Sage Journals: Discover world-class research

Abstract

Objective

To evaluate clinicopathological differences between screen-detected (SD) and interval (IC) breast cancers diagnosed in women enrolled in an organized breast screening programme in 2000--2007.

Setting

Breast Cancer Screening Programme of the north region of Portugal.

Methods

Using data from the screening programme and from the population-based North Region Cancer Registry, SD and IC were identified. Information on screening history, age, date of diagnosis, tumour size, histological type and grade, lymph node status, tumour stage, biomarkers, and treatment was obtained from the cancer registry and from clinical and pathological reports. Association between mode of detection and these clinicopathological characteristics was estimated by unconditional logistic regression.

Results

A total of 442 SD and 112 IC were identified in women aged 50--69. Compared with SD, IC were diagnosed in younger women (60.0 ± 5.8 years and 58.4 ± 6.0 years, respectively), were larger (tumour size >20 mm: 60.2% versus 25.1%), lobular (6.3% versus 16.1%), with a higher differentiation grade (grade 3: 17.7% versus 38.9%), had more lymph node metastases, more advanced stage, and oestrogen receptor (ER) negative (12.9% versus 29.0%) and progesterone negative, and HER2 positive. After multivariable analysis, compared with SD, IC were more likely to be larger than 20 mm, lobular, of grade 3 and negative for ER.

Conclusion

Our results are consistent with other studies. IC’s have a more aggressive biology than SDs. Our findings did not show any unexpected pattern requiring changes to our screening procedures, but continuous identification and characterization of IC is advisable.

Keywords

interval cancer organized screening breast cancer

Introduction

Good practices in organized breast cancer screening include the identification and characterization of interval breast cancers (IC), those cancers that occur after a negative mammogram and before the scheduled date for the next examination.¹ In comparison with the evaluation of other routine performance indicators of screening, the identification of IC is a more complex task, relying on the availability of a population-based cancer registry with high coverage, operating continuously over time.^2–4 In the absence of a cancer registry, it is also possible to consult hospital data (including that from pathology units) in order to identify and characterize all breast cancers in the screened population.^2,5,6

IC are related both to the sensitivity of the screening programme and the length of interval between screening rounds, and are expected events in any screening programme.^1,2,7,8 In general, compared with screen-detected cancers (SD), IC consist of larger tumours, with lymph node invasion, more advanced stage, and higher grade^9,10, but not all studies are consistent and some have shown different patterns.^5,11,12 Although there may be some controversy on the interpretation of the findings, IC are an heterogeneous group of tumours, the clinical and pathological features of which should be characterized in order to assess their relative contribution to the efficacy of breast cancer screening.^13–15

The organized population-based breast cancer screening programme, conducted by the north branch of the Portuguese Cancer League (Liga Portuguesa Contra o Cancro – LPCC), was initiated in the Northern Region of Portugal in 1999. Preliminary results from the programme have been published, and considered satisfactory.¹⁶

The aim of this study was to evaluate the clinicopathological differences between SD and IC diagnosed in the northern region of Portugal, in women enrolled in the screening programme between 2000 and 2007, within the context of evaluation of the screening programme.

Methods

By the end of 2009 the screening programme had expanded to 43 municipalities, covering 17% of the 454666 eligible women resident in the Northern Region of Portugal.¹⁷ Women aged between 45 and 69 were offered biennial screening with two-view mammography, with double independent reading of the examinations by trained radiologists and, in case of discrepancy, a final reading by a third independent and experienced radiologist.

SD was defined as a breast cancer diagnosed after a positive screening result. IC was defined as a breast cancer diagnosed in a woman who had a screening test, with or without further assessment, which was negative either before the next screening invitation or within a time period equal to a screen interval for women who have reached the upper age limit.¹ IC diagnosis occurred as a result of opportunistic screening examinations, or in the course of a medical examination sought by women due to symptoms (eg. a lump, nipple discharge, pain).

To identify SD cases in women with a positive mammogram, we conducted a search for diagnosed breast cancers from hospital clinical files, pathology reports, and the North Region Cancer Registry (Registo Oncológico Regional do Norte – RORENO), a population-based registry with high completeness.¹⁸ Information on identified SD cases was added to the Breast Cancer Screening Programme database. To be included in this study, screening must have occurred in the period 2000–2007.

