Breast cancer patients participating in the Swiss mammography screening program "donna" exhibit extended survival

doi:10.21203/rs.3.rs-3888314/v1

Study goal: We compared the survival rates of women with breast cancer (BC) within vs. outside the mammography screening program (MSP) “donna”.

Methods: We merged data from the “donna” MSP with the data from corresponding cancer registries to categorize BC cases as within MSP (screen-detected and interval carcinomas) and outside the MSP. We analyzed the tumor stage distribution, tumor characteristics and the survival of the women. We further estimated hazard ratios using Cox-regressions to account for different characteristics between groups and corrected the survival rates for lead-time bias.

Results: We identified 1,057 invasive (ICD-10: C50) and in-situ (D05) BC cases within the MSP and 1,501 outside the MSP between 2010 and 2019 in the Swiss cantons of St. Gallen and Grisons. BC within the MSP had a higher share of stage I carcinoma (46.5% vs. 33.0%, p<0.01), fewer metastases at diagnosis (2.7% vs. 11.6%, p<0.01), and a smaller (mean) tumor size (19.1mm vs. 24.9mm, p<0.01). The 10-year survival rates were 92.2% for women within and 73.9% for women outside the MSP (p<0.05). A significant survival difference persisted when the same tumor stages were compared. Lead-time corrected hazard ratios were 0.391 (p<0.01) for overall survival and 0.293 (p<0.01) for BC related survival.

Conclusion: Women participating in the “donna” MSP had a significantly higher overall and BC related survival rate than women outside the program. Detection of BC at an earlier tumor stage only partially explains the observed differences.

Mammography screening program

breast cancer

survival

mastectomy

Switzerland

The goal of mammography screening programs (MSP) is to detect breast cancer (BC) at a prognostically favorable early stage, thus reducing the burden of disease by less aggressive treatment as well as increasing survival rates [1]. Randomized controlled trials (RCT) from the 1970’s and 1980’s demonstrated a reduction of BC related mortality rates by 15 to 20% through MSPs [2]. Consequently, national screening programs were introduced in an increasing number of countries. However, following a review by Gøtzsche et al. [3,4] which concluded that mammography screening programs were not cost-effective, in 2013, the Swiss Medical Board [5] officially advised against the introduction of new screening programs and recommended fading out existing ones. This recommendation was controversially discussed due to contradicting evidence from other studies [6].

While the effectiveness of MSPs is often evaluated based on studies from the past, the underlying decades-old data are only partially applicable to today’s situation. Mortality of BC has about halved in Switzerland [7], the USA [8], and many other Western countries [9]. More advanced treatment options, better technology for digital mammographies, the use of magnetic resonance imaging (MRI), vacuum assisted biopsies and wider use of screening programs as well as a higher awareness of women and the availability of specialized breast centers have all contributed to this favorable change.

To account for these improvements in diagnosis and treatment, updated studies are required using different methodologies, as conducting RCTs for the evaluation of MSPs is nowadays ethically not acceptable. Employing various methodological designs, a series of more recent international studies has shown decreased mortality and less disability for women with BC diagnosed via MSPs [10–12]. For Switzerland, Herrmann [7] had found a strong decline of overall BC mortality from 1992 to 2012, but, by comparing cantons with and without MSP, he did not detect significant regional differences as an effect of organized MSPs. Reports of several Swiss MSPs do not comment on mortality reduction due to screening. To the best of our knowledge a scientific evaluation on patient level of the effect on survival rates of MSPs compared to opportunistic screening was not performed in Switzerland, yet.

Therefore, we merged data from the MSP “donna” with a comprehensive dataset from the Cancer Registry of Eastern Switzerland and Grisons-Glarus to evaluate the impact of the MSP “donna” on the stage distribution of detected BC, on the survival of the affected women and the applied cancer treatment. While other studies focused on effect differences between screen-detected vs. symptomatic BC cases [13,14], we deliberately investigated the effect that participation in an MSP had on women diagnosed with BC.

Study population

The study included all 50- to 69-year-old women diagnosed with invasive and in-situ BC (ICD-10: C50 and D05) between 2010 and 2019 in the Swiss cantons of St. Gallen and Grisons. BC was either diagnosed within or outside the “donna” MSP. Cancer data was retrieved from the databases of the Cancer Registries of Eastern Switzerland and Grisons-Glarus [15], which document all cancer cases diagnosed in the two cantons and follow-up the vital status of each patient via the cantonal administrations and record the date of death or the date of last contact.

