Breast cancer precursors: lessons learned from molecular genetics

Risk factors for breast cancer development by tumor characteristics among women with benign breast disease

Background

Among women diagnosed with invasive breast cancer, 30% have a prior diagnosis of benign breast disease (BBD). Thus, it is important to identify factors among BBD patients that elevate invasive cancer risk. In the general population, risk factors differ in their associations by clinical pathologic features; however, whether women with BBD show etiologic heterogeneity in the types of breast cancers they develop remains unknown.

Methods

Using a nested case-control study of BBD and breast cancer risk conducted in a community healthcare plan (Kaiser Permanente Northwest), we assessed relationships of histologic features in BBD biopsies and patient characteristics with subsequent breast cancer risk and tested for heterogeneity of associations by estrogen receptor (ER) status, tumor grade, and size. The study included 514 invasive breast cancer cases (median follow-up of 9 years post-BBD diagnosis) and 514 matched controls, diagnosed with proliferative or non-proliferative BBD between 1971 and 2006, with follow-up through mid-2015. Odds ratios (ORs) and 95% confidence intervals (CIs) were obtained using multivariable polytomous logistic regression models.

Results

Breast cancers were predominantly ER-positive (86%), well or moderately differentiated (73%), small (74% < 20 mm), and stage I/II (91%). Compared to patients with non-proliferative BBD, proliferative BBD with atypia conferred increased risk for ER-positive cancer (OR = 5.48, 95% CI = 2.14–14.01) with only one ER-negative case, P-heterogeneity = 0.45. The presence of columnar cell lesions (CCLs) at BBD diagnosis was associated with a 1.5-fold increase in the risk of both ER-positive and ER-negative tumors, with a 2-fold increase (95% CI = 1.21–3.58) observed among postmenopausal women (56%), independent of proliferative BBD status with and without atypia. We did not identify statistically significant differences in risk factor associations by tumor grade or size.

Conclusion

Most tumors that developed after a BBD diagnosis in this cohort were highly treatable low-stage ER-positive tumors. CCL in BBD biopsies may be associated with moderately increased risk, independent of BBD histology, and irrespective of ER status.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13058-021-01410-1.

The Emerging Roles of Steroid Hormone Receptors in Ductal Carcinoma in Situ (DCIS) of the Breast

Ductal carcinoma in situ (DCIS) is a non-obligate precursor to most types of invasive breast cancer (IBC). Although it is estimated only one third of untreated patients with DCIS will progress to IBC, standard of care for treatment is surgery and radiation. This therapeutic approach combined with a lack of reliable biomarker panels to predict DCIS progression is a major clinical problem. DCIS shares the same molecular subtypes as IBC including estrogen receptor (ER) and progesterone receptor (PR) positive luminal subtypes, which encompass the majority (60-70%) of DCIS. Compared to the established roles of ER and PR in luminal IBC, much less is known about the roles and mechanism of action of estrogen (E2) and progesterone (P4) and their cognate receptors in the development and progression of DCIS. This is an underexplored area of research due in part to a paucity of suitable experimental models of ER+/PR + DCIS. This review summarizes information from clinical and observational studies on steroid hormones as breast cancer risk factors and ER and PR as biomarkers in DCIS. Lastly, we discuss emerging experimental models of ER+/PR+ DCIS.

Analysis of gene copy number alterations by multiplex ligation-dependent probe amplification in columnar cell lesions of the breast

BACKGROUND

Columnar cell lesions (CCLs) are possible precursors of breast cancer, but little is known about the role of breast cancer-related genes in the progression of CCL to invasive breast cancer.

METHODS

Gene copy numbers of 17 breast cancer-related genes were analyzed using Multiplex Ligation-dependent Probe Amplification (MLPA) in CCL (N = 28), ductal carcinoma in situ (DCIS) grade I likely originating from CCL (N = 5), and paired CCL (N = 14/28) with DCIS (N = 7) and/or invasive carcinoma (N = 13). The genes included were BIRC5, C11orf30, CCND1, CCNE1, CDH1, CPD, EGFR, ERBB2, ESR1, FGFR1, IKBKB, MAPT, MED1, MTDH, MYC, TOP2A and TRAF4.

RESULTS

No high level gene amplifications were observed in CCL, but copy number gains were encountered for the C11orf30 (3/28), MYC, CPD, MTDH (2/28), and CCND1, CCNE1, ESR1 and TOP2A genes (1/28). In addition, CDH1 showed loss in 2/28 and TOP2A in 1/28 cases. CCLs with or without atypia exhibited comparable numbers of copy number changes (p = 0.312). Overall, the frequency of gene copy number changes increased from CCL towards DCIS and invasive carcinoma (p = 0.004). Also in the cases with synchronous lesions, the CCLs exhibited fewer copy number changes than the DCIS/invasive carcinomas.