Information on IC was gathered by merging the RORENO and Breast Cancer Screening Programme databases.The woman’s name, date of birth, and national health service number were used to link records. Inconsistencies and mismatching were checked manually. IC were those breast cancers registered in RORENO which matched with negative screening results in the previous two year period. This ascertainment was limited to women participating in the screening programme from 2000–2007. Because of the IC definition, breast cancers gathered might have been diagnosed in the period 2000–2009. As RORENO completeness is limited to invasive cancers, analysis was restricted to invasive SD and IC. Ductal carcinomas in situ were not considered. Following European Guidelines,¹ only women aged 50–69 were considered in the analysis.

The data collected was used to build a specific database for this study, including only cases of SD and IC. Information extracted from the screening programme database included variables such as date and round of screening, outcome of screening, woman's age at last screening examination, and residence. Information was also extracted from RORENO. To obtain additional data not routinely collected at the cancer registry, hospitals were contacted and clinical files and pathology reports were accessed. Overall, the following variables were collected: date of diagnosis, tumour size, histological type (using the International Classification of Diseases for Oncology-3rd edition;¹⁹ 8500, 8521 coded as ductal; 8520, 8522, 8524 coded as lobular; 8211, 8480, 8510, 8530, 8540 coded as other), histological grade (according to Nottingham Grading System²⁰), lymph node status, tumour stage (TNM classification - AJCC²¹), and first treatment (mastectomy, breast conserving surgery, chemotherapy). IC and SD tumour size were classified in two categories (≤20 mm or >20 mm; the cut-off between T1 and T2, T3, T4²¹). Information on biomarkers such as oestrogen (ER) and progesterone (PR) receptor status, and detection of overexpression and/or amplification of the human epidermal growth factor receptor 2 (HER2) were registered according to the pathology reports.

The two types of breast cancer (IC versus SD) were compared for each of these variables. To analyse the contribution of the different parameters to the risk of IC versus SD, an unconditional logistic regression was performed to calculate crude odds ratios (ORs) with corresponding 95% confidence intervals (CI-95%). T-test and χ² test were also used to compare differences between IC and SD, depending on the type of variables considered. After multivariable analysis (with all variables), the final model included only parameters significant to the risk of IC versus SD. HER2 status and triple negative assessment variables were not used in the multivariable analysis, due to the small number of cases. A P < 0.05 was considered to be statistically significant.

Results

In total, 554 invasive breast cancers were diagnosed in the study population, 442 (79.8%) classified as SD, and 112 (20.2%) as IC.

At last screening examination, women from the SD group were significantly older than the women with IC (mean age 60.0 ± 5.8 years and 58.4 ± 6.0 years, p = 0.009). IC were larger than SD (tumour size > 20 mm: 60.2% versus 25.1%). Results in table 1 are given as crude ORs, CIs and P values for each independent variable assessed.

Table 1.

Odds ratio of interval breast cancers relative to screen detected cancers according to clinicopathological characteristics.

Parameters	Interval cancers n = 112 (%*)	Screen-detected cancers n = 442 (%*)	Crude OR (95% CI)	P value^†
Age group
50–59	64 (57.1)	199 (45.0)	1	0.022
60–69	48 (42.9)	243 (55.0)	0.61 (0.40-0.93)	0.022
Histology
Ductal	84 (75.0)	371 (83.9)	1	0.004
Lobular	18 (16.1)	28 (6.3)	2.84 (1.50-5.37)
Other	10 (8.9)	43 (9.7)	1.03 (0.50-2.13)
Tumour grade
grade 1	14 (13.0)	119 (28.4)	1	<0.001
grade 2	52 (48.1)	226 (53.9)	1.96 (1.04-3.67)
grade 3	42 (38.9)	74 (17.7)	4.82 (2.47-9.44)
Missing	4	23
Tumour size
≤20 mm	43 (39.8)	329 (74.9)	1	<0.001
>20 mm	65 (60.2)	110 (25.1)	4.52 (2.91-7.03)
Missing	4	3
Lymph nodes
node negative	48 (44.0)	303 (69.2)	1	<0.001
node positive	61 (56.0)	135 (30.8)	2.85 (1.86-4.38)
Missing	3	4
Stage
I	32 (28.8)	260 (59.1)	1	<0.001
II	45 (40.5)	128 (29.1)	2.86 (1.73–4.71)
III	26 (23.4)	49 (11.1)	4.31 (2.36–7.86)
IV	8 (7.2)	3 (0.7)	21.67 (5.47–85.85)
Missing	1	2
ER status
Positive	76 (71.0)	386 (88.9)	1	<0.001
Negative	31 (29.0)	48 (11.1)	3.28 (1.96–5.49)
Missing	5	8
PR status
Positive	65 (60.7)	328 (75.6)	1	0.002
Negative	42 (39.3)	106 (24.4)	2.00 (1.28–3.12)
Missing	5	8
HER2 status
Negative	77 (78.6)	171 (89.5)	1	0.011
Positive	21 (21.4)	20 (10.5)	2.33 (1.20–4.55)
Missing	14	251
Triple negative
no	78 (79.6)	178 (93.2)	1	0.001
yes	20 (20.4)	13 (6.8)	3.51 (1.66–7.41)
Missing	14	251

