy only revealed over 30 years of long-term follow-up. Cancer2005;103:2481-4. doi:10.1002/cncr.21069 pmid:15884091
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      National Comprehensive Cancer Network. DCIS Breast Cancer (version 2020). https://www.nccn.org/patients/guidelines/content/PDF/stage_0_breast-patient.pdf.
    13. ↵
      1. Narod SA,
      2. Iqbal J,
      3. Giannakeas V,
      4. Sopik V,
      5. Sun P
      . Breast Cancer Mortality After a Diagnosis of Ductal Carcinoma In Situ. JAMA Oncol2015;1:888-96. doi:10.1001/jamaoncol.2015.2510 pmid:26291673
      OpenUrlCrossRefPubMed
    14. ↵
      1. van Seijen M,
      2. Lips EH,
      3. Thompson AM,
      4. et al.,
      5. PRECISION team
      . Ductal carcinoma in situ: to treat or not to treat, that is the question. Br J Cancer2019;121:285-92. doi:10.1038/s41416-019-0478-6 pmid:31285590
      OpenUrlCrossRefPubMed
    15. ↵
      1. Gierisch JM,
      2. Myers ER,
      3. Schmit KM,
      4. et al
      . Prioritization of research addressing management strategies for ductal carcinoma in situ. Ann Intern Med2014;160:484-91. doi:10.7326/M13-2548 pmid:24567146
      OpenUrlCrossRefPubMed
    16. ↵
      1. Maxwell AJ,
      2. Clements K,
      3. Hilton B,
      4. et al.,
      5. Sloane Project Steering Group
      . Risk factors for the development of invasive cancer in unresected ductal carcinoma in situ. Eur J Surg Oncol2018;44:429-35. doi:10.1016/j.ejso.2017.12.007 pmid:29398324
      OpenUrlCrossRefPubMed
    17. ↵
      1. Grand Challenge PRECISION consortium
      . https://www.dcisprecision.org/.
    18. ↵
      1. Kantor O,
      2. Winchester DJ
      . Breast conserving therapy for DCIS--does size matter?J Surg Oncol2014;110:75-81. doi:10.1002/jso.23657 pmid:24861481
      OpenUrlCrossRefPubMed
    19. ↵
      1. Mannu GS,
      2. Wang Z,
      3. Broggio J,
      4. et al
      . Invasive breast cancer and breast cancer mortality after ductal carcinoma in situ in women attending for breast screening in England, 1988-2014: population based observational cohort study. BMJ2020;369:m1570. doi:10.1136/bmj.m1570 pmid:32461218
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Silverstein MJ,
      2. Lagios MD,
      3. Craig PH,
      4. et al
      . A prognostic index for ductal carcinoma in situ of the breast. Cancer1996;77:2267-74. doi:10.1002/(SICI)1097-0142(19960601)77:11<2267::AID-CNCR13>3.0.CO;2-V pmid:8635094
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. MacDonald HR,
      2. Silverstein MJ,
      3. Mabry H,
      4. et al
      . Local control in ductal carcinoma in situ treated by excision alone: incremental benefit of larger margins. Am J Surg2005;190:521-5. doi:10.1016/j.amjsurg.2005.06.005 pmid:16164913
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. Solin LJ,
      2. Gray R,
      3. Hughes LL,
      4. et al
      . Surgical Excision Without Radiation for Ductal Carcinoma in Situ of the Breast: 12-Year Results From the ECOG-ACRIN E5194 Study. J Clin Oncol2015;33:3938-44. doi:10.1200/JCO.2015.60.8588 pmid:26371148
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Collins LC,
      2. Achacoso N,
      3. Haque R,
      4. et al
      . Risk factors for non-invasive and invasive local recurrence in patients with ductal carcinoma in situ. Breast Cancer Res Treat2013;139:453-60. doi:10.1007/s10549-013-2539-5 pmid:23624751
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Wai ES,
      2. Lesperance ML,
      3. Alexander CS,
