Stromal collagen IV expression and risk of breast cancer death in ductal carcinoma in situ

Stromal collagen IV expression and risk of breast cancer death in ductal carcinoma in situ | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Stromal collagen IV expression and risk of breast cancer death in ductal carcinoma in situ Gunilla Rask, Malin Jansson, Johan Svensson, Rebecca Wiberg, Fredrik Wärnberg, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6714442/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background . Current treatment for ductal carcinoma in situ (DCIS) of the breast is generic, due to lack of risk stratification tools. We investigate the correlation between expression of collagen IV in the breast and risk of breast cancer death. We also explore the effect of collagen IV in vitro. Methods . Tissue microarrays from a cohort of women treated for DCIS who later died from breast cancer (n=43) or were still alive (n=119), were analysed for collagen IV by immunohistochemistry. Oestrogen receptor positive (ER+), triple negative and human epidermal growth factor receptor 2 amplified (HER2+) cell lines were cultured with and without collagen IV. Results . High expression of stromal collagen IV correlated with increased odds of dying of breast cancer (OR 2.50; 95% CI 1.16-5.39). This association remained when adjusting for tumour size, margin status, comedo necrosis and progesterone receptor negativity (PR-) (OR 4.27; 95% CI 1.64 – 11.1). Triple negative breast cancer cell lines migrated quicker on collagen IV-coated than on uncoated surfaces. By contrast, collagen IV coating did not affect ER+ and HER2+ cell lines. Conclusions . Abundance of stromal collagen IV increases risk of dying in breast cancer after DCIS, and collagen IV can promote cell motility in vitro. Figures Figure 1 Figure 2 Figure 3 Background As screening programmes are widely implemented and radiological methods become ever more refined, the proportion of women with breast cancer that are diagnosed with ductal carcinoma in situ (DCIS) is increasing and now comprises about 15% of new breast cancers ( 1 ). Current treatment of DCIS is surgery with or without radiotherapy, with possible addition of endocrine treatment for women with oestrogen receptor positive (ER+) DCIS. As survival after DCIS treatment is excellent, there is rising concern that several women are overtreated, meaning that they would not have developed any symptomatic invasive carcinoma if left untreated ( 2 ). Methods for robust risk stratification of patients is therefore in high demand. Traditional histological factors, and those recommended in the EUSOMA (European Society of Breast Cancer specialists) guidelines are lesion size, nuclear grade, growth pattern, presence of comedo necrosis, calcifications and expression of ER ( 3 ). Nuclear grade is at best however moderately reproducible between pathologists ( 4 ) and that there are conflicting results as to whether grade, comedo necrosis, histological pattern and ER-status are related to patient outcome ( 5 – 9 ). Assuming that old or novel biomarkers may be used to predict risk begs the question of which risk specifically. Is it risk of local DCIS recurrence, of invasive recurrence or of breast cancer death? All treatments come with side effects, and the benefit of risk reduction must be substantial enough to motivate treatment. Features which predict DCIS recurrence are not the same as those that predict recurrence as invasive disease ( 9 ). Invasive recurrence increases risk of breast cancer death, and the addition of radiotherapy is routinely used to lower risk of local recurrence after breast conserving surgery. However, neither radiotherapy nor addition of endocrine therapy have been shown to affect the risk of breast cancer death ( 10 – 12 ). To refine the traditional risk stratification of DCIS there is a need for new biomarkers that are related to the risk of distant metastasis and breast cancer death. Surrounding the DCIS cells is the myoepithelium, resting in turn on the basement membrane (BM) which is composed almost entirely of collagen IV. Outside of this structure is the stroma, normally consisting of many other types of collagens and smaller amounts of other molecules ( 13 ). Cancer-associated stroma, however, often contains collagen IV also beyond the BM ( 14 – 16 ). Our group has previously linked this stromal collagen IV expression to an increased risk of distant metastasis and worse survival ( 17 ). Circulating collagen IV has also been linked to an increased risk of distant metastasis and worse survival in both breast ( 18 ) and colorectal ( 19 ) cancer patients. A previous study using mouse and in vitro models demonstrated that breast cancer cells of a predominantly triple-negative phenotype invade quicker and further in collagen IV-rich decellularized matrix scaffolds compared to scaffolds with low collagen IV-content ( 20 ). This study aimed to determine 1) if stromal collagen IV-expression in DCIS correlated with risk of breast cancer death in a case-control cohort, and 2) if collagen IV could affect the motility of breast cancer cell lines in vitro. Materials and method Patient cohort and TMA-contruction We have previously described a nested case-control cohort of women with DCIS with clinicopathological data available ( 21 ). Briefly, the cancer registries of three health care regions in Sweden were used to identify women treated for DCIS between 1992 and 2012 (n = 6 964). Cross-linking with the National Cause of Death Registry identified women who later died from breast cancer (cases, n = 95). Controls were selected using incidence density sampling, from the whole population of women with DCIS. Controls had to be alive and without distant metastasis of breast cancer at the time of death of the corresponding case (n = 318). Clinical data was gathered from medical records. In this cohort, tumour size larger than 25 mm and positive margin status were associated with increased risk of breast cancer death, while surgical treatment and radiotherapy were not. From the surgical specimens of the women in the cohort, tumour tissue was obtained for whole slide sectioning and construction of tissue microarrays (TMA) for 65 of the 95 cases (68%) and 195 of 318 controls (61%). Whole slides and TMA-slides were centrally evaluated by a specialized breast pathologist (G.R) for DCIS grade, presence of comedo necrosis, amount of tumour infiltrating lymphocytes (TILs), as well as expression of ER, progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and Ki67 ( 22 ). For the present study there was sufficient tumour tissue in the TMAs for analysis of collagen IV in 43 cases and 119 controls. (Fig. 1 ) Immunohistochemistry for collagen IV Tissue sections were stained with rabbit anti-collagen IV (AB748, Millipore, Billerica, United States) diluted 1:50 in blocking buffer for 2 hours in room temperature. Sections were then washed in phosphate buffered saline (PBS) and incubated with the biotinylated secondary antibody, diluted 1:200 in blocking buffer, for 30 minutes in room temperature. Finally, slides were washed in PBS followed by addition of diaminobenzidine tetrahydrochloride (DAB) as chromogen. Expression of collagen IV was scored separately in the compartment closest to the myoepithelial/stromal interface, defined as within 10 µm of the interface and dubbed periductal collagen IV, and in the rest of the stroma, henceforth called stromal collagen IV. Each compartment was scored on a scale 0–3 (0 = no, 1 = minimal, 2 = moderate and 3 = strong expression). Scoring was done by two independent researchers (G.R and M.J), blinded to outcome at the time of scoring. Discrepant cases were reviewed jointly, and a consensus score was set. There were one to six biopsies for each patient in the TMA, and in cases were the expression of collagen IV within a compartment differed between biopsies the maximum score was used for the final analyses. Cell cultures and migration experiments Three different breast cancer cell lines were used. An ER+/HER2-cell line (MCF-7; ATCC HTB-22), an ER-/HER2- cell line (MDA-MB-231; ATCC HTB-26) and a ER-/HER2 + cell line (JIMT-1; RRID:CVCL_2077). Cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) 4,5% glucose with 10% foetal bovine serum (FBS) and 1% penicillin/streptomycin. For migration experiments 96-well plates were coated with either DMEM or collagen IV (ab7536 Abcam; 1 mg/mL) diluted in DMEM to a concentration of 29 ug/mL. Cells were seeded at a density of 5000 cells/well (MCF-7), or 2500 cells/well (MDA-MB-231 and JIMT-1). After seeding the cells were left to adhere for 4–6 hours, then washed with PBS. New DMEM was added, and the plates were inserted to the Holomonitor M4, which uses quantitative phase imaging to automatically detect and measure live cells without labelling.( 23 ) We used the adaptive Gaussian algorithm of detection to minimize background noise from the coating. Images were captured every 15 minutes for 24 hours. In each experiment, six wells per breast cancer cell line were used, three with and three without collagen IV, respectively. Three fields of view were analysed for each well, each field of view containing approximately 50–200 cells. The speed of individual cells between each consecutive captured image frame (distance travelled/time between frames) was calculated by the built-in software of the HoloMonitor, as well as the average speed of all cells of each type between consecutive frames. The experiment was repeated thrice and the values of the resulting averages for all experiments were used for the final analyses. After all experiments were finished, all three cell lines tested negative for potential contamination of various Mycoplasma species. Statistical analysis Odds ratios (OR) for dying of breast cancer in the nested case control cohort were calculated using conditional logistic regression and all analyses adjusted for time at risk. The following clinical variables were used: Age, tumour size (< 25 mm or ≥ 25 mm, as in previous studies on this cohort ( 21 )), margin status (negative vs positive, uncertain or missing), mode of detection (screening vs non-screening) and surgical treatment (breast conserving therapy only (BCS), BCS with radiotherapy (BCS + RT) or mastectomy). Moreover the following biomarkers were used: Nuclear grade according to Holland ( 24 ), comedo necrosis (absent vs. present) and ER and PR expression (positive if ≥ 10% of nuclei stained, in keeping with national guidelines for invasive carcinoma) ( 25 ). Periductal TILs were categorized as 0–5%, 10% or ≥ 20% according to international guidelines ( 26 ). The expression of both stromal and periductal collagen IV as described above was simplified to a two-grade scale for the final analyses with 0–1 termed “low” and 2–3 termed “high”. Pearson Chi-square was used to test for independence between variables. The test for interaction between ER-status and outcome was conducted by adding an interaction term in the conditional logistic regression model. Loss of material in sectioning resulted in missing values of the case response variable, making some strata entirely unused. To maximise data use, strata with cases treated at similar timepoints were fused. This gave a total of 32 strata, compared to the original 66. For the adjusted analyses, missing values for variables other than either collagen IV variable were imputed using multiple imputation in 20 datasets estimated with chained equations. For the imputation model, the variables in the conditional logistic regression models were used, including the responses