the percentages were calculated excluding those cancers with value unknown; ^†P value corresponds to χ² heterogeneity test; OR, odds ratio; CI, confidence interval; ER, oestrogen receptor; PR, progesterone receptor; HER2, epidermal growth factor receptor 2.

The predominant histological type in both IC and SD cancers was ductal invasive, but the relative proportions of ductal invasive and lobular were significantly different; lobular type was more likely to occur among IC (OR = 2.84, table 1). In this series of cases there were four inflammatory cancers, three in the IC group and one in the SD. In 27 cases, histological grade was not classified (table 1), either because the tumour had micro-invasion or because grade was not reported by the pathologist. In the remaining cancers, grade 3 was reported for 17.7% of the SD and for 38.9% of IC.

Women with IC presented more frequently than women with SD at diagnosis with lymph node invasion, and also with stage IV presentation (7.2% versus 0.7%). Evaluation of ER and PR status showed that SD more often had positive receptors than IC.

HER2 expression was evaluated in only 52.2% (n = 289) of the studied cancers; the proportion of HER2 positivity was higher in IC than in SD (21.4% and 10.5%, respectively) and the difference was significant. The phenotype triple negative for oestrogen, progesterone, and HER2 expression was more frequent in IC.

A multivariable logistic regression model was built to compute adjusted odds ratios of having IC (table 2). Only significant variables were considered in the final model: tumour size, histology, tumour grade, and ER status. Compared with SD, IC were more likely to be larger than 20 mm, be lobular, of grade 3 and negative for ER.

Table 2.

Odds ratio of developing an interval cancer relative to screen detected cancer. Final model, after multivariable analysis.

Parameters	OR adjusted for covariates*	95% Confidence Interval
Tumour size
≤20 mm	1
>20 mm	3.42	2.09–5.59
Histology
Ductal	1
Lobular	3.90	1.88–8.10
Other	1.23	0.55–2.74
Tumour grade
grade 1	1
grade 2	1.36	0.68–2.74
grade 3	2.58	1.17–5.71
ER status
Positive	1
Negative	2.50	1.34–4.66

*tumour size (categorical), histology, tumour grade, oestrogen receptors status.

OR, odds ratio; ER, oestrogen receptor.

First treatment modality was significantly different between the two groups. Only 1.1% of SD cases had upfront chemotherapy compared with 13.8% of IC cases (p < 0.001). For those 525 women whose first treatment was surgery, breast conserving surgery was more commonly used in SD (261 in 431 cases, 60.6%) than IC (36 in 94 surgeries, 38.3%).

Discussion

This is the first report on the clinicopathological characteristics of IC versus SD within this screening programme.