The MSP “donna” in the Swiss cantons of St.Gallen and Grisons

Each woman between 50 and 69 years of age in the cantons of St. Gallen (since 2010) and Grisons (since 2011) is invited bi-annually to voluntarily participate in the MSP “donna”, free of charge except for a small deductible. Each year around 18,500 mammographies are done within the MSP, with a participation rate oscillating around 50%.

Sample generation and data cleansing

Figure 1 shows the derivation of the final data sample for analyses. The data from the Cancer Registries and from the MSP were merged using a unique screening-ID. A recorded tumor from the Cancer Registries was classified as a BC within MSP if the woman took part in the MSP “donna” and the carcinoma was either screen-detected or symptomatically detected within 2 years after the last mammography (see definition for interval carcinoma [16]). The remaining women with BC were labelled as outside the MSP. A total of 35 lobular in situ (LCIS) cases were excluded from the analyses. Further, six cases were excluded due to insufficient information regarding their UICC-TNM tumor stage. 100 women had two primary breast carcinomas within the study period. Only the first tumor was included in the sample. If two primary tumors were diagnosed on the same day, the tumor with the highest stage was considered. Hence, the final data set comprised 1,057 BC cases within and 1,501 BC cases outside the MSP. The carcinomas within the MSP were either screen-detected carcinomas following a positive mammography (n=840) or interval cancer cases (n=217).

Variables of interest

The final sample includes variables on patient characteristics, tumor characteristics, treatments, and survival. Patient characteristics include the age at diagnosis, canton of living and birthplace. Tumor characteristics include information about the biology of the tumor, i.e., tumor stage, tumor size, hormone receptor status, histologic grading, the human epidermal growth factor receptor 2 (HER2) status, and the Ki-67 proliferation index (Ki-67). Data for the patient’s primary treatments include information about adjuvant chemotherapies, surgical interventions, radiation therapy and whether the treatment was done in a certified breast center. Finally, follow-up data includes information about any occurred metastasis or recurrence, and the date of the last updated vital status. Patients from the canton of St. Gallen were followed up until mid-November 2022, from the canton of Grisons until the end of August 2023.

Methods and statistical analyses

Carcinomas within the MSP were compared to those outside the MSP with respect to their patient and tumor characteristics, primary treatments, and their survival. Wilcoxon-Mann Whitney rank sum tests or student’s t-tests, as appropriate, were used to test for significance in differences. The survival of the two groups was compared using Kaplan-Meier curves. Greenwood Point estimators for the Kaplan Meier estimates were used to test for significance of the results.

Furthermore, to account for confounders, Cox proportional hazards regression models were calculated to estimate hazard ratios (HR). We estimated models controlling for age at diagnosis, tumor stages, Ki-67 and tumor size to investigate potential survival differences. Moreover, we correct results from Kaplan-Meier estimates and Cox proportional hazards regressions for lead-time bias, by adjusting survival time of women whose BC was screen-detected. Corrections are based on the methodological approach of Duffy et al. [13], using mean sojourn times from Tabar et al. [17]. All statistical analyses were done in Stata 18.

Ethics approval

This retrospective cohort study was submitted to the Ethics Commission of Eastern Switzerland (EKOS - 23/061). It was ruled to be a quality control covered by the informed patient consent of the MSP. All data was pseudonymized by the Cancer Registry of Eastern Switzerland prior to analyses.

Descriptives and comparative analyses

The study included 2,558 women aged 50 to 69 with a BC diagnosis between 2010 and 2019. The average age at diagnosis was 59.3 years for women within and 60.7 years outside the MSP (see table 1). Geographically, more BC cases are within the MSP in the canton of St.Gallen, compared to the more rural canton of Grisons (p<0.05).