      4. et al
      . Predictors of local recurrence in a population-based cohort of women with ductal carcinoma in situ treated with breast conserving surgery alone. Ann Surg Oncol2011;18:119-24. doi:10.1245/s10434-010-1214-x pmid:20645008
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Bijker N,
      2. Meijnen P,
      3. Peterse JL,
      4. et al.,
      5. EORTC Breast Cancer Cooperative Group,
      6. EORTC Radiotherapy Group
      . Breast-conserving treatment with or without radiotherapy in ductal carcinoma-in-situ: ten-year results of European Organisation for Research and Treatment of Cancer randomized phase III trial 10853--a study by the EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. J Clin Oncol2006;24:3381-7. doi:10.1200/JCO.2006.06.1366 pmid:16801628
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Wapnir IL,
      2. Dignam JJ,
      3. Fisher B,
      4. et al
      . Long-term outcomes of invasive ipsilateral breast tumor recurrences after lumpectomy in NSABP B-17 and B-24 randomized clinical trials for DCIS. J Natl Cancer Inst2011;103:478-88. doi:10.1093/jnci/djr027 pmid:21398619
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Ringberg A,
      2. Nordgren H,
      3. Thorstensson S,
      4. et al
      . Histopathological risk factors for ipsilateral breast events after breast conserving treatment for ductal carcinoma in situ of the breast--results from the Swedish randomised trial. Eur J Cancer2007;43:291-8. doi:10.1016/j.ejca.2006.09.018 pmid:17118648
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Kerlikowske K,
      2. Molinaro A,
      3. Cha I,
      4. et al
      . Characteristics associated with recurrence among women with ductal carcinoma in situ treated by lumpectomy. J Natl Cancer Inst2003;95:1692-702. doi:10.1093/jnci/djg097 pmid:14625260
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Morrow M,
      2. Van Zee KJ,
      3. Solin LJ,
      4. et al
      . Society of Surgical Oncology-American Society for Radiation Oncology-American Society of Clinical Oncology Consensus Guideline on Margins for Breast-Conserving Surgery With Whole-Breast Irradiation in Ductal Carcinoma In Situ. J Clin Oncol2016;34:4040-6. doi:10.1200/JCO.2016.68.3573 pmid:27528719
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Amin MB,
      2. Greene FL,
      3. Edge SB,
      4. et al
      . The Eighth Edition AJCC Cancer Staging Manual: Continuing to build a bridge from a population-based to a more “personalized” approach to cancer staging. CA Cancer J Clin2017;67:93-9. doi:10.3322/caac.21388 pmid:28094848
      OpenUrlCrossRefPubMed
    31. ↵
      1. van Buuren S,
      2. Boshuizen HC,
      3. Knook DL
      . Multiple imputation of missing blood pressure covariates in survival analysis. Stat Med1999;18:681-94. doi:10.1002/(SICI)1097-0258(19990330)18:6<681::AID-SIM71>3.0.CO;2-R pmid:10204197
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Raghunathan TE,
      2. Lepkowski JM,
      3. Hoewyk JV,
      4. Solenberger PW
      . A multivariate technique for multiply imputing missing values using a sequence of regression models. Surv Methodol2001;27:85-95.
      OpenUrl
    33. ↵
      1. White IR,
      2. Royston P
      . Imputing missing covariate values for the Cox model. Stat Med2009;28:1982-98. doi:10.1002/sim.3618 pmid:19452569
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Rubin DB
      . Multiple Imputation for Nonresponse in Surveys.John Wiley & Sons, 2004.