At screening examination, women with IC were younger than women with SD, as verified in other studies.^14,22–25 This has also been associated with higher prevalence of mammographic density in younger cases.^10,23,26

Size of the breast cancer has been interpreted as reflecting the chronological age of the tumour.^14,27 In this study, IC were significantly larger than SD, in agreement with the findings from other studies.^{10,12,28–30}

Lobular histological type was more frequent among IC than SD breast cancers. This has also been observed in some studies,^5,14,22,31 but others have found no difference in the histology type between IC and SD.^10,11,28 For the purposes of our analysis, “lobular” carcinoma and “mixed lobular and ductal” carcinoma were considered the same entity, as their characteristics and outcomes have been considered similar.³² The growth patterns detected in lobular tumours make these more difficult to detect with mammography.¹⁴

Lymph node involvement in breast cancer has been considered to reflect both tumour chronology and biology.²⁷ In our study, after adjustment for the other variables, lymph node metastases were not significantly more frequent in IC than in SD. The small size of our studied population may explain this finding, but this lack of association has also been observed in other studies, after adjustment for size of the tumour.^7,14 This excess of lymph node metastases among IC depended not only on the chronological age and biology of the tumour, but also on other unexplained variables.^22,27

Grade is an important indicator of biology and prognosis, considered by some as time independent,^13,33,34 but others have shown a phenotypic drift in cancer grade as tumour progresses.^35,36 With mammographic screening, a high proportion of tumours is stage T1N0M0 and grade is considered of particular utility in the evaluation of small tumours.³³ In our dataset the distribution of high grade tumours was more frequent in IC than SD, and this association was independent of the size of the tumour. This result is consistent with previous studies reporting a more aggressive behaviour of IC.^{10,11,14,22,24,30,31}

This pattern of a more aggressive behaviour associated with IC was also expressed in the biomarkers results. The proportion of cases that were negative for RE or triple negative for RE, RP, and HER2, was higher in IC compared with SD.^7,24,37

In the final model (after multivariable analysis), tumours larger than 20 mm, histological type lobular, high grade histology and RE negative were the independent predictors of the presence of interval cancers.

Primary treatment for women with IC was more likely to be extensive surgery, or systemic therapy, compared with SD cases. This is also in accordance with other published results, and reflects the more advanced stage at diagnosis for IC cases.³⁰

Radiological review of last screening/assessment imaging and diagnostic mammography of the IC was not performed; this may be considered a limitation. However, although some authors have used a radiological category to distinguish between true interval and missed interval breast cancers,^28,38–41 others have not considered this assessment essential, and some have even thought it to be misleading.¹¹ Mammographic density has been considered a risk factor for IC,^23,26 but in this study the retrospective analysis of the mammographic pattern was not performed.

The sample size of our study may have hampered the detection of some associations of smaller magnitude. As mentioned in other studies, the existence of unregistered opportunistic screening can introduce some misclassification on the detection mode (SD versus IC).^29,42,43 In the northern region of Portugal, opportunistic screening coexists with the organized screening programme, but it is difficult to quantify with precision.⁴⁴

We believe that limitations mentioned above do not seriously affect the internal validity of our study. Furthermore, the methodology used in data collection and management support the validity of the associations identified. Data were retrieved from the population-based breast cancer screening organized programme, ascertainment of IC cases was made through a population-based cancer registry (RORENO) to ensure complete capture of breast cancer cases in the population served by the screening programme, additional data not routinely collected at the cancer registry was actively sought in hospital clinical files and pathology reports, and this procedure was identical for SD and IC.

Variable by variable, these results are consistent with other studies conducted in different settings and with different screening methodologies. IC constitute a subset of breast cancers with a more aggressive biology than SD. They occur in any screening programme, but our findings did not show any unexpected pattern leading us to changes in the screening procedures. Continuous ongoing identification and characterization of IC, as we have carried out in this study, is advisable.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Declaration of conflicting interests

Ana Aguiar is the Head of the Breast Cancer Screening Programme in the North Region of Portugal and Vítor Veloso is the President of the Portuguese League Against Cancer, North Branch. The remaining authors have no conflicts of interest to disclose.

Acknowledgements

The authors thank Manuel António Silva and Joana Bastos from the Centre Region Cancer Registry, Alice Coelho from São Sebastião Hospital, Isabel Carvalho from S. João Hospital and pathologists Vicente Gonçalves, Eduardo Ferreira and Fernando Pardal for their valuable contribution to data collection.

References

Perry

Broeders

de Wolf

Tornberg

von Holland

Karsa

. European guidelines for quality assurance in breast cancer screening and diagnosis, 4th ed. Luxembourg: Office for Official Publications of the European Communities, 2006.