periductal collagen IV and stromal collagen IV as well as time at risk and year of diagnosis. Proportion of imputed values were 16% for tumour size, 6% for ER-status and less than 5% for all other variables. For comparison of means of average migration speed in the in vitro experiments, independent samples t-test was used. Conditional logistic regression analyses, imputation and tests for interaction were performed in StataIC15.1 and all other analyses in SPSS® version 23 (IBM, Armonk, NY, USA). Results Study population Median age at diagnosis was 54 years, with no significant difference between cases and controls. A majority of the women (63.6%) were diagnosed within the breast cancer screening programme. Most were treated with breast conserving surgery (BCS) with or without subsequent radiotherapy (27.8% and 32.7% respectively) while 39.5% underwent mastectomy. The proportion of tumours < 25 mm was approximately equal to those ≥ 25 mm (43.8% and 40.1%, with missing data on 16%). Approximately 10% of the surgical specimens had resection margins that were either positive or uncertain. Risk factors for breast cancer death Several characteristics associated with increased risk of breast cancer death in the unadjusted analysis: Tumour size ≥ 25 mm (OR 3.48; 95% CI 1.42–8.56), presence of comedo necrosis (OR 2.68; 95% CI 1.10–6.53), negative PR expression (OR 2.45; 95% CI 1.08–5.57), and high expression of stromal collagen IV (OR 2.50; 95% CI 1.16–5.39). Neither grade nor ER-expression had a statistically significant correlation with risk of breast cancer death. Table 1 summarizes the characteristics of the study cohort and ORs for breast cancer death. Table 1 Clinicopathological factors and odds ratios for breast cancer death after treatment for ductal carcinoma in situ. Study cohort (n = 162) Cases (n = 43) Controls (n = 119) OR (95% CI) for breast cancer death a Age yrs, median (IQR) 54 (48–62) 52 (46–62) 54 (49–62) 0.99 (0.96–1.02) Tumour size* n (%) < 25 mm 71 (43.8) 11 (25.6) 60 (50.4) 1.0 (ref.) ≥ 25 mm 65 (40.1) 24 (55.8) 41 (34.5) 3.48 (1.42–8.56) Missing 26 (16.0) 8 (18.6) 18 (15.1) - Margin status n (%) Negative 145 (89.5) 34 (79.1) 111 (93.3) 1.0 (ref.) Positive/uncertain/missing 17 (10.5) 9 (20.9) 8 (6.7) 2.72 (0.94–7.87) Mode of detection n (%) Screening 103 (63.6) 22 (51.2) 81 (68.1) 1.0 (ref.) Non-screening 50 (30.9) 14 (32.6) 36 (30.3) 1.50 (0.68–3.32) Missing 9 (5.6%) 7 (16.3) 2 (1.7) - Breast surgery n (%) BCS 45 (27.8) 12 (27.9) 33 (27.7) 1.0 (ref.) BCS + RT 53 (32.7) 12 (27.9) 41 (34.5) 0.86 (0.33–2.25) Mastectomy 64 (39.5) 19 (44.2) 45 (37.8) 1.31 (0.55–3.18) Grade n (%) 1 26 (16.0) 5 (11.6) 21 (17.6) 1.0 (ref.) 2 78 (48.1) 23 (53.5) 55 (46.2) 1.84 (0.63–5.33) 3 55 (34..0) 14 (32.6) 41 (34.5) 1.58 (0.49–5.10) Missing 3 (1.9) 1 (2.3) 2 (1.7) - Comedonecrosis* n (%) Absent 51 (31.5) 8 (18.6) 43 (36.1) 1.0 (ref.) Present 108 (66.7) 34 (79.1) 74 (62.2) 2.68 (1.10–6.53) Missing 3 (1.9) 1 (2.3) 2 (1.7) - ER n (%) ≥ 10% 112 (69.1) 27 (62.8) 85 (71.4) 1.0 (ref.) < 10% 41 (25.3) 13 (30.2) 28 (23.5) 1.63 (0.73–3.67) Missing 9 (5.5) 3 (7.0) 6 (5.0) - PR* n (%) ≥ 10% 93 (57.4) 20 (46.5) 73 (61.3) 1.0 (ref.) < 10% 63 (38.9) 21 (48.8) 42 (35.3) 2.45 (1.08–5.57) Missing 6 (3.7) 2 (4.7) 4 (3.3) - Periductal TILs n (%) 0–5% 98 (60.5) 23 (53.5) 75 (63.0) 1.0 (ref.) 10% 31 (19.1) 6 ( 14 ) 25 ( 21 ) 0.83 (0.30–2.32) ≥ 20% 30 (18.5) 13 (30.2) 17 (14.3) 2.34 (0.95–5.76) Missing 3 (1.9) 1 (2.3) 2 (1.7) - Stromal collagen IV* n (%) Low (0–1) 110 (67.9) 23 (53.5) 87 (73.1) 1.0 (ref.) High ( 2 – 3 ) 50 (30.8) 19 (44.2) 31 (26.1) 2.50 (1.16–5.39) Missing 2 (1.2) 1 (2.3) 1 (0.8) - Periductal collagen IV n (%) Low (0–1) 81 (50.0) 17 (39.5) 64 (53.8) 1.0 (ref.) High ( 2 – 3 ) 81 (50.0) 26 (60.5) 55 (46.2) 2.00 (0.96–4.15) OR – odds ratio; ER – oestrogen receptor; PR – progesterone receptor; TILs – tumour infiltrating lymphocytes. * denotes variables with statistically significant difference between cases and controls. a – all analyses adjusted for time at risk In the present cohort, a positive or uncertain margin status only showed a non-significant trend towards increased risk for breast cancer-related death (OR 2.72; 95% CI 0.94–7.87). The dropout analysis showed fewer small (< 25 mm) tumours remaining in the present cohort compared to tumours not analysed for collagen IV (43.8% vs 70.2%). The remaining cases were in all other respects similar to the cases that were lost. By contrast, the remaining controls were less often screening detected (68.1% vs 76.1%), had fewer negative surgical margins (93.3% vs 98.9%), more often had a suspicion of microinvasion (7.6% vs 0.99%) and were more often treated by mastectomy (37.8% vs 25.1%) than the controls that could not be analysed for collagen IV. Comparisons between the original cohort and the present one are summarised in supplementary table S1 (included at the bottom of this paper) Collagen IV expression and other clinicopathological variables Interobserver agreement between the two assessors for collagen IV expression in the TMA-biopsies, when using the 0 to 3 scale, was moderate for both periductal (kappa 0.52) and stromal (kappa 0.43) expression. Grouping the variables as low (0 or 1) and high (2 or 3) resulted in substantial agreement with kappa 0.65 for stromal and 0.76 for periductal expression. Representative images of TMA-biopsies with different expressions of collagen IV are shown in Fig. 2 . There was an association between expression of periductal and stromal collagen (p < 0.001) i.e. they were not independent variables. There was no dependency between collagen IV expression and ER or PR expression. There was a borderline significant correlation between comedo necrosis and stromal collagen IV (p = 0.06), where DCIS with necrosis was more likely to have a low stromal collagen expression (73% vs 58%). Necrosis did not correlate with periductal collagen IV (p = 0.73). There was no statistically significant association between stromal collagen IV and tumour size (p = 0.08). However, smaller tumours (< 25 mm) trended towards higher stromal collagen IV expression (39% vs. 20%). Collagen IV expression predicts breast cancer-related death in adjusted models. In the previous unadjusted analysis, high stromal collagen IV expression correlated with increased risk of dying of breast cancer. This association remained significant when adjusting for tumour size and margin status (OR 3.62; 95% CI 1.45-9.00), and even after adjusting for presence of comedonecrosis and PR negativity (OR 4.27; 95% CI 1.64–11.1). ER expression did not interact with stromal collagen IV for the risk of breast cancer death (p = 0.19). Periductal expression of collagen IV was not statistically significantly associated with increased risk of breast cancer death in the unadjusted analysis (OR 2.00; 95%CI 0.96–4.15). However, it showed a similar association as stromal collagen IV when adjusting for tumour size and margin status (OR 2.50; 95%CI 1.10–5.64) as well as with further adjusting for comedonecrosis and PR (OR 2.55; 95% CI1.11-5.86). The results are summarized in Table 2 . Table 2 Collagen IV expression and risk of breast cancer death OR for breast cancer death (95% CI)* Unadjusted Adjusted for tumour size and margin status Adjusted for tumour size, margin, PR and necrosis Stromal collagen IV Low (0–1) 1.0 (ref.) 1.0 (ref.) 1.0 (ref.) High ( 2 – 3 ) 2.50 (1.16–5.39) 3.62 (1.45-9.00) OR 4.27 (1.64–11.1) Periductal collagen IV Low (0–1) 1.0 (ref.) 1.0 (ref.) 1.0 (ref.) High ( 2 – 3 ) 2.00 (0.96–4.15) 2.50 (1.10–5.64) 2.55 (1.11–5.86) OR – odds ratio; CI – confidence interval; PR – progesterone receptor * All analyses adjusted for time at risk Collagen IV and cell migration As collagen IV has been shown to increase motility in vitro only in triple negative but not other types of breast cancer cells, experiments were performed using triple negative, ER+/HER2- and ER-/HER2 + cell lines. By light microscopy, the cells displayed different phenotypes. The triple negative cells (MDA-MB-231) were elongated and mesenchymal-like, growing evenly dispersed across the plates. The ER + and ER-/HER2 + cells (MCF-7 and JIMT-1) were rounder, plumper and tended to grow in aggregates rather than spreading out. The appearances were the same in both collagen IV-coated and uncoated plates. Effects of collagen IV-coating on cell migration differed between cell types (Fig. 3 ). Collagen IV coating resulted in an increased average migration speed in triple negative cells (5.7 vs. 7.7 µm/h, p < 0.001), whereas the opposite was observed for ER+/HER2- cells (7.8 vs. 6.6 µm/h, p < 0.001). No effect on migration speed was seen in the ER-/HER2 + cell line (7.3 vs. 7.1 µm/h, p = 0.68). Discussion Stromal components and other factors related to the tumour microenvironment are gaining increasing attention as potential prognostic markers and treatment targets in many cancer types ( 27 ). We demonstrate for the first time that high levels of collagen IV, both in the stromal and periductal compartment in DCIS correlate with increased odds of breast cancer death. This correlation became even more pronounced when adjusting for established clinical risk factors, such as tumour size and positive margins, as well as tentative histological risk factors such as comedonecrosis and PR-negativity. Moreover, we confirm previous findings that collagen IV promotes migration of triple negative breast cancer cells( 20 ), but the same was not shown for either ER+/HER2- or ER-/HER2 + breast cancer cells. In the breast, collagen IV is normally secreted by the myoepithelial cells (MECs), but in invasive breast cancer the collagen IV in the extracellular matrix (ECM) can originate from the cancer cells themselves, the surrounding stromal cells or both ( 28 ). It is thus conceivable that the periductal and stromal collagen IV in DCIS may be produced by the DCIS cells, the myoepithelial cells, the stromal cells or any combination of these. Production by the DCIS cells seems less likely, since the MEC layer limits physical access to both periductal and stromal compartments. Considering the MECs, it is known that tumour associated MECs differ in many ways from native MECs in gene expression patterns and phenotype ( 29 – 31 ). To the best of our knowledge though, an increased production of collagen IV by cancer associated MECs has not yet been demonstrated, and this would be worth to explore in future studies. That at least part of the collagen IV is produced by stromal cells is highly likely. Collagen IV self-polymerizes and cannot diffuse easily ( 32 ) but our analyses show stromal collagen IV expression as far as one millimetre from the DCIS (the limit of the diameter of our TMA-cores). If it is produced in the stroma, it also means that direct contact between DCIS cells and collagen IV-producing cells is not feasible as the only mode of communication. Either the signal needs to be endocrine rather than paracrine, or the DCIS cells need to inform the stromal cells to feed the signal forward. The stronger correlation between stromal collagen IV and breast cancer death compared to periductal collagen IV in both crude and adjusted analyses, is in keeping with either notion. It has been shown that the DCIS-induced PDGFRα (low) /PDGFRβ (high) phenotype of stromal fibroblasts is dependent on direct cell-to cell or paracrine signalling ( 33 ); investigating the levels of collagen IV in healthy breast tissue several centimetres away from the DCIS could show if the same is true for collagen IV-expression. Why increased collagen IV would lead to increased risk of breast cancer death remains unclear. Almost