Table 1: Patient and tumor characteristics

	All observations		within MSP		outside MSP		Test for differences within vs. outside MSP
	N	Mean (SD) / Proportion %	N	Mean (SD) / Proportion %	N	Mean (SD) / Proportion %	p-value
Age	2,558	60.1 (6.0)	1,057	59.3 (6.2)	1,501	60.7 (5.9)	p < 0.05
Birthplace (%)							p = 0.30
outside Switzerland	669	26.2%	265	25.1%	404	26.9%
in Switzerland	1,889	73.8%	792	74.9%	1,097	73.1%
Canton (%)							p < 0.05
Grisons	726	28.4%	258	24.4%	468	31.2%
St. Gallen	1,832	71.6%	799	75.6%	1,033	68.8%
Stage distribution (%)							p < 0.05
In-Situ	316	12.4%	185	17.5%	131	8.7%
I	987	38.6%	491	46.5%	496	33.0%
II	821	32.1%	290	27.4%	531	35.4%
III	231	9.0%	66	6.2%	165	11.0%
IV	203	7.9%	25	2.4%	178	11.9%
Mean tumor size	2,390	22.7 (18.9)	1,012	19.3 (16.4)	1,378	25.3 (20.3)	p < 0.05
Tumor Size < 2cm	1,307	54.7%	659	65.1%	648	47.0%	p < 0.05
Tumor Size < 5cm	2,208	92.4%	962	95.1%	1,246	90.4%	p < 0.05
Lymph nodes involvement							p < 0.05
positive (N1+)	907	35.7%	275	26.1%	632	42.4%
negative (N0)	1,637	64.3%	778	73.9%	859	57.6%
Hormone receptor							p = 0.21
positive	2,146	90.8%	872	91.7%	1,274	90.2%
negative	218	9.2%	79	8.3%	139	9.8%
Estrogen receptors							p = 0.19
positive	2,135	90.5%	868	91.5%	1,267	89.9%
negative	224	9.5%	81	8.5%	143	10.1%
Progesterone receptors							p < 0.05
positive	1,774	76.5%	740	78.6%	1,034	75.0%
negative	546	23.5%	201	21.4%	345	25.0%
Ki-67 proliferation index							p < 0.05
high (i.e. ≥ 25%)	825	38.7%	293	34.4%	532	41.5%
low (i.e. < 25%)	1,307	61.3%	558	65.6%	749	58.5%
Grading							p < 0.05
Grade I	367	16.7%	200	23.0%	167	12.6%
Grade II	1,170	53.3%	464	53.3%	706	53.2%
Grade III	660	30.0%	206	23.7%	454	34.2%
HER2 over expression							p = 0.58
positive	341	15.5%	128	14.9%	213	15.8%
negative	1,863	84.5%	729	85.1%	1,134	84.2%

Tumor stage distribution of women within the MSP is significantly different from women outside the MSP (p<0.05). Figure 2 displays the corresponding distributions and shows that BC within the MSP are diagnosed at earlier stages with fewer locally advanced and metastatic characteristics. In particular, within the MSP 17.5% of BC are in the in-situ stage (vs. 8.7% outside the MSP) and further 46.5% are detected in stage I (vs. 33.0%). Finally, only 6.2 and 2.4% of BC cases within the MSP have stages III and IV (vs. 11.0 and 11.9%). Moreover, lymph node involvement (26.1 vs. 42.4%) and tumor sizes (19.3 vs. 25.3mm) are significantly different. Other significant differences between the two groups can be found for positivity of progesterone receptors (78.6% vs. 75.0%), Ki-67 (34.4% vs. 41.5%) and the distribution of the histologic grading – with a more favorable grading for BC within compared to outside the MSP.

Treatment and survival analyses

As can be seen in table 2, treatments of women within the MSP were less aggressive and less burdensome than for women outside the MSP, which is reflected by half the rate of radical mastectomies (8.9% vs. 17.7%, p<0.05) and a significantly lower proportion of patients undergoing adjuvant chemotherapy (27.6% vs 39.9%, p<0.05). Treatment was significantly more often performed within certified breast centers for women within the MSP. Noteworthy, the share of treated women in certified breast centers has changed over time. For instance, the first breast centers were certified in 2012 and the overall fraction of patients treated in breast centers has increased from 3.2% to 69.9% in 2019.

Table 2: Comparative analyses of treatment and survival data

	All observations		within MSP		outside MSP		Test for differences within vs. outside MSP
	N	Proportion in %	N	Proportion in %	N	Proportion in %	p-value

Certified Breast Centers	590	23.1%	318	30.1%	272	18.1%	p < 0.05
Radical Mastectomy	358	14.0%	92	8.7%	266	17.7%	p < 0.05
Adjuvant Chemotherapy	899	35.1%	294	27.8%	605	40.3%	p < 0.05

Recurrence or metastasis in the follow-up period	281	11.0%	63	6.0%	218	14.5%	p < 0.05

Overall survival
5-year	1839	90.6%	789	97.1%	1050	86.2%	p < 0.05
10-year	651	80.9%	202	92.2%	449	73.9%	p < 0.05

In-situ
5-year	255	98.4%	148	98.9%	108	97.7%	Not significant
10-year	94	94.6%	45	95.5%	50	93.3%	Not significant

Stage I
5-year	743	97.0%	370	98.7%	373	95.4%	p < 0.05
10-year	261	90.6%	94	96.3%	168	85.8%	p < 0.05

Stage II
5-year	622	94.6%	221	98.6%	402	92.5%	p < 0.05
10-year	223	83.6%	55	90.7%	169	80.4%	Not significant

Stage III
5-year	162	86.5%	44	91.3%	120	84.7%	Not significant
10-year	59	68.8%	9	76.6%	51	66.3%	Not significant

Stage IV
5-year	60	36.3%	10	49.1%	51	34.5%	Not significant
10-year	17	18.1%	3	36.4%	15	16.2%	Not significant
Note: Survival rates are derived from Kaplan-Meier estimates.