Stockton

McCann

Cancer registries in monitoring, evaluating and planning breast cancer screening programmes. In: Sankila

Démaret

Hakama

Lynge

Shouten

Parkin

(eds). Evaluation and Monitoring of Screening Programmes, Luxembourg: Office for Official Publications of the European Communities, 2001, pp. 181–94.

Bennett

Sellars

Moss

. Interval cancers in the NHS breast cancer screening programme in England, Wales and Northern Ireland. Br J Cancer 2011; 104: 571–7.

Tornberg

Kemetli

Ascunce

. A pooled analysis of interval cancer rates in six European countries. Eur J Cancer Prev 2010; 19: 87–93.

Baré

Bonfill

Andreu

. on behalf of the Breast Cancer Screening Programme of Sabadell-Cerdanyola Research Group. Relationship between the method of detection and prognostic factors for breast cancer in a community with a screening programme. J Med Screen 2006; 13: 183–91.

Bare

Sentis

Galceran

. Interval breast cancers in a community screening programme: frequency, radiological classification and prognostic factors. Eur J Cancer Prev 2008; 17: 414–21.

Domingo

Sala

Servitja

. Phenotypic characterization and risk factors for interval breast cancers in a population-based breast cancer screening program in Barcelona, Spain. Cancer Causes Control 2010; 21: 1155–64.

Houssami

Irwig

Ciatto

. Radiological surveillance of interval breast cancers in screening programmes. Lancet Oncol 2006; 7: 259–65.

Burrell

Sibbering

Wilson

. Screening interval breast cancers: mammographic features and prognosis factors. Radiology 1996; 199: 811–7.

10.

Kirsh

Chiarelli

Edwards

. Tumor characteristics associated with mammographic detection of breast cancer in the Ontario breast screening program. J Natl Cancer Inst 2011; 103: 942–50.

11.

Gilliland

Joste

Stauber

. Biologic characteristics of interval and screen-detected breast cancers. J Natl Cancer Inst 2000; 92: 743–9.

12.

Caldarella

Puliti

Crocetti

. Biological characteristics of interval cancers: a role for biomarkers in the breast cancer screening. J Cancer Res Clin Oncol 2013; 139: 181–5.

13.

Webster

Bilous

Willis

. Histopathologic indicators of breast cancer biology: insights from population mammographic screening. Br J Cancer 2005; 92: 1366–71.

14.

Porter

El-Bastawissi

Mandelson

. Breast tumor characteristics as predictors of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 1999; 91: 2020–8.

15.

Dawson

Duffy

Blows

. Molecular characteristics of screen-detected vs symptomatic breast cancers and their impact on survival. Br J Cancer 2009; 101: 1338–44.

16.

Giordano

von Karsa

Tomatis

. Mammographic screening programmes in Europe: organization, coverage and participation. J Med Screen 2012; 19(Suppl 1): 72–82.

17.

Statistics Portugal. Annual estimates of resident population 2009 Lisboa2013 [Access Date: 2013/12/14]. Available from: http://www.ine.pt/.

18.

Castro

Bento

Lunet

Campos

. Assessing the completeness of cancer registration using suboptimal death certificate information. Eur J Cancer Prev 2012; 21: 478–9.

19.

World Health Organization. International Classification of Diseases for Oncology. 3rd ed. Fritz A, Percy C, Jack A, et al., editors. Geneva, 2000.

20.

Elston

Ellis

. Pathological prognostic factors in breast cancer. I. . The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 1991; 19: 403–10.

21.

AJCC Cancer Staging Handbook. Greene FL, Page DL, Fleming ID, editors. New York, NY: Springer; 2002.

22.

Bucchi

Puliti

Ravaioli

. Breast screening: axillary lymph node status of interval cancers by interval year. Breast 2008; 17: 477–83.

23.

Mandelson

Oestreicher

Porter

. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 2000; 92: 1081–7.

24.

Musolino

Michiara

Conti

. Human epidermal growth factor receptor 2 status and interval breast cancer in a population-based cancer registry study. J Clin Oncol 2012; 30: 2362–8.

25.

Bucchi

Ravaioli

Foca

Colamartini

Falcini

Naldoni

. Incidence of interval breast cancers after 650,000 negative mammographies in 13 Italian health districts. J Med Screen 2008; 15: 30–5.