all in vitro or animal models exploring the effects of collagen IV have used triple negative cell lines ( 20 , 34 , 35 ), and all these show increased migration in response to collagen IV exposure. One study also found similar results in non-neoplastic mammary epithelial cells ( 36 ) but effects on ER + or HER2 + breast cancer cells have not been reported previously. Surprisingly, we found that a collagen IV coating decreased the migration speed of ER+/HER2- cells. This finding needs to be validated in additional cell lines and in other models. Cells may migrate by different means, either as single cells without any cell-to cell interaction, or as loosely or non-cohesive cell groups that move along the same path (streaming), or as cohesive units along a broad front (collective migration)( 37 ). One explanation to our findings may be that the mode of migration differs between the two cell types and that collagen IV, when used as coating, mainly increases one or two of these migration modes. Triple negative breast cancer cells (MDA-MB-231) migrate by single cell or streaming ( 37 ), while ER + cells (MCF7) are less investigated also in this respect. We found no effect of collagen IV coating on the migration of ER-/HER2 + cells when looking at the sum of all experiments. This cell line was, however, somewhat more difficult to get good readings of in the HoloMonitor, which might have been due to the conformation of the cells or how they affected the coating during incubation, which is a limitation of these experiments. Other limitations of the present study include the use of TMA rather than whole tissue sections since the field of vision is limited to stroma within 1 mm from the DCIS. This prevents observation of collagen IV in healthy tissue further from the DCIS as well as observations regarding spatial heterogeneity in the tumour stroma. The case-control cohort is small in numbers, preventing further subgroup analysis. This is however almost inevitable when dealing with rare outcomes such as breast cancer death after DCIS, and the cohort is in this respect larger or equal to others. In conclusion we show for the first time that stromal collagen IV correlates with risk of breast cancer death already at the in situ -stage of disease. This means that there may be aggressive types of DCIS with the potential to create conditions necessary for metastatic spread even before invasion. We further confirm that collagen IV increases migration of triple negative breast cancer cells, but not necessarily of any other type of breast cancer subtypes. Further studies should focus on collagen IV expression both in healthy tissue of breast cancer patients and cancer-associated MECs, as well as on how collagen IV affects different modes of migration. Declarations Acknowledgements Anette Berglund – for sectioning, staining and scanning of tissue microarrays. Christina Lundin – for guidance and help with cell cultures and collagen coating. Authors' contributions G.R – scored collagen IV staining, designed and performed cell culture experiments, processed HoloMonitor output, analysed data, drafted manuscript. M.J – scored collagen IV staining, analysed data, reviseed manuscript. J.S – analysed data, revised manuscript. R.W – conceived cell culture experiments, revised draft. F.W – conceived original cohort design, revised draft. O.B – designed and performed cell culture experiments, revised draft. C.W – acquired original cohort data and biomaterial, revised draft. M.S – conceived study design, revised draft. Ethics approval and consent to participate Ethical approval was granted from the regional ethics committee in Umeå (2014/45- 31). The need for informed consent was waived. The study was performed in accordance with the Declaration of Helsinki. Consent for publication The manuscript does not contain any individual person’s data. Data availability Data is not uploaded to a publicly available platform. Researchers have access to data through application to the study PI ( [email protected] ) or the corresponding author under standard rules of protecting data integrity and existing ethics permissions. Competing interests The authors declare no competing interests. Funding information This study was supported through research grants by the Swedish Breast Cancer Association, and VISARE Norr (grant no. 993966). The funding agents had no influence on the content of this manuscript. References van Seijen M, Lips EH, Thompson AM, Nik-Zainal S, Futreal A, Hwang ES, et al. Ductal carcinoma in situ: to treat or not to treat, that is the question. British Journal of Cancer. 2019;121(4):285–92. Delaloge S, Khan SA, Wesseling J, Whelan T. Ductal carcinoma in situ of the breast: finding the balance between overtreatment and undertreatment. Lancet. 2024;403(10445):2734–46. Rubio IT, Marotti L, Biganzoli L, Aristei C, Athanasiou A, Campbell C, et al. EUSOMA quality indicators for non-metastatic breast cancer: An update. European Journal of Cancer. 2024;198. Cserni G, Sejben A. Grading Ductal Carcinoma In Situ (DCIS) of the Breast - What's Wrong with It? Pathol Oncol Res. 2020;26(2):665–71. Toss MS, Miligy I, Al-Kawaz A, Alsleem M, Khout H, Rida PC, et al. Prognostic significance of tumor-infiltrating lymphocytes in ductal carcinoma in situ of the breast. Mod Pathol. 2018;31(8):1226–36. Fisher B, Dignam J, Wolmark N, Mamounas E, Costantino J, Poller W, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol. 1998;16(2):441–52. Bijker N, Peterse JL, Duchateau L, Julien JP, Fentiman IS, Duval C, et al. Risk factors for recurrence and metastasis after breast-conserving therapy for ductal carcinoma-in-situ: analysis of European Organization for Research and Treatment of Cancer Trial 10853. J Clin Oncol. 2001;19(8):2263–71. Rakovitch E, Nofech-Mozes S, Hanna W, Narod S, Thiruchelvam D, Saskin R, et al. HER2/neu and Ki-67 expression predict non-invasive recurrence following breast-conserving therapy for ductal carcinoma in situ. Br J Cancer. 2012;106(6):1160–5. Gorringe KL, Fox SB. Ductal Carcinoma In Situ Biology, Biomarkers, and Diagnosis. Front Oncol. 2017;7:248. Correa C, McGale P, Taylor C, Wang Y, Clarke M, Davies C, et al. Overview of the randomized trials of radiotherapy in ductal carcinoma in situ of the breast. J Natl Cancer Inst Monogr. 2010;2010(41):162–77. Staley H, McCallum I, Bruce J. Postoperative Tamoxifen for ductal carcinoma in situ: Cochrane systematic review and meta-analysis. Breast. 2014;23(5):546–51. Houghton J, George WD, Cuzick J, Duggan C, Fentiman IS, Spittle M. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet. 2003;362(9378):95–102. Kaushik S, Pickup MW, Weaver VM. From transformation to metastasis: deconstructing the extracellular matrix in breast cancer. Cancer Metastasis Rev. 2016;35(4):655–67. Galbavý S, Lukác L, Porubský J, Cerná M, Labuda M, Kmet'ová J, et al. Collagen type IV in epithelial tumours of colon. Acta Histochem. 2002;104(4):331–4. Kusunoki T, Nishida S, Kimoto-Kinoshita S, Murata K, Satou T, Tomura T. Type IV collagenase and immunostaining of type IV collagen in human thyroid tumors. Auris Nasus Larynx. 2000;27(2):161–5. Imamura T, Manabe T, Ohshio G, Wang ZH, Yamaki K, Yoshimura T, et al. Immunohistochemical staining for type IV collagen and laminin in the stroma of human pancreatic cancer. Int J Pancreatol. 1995;18(2):95–9. Jansson M, Lindberg J, Rask G, Svensson J, Billing O, Nazemroaya A, et al. Prognostic Value of Stromal Type IV Collagen Expression in Small Invasive Breast Cancers. Frontiers in Molecular Biosciences. 2022;9. Lindgren M, Jansson M, Tavelin B, Dirix L, Vermeulen P, Nyström H. Type IV collagen as a potential biomarker of metastatic breast cancer. Clin Exp Metastasis. 2021;38(2):175–85. Nyström H, Naredi P, Hafström L, Sund M. Type IV collagen as a tumour marker for colorectal liver metastases. Eur J Surg Oncol. 2011;37(7):611–7. Fatherree JP, Guarin JR, McGinn RA, Naber SP, Oudin MJ. Chemotherapy-Induced Collagen IV Drives Cancer Cell Motility through Activation of Src and Focal Adhesion Kinase. Cancer Res. 2022;82(10):2031–44. Wadsten C, Garmo H, Fredriksson I, Sund M, Warnberg F. Risk of death from breast cancer after treatment for ductal carcinoma in situ. Br J Surg. 2017;104(11):1506–13. Rask G, Wadsten C, Acs B, Hartman J, Fredriksson I, Garmo H, et al. Immune cell infiltrate in ductal carcinoma in situ and the risk of dying from breast cancer: case-control study. Br J Surg. 2024;111(2). Sebesta M, Egelberg P, Langberg A, Lindskov J-H, Alm K, Janicke B. HoloMonitor M4: holographic imaging cytometer for real-time kinetic label-free live-cell analysis of adherent cells: SPIE; 2016. Holland R, Peterse JL, Millis RR, Eusebi V, Faverly D, van de Vijver MJ, et al. Ductal carcinoma in situ: a proposal for a new classification. Semin Diagn Pathol. 1994;11(3):167–80. Nationellt vårdprogram bröstcancer - RCC Kunskapsbanken 2023 [Available from: https://kunskapsbanken.cancercentrum.se/diagnoser/brostcancer/vardprogram/ . Dieci MV, Radosevic-Robin N, Fineberg S, van den Eynden G, Ternes N, Penault-Llorca F, et al. Update on tumor-infiltrating lymphocytes (TILs) in breast cancer, including recommendations to assess TILs in residual disease after neoadjuvant therapy and in carcinoma in situ: A report of the International Immuno-Oncology Biomarker Working Group on Breast Cancer. Seminars in cancer biology. 2018;52(Pt 2):16–25. Valkenburg KC, de Groot AE, Pienta KJ. Targeting the tumour stroma to improve cancer therapy. Nat Rev Clin Oncol. 2018;15(6):366–81. Naba A, Clauser KR, Lamar JM, Carr SA, Hynes RO. Extracellular matrix signatures of human mammary carcinoma identify novel metastasis promoters. eLife. 2014;3:e01308. Gudjonsson T, Rønnov-Jessen L, Villadsen R, Rank F, Bissell MJ, Petersen OW. Normal and tumor-derived myoepithelial cells differ in their ability to interact with luminal breast epithelial cells for polarity and basement membrane deposition. J Cell Sci. 2002;115(Pt 1):39–50. Hilson JB, Schnitt SJ, Collins LC. Phenotypic alterations in ductal carcinoma in situ-associated myoepithelial cells: biologic and diagnostic implications. Am J Surg Pathol. 2009;33(2):227–32. Allinen M, Beroukhim R, Cai L, Brennan C, Lahti-Domenici J, Huang H, et al. Molecular characterization of the tumor microenvironment in breast cancer. Cancer Cell. 2004;6(1):17–32. Delage S, Zadhoosh A, You W, Brown TJ, Ringuette MJ. Drosophila SPARC collagen IV chaperone-like activity essential for development is unique to the fat body. iScience. 2025;28(4):112111. Strell C, Paulsson J, Jin SB, Tobin NP, Mezheyeuski A, Roswall P, et al. Impact of Epithelial-Stromal Interactions on Peritumoral Fibroblasts in Ductal Carcinoma in Situ. J Natl Cancer Inst. 2019;111(9):983–95. JingSong H, Hong G, Yang J, Duo Z, Li F, WeiCai C, et al. siRNA-mediated suppression of collagen type iv alpha 2 (COL4A2) mRNA inhibits triple-negative breast cancer cell proliferation and migration. Oncotarget. 2017;8(2):2585–93. Castro-Sanchez L, Soto-Guzman A, Navarro-Tito N, Martinez-Orozco R, Salazar EP. Native type IV collagen induces cell migration through a CD9 and DDR1-dependent pathway in MDA-MB-231 breast cancer cells. European Journal of Cell Biology. 2010;89(11):843–52. Espinosa Neira R, Salazar EP. Native type IV collagen induces an epithelial to mesenchymal transition-like process in mammary epithelial cells MCF10A. The International Journal of Biochemistry & Cell Biology. 2012;44(12):2194–203. Clark AG, Vignjevic DM. Modes of cancer cell invasion and the role of the microenvironment. Curr Opin Cell Biol. 2015;36:13–22. Additional Declarations No competing interests reported. Supplementary Files REMARKchecklistcollagenIV.