Furthermore, table 2 shows that the survival rates were significantly higher for women within the MSP. The difference in the survival rates increased over time, where 1-year survival rates differed by 2.6 percentage points (pp) (99.7% vs 97.1%), the 5-year survival rate already differed by 10.9pp (97.1% vs. 86.2%) and the 10-year survival rate by 18.2pp (92.2% vs. 73.9%). Similar differences were also observed for the subgroups of tumor stage I and II. Other stages have a fairly low number of cases but show similar, albeit insignificant patterns.

Treatment and survival analyses (comparative, Kaplan Meier curves, Cox proportional hazards regressions)

Figure 3 depicts the survival probabilities over time. While the difference in survival increases over time, the differences are at almost all points in time significant. The overall Kaplan Meier hazard ratio (HR) of within vs. outside the MSP is 0.250 for all BC cases. These significant better outcomes for women within the MSP persist, albeit to a fewer magnitude, for the subgroups of stage I and II, that account for 71% of diagnosed BC cases. Moreover, Figure 3b shows the lead-time corrected Kaplan Meier curve for all tumor groups. The lead-time correction moved down the survival curve for women within the MSP, but survival was still significantly better for women within the MSP. Confidence intervals become large over time as sample size within MSP shrinks due to the lead-time correction.

Furthermore, figure 4 shows, that the survival curve of BC specific survival is higher than for overall survival. For instance, 10-year BC survival lies at 96.5% for women within the MSP vs at 92.5% for overall survival (82.5% vs 74.0% for BC outside MSP). The Kaplan Meier HR of BC related mortality of within MSP vs outside MSP is lower (0.197) than that of overall mortality (0.250) between these groups. This pattern is also true for the lead-time corrected Kaplan-Meier curve.

Table 3 displays HR for overall mortality from six specifications of the Cox proportional hazards regression model, from a simple specification (1) only including whether the woman participated in the MSP or not, to controlling for different variables (3) and (5). Specifications (2), (4), and (6) show results for lead-time corrected survival times. Specification (1) shows that women within the MSP had a 75.0% lower hazard of mortality (HR: 0.250, p<0.01). When only looking at BC-related mortality, the HR was 0.197 (p<0.01) (see table A1 in the Appendix). The HR of the program remained similar when accounted for age (3). When additionally controlled for tumor stages (5), the effect of the MSP decreased (HR: 0.445; p<0.01) but was still large and significant. The higher the tumor stage, the higher was the hazard of mortality. Women diagnosed with stage IV had a ~27-times higher hazard of mortality than women with in-situ carcinomas. Analogue to the Kaplan-Meier estimators, specifications (2), (4) and (6) show that HR were higher for lead-time corrected estimates. Nevertheless, whether controlled for age, or for age and tumor stage distribution, the MSP still had a significant favorable effect on survival (HR between 0.414 and 0.677, p<0.01). For most models, proportional hazard assumptions did not hold, as survival effects were increasing with time (see figure 3). HRs for BC-related mortality were shown in table A1 in the Appendix. Additionally, table A2 in the Appendix shows, that when accounting for age, Ki-67 proliferation index, and tumor size, hazard of women within the MSP was still significantly lower than for women outside the MSP (HR: 0.332, p<0.01). This was also true for lead-time corrected survival (HR: 0.553, p<0.01) and controlled for tumor stages (HR: 0.708, p<0.05).

Table 3: Hazard ratios for overall mortality calculated through cox proportional hazards regression models

	Models
*Hazard ratios*	(1)	(2)	(3)	(4)	(5)	(6)
Screening program	0.250*** (0.188, 0.333)	0.391*** (0.293, 0.523)	0.263*** (0.198, 0.349)	0.414*** (0.310, 0.553)	0.445*** (0.332, 0.597)	0.677*** (0.504, 0.908)
Age			1.046*** (1.028, 1.064)	1.046*** (1.028, 1.064)	1.040*** (1.021, 1.059)	1.039*** (1.021, 1.058)

UICC TMN tumor stage (baseline in-situ)
I					1.464 (0.806, 2.658)	1.446 (0.796, 2.625)
II					2.358*** (1.318, 4.217)	2.262*** (1.265, 4.041)
III					5.167*** (2.821, 9.463)	4.810*** (2.630, 8.797)
IV					27.507*** (15.477, 48.885)	25.667*** (14.478, 45.504)