    35. ↵
      1. Xue X,
      2. Kim MY,
      3. Gaudet MM,
      4. et al
      . A comparison of the polytomous logistic regression and joint cox proportional hazards models for evaluating multiple disease subtypes in prospective cohort studies. Cancer Epidemiol Biomarkers Prev2013;22:275-85. doi:10.1158/1055-9965.EPI-12-1050 pmid:23292084
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Pinsky RW,
      2. Rebner M,
      3. Pierce LJ,
      4. et al
      . Recurrent cancer after breast-conserving surgery with radiation therapy for ductal carcinoma in situ: mammographic features, method of detection, and stage of recurrence. AJR Am J Roentgenol2007;189:140-4. doi:10.2214/AJR.06.1281 pmid:17579163
      OpenUrlCrossRefPubMed
    37. ↵
      1. Casparie M,
      2. Tiebosch AT,
      3. Burger G,
      4. et al
      . Pathology databanking and biobanking in The Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Cell Oncol2007;29:19-24.pmid:17429138
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Alaeikhanehshir S,
      2. Engelhardt EG,
      3. van Duijnhoven FH,
      4. et al.,
      5. PRECISION
      . The impact of patient characteristics and lifestyle factors on the risk of an ipsilateral event after a primary DCIS: A systematic review. Breast2020;50:95-103. doi:10.1016/j.breast.2020.02.006 pmid:32120064
      OpenUrlCrossRefPubMed
    39. ↵
      1. Hwang ES,
      2. Hyslop T,
      3. Lynch T,
      4. et al
      . The COMET (Comparison of Operative versus Monitoring and Endocrine Therapy) trial: a phase III randomised controlled clinical trial for low-risk ductal carcinoma in situ (DCIS). BMJ Open2019;9:e026797. doi:10.1136/bmjopen-2018-026797 pmid:30862637
      OpenUrlAbstract/FREE Full Text
    40. ↵
      1. Francis A,
      2. Fallowfield L,
      3. Rea D
      . The LORIS Trial: Addressing overtreatment of ductal carcinoma in situ. Clin Oncol (R Coll Radiol)2015;27:6-8. doi:10.1016/j.clon.2014.09.015 pmid:25445552
      OpenUrlCrossRefPubMed
    41. ↵
      1. Elshof LE,
      2. Tryfonidis K,
      3. Slaets L,
      4. et al
      . Feasibility of a prospective, randomised, open-label, international multicentre, phase III, non-inferiority trial to assess the safety of active surveillance for low risk ductal carcinoma in situ - The LORD study. Eur J Cancer2015;51:1497-510. doi:10.1016/j.ejca.2015.05.008 pmid:26025767
      OpenUrlCrossRefPubMed
    42. ↵
      1. Van Zee KJ,
      2. Subhedar P,
      3. Olcese C,
      4. Patil S,
      5. Morrow M
      . Relationship Between Margin Width and Recurrence of Ductal Carcinoma In Situ: Analysis of 2996 Women Treated With Breast-conserving Surgery for 30 Years. Ann Surg2015;262:623-31. doi:10.1097/SLA.0000000000001454 pmid:26366541
      OpenUrlCrossRefPubMed
    43. ↵
      1. Wärnberg F,
      2. Garmo H,
      3. Emdin S,
      4. et al
      . Effect of radiotherapy after breast-conserving surgery for ductal carcinoma in situ: 20 years follow-up in the randomized SweDCIS Trial. J Clin Oncol2014;32:3613-8. doi:10.1200/JCO.2014.56.2595 pmid:25311220
      OpenUrlAbstract/FREE Full Text
    44. ↵
      Cancer Statistics Review SEER. 1975-2017, National Cancer Institute. Bethesda, MD, USA. 2020. https://seer.cancer.gov/csr/1975_2017/.
    45. ↵
      Institute NC. Breast Cancer Risk in American Women 2020 (updated December 16, 2020). https://www.cancer.gov/types/breast/risk-fact-sheet.
    46. ↵
      1. Lips EH,
      2. Kumar T,
      3. Megalios A,
      4. et al.,
      5. Grand Challenge PRECISION consortium
      . Genomic analysis defines clonal relationships of ductal carcinoma in situ and recurrent invasive breast cancer. Nat Genet2022;54:850-60. doi:10.1038/s41588-022-01082-3 pmid:35681052
      OpenUrlCrossRefPubMed
    47. ↵
      1. Gorringe KL,
      2. Hunter SM,
      3. Pang JM,
      4. et al
      . Copy number analysis of ductal carcinoma in situ with and without recurrence. Mod Pathol2015;28:1174-84. doi:10.1038/modpathol.2015.75 pmid:26321097
      OpenUrlCrossRefPubMed
    48. ↵
      1. Fisher B,
      2. Dignam J,
      3. Wolmark N,
      4. et al
      . Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet1999;353:1993-2000. doi:10.1016/S0140-6736(99)05036-9 pmid:10376613
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. Cuzick J,
      2. Sestak I,
      3. Pinder SE,
      4. et al
      . Effect of tamoxifen and radiotherapy in women with locally excised ductal carcinoma in situ: long-term results from the UK/ANZ DCIS trial. Lancet Oncol2011;12:21-9. doi:10.1016/S1470-2045(10)70266-7 pmid:21145284
      OpenUrlCrossRefPubMedWeb of Science
    Back to top