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 18 Jul, 2025 Reviews received at journal 12 Jul, 2025 Reviews received at journal 30 Jun, 2025 Reviews received at journal 23 Jun, 2025 Reviewers agreed at journal 23 Jun, 2025 Reviewers agreed at journal 23 Jun, 2025 Reviewers agreed at journal 17 Jun, 2025 Reviewers invited by journal 16 Jun, 2025 Editor assigned by journal 27 May, 2025 Submission checks completed at journal 22 May, 2025 First submitted to journal 21 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6714442","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":472433049,"identity":"66bb0875-8305-405c-b8d6-974fa2c40724","order_by":0,"name":"Gunilla Rask","email":"data:image/png;base64,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","orcid":"","institution":"Umeå University","correspondingAuthor":true,"prefix":"","firstName":"Gunilla","middleName":"","lastName":"Rask","suffix":""},{"id":472433051,"identity":"d848bb66-6eee-478c-a3e9-bb6d38086bb0","order_by":1,"name":"Malin Jansson","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Malin","middleName":"","lastName":"Jansson","suffix":""},{"id":472433053,"identity":"95dc0acb-ae4a-4ac1-836d-e8b369f90b76","order_by":2,"name":"Johan Svensson","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Johan","middleName":"","lastName":"Svensson","suffix":""},{"id":472433056,"identity":"4fea9529-a720-4304-9ced-dbf634bd0b8d","order_by":3,"name":"Rebecca Wiberg","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Rebecca","middleName":"","lastName":"Wiberg","suffix":""},{"id":472433057,"identity":"05ba81a3-9909-4a03-b521-6f3e445b142f","order_by":4,"name":"Fredrik Wärnberg","email":"","orcid":"","institution":"Sahlgrenska University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Fredrik","middleName":"","lastName":"Wärnberg","suffix":""},{"id":472433059,"identity":"a509e1c0-6c27-43f5-966d-400f16138ff5","order_by":5,"name":"Ola Billing","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Ola","middleName":"","lastName":"Billing","suffix":""},{"id":472433061,"identity":"8c1675c7-d8a9-4635-88b4-029e13330390","order_by":6,"name":"Charlotta Wadsten","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Charlotta","middleName":"","lastName":"Wadsten","suffix":""},{"id":472433062,"identity":"cf3afdb5-c690-439a-aaf9-15fd31693481","order_by":7,"name":"Malin Sund","email":"","orcid":"","institution":"Umeå University","correspondingAuthor":false,"prefix":"","firstName":"Malin","middleName":"","lastName":"Sund","suffix":""}],"badges":[],"createdAt":"2025-05-21 08:38:33","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6714442/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6714442/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84883252,"identity":"eaee7364-7920-48d0-a99d-b43ffda39ecc","added_by":"auto","created_at":"2025-06-18 11:15:52","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":952543,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSelection of the study cohort\u003cbr\u003e\n\u003c/strong\u003e\u003cem\u003eDCIS – ductal carcinoma in situ, TMA – tissue micro array, BC - breast cancer\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6714442/v1/2a1272bea7f3074c10d6e812.png"},{"id":84882079,"identity":"df280c31-b456-4993-bfe7-a7e67f3f60df","added_by":"auto","created_at":"2025-06-18 11:07:52","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":8382190,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExamples of collagen IV staining\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eOpen arrow – periductal collagen IV, filled arrows – normal collagen IV in blood vessels\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6714442/v1/87be970031f237fa589cf56a.png"},{"id":84883251,"identity":"ea910eee-57dd-4037-b1e7-9a2eeaddad9a","added_by":"auto","created_at":"2025-06-18 11:15:52","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":548501,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMigration speed per cell type with and without collagen IV coating.\u003c/strong\u003e\u003cbr\u003e\n\u003cem\u003eMean speed of all cells of each subtype between each consecutive recorded image.\u003cbr\u003e\nER – oestrogen receptor; HER2 – human epidermal growth factor receptor 2; Asterisk (*) denotes significantly different mean average migration speed (p\u0026lt;0.001); n.s. – not statistically significant\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6714442/v1/a36eb2ff7573cc1f5cac4ebf.png"},{"id":84884214,"identity":"024ead82-147a-4ddf-8846-5aefccc6397a","added_by":"auto","created_at":"2025-06-18 11:23:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":9573915,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6714442/v1/be046f73-73b7-4923-b9c7-39fef9f5af36.pdf"},{"id":84882069,"identity":"456058fc-d8bc-4d08-b965-fcda95dabb18","added_by":"auto","created_at":"2025-06-18 11:07:52","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":23808,"visible":true,"origin":"","legend":"","description":"","filename":"REMARKchecklistcollagenIV.docx","url":"https://assets-eu.researchsquare.com/files/rs-6714442/v1/dc74a5f986c7394b4a091939.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Stromal collagen IV expression and risk of breast cancer death in ductal carcinoma in situ","fulltext":[{"header":"Background","content":"\u003cp\u003eAs screening programmes are widely implemented and radiological methods become ever more refined, the proportion of women with breast cancer that are diagnosed with ductal carcinoma in situ (DCIS) is increasing and now comprises about 15% of new breast cancers (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Current treatment of DCIS is surgery with or without radiotherapy, with possible addition of endocrine treatment for women with oestrogen receptor positive (ER+) DCIS. As survival after DCIS treatment is excellent, there is rising concern that several women are overtreated, meaning that they would not have developed any symptomatic invasive carcinoma if left untreated (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Methods for robust risk stratification of patients is therefore in high demand.\u003c/p\u003e \u003cp\u003eTraditional histological factors, and those recommended in the EUSOMA (European Society of Breast Cancer specialists) guidelines are lesion size, nuclear grade, growth pattern, presence of comedo necrosis, calcifications and expression of ER (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Nuclear grade is at best however moderately reproducible between pathologists (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) and that there are conflicting results as to whether grade, comedo necrosis, histological pattern and ER-status are related to patient outcome (\u003cspan additionalcitationids=\"CR6 CR7 CR8\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Assuming that old or novel biomarkers may be used to predict risk begs the question of which risk specifically. Is it risk of local DCIS recurrence, of invasive recurrence or of breast cancer death? All treatments come with side effects, and the benefit of risk reduction must be substantial enough to motivate treatment. Features which predict DCIS recurrence are not the same as those that predict recurrence as invasive disease (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Invasive recurrence increases risk of breast cancer death, and the addition of radiotherapy is routinely used to lower risk of local recurrence after breast conserving surgery. However, neither radiotherapy nor addition of endocrine therapy have been shown to affect the risk of breast cancer death (\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo refine the traditional risk stratification of DCIS there is a need for new biomarkers that are related to the risk of distant metastasis and breast cancer death. Surrounding the DCIS cells is the myoepithelium, resting in turn on the basement membrane (BM) which is composed almost entirely of collagen IV. Outside of this structure is the stroma, normally consisting of many other types of collagens and smaller amounts of other molecules (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Cancer-associated stroma, however, often contains collagen IV also beyond the BM (\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Our group has previously linked this stromal collagen IV expression to an increased risk of distant metastasis and worse survival (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Circulating collagen IV has also been linked to an increased risk of distant metastasis and worse survival in both breast (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e) and colorectal (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e) cancer patients. A previous study using mouse and \u003cem\u003ein vitro\u003c/em\u003e models demonstrated that breast cancer cells of a predominantly triple-negative phenotype invade quicker and further in collagen IV-rich decellularized matrix scaffolds compared to scaffolds with low collagen IV-content (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis study aimed to determine 1) if stromal collagen IV-expression in DCIS correlated with risk of breast cancer death in a case-control cohort, and 2) if collagen IV could affect the motility of breast cancer cell lines in vitro.\u003c/p\u003e"},{"header":"Materials and method","content":"\u003cp\u003ePatient cohort and TMA-contruction\u003c/p\u003e \u003cp\u003eWe have previously described a nested case-control cohort of women with DCIS with clinicopathological data available (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Briefly, the cancer registries of three health care regions in Sweden were used to identify women treated for DCIS between 1992 and 2012 (n\u0026thinsp;=\u0026thinsp;6 964). Cross-linking with the National Cause of Death Registry identified women who later died from breast cancer (cases, n\u0026thinsp;=\u0026thinsp;95). Controls were selected using incidence density sampling, from the whole population of women with DCIS. Controls had to be alive and without distant metastasis of breast cancer at the time of death of the corresponding case (n\u0026thinsp;=\u0026thinsp;318). Clinical data was gathered from medical records. In this cohort, tumour size larger than 25 mm and positive margin status were associated with increased risk of breast cancer death, while surgical treatment and radiotherapy were not. From the surgical specimens of the women in the cohort, tumour tissue was obtained for whole slide sectioning and construction of tissue microarrays (TMA) for 65 of the 95 cases (68%) and 195 of 318 controls (61%). Whole slides and TMA-slides were centrally evaluated by a specialized breast pathologist (G.R) for DCIS grade, presence of comedo necrosis, amount of tumour infiltrating lymphocytes (TILs), as well as expression of ER, progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and Ki67 (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). For the present study there was sufficient tumour tissue in the TMAs for analysis of collagen IV in 43 cases and 119 controls. (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e)\u003c/p\u003e \u003cp\u003eImmunohistochemistry for collagen IV\u003c/p\u003e \u003cp\u003eTissue sections were stained with rabbit anti-collagen IV (AB748, Millipore, Billerica, United States) diluted 1:50 in blocking buffer for 2 hours in room temperature. Sections were then washed in phosphate buffered saline (PBS) and incubated with the biotinylated secondary antibody, diluted 1:200 in blocking buffer, for 30 minutes in room temperature. Finally, slides were washed in PBS followed by addition of diaminobenzidine tetrahydrochloride (DAB) as chromogen.