Control factors
Age			x	x	x	x
UICC-TNM tumor stage					x	x
Lead-time corrected		x		x		x
Model statistics
No. of observations	2,558	2,558	2,558	2,558	2,558	2,558
Proportional-hazards assumption holds	no	no	yes	no	no	no
Prob > Chi2	0.000	0.000	0.000	0.000	0.000	0.000

There was a statistically more favorable tumor stage distribution within compared to outside the MSP. Tumor stage is predictive for survival and women with lower stages had less aggressive treatment options available. We found, furthermore, that the proportion of Ki-67 levels was significantly higher for BC diagnosed outside the MSP. High Ki-67 levels are associated with higher grading and larger tumor size [18], and correlate to late endocrine resistance [19]. Consequently, we found that the number of mastectomies for cases within the MSP was about half compared to outside the MSP (8.9% vs 17.7%). These results are in line with Hermann et al. [20] in a partially overlapping population covering a different time period. Less invasive and aggressive treatments directly translate into an improved quality of life for patients, as reported by Jay et al. [21] who showed better patient reported outcomes with respect to breast, psychosocial and sexual well-being of breast conserving therapy compared to mastectomy. Moreover, less lymph node infiltrations lead to fewer axillary dissections, potentially preventing possible burdensome side effects, such as lymphedema.

Most importantly, we found a significant survival difference for women within the MSP compared to women outside the MSP, starting in the first year after diagnosis with a continuously widening survival difference in absolute terms. At ten years after diagnosis the survival difference amounted to 18.2pp (92.2% vs 74.0%, HR: 0.263, p < 0.01).

This difference in survival can have multiple explanations. Firstly, we observed different patient characteristics of women depending on MSP participation. For instance, a more favorable tumor stage distribution is a well-recognized prognostic factor [22] and an acknowledged advantage of screening [23], and might explain the survival differences. While we showed that the detection of BC at an earlier stage had a significant effect on survival, it only partially explained the observed effects. Significant survival differences persisted when women of the same tumor stages were compared, and when HR were controlled for tumor stage distribution. Moreover, a different age distribution between the groups could affect overall survival, due to non-BC related mortality. Accounting for age only marginally affected survival. Further, differences in other characteristics such as Ki-67 and tumor size – potential determinants for a length bias (screening detects disproportionately more slowly growing tumors) [24], could lead to an overestimation of the effectiveness of MSPs. HR controlled for age, Ki-67 and tumor size still showed significantly positive effects (HR: 0.332, p < 0.01).

Secondly, estimations for survival effects of MSPs in observational studies are prone to lead-time bias [25]. Lead-time bias happens when screening increases the perceived survival time without affecting the course of the disease [26]. When correcting the survival of women with screen-detected BC for lead-time [13] the positive effect of MSPs on survival was still highly significant (HR: 0.391, p < 0.01). HR of the MSP, uncorrected and corrected for lead-time were in line with other studies [13, 27].

Thirdly, as this was an observational study, there might have been unobserved differences in the characteristics of women participating in MSPs compared to those not participating. For instance, the screened population might differ in socioeconomic status (SES) and simultaneously in health-related factors as shown in several studies [28, 29], which can affect survival. A study from the Netherlands [28] showed that women of low SES were less likely to participate in MSPs than women with high SES (odds ratio of 1.8), and further had less favorable tumor stages – and thus were more likely to die from BC. Similar studies from Switzerland demonstrated significant differences of the tumor stage distribution of BC diagnosed in patients of lower compared to higher SES [30] as well as higher screening uptake of women with higher SES [31]. While we controlled for tumor stage distribution in our analyses, other unobserved differences in survival relevant factors between the groups might have affected the differences in survival.

Lastly, outcome differences might occur if groups are treated in different medical settings. We found that women within the MSP were more often treated in certified breast centers. Treatment was different with less mastectomies and fewer chemotherapy in breast centers. These differences are supported by evidence from other studies. For example, a Dutch study found that non-screeners more often stopped taking adjuvant antihormonal treatments [32]. Moreover, if a suspicious finding appears in a screening mammography, the warranted sequence of further examinations, biopsy, and interdisciplinary tumor boards is still part of the organized MSP. Most of the time surgery and adjuvant treatment are given in one of the certified breast centers according to strict quality-controlled guidelines. The fact that women, whose BC was diagnosed outside the program, were more often treated outside of a certified breast center, could contribute to the different treatment outcome and is a possible explanation for the unexpected finding that large survival differences persisted in favor of the MSP even if corrected for stage distribution. This difference will likely disappear in the future, as today most women are treated in certified breast centers, irrespective of participation in MSPs.