26.

Porter

Evans

Cornford

. Influence of mammographic parenchymal pattern in screening-detected and interval invasive breast cancers on pathologic features, mammographic features, and patient survival. AJR 2007; 188: 676–83.

27.

Smeets

Ryckx

Belmans

. Impact of tumor chronology and tumor biology on lymph node metastasis in breast cancer. Springerplus 2013; 2: 480–480.

28.

Raja

Hubbard

Salman

. Interval breast cancer: is it a different type of breast cancer? Breast 2001; 10: 100–8.

29.

Domingo

Jacobsen

von Euler-Chelpin

. Seventeen-years overview of breast cancer inside and outside screening in Denmark. Acta Oncol 2013; 52: 48–56.

30.

Olsson

Borgquist

Butt

Zackrisson

Landberg

Manjer

. Tumour-related factors and prognosis in breast cancer detected by screening. Br J Surg 2012; 99: 78–87.

31.

Crane

Luke

Rogers

Playford

Roder

. An analysis of factors associated with interval as opposed to screen-detected breast cancers, including hormone therapy and mammographic density. Breast 2002; 11: 131–6.

32.

Bharat

Gao

Margenthaler

. Tumor characteristics and patient outcomes are similar between invasive lobular and mixed invasive ductal/lobular breast cancers but differ from pure invasive ductal breast cancers. Am J Surg 2009; 198: 516–9.

33.

Rakha

Reis-Filho

Baehner

. Breast cancer prognostic classification in the molecular era: the role of histological grade. Breast Cancer Res 2010; 12: 207–207.

34.

Engstrom

Opdahl

Hagen

. Molecular subtypes, histopathological grade and survival in a historic cohort of breast cancer patients. Breast Cancer Res Treat 2013; 140: 463–73.

35.

Anderson

Waller

Ellis

Bobrow

Moss

. Influence of annual mammography from age 40 on breast cancer pathology. Hum Pathol 2004; 35: 1252–9.

36.

Tabar

Duffy

Vitak

Chen

Prevost

. The natural history of breast carcinoma: what have we learned from screening? Cancer 1999; 86: 449–62.

37.

Nagtegaal

Allgood

Duffy

. Prognosis and pathology of screen-detected carcinomas: how different are they? Cancer 2011; 117: 1360–8.

38.

Hofvind

Geller

Skaane

. Mammographic features and histopathological findings of interval breast cancers. Acta Radiol 2008; 49: 975–81.

39.

Britton

McCann

O'Driscoll

Hunnam

Warren

. Interval cancer peer review in East Anglia: implications for monitoring doctors as well as the NHS breast screening programme. Clin Radiol 2001; 56: 44–9.

40.

Hoff

Abrahamsen

A-L

Samset

Vigeland

Klepp

Hofvind

. Breast Cancer: Missed Interval and Screening-detected Cancer at Full-Field Digital Mammography and Screen-Film Mammography— Results from a Retrospective Review. Radiology 2012; 264: 378–86.

41.

Vitak

Olsen

Manson

Arnesson

Stal

. Tumour characteristics and survival in patients with invasive interval breast cancer classified according to mammographic findings at the latest screening: a comparison of true interval and missed interval cancers. Eur Radiol 1999; 9: 460–9.

42.

Heidinger

Batzler

Krieg

. The incidence of interval cancers in the German mammography screening program: results from the population-based cancer registry in North Rhine-Westphalia. Dtsch Arztebl Int 2012; 109: 781–7.

43.

Bulliard

Sasieni

Klabunde

De Landtsheer

Yankaskas

Fracheboud

. Methodological issues in international comparison of interval breast cancers. Int J Cancer 2006; 119: 1158–63.

44.

Departamento de Estudos e Planeamento. Relatório de Actividades 2009. Porto: Administração Regional de Saúde do Norte, 2010.

Clinicopathological differences between interval and screen-detected breast cancers diagnosed within a screening programme in Northern Portugal

Abstract

Objective

Setting

Methods

Results

Conclusion

Keywords

Introduction

Methods

Results

Discussion

Footnotes

Funding

Declaration of conflicting interests

Acknowledgements

References