\u003c/p\u003e \u003cp\u003eExpression of collagen IV was scored separately in the compartment closest to the myoepithelial/stromal interface, defined as within 10 \u0026micro;m of the interface and dubbed periductal collagen IV, and in the rest of the stroma, henceforth called stromal collagen IV. Each compartment was scored on a scale 0\u0026ndash;3 (0\u0026thinsp;=\u0026thinsp;no, 1\u0026thinsp;=\u0026thinsp;minimal, 2\u0026thinsp;=\u0026thinsp;moderate and 3\u0026thinsp;=\u0026thinsp;strong expression). Scoring was done by two independent researchers (G.R and M.J), blinded to outcome at the time of scoring. Discrepant cases were reviewed jointly, and a consensus score was set. There were one to six biopsies for each patient in the TMA, and in cases were the expression of collagen IV within a compartment differed between biopsies the maximum score was used for the final analyses.\u003c/p\u003e \u003cp\u003eCell cultures and migration experiments\u003c/p\u003e \u003cp\u003eThree different breast cancer cell lines were used. An ER+/HER2-cell line (MCF-7; ATCC HTB-22), an ER-/HER2- cell line (MDA-MB-231; ATCC HTB-26) and a ER-/HER2\u0026thinsp;+\u0026thinsp;cell line (JIMT-1; RRID:CVCL_2077). Cells were cultured in Dulbecco\u0026rsquo;s Modified Eagle\u0026rsquo;s Medium (DMEM) 4,5% glucose with 10% foetal bovine serum (FBS) and 1% penicillin/streptomycin. For migration experiments 96-well plates were coated with either DMEM or collagen IV (ab7536 Abcam; 1 mg/mL) diluted in DMEM to a concentration of 29 ug/mL. Cells were seeded at a density of 5000 cells/well (MCF-7), or 2500 cells/well (MDA-MB-231 and JIMT-1). After seeding the cells were left to adhere for 4\u0026ndash;6 hours, then washed with PBS. New DMEM was added, and the plates were inserted to the Holomonitor M4, which uses quantitative phase imaging to automatically detect and measure live cells without labelling.(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) We used the adaptive Gaussian algorithm of detection to minimize background noise from the coating. Images were captured every 15 minutes for 24 hours. In each experiment, six wells per breast cancer cell line were used, three with and three without collagen IV, respectively. Three fields of view were analysed for each well, each field of view containing approximately 50\u0026ndash;200 cells. The speed of individual cells between each consecutive captured image frame (distance travelled/time between frames) was calculated by the built-in software of the HoloMonitor, as well as the average speed of all cells of each type between consecutive frames. The experiment was repeated thrice and the values of the resulting averages for all experiments were used for the final analyses. After all experiments were finished, all three cell lines tested negative for potential contamination of various Mycoplasma species.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eOdds ratios (OR) for dying of breast cancer in the nested case control cohort were calculated using conditional logistic regression and all analyses adjusted for time at risk. The following clinical variables were used: Age, tumour size (\u0026lt;\u0026thinsp;25 mm or \u0026ge;\u0026thinsp;25 mm, as in previous studies on this cohort (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e)), margin status (negative vs positive, uncertain or missing), mode of detection (screening vs non-screening) and surgical treatment (breast conserving therapy only (BCS), BCS with radiotherapy (BCS\u0026thinsp;+\u0026thinsp;RT) or mastectomy). Moreover the following biomarkers were used: Nuclear grade according to Holland (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e), comedo necrosis (absent vs. present) and ER and PR expression (positive if\u0026thinsp;\u0026ge;\u0026thinsp;10% of nuclei stained, in keeping with national guidelines for invasive carcinoma) (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). Periductal TILs were categorized as 0\u0026ndash;5%, 10% or \u0026ge;\u0026thinsp;20% according to international guidelines (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). The expression of both stromal and periductal collagen IV as described above was simplified to a two-grade scale for the final analyses with 0\u0026ndash;1 termed \u0026ldquo;low\u0026rdquo; and 2\u0026ndash;3 termed \u0026ldquo;high\u0026rdquo;. Pearson Chi-square was used to test for independence between variables. The test for interaction between ER-status and outcome was conducted by adding an interaction term in the conditional logistic regression model.\u003c/p\u003e \u003cp\u003eLoss of material in sectioning resulted in missing values of the case response variable, making some strata entirely unused. To maximise data use, strata with cases treated at similar timepoints were fused. This gave a total of 32 strata, compared to the original 66. For the adjusted analyses, missing values for variables other than either collagen IV variable were imputed using multiple imputation in 20 datasets estimated with chained equations. For the imputation model, the variables in the conditional logistic regression models were used, including the responses periductal collagen IV and stromal collagen IV as well as time at risk and year of diagnosis. Proportion of imputed values were 16% for tumour size, 6% for ER-status and less than 5% for all other variables.\u003c/p\u003e \u003cp\u003eFor comparison of means of average migration speed in the in vitro experiments, independent samples t-test was used.\u003c/p\u003e \u003cp\u003eConditional logistic regression analyses, imputation and tests for interaction were performed in StataIC15.1 and all other analyses in SPSS\u0026reg; version 23 (IBM, Armonk, NY, USA).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eStudy population\u003c/p\u003e\n\u003cp\u003eMedian age at diagnosis was 54 years, with no significant difference between cases and controls. A majority of the women (63.6%) were diagnosed within the breast cancer screening programme. Most were treated with breast conserving surgery (BCS) with or without subsequent radiotherapy (27.8% and 32.7% respectively) while 39.5% underwent mastectomy. The proportion of tumours\u0026thinsp;\u0026lt;\u0026thinsp;25 mm was approximately equal to those\u0026thinsp;\u0026ge;\u0026thinsp;25 mm (43.8% and 40.1%, with missing data on 16%). Approximately 10% of the surgical specimens had resection margins that were either positive or uncertain.\u003c/p\u003e\n\u003cp\u003eRisk factors for breast cancer death\u003c/p\u003e\n\u003cp\u003eSeveral characteristics associated with increased risk of breast cancer death in the unadjusted analysis: Tumour size\u0026thinsp;\u0026ge;\u0026thinsp;25 mm (OR 3.48; 95% CI 1.42\u0026ndash;8.56), presence of comedo necrosis (OR 2.68; 95% CI 1.10\u0026ndash;6.53), negative PR expression (OR 2.45; 95% CI 1.08\u0026ndash;5.57), and high expression of stromal collagen IV (OR 2.50; 95% CI 1.16\u0026ndash;5.39).\u003c/p\u003e\n\u003cp\u003eNeither grade nor ER-expression had a statistically significant correlation with risk of breast cancer death. Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e summarizes the characteristics of the study cohort and ORs for breast cancer death.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eClinicopathological factors and odds ratios for breast cancer death after treatment for ductal carcinoma in situ.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eStudy cohort (n\u0026thinsp;=\u0026thinsp;162)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCases\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;43)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eControls\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;119)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOR (95% CI) for breast cancer death\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e yrs, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54 (48\u0026ndash;62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e52 (46\u0026ndash;62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e54 (49\u0026ndash;62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.99 (0.96\u0026ndash;1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eTumour size*\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;25 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e71 (43.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11 (25.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e60 (50.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026ge;\u0026thinsp;25 mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e65 (40.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e24 (55.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41 (34.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.48 (1.42\u0026ndash;8.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e26 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (18.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18 (15.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eMargin status\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNegative\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e145 (89.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34 (79.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e111 (93.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePositive/uncertain/missing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17 (10.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9 (20.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (6.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.72 (0.94\u0026ndash;7.87)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eMode of detection\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eScreening\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e103 (63.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22 (51.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e81 (68.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-screening\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50 (30.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14 (32.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e36 (30.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.50 (0.68\u0026ndash;3.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9 (5.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7 (16.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eBreast surgery\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBCS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45 (27.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (27.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e33 (27.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBCS\u0026thinsp;+\u0026thinsp;RT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e53 (32.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (27.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41 (34.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.86 (0.33\u0026ndash;2.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMastectomy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e64 (39.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19 (44.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e45 (37.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.31 (0.55\u0026ndash;3.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eGrade\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e26 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (11.