Comparing our results to other studies shows that various incidence based mortality studies could find a 26% reduction of deaths of BC comparing to historical controls or areas without screening [11]. A more current survey of European screening programs showed a wide variation but significant reductions of BC mortality when comparing screened to non-screened women [12]. Duffy et al. [33] found that in the population of women regularly undergoing mammography screening the proportion of BC fatal after 10 years was reduced by 50%, somewhat less so in in intermittent participants (-36%). For women participating at five consecutive screening rounds the risk of dying from BC was reduced by 74%, and after statistical adjustments for self-selection to 66% (HR: 0,34) [34]. These HR are in the range of our findings.

Limitations

Our study comes with limitations. Firstly, the overall mortality rate of age-standardized BC has significantly shrunk over the last 30 years not only internationally, but also in our local data from 34.3 to 18.7 per 100,000/year, whereas the age-standardized incidence has increased from 87.0 to 103.5 cases per 100,000/year [35]. Improvements in treatment had an important and beneficial effect on BC mortality and might be more relevant for decreasing overall mortality than screening [36]. In our study we do not compare with historical controls before the introduction of screening. Using incidence-based mortality data of the two groups diagnosed and treated at the same time in the same area eliminates a bias caused by considerably better treatment regimens over time. Secondly, the correction of lead-time bias can only approximate the effect of sojourn time and is based on a multi-state Markov model [13]. Abrahamsson et al. [37] in their simulation based on a tumor growth model showed that lead times can be shorter when conditional to tumor characteristics. This could potentially result in an underestimation of lead-time corrected survival effects from MSPs. Thirdly, adjustment of length bias is usually based on many assumptions [37]. By estimating models controlling for Ki-67 and tumor size, potential indicators for length bias, we showed that survival differences persisted. At the same time, differences in tumor size and Ki-67 are also direct advantages of MSPs – and controlled results do not fully reflect length bias. Lastly, bias might arise due to an overdiagnosis of tumors within the MSP [38]. By estimating HR for BC related mortality, and Kaplan Meier curves for BC cases excluding in-situ carcinoma (see figure A1 in the Appendix) we address the overdiagnosis bias. The former ensures that we do not include women who died of other causes in our analysis. Surprisingly, we find that survival effect from MSP increases, and simultaneously make age an insignificant estimate. The latter eliminates in-situ carcinoma, which most often suffer from overdiagnosis bias, from the analysis.

To summarize, our results strongly support findings from other countries that participation in MSPs significantly increases survival of women with BC. The advantages of MSPs persist across tumor stages, and when corrected for lead-time bias. Moreover, diagnosis of BC in earlier tumor stages within MSPs allows for significantly less aggressive treatments contributing to better quality of life.

Ethics approval and consent to participate

This retrospective cohort study was submitted to the Ethics Commission of Eastern Switzerland (EKOS - 23/061). It was ruled to be a quality control covered by the informed patient consent of the MSP. All data was pseudonymized by the Cancer Registry of Eastern Switzerland prior to analyses.

Consent for publication

Not applicable

Availability of data and materials

Raw data for the mammography screening program “donna” and the Cancer Registry of Eastern Switzerland are not publicly available to preserve individuals’/patients´ privacy.

Competing interests

RM, MB, and AE work for the Cancer League of Eastern Switzerland, which is responsible for the mammography screening program “donna” and the Cancer Registry of Eastern Switzerland. All other authors do not declare any conflicts of interest.

Funding

This research was funded by the Cancer League of Eastern Switzerland.

Authors' contributions

All authors contributed to conceptualization. DK and MB developed the methodology. DK, JS and MB were responsible for data curation and DK and MB performed all analyses. All authors interpreted the results. DK wrote the original draft. All authors reviewed and edited the manuscript. RM supervised.