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 (17.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e78 (48.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e23 (53.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.84 (0.63\u0026ndash;5.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55 (34..0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e14 (32.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41 (34.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.58 (0.49\u0026ndash;5.10)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (1.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (2.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eComedonecrosis*\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAbsent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e51 (31.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (18.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e43 (36.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePresent\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e108 (66.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e34 (79.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e74 (62.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.68 (1.10\u0026ndash;6.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (1.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (2.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eER\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026ge;\u0026thinsp;10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e112 (69.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e27 (62.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e85 (71.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41 (25.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (30.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e28 (23.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.63 (0.73\u0026ndash;3.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9 (5.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (7.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (5.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePR*\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026ge;\u0026thinsp;10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e93 (57.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20 (46.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e73 (61.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e63 (38.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e21 (48.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e42 (35.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.45 (1.08\u0026ndash;5.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (3.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (4.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4 (3.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePeriductal TILs\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0\u0026ndash;5%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e98 (60.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e23 (53.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e75 (63.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e31 (19.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (\u003cspan class=\"CitationRef\"\u003e14\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25 (\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.83 (0.30\u0026ndash;2.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026ge;\u0026thinsp;20%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e30 (18.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13 (30.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17 (14.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.34 (0.95\u0026ndash;5.76)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3 (1.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (2.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStromal collagen IV*\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLow (0\u0026ndash;1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e110 (67.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e23 (53.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e87 (73.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHigh (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e50 (30.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e19 (44.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e31 (26.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.50 (1.16\u0026ndash;5.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMissing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2 (1.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (2.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1 (0.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePeriductal collagen IV\u003c/strong\u003e n (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLow (0\u0026ndash;1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e81 (50.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17 (39.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e64 (53.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHigh (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e81 (50.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e26 (60.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e55 (46.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.00 (0.96\u0026ndash;4.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cem\u003eOR \u0026ndash; odds ratio; ER \u0026ndash; oestrogen receptor; PR \u0026ndash; progesterone receptor; TILs \u0026ndash; tumour infiltrating lymphocytes. * denotes variables with statistically significant difference between cases and controls. a \u0026ndash; all analyses adjusted for time at risk\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn the present cohort, a positive or uncertain margin status only showed a non-significant trend towards increased risk for breast cancer-related death (OR 2.72; 95% CI 0.94\u0026ndash;7.87). The dropout analysis showed fewer small (\u0026lt;\u0026thinsp;25 mm) tumours remaining in the present cohort compared to tumours not analysed for collagen IV (43.8% vs 70.2%). The remaining cases were in all other respects similar to the cases that were lost. By contrast, the remaining controls were less often screening detected (68.1% vs 76.1%), had fewer negative surgical margins (93.3% vs 98.9%), more often had a suspicion of microinvasion (7.6% vs 0.99%) and were more often treated by mastectomy (37.8% vs 25.1%) than the controls that could not be analysed for collagen IV. Comparisons between the original cohort and the present one are summarised in supplementary table \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e (included at the bottom of this paper)\u003c/p\u003e\n\u003cp\u003eCollagen IV expression and other clinicopathological variables\u003c/p\u003e\n\u003cp\u003eInterobserver agreement between the two assessors for collagen IV expression in the TMA-biopsies, when using the 0 to 3 scale, was moderate for both periductal (kappa 0.52) and stromal (kappa 0.43) expression. Grouping the variables as low (0 or 1) and high (2 or 3) resulted in substantial agreement with kappa 0.65 for stromal and 0.76 for periductal expression. Representative images of TMA-biopsies with different expressions of collagen IV are shown in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003cp\u003eThere was an association between expression of periductal and stromal collagen (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) i.e. they were not independent variables. There was no dependency between collagen IV expression and ER or PR expression. There was a borderline significant correlation between comedo necrosis and stromal collagen IV (p\u0026thinsp;=\u0026thinsp;0.06), where DCIS with necrosis was more likely to have a low stromal collagen expression (73% vs 58%). Necrosis did not correlate with periductal collagen IV (p\u0026thinsp;=\u0026thinsp;0.73). There was no statistically significant association between stromal collagen IV and tumour size (p\u0026thinsp;=\u0026thinsp;0.08). However, smaller tumours (\u0026lt;\u0026thinsp;25 mm) trended towards higher stromal collagen IV expression (39% vs. 20%).\u003c/p\u003e\n\u003cp\u003eCollagen IV expression predicts breast cancer-related death in adjusted models.\u003c/p\u003e\n\u003cp\u003eIn the previous unadjusted analysis, high stromal collagen IV expression correlated with increased risk of dying of breast cancer. This association remained significant when adjusting for tumour size and margin status (OR 3.62; 95% CI 1.45-9.00), and even after adjusting for presence of comedonecrosis and PR negativity (OR 4.27; 95% CI 1.64\u0026ndash;11.1). ER expression did not interact with stromal collagen IV for the risk of breast cancer death (p\u0026thinsp;=\u0026thinsp;0.19). Periductal expression of collagen IV was not statistically significantly associated with increased risk of breast cancer death in the unadjusted analysis (OR 2.00; 95%CI 0.96\u0026ndash;4.15). However, it showed a similar association as stromal collagen IV when adjusting for tumour size and margin status (OR 2.50; 95%CI 1.10\u0026ndash;5.64) as well as with further adjusting for comedonecrosis and PR (OR 2.55; 95% CI1.11-5.86). The results are summarized in Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003cdiv class=\"colspec\" align=\"left\"\u003e\u0026nbsp;\u003c/div\u003e\n \u003ctable id=\"Tab3\" style=\"width: 774px;\" border=\"1\"\u003e\n \u003ccaption\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eCollagen IV expression and risk of breast cancer death\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth style=\"width: 10px;\" align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth style=\"width: 740px;\" colspan=\"3\" align=\"left\"\u003e\n \u003cp\u003eOR for breast cancer death (95% CI)*\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\n \u003cp\u003eUnadjusted\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\n \u003cp\u003eAdjusted for tumour size and margin status\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\n \u003cp\u003eAdjusted for tumour size, margin, PR and necrosis\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStromal collagen IV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003eLow (0\u0026ndash;1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003eHigh (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\n \u003cp\u003e2.50 (1.16\u0026ndash;5.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\n \u003cp\u003e3.62 (1.45-9.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\n \u003cp\u003eOR 4.27 (1.64\u0026ndash;11.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePeriductal collagen IV\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003eLow (0\u0026ndash;1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\n \u003cp\u003e1.0 (ref.)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 10px;\" align=\"left\"\u003e\n \u003cp\u003eHigh (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 221px;\" align=\"left\"\u003e\n \u003cp\u003e2.00 (0.96\u0026ndash;4.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 241px;\" align=\"left\"\u003e\n \u003cp\u003e2.50 (1.10\u0026ndash;5.64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 278px;\" align=\"left\"\u003e\n \u003cp\u003e2.55 (1.11\u0026ndash;5.