Bulliard JL, Braendle K, Fracheboud J, Zwahlen M.: Breast cancer screening programmes in Switzerland, 2010-2018.; 2021.
Marmot MG, Altman DG, Cameron DA, Dewar JA, Thompson SG, Wilcox M: The benefits and harms of breast cancer screening: an independent review. British Journal of Cancer 2013, 108:2205–2240
Gøtzsche PC, Olsen O: Is screening for breast cancer with mammography justifiable? The Lancet 2000, 355:129–134
Gøtzsche PC: Relation between breast cancer mortality and screening effectiveness: systematic review of the mammography trials. Danish medical bulletin 2011, 58:A4246
Swiss Medical Board: Systematisches Mammographie-Screening; 2013.
Duffy SW, Chen TH-H, Smith RA, Yen AM-F, Tabar L: Real and artificial controversies in breast cancer screening. Breast Cancer Management 2013, 2:519–528
Herrmann C, Vounatsou P, Thürlimann B, Probst-Hensch N, Rothermundt C, Ess S: Impact of mammography screening programmes on breast cancer mortality in Switzerland, a country with different regional screening policies. BMJ open 2018, 8:e017806
Giaquinto AN, Sung H, Miller KD, Kramer JL, Newman LA, Minihan A, Jemal A, Siegel RL: Breast Cancer Statistics, 2022. CA: a cancer journal for clinicians 2022, 72:524–541
Lei S, Zheng R, Zhang S, Wang S, Chen R, Sun K, Zeng H, Zhou J, Wei W: Global patterns of breast cancer incidence and mortality: A population-based cancer registry data analysis from 2000 to 2020. Cancer communications (London, England) 2021, 41:1183–1194
Lim ZL, Ho PJ, Khng AJ, Yeoh YS, Ong ATW, Tan BKT, Tan EY, Tan S-M, Lim GH, Lee JA, Tan VK-M, Hu J, Li J, Hartman M: Mammography screening is associated with more favourable breast cancer tumour characteristics and better overall survival: case-only analysis of 3739 Asian breast cancer patients. BMC medicine 2022, 20:239
Njor S, Nyström L, Moss S, Paci E, Broeders M, Segnan N, Lynge E: Breast cancer mortality in mammographic screening in Europe: a review of incidence-based mortality studies. Journal of medical screening 2012, 19 Suppl 1:33–41
Molassiotis A, Tyrovolas S, Giné-Vázquez I, Yeo W, Aapro M, Herrstedt J: Organized breast cancer screening not only reduces mortality from breast cancer but also significantly decreases disability-adjusted life years: analysis of the Global Burden of Disease Study and screening programme availability in 130 countries. ESMO open 2021, 6:100111
Duffy SW, Nagtegaal ID, Wallis M, Cafferty FH, Houssami N, Warwick J, Allgood PC, Kearins O, Tappenden N, O'Sullivan E, Lawrence G: Correcting for lead time and length bias in estimating the effect of screen detection on cancer survival. American journal of epidemiology 2008, 168:98–104
Munck L de, Eijkelboom AH, Otten JDM, Broeders MJM, Siesling S: Method of primary breast cancer detection and the disease-free interval, adjusting for lead time. Journal of the National Cancer Institute 2023.
Swiss Cancer Regsitries: Description of the legally binding data dictionary for all cancer registries under the new law on cancer registrations; 2019.
Perry N: Guidelines for quality assurance in breast cancer screening and diagnosis. Luxembourg: Office for Official Publications of the European Communities; 2006.
Tabár L, Vitak B, Chen HH, Duffy SW, Yen MF, Chiang CF, Krusemo UB, Tot T, Smith RA: The Swedish Two-County Trial twenty years later. Updated mortality results and new insights from long-term follow-up. Radiologic Clinics of North America 2000, 38:625–651
Marwah N, Batra A, Marwah S, Gupta V, Shakya S, Sen R: Correlation of proliferative index with various clinicopathologic prognostic parameters in primary breast carcinoma: A study from North India. Journal of cancer research and therapeutics 2018, 14:537–542
Lee J, Lee Y-J, Bae SJ, Baek SH, Kook Y, Cha YJ, Lee JW, Son BH, Ahn SH, Lee HJ, Gong G, Jeong J, Lee SB, Ahn SG: Ki-67, 21-Gene Recurrence Score, Endocrine Resistance, and Survival in Patients With Breast Cancer. JAMA network open 2023, 6:e2330961
Herrmann C, Morant R, Walser E, Mousavi M, Thürlimann B: Screening is associated with lower mastectomy rates in eastern Switzerland beyond stage effects. BMC Cancer 2021, 21:229
Jay M, Creelman B, Baliski C: Patient reported outcomes associated with surgical intervention for breast cancer. American journal of surgery 2020, 219:816–822
Ellison LF, Saint-Jacques N: Five-year cancer survival by stage at diagnosis in Canada. Health reports 2023, 34:3–15