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eOR \u0026ndash; odds ratio; CI \u0026ndash; confidence interval; PR \u0026ndash; progesterone receptor\u003c/p\u003e\n\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003cp\u003e* All analyses adjusted for time at risk\u003c/p\u003e\n \u003cp\u003eCollagen IV and cell migration\u003c/p\u003e\n \u003cp\u003eAs collagen IV has been shown to increase motility \u003cem\u003ein vitro\u003c/em\u003e only in triple negative but not other types of breast cancer cells, experiments were performed using triple negative, ER+/HER2- and ER-/HER2\u0026thinsp;+\u0026thinsp;cell lines. By light microscopy, the cells displayed different phenotypes. The triple negative cells (MDA-MB-231) were elongated and mesenchymal-like, growing evenly dispersed across the plates. The ER\u0026thinsp;+\u0026thinsp;and ER-/HER2\u0026thinsp;+\u0026thinsp;cells (MCF-7 and JIMT-1) were rounder, plumper and tended to grow in aggregates rather than spreading out. The appearances were the same in both collagen IV-coated and uncoated plates. Effects of collagen IV-coating on cell migration differed between cell types (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Collagen IV coating resulted in an increased average migration speed in triple negative cells (5.7 vs. 7.7 \u0026micro;m/h, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), whereas the opposite was observed for ER+/HER2- cells (7.8 vs. 6.6 \u0026micro;m/h, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). No effect on migration speed was seen in the ER-/HER2\u0026thinsp;+\u0026thinsp;cell line (7.3 vs. 7.1 \u0026micro;m/h, p\u0026thinsp;=\u0026thinsp;0.68).\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eStromal components and other factors related to the tumour microenvironment are gaining increasing attention as potential prognostic markers and treatment targets in many cancer types (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). We demonstrate for the first time that high levels of collagen IV, both in the stromal and periductal compartment in DCIS correlate with increased odds of breast cancer death. This correlation became even more pronounced when adjusting for established clinical risk factors, such as tumour size and positive margins, as well as tentative histological risk factors such as comedonecrosis and PR-negativity. Moreover, we confirm previous findings that collagen IV promotes migration of triple negative breast cancer cells(\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e), but the same was not shown for either ER+/HER2- or ER-/HER2\u0026thinsp;+\u0026thinsp;breast cancer cells.\u003c/p\u003e \u003cp\u003eIn the breast, collagen IV is normally secreted by the myoepithelial cells (MECs), but in invasive breast cancer the collagen IV in the extracellular matrix (ECM) can originate from the cancer cells themselves, the surrounding stromal cells or both (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). It is thus conceivable that the periductal and stromal collagen IV in DCIS may be produced by the DCIS cells, the myoepithelial cells, the stromal cells or any combination of these. Production by the DCIS cells seems less likely, since the MEC layer limits physical access to both periductal and stromal compartments. Considering the MECs, it is known that tumour associated MECs differ in many ways from native MECs in gene expression patterns and phenotype (\u003cspan additionalcitationids=\"CR30\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). To the best of our knowledge though, an increased production of collagen IV by cancer associated MECs has not yet been demonstrated, and this would be worth to explore in future studies.\u003c/p\u003e \u003cp\u003eThat at least part of the collagen IV is produced by stromal cells is highly likely. Collagen IV self-polymerizes and cannot diffuse easily (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e) but our analyses show stromal collagen IV expression as far as one millimetre from the DCIS (the limit of the diameter of our TMA-cores). If it is produced in the stroma, it also means that direct contact between DCIS cells and collagen IV-producing cells is not feasible as the only mode of communication. Either the signal needs to be endocrine rather than paracrine, or the DCIS cells need to inform the stromal cells to feed the signal forward. The stronger correlation between stromal collagen IV and breast cancer death compared to periductal collagen IV in both crude and adjusted analyses, is in keeping with either notion. It has been shown that the DCIS-induced PDGFRα\u003csup\u003e(low)\u003c/sup\u003e/PDGFRβ\u003csup\u003e(high)\u003c/sup\u003e phenotype of stromal fibroblasts is dependent on direct cell-to cell or paracrine signalling (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e); investigating the levels of collagen IV in healthy breast tissue several centimetres away from the DCIS could show if the same is true for collagen IV-expression.\u003c/p\u003e \u003cp\u003eWhy increased collagen IV would lead to increased risk of breast cancer death remains unclear. Almost all in vitro or animal models exploring the effects of collagen IV have used triple negative cell lines (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e), and all these show increased migration in response to collagen IV exposure. One study also found similar results in non-neoplastic mammary epithelial cells (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e) but effects on ER\u0026thinsp;+\u0026thinsp;or HER2\u0026thinsp;+\u0026thinsp;breast cancer cells have not been reported previously. Surprisingly, we found that a collagen IV coating decreased the migration speed of ER+/HER2- cells. This finding needs to be validated in additional cell lines and in other models. Cells may migrate by different means, either as single cells without any cell-to cell interaction, or as loosely or non-cohesive cell groups that move along the same path (streaming), or as cohesive units along a broad front (collective migration)(\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). One explanation to our findings may be that the mode of migration differs between the two cell types and that collagen IV, when used as coating, mainly increases one or two of these migration modes. Triple negative breast cancer cells (MDA-MB-231) migrate by single cell or streaming (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e), while ER\u0026thinsp;+\u0026thinsp;cells (MCF7) are less investigated also in this respect. We found no effect of collagen IV coating on the migration of ER-/HER2\u0026thinsp;+\u0026thinsp;cells when looking at the sum of all experiments. This cell line was, however, somewhat more difficult to get good readings of in the HoloMonitor, which might have been due to the conformation of the cells or how they affected the coating during incubation, which is a limitation of these experiments.\u003c/p\u003e \u003cp\u003eOther limitations of the present study include the use of TMA rather than whole tissue sections since the field of vision is limited to stroma within 1 mm from the DCIS. This prevents observation of collagen IV in healthy tissue further from the DCIS as well as observations regarding spatial heterogeneity in the tumour stroma. The case-control cohort is small in numbers, preventing further subgroup analysis. This is however almost inevitable when dealing with rare outcomes such as breast cancer death after DCIS, and the cohort is in this respect larger or equal to others.\u003c/p\u003e \u003cp\u003eIn conclusion we show for the first time that stromal collagen IV correlates with risk of breast cancer death already at the in situ -stage of disease. This means that there may be aggressive types of DCIS with the potential to create conditions necessary for metastatic spread even before invasion. We further confirm that collagen IV increases migration of triple negative breast cancer cells, but not necessarily of any other type of breast cancer subtypes. Further studies should focus on collagen IV expression both in healthy tissue of breast cancer patients and cancer-associated MECs, as well as on how collagen IV affects different modes of migration.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnette Berglund – for sectioning, staining and scanning of tissue microarrays.\u003c/p\u003e\n\u003cp\u003eChristina Lundin – for guidance and help with cell cultures and collagen coating.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eG.R – scored collagen IV staining, designed and performed cell culture experiments, processed HoloMonitor output, analysed data, drafted manuscript.\u003c/p\u003e\n\u003cp\u003eM.J – scored collagen IV staining, analysed data, reviseed manuscript.\u003c/p\u003e\n\u003cp\u003eJ.S – analysed data, revised manuscript.\u003c/p\u003e\n\u003cp\u003eR.W – conceived cell culture experiments, revised draft.\u003c/p\u003e\n\u003cp\u003eF.W – conceived original cohort design, revised draft.\u003c/p\u003e\n\u003cp\u003eO.B – designed and performed cell culture experiments, revised draft.\u003c/p\u003e\n\u003cp\u003eC.W – acquired original cohort data and biomaterial, revised draft.\u003c/p\u003e\n\u003cp\u003eM.S – conceived study design, revised draft.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEthical approval was granted from the regional ethics committee in Umeå (2014/45- 31). The need for informed consent was waived. The study was performed in accordance with the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003cbr\u003e\u003c/strong\u003eThe manuscript does not contain any individual person’s data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003cbr\u003e\u003c/strong\u003eData is not uploaded to a publicly available platform. Researchers have access to data through application to the study PI ([email protected]) or the corresponding author under standard rules of protecting data integrity and existing ethics permissions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003cbr\u003e\u003c/strong\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported through research grants by the Swedish Breast Cancer Association, and VISARE Norr (grant no. 993966). The funding agents had no influence on the content of this manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003evan Seijen M, Lips EH, Thompson AM, Nik-Zainal S, Futreal A, Hwang ES, et al. Ductal carcinoma in situ: to treat or not to treat, that is the question. British Journal of Cancer. 2019;121(4):285\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelaloge S, Khan SA, Wesseling J, Whelan T. Ductal carcinoma in situ of the breast: finding the balance between overtreatment and undertreatment. Lancet. 2024;403(10445):2734\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRubio IT, Marotti L, Biganzoli L, Aristei C, Athanasiou A, Campbell C, et al. EUSOMA quality indicators for non-metastatic breast cancer: An update. European Journal of Cancer. 2024;198.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCserni G, Sejben A. Grading Ductal Carcinoma In Situ (DCIS) of the Breast - What's Wrong with It? Pathol Oncol Res. 2020;26(2):665\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eToss MS, Miligy I, Al-Kawaz A, Alsleem M, Khout H, Rida PC, et al. Prognostic significance of tumor-infiltrating lymphocytes in ductal carcinoma in situ of the breast. Mod Pathol. 2018;31(8):1226\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFisher B, Dignam J, Wolmark N, Mamounas E, Costantino J, Poller W, et al. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol. 1998;16(2):441\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBijker N, Peterse JL, Duchateau L, Julien JP, Fentiman IS, Duval C, et al. Risk factors for recurrence and metastasis after breast-conserving therapy for ductal carcinoma-in-situ: analysis of European Organization for Research and Treatment of Cancer Trial 10853. J Clin Oncol. 2001;19(8):2263\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRakovitch E, Nofech-Mozes S, Hanna W, Narod S, Thiruchelvam D, Saskin R, et al. HER2/neu and Ki-67 expression predict non-invasive recurrence following breast-conserving therapy for ductal carcinoma in situ. Br J Cancer. 2012;106(6):1160\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGorringe KL, Fox SB. Ductal Carcinoma In Situ Biology, Biomarkers, and Diagnosis. Front Oncol. 2017;7:248.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorrea C, McGale P, Taylor C, Wang Y, Clarke M, Davies C, et al. Overview of the randomized trials of radiotherapy in ductal carcinoma in situ of the breast. J Natl Cancer Inst Monogr. 2010;2010(41):162\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStaley H, McCallum I, Bruce J. Postoperative Tamoxifen for ductal carcinoma in situ: Cochrane systematic review and meta-analysis. Breast. 2014;23(5):546\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoughton J, George WD, Cuzick J, Duggan C, Fentiman IS, Spittle M. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet. 2003;362(9378):95\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaushik S, Pickup MW, Weaver VM. From transformation to metastasis: deconstructing the extracellular matrix in breast cancer. Cancer Metastasis Rev. 2016;35(4):655\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGalbav\u0026yacute; S, Luk\u0026aacute;c L, Porubsk\u0026yacute; J, Cern\u0026aacute; M, Labuda M, Kmet'ov\u0026aacute; J, et al. Collagen type IV in epithelial tumours of colon. Acta Histochem. 2002;104(4):331\u0026ndash;4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKusunoki T, Nishida S, Kimoto-Kinoshita S, Murata K, Satou T, Tomura T. Type IV collagenase and immunostaining of type IV collagen in human thyroid tumors. Auris Nasus Larynx. 2000;27(2):161\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eImamura T, Manabe T, Ohshio G, Wang ZH, Yamaki K, Yoshimura T, et al. Immunohistochemical staining for type IV collagen and laminin in the stroma of human pancreatic cancer. Int J Pancreatol. 1995;18(2):95\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJansson M, Lindberg J, Rask G, Svensson J, Billing O, Nazemroaya A, et al. Prognostic Value of Stromal Type IV Collagen Expression in Small Invasive Breast Cancers. Frontiers in Molecular Biosciences. 2022;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindgren M, Jansson M, Tavelin B, Dirix L, Vermeulen P, Nystr\u0026ouml;m H. Type IV collagen as a potential biomarker of metastatic breast cancer. Clin Exp Metastasis. 2021;38(2):175\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNystr\u0026ouml;m H, Naredi P, Hafstr\u0026ouml;m L, Sund M. Type IV collagen as a tumour marker for colorectal liver metastases. Eur J Surg Oncol. 2011;37(7):611\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFatherree JP, Guarin JR, McGinn RA, Naber SP, Oudin MJ. Chemotherapy-Induced Collagen IV Drives Cancer Cell Motility through Activation of Src and Focal Adhesion Kinase. Cancer Res. 2022;82(10):2031\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWadsten C, Garmo H, Fredriksson I, Sund M, Warnberg F. Risk of death from breast cancer after treatment for ductal carcinoma in situ. Br J Surg. 2017;104(11):1506\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRask G, Wadsten C, Acs B, Hartman J, Fredriksson I, Garmo H, et al. Immune cell infiltrate in ductal carcinoma in situ and the risk of dying from breast cancer: case-control study. Br J Surg. 2024;111(2).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSebesta M, Egelberg P, Langberg A, Lindskov J-H, Alm K, Janicke B. HoloMonitor M4: holographic imaging cytometer for real-time kinetic label-free live-cell analysis of adherent cells: SPIE; 2016.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolland R, Peterse JL, Millis RR, Eusebi V, Faverly D, van de Vijver MJ, et al. Ductal carcinoma in situ: a proposal for a new classification. Semin Diagn Pathol. 1994;11(3):167\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNationellt v\u0026aring;rdprogram br\u0026ouml;stcancer - RCC Kunskapsbanken 2023 [Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://kunskapsbanken.cancercentrum.se/diagnoser/brostcancer/vardprogram/\u003c/span\u003e\u003cspan address=\"https://kunskapsbanken.cancercentrum.se/diagnoser/brostcancer/vardprogram/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDieci MV, Radosevic-Robin N, Fineberg S, van den Eynden G, Ternes N, Penault-Llorca F, et al. Update on tumor-infiltrating lymphocytes (TILs) in breast cancer, including recommendations to assess TILs in residual disease after neoadjuvant therapy and in carcinoma in situ: A report of the International Immuno-Oncology Biomarker Working Group on Breast Cancer. Seminars in cancer biology. 2018;52(Pt 2):16\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValkenburg KC, de Groot AE, Pienta KJ. Targeting the tumour stroma to improve cancer therapy. Nat Rev Clin Oncol. 2018;15(6):366\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaba A, Clauser KR, Lamar JM, Carr SA, Hynes RO. Extracellular matrix signatures of human mammary carcinoma identify novel metastasis promoters. eLife. 2014;3:e01308.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGudjonsson T, R\u0026oslash;nnov-Jessen L, Villadsen R, Rank F, Bissell MJ, Petersen OW. Normal and tumor-derived myoepithelial cells differ in their ability to interact with luminal breast epithelial cells for polarity and basement membrane deposition. J Cell Sci. 2002;115(Pt 1):39\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHilson JB, Schnitt SJ, Collins LC. Phenotypic alterations in ductal carcinoma in situ-associated myoepithelial cells: biologic and diagnostic implications. Am J Surg Pathol. 2009;33(2):227\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllinen M, Beroukhim R, Cai L, Brennan C, Lahti-Domenici J, Huang H, et al. Molecular characterization of the tumor microenvironment in breast cancer. Cancer Cell. 2004;6(1):17\u0026ndash;32.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelage S, Zadhoosh A, You W, Brown TJ, Ringuette MJ. Drosophila SPARC collagen IV chaperone-like activity essential for development is unique to the fat body. iScience. 2025;28(4):112111.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrell C, Paulsson J, Jin SB, Tobin NP, Mezheyeuski A, Roswall P, et al. Impact of Epithelial-Stromal Interactions on Peritumoral Fibroblasts in Ductal Carcinoma in Situ. J Natl Cancer Inst. 2019;111(9):983\u0026ndash;95.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJingSong H, Hong G, Yang J, Duo Z, Li F, WeiCai C, et al. siRNA-mediated suppression of collagen type iv alpha 2 (COL4A2) mRNA inhibits triple-negative breast cancer cell proliferation and migration. Oncotarget. 2017;8(2):2585\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastro-Sanchez L, Soto-Guzman A, Navarro-Tito N, Martinez-Orozco R, Salazar EP. Native type IV collagen induces cell migration through a CD9 and DDR1-dependent pathway in MDA-MB-231 breast cancer cells. European Journal of Cell Biology. 2010;89(11):843\u0026ndash;52.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEspinosa Neira R, Salazar EP. Native type IV collagen induces an epithelial to mesenchymal transition-like process in mammary epithelial cells MCF10A. The International Journal of Biochemistry \u0026amp; Cell Biology. 2012;44(12):2194\u0026ndash;203.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClark AG, Vignjevic DM. Modes of cancer cell invasion and the role of the microenvironment. Curr Opin Cell Biol. 2015;36:13\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bjc-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BJC Reports](https://www.springer.com/journal/44276) ","snPcode":"44276","submissionUrl":"https://submission.springernature.com/new-submission/44276/3","title":"BJC Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6714442/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6714442/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e. Current treatment for ductal carcinoma in situ (DCIS) of the breast is generic, due to lack of risk stratification tools. We investigate the correlation between expression of collagen IV in the breast and risk of breast cancer death. We also explore the effect of collagen IV \u003cem\u003ein vitro.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e. Tissue microarrays from a cohort of women treated for DCIS who later died from breast cancer (n=43) or were still alive (n=119), were analysed for collagen IV by immunohistochemistry. Oestrogen receptor positive (ER+), triple negative and human epidermal growth factor receptor 2 amplified (HER2+) cell lines were cultured with and without collagen IV.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e. High expression of stromal collagen IV correlated with increased odds of dying of breast cancer (OR 2.50; 95% CI 1.16-5.39). This association remained when adjusting for tumour size, margin status, comedo necrosis and progesterone receptor negativity (PR-) (OR 4.27; 95% CI 1.64 – 11.1).\u003c/p\u003e\n\u003cp\u003eTriple negative breast cancer cell lines migrated quicker on collagen IV-coated than on uncoated surfaces. By contrast, collagen IV coating did not affect ER+ and HER2+ cell lines.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e. Abundance of stromal collagen IV increases risk of dying in breast cancer after DCIS, and collagen IV can promote cell motility in vitro.\u003c/p\u003e","manuscriptTitle":"Stromal collagen IV expression and risk of breast cancer death in ductal carcinoma in situ","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-18 11:07:47","doi":"10.21203/rs.3.rs-6714442/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-18T17:44:22+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-12T15:49:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-30T11:58:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-23T22:21:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"334175298760770132105299940703917834280","date":"2025-06-23T21:45:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294950955966858448438898171230971401750","date":"2025-06-23T15:03:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"16090059129329578909353912504230121134","date":"2025-06-17T08:42:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-16T13:51:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-27T18:22:15+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-22T09:58:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"BJC Reports","date":"2025-05-21T08:31:44+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bjc-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [BJC Reports](https://www.springer.com/journal/44276) ","snPcode":"44276","submissionUrl":"https://submission.springernature.com/new-submission/44276/3","title":"BJC Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"599101a0-373b-4e56-a8e2-1e78411a22a8","owner":[],"postedDate":"June 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2025-09-30T16:38:31+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-18 11:07:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6714442","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6714442","identity":"rs-6714442","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}