Wilkinson AN, Billette J-M, Ellison LF, Killip MA, Islam N, Seely JM: The Impact of Organised Screening Programs on Breast Cancer Stage at Diagnosis for Canadian Women Aged 40-49 and 50-59. Current oncology (Toronto, Ont.) 2022, 29:5627–5643
Marcus PM, Prorok PC, Miller AB, DeVoto EJ, Kramer BS: Conceptualizing overdiagnosis in cancer screening. J Natl Cancer Inst 2015, 107.
Shapiro S, Goldberg JD, Hutchison GB: Lead time in breast cancer detection and implications for periodicity of screening. American journal of epidemiology 1974, 100:357–366
Cox B, Sneyd MJ: Bias in breast cancer research in the screening era. Breast (Edinburgh, Scotland) 2013, 22:1041–1045
Shen Y, Yang Y, Inoue LYT, Munsell MF, Miller AB, Berry DA: Role of detection method in predicting breast cancer survival: analysis of randomized screening trials. Journal of the National Cancer Institute 2005, 97:1195–1203
Aarts MJ, Voogd AC, Duijm LEM, Coebergh JWW, Louwman WJ: Socioeconomic inequalities in attending the mass screening for breast cancer in the south of the Netherlands--associations with stage at diagnosis and survival. Breast cancer research and treatment 2011, 128:517–525
Pokora RM, Büttner M, Schulz A, Schuster AK, Merzenich H, Teifke A, Michal M, Lackner K, Münzel T, Zeissig SR, Wild PS, Singer S, Wollschläger D: Determinants of mammography screening participation-a cross-sectional analysis of the German population-based Gutenberg Health Study (GHS). PloS one 2022, 17:e0275525
Feller A, Schmidlin K, Bordoni A, Bouchardy C, Bulliard J-L, Camey B, Konzelmann I, Maspoli M, Wanner M, Clough-Gorr KM: Socioeconomic and demographic disparities in breast cancer stage at presentation and survival: A Swiss population-based study. International journal of cancer 2017, 141:1529–1539
Jolidon V, Prez V de, Bracke P, Bell A, Burton-Jeangros C, Cullati S: Revisiting the Effects of Organized Mammography Programs on Inequalities in Breast Screening Uptake: A Multilevel Analysis of Nationwide Data From 1997 to 2017. Frontiers in public health 2022, 10:812776
He W, Eriksson L, Törnberg S, Strand F, Hall P, Czene K: Discontinuation of adjuvant hormone therapy among breast cancer patients not previously attending mammography screening. BMC medicine 2019, 17:24
Duffy SW, Tabár L, Yen AM-F, Dean PB, Smith RA, Jonsson H, Törnberg S, Chen SL-S, Chiu SY-H, Fann JC-Y, Ku MM-S, Wu WY-Y, Hsu C-Y, Chen Y-C, Svane G, Azavedo E, Grundström H, Sundén P, Leifland K, Frodis E, Ramos J, Epstein B, Åkerlund A, Sundbom A, Bordás P, Wallin H, Starck L, Björkgren A, Carlson S, Fredriksson I et al.: Mammography screening reduces rates of advanced and fatal breast cancers: Results in 549,091 women. Cancer 2020, 126:2971–2979
Smith RA, Duffy SW, Yen AM-F, Tabár L, Lin AT-Y, Chen SL-S, Hsu C-Y, Dean P, Chen TH-H: Beneficial effect of repeated participation in breast cancer screening upon survival. JCO 2023, 41:552
Katalinic A, Eisemann N, Kraywinkel K, Noftz MR, Hübner J: Breast cancer incidence and mortality before and after implementation of the German mammography screening program. International journal of cancer 2020, 147:709–718
Plevritis SK, Munoz D, Kurian AW, Stout NK, Alagoz O, Near AM, Lee SJ, van den Broek JJ, Huang X, Schechter CB, Sprague BL, Song J, Koning HJ de, Trentham-Dietz A, van Ravesteyn NT, Gangnon R, Chandler Y, Li Y, Xu C, Ergun MA, Huang H, Berry DA, Mandelblatt JS: Association of Screening and Treatment With Breast Cancer Mortality by Molecular Subtype in US Women, 2000-2012. JAMA 2018, 319:154–164
Abrahamsson L, Isheden G, Czene K, Humphreys K: Continuous tumour growth models, lead time estimation and length bias in breast cancer screening studies. Statistical methods in medical research 2020, 29:374–395
The benefits and harms of breast cancer screening: an independent review.Lancet (London, England) 2012, 380:1778–1786

Competing interest reported. RM, MB, and AE work for the Cancer League of Eastern Switzerland, which is responsible for the mammography screening program “donna” and the Cancer Registry of Eastern Switzerland. The other authors declare that they have no conflicts of interest.

Appendix.docx

Breast cancer patients participating in the Swiss mammography screening program "donna" exhibit extended survival

Status:

Version 1

Abstract

Figures

Introduction

Patients and methods

Results

Discussion

Limitations

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1