CONCLUSIONS

CCLs carry copy number changes of several known breast cancer-related genes, thereby substantiating their role in breast carcinogenesis. Among them, CCND1 and ESR1 copy number gains and CDH1 copy number losses are of particular interest. Since the copy number changes observed were more prevalent in DCIS and invasive carcinoma than in CCL, the corresponding gene alterations may represent rather late occurring events in low nuclear grade breast carcinogenesis.

Columnar cell lesions without atypia initially diagnosed on breast needle biopsies: is imaging follow-up enough?

OBJECTIVE

The purpose of this study was to evaluate the underestimation rate and predictive factor of underestimation of columnar cell lesions (CCLs) without atypia diagnosed through breast core needle biopsies (CNBs).

MATERIALS AND METHODS

From January 2007 through December 2011, 141 CCLs without atypia, including columnar cell change and columnar cell hyperplasia, were diagnosed in 138 women by CNB. Excisional (n = 16) or imaging follow-up (n = 125) findings were available in all cases. On a per-lesion basis, the underestimation rate and predictive factor of underestimation were evaluated.

RESULTS

Among the 16 surgically excised lesions, there were two malignancies (one ductal carcinoma in situ and one invasive ductal carcinoma) and one lobular carcinoma in situ. Overall, the pooled underestimation rate of malignancy was 1.4% (2/141). With regard to lesion variables, the mean lesion size was significantly larger in the underestimation group of CCLs (p = 0.007). Fine pleomorphic morphology of microcalcifications (p < 0.001), the distribution of the microcalcifications (p = 0.007), BI-RADS final assessment (p = 0.001), and imaging-pathologic correlation (p < 0.001) were significantly associated with underestimation. Multivariate analysis showed that fine pleomorphic morphology of microcalcifications (p < 0.0001) was an independent predictor of underestimation in 58 lesions with microcalcifications on mammography.

CONCLUSION

The overall underestimation rate of malignancy was 1.4%. Imaging follow-up is reasonable for CCLs without atypia at CNB, especially in small lesions with less suspicious imaging findings. Fine pleomorphic microcalcifications and higher BI-RADS category might be helpful in the prediction of underestimation of a high-risk lesion or malignancy.

Modelling the overdiagnosis of breast cancer due to mammography screening in women aged 40 to 49 in the United Kingdom

INTRODUCTION

Overdiagnosis of breast cancer due to mammography screening, defined as the diagnosis of screen-detected cancers that would not have presented clinically in a women's lifetime in the absence of screening, has emerged as a highly contentious issue, as harm caused may question the benefit of mammographic screening. Most studies included women over 50 years old and little information is available for younger women.

METHODS

We estimated the overdiagnosis of breast cancer due to screening in women aged 40 to 49 years using data from a randomised trial of annual mammographic screening starting at age 40 conducted in the UK. A six-state Markov model was constructed to estimate the sensitivity of mammography for invasive and in situ breast cancer and the screen-detectable mean sojourn time for non-progressive in situ, progressive in situ, and invasive breast cancer. Then, a 10-state simulation model of cancer progression, screening, and death, was developed to estimate overdiagnosis attributable to screening.

RESULTS

The sensitivity of mammography for invasive and in situ breast cancers was 90% (95% CI, 72 to 99) and 82% (43 to 99), respectively. The screen-detectable mean sojourn time of preclinical non-progressive and progressive in situ cancers was 1.3 (0.4 to 3.4) and 0.11 (0.05 to 0.19) years, respectively, and 0.8 years (0.6 to 1.2) for preclinical invasive breast cancer. The proportion of screen-detected in situ cancers that were non-progressive was 55% (25 to 77) for the first and 40% (22 to 60) for subsequent screens. In our main analysis, overdiagnosis was estimated as 0.7% of screen-detected cancers. A sensitivity analysis, covering a wide range of alternative scenarios, yielded a range of 0.5% to 2.9%.

CONCLUSION

Although a high proportion of screen-detected in situ cancers were non-progressive, a majority of these would have presented clinically in the absence of screening. The extent of overdiagnosis due to screening in women aged 40 to 49 was small. Results also suggest annual screening is most suitable for women aged 40 to 49 in the United Kingdom due to short cancer sojourn times.

Epigenetic progression of columnar cell lesions of the breast to invasive breast cancer

Promoter hypermethylation of several tumour suppressor genes often occurs during breast carcinogenesis, but little is known about epigenetic silencing in the possible precursor columnar cell lesion (CCL). Promoter hypermethylation of 50 different tumour suppressor genes was assessed in normal breast tissue (N = 10), CCL (N = 15), ductal carcinoma in situ (DCIS) grade I originating in CCL (N = 5) and paired CCL (N = 15) with DCIS (N = 7) and/or invasive carcinoma (N = 14) by Methylation-specific multiplex ligation-dependent probe amplification. Increasing mean cumulative methylation levels were found from normal breast tissue to CCL to DCIS and invasive carcinoma (P < 0.001) with similar methylation levels in DCIS and invasive carcinoma. Methylation levels and frequencies (in the overall analysis and analysis of only the synchronous lesions) were the highest for RASSF1, CCND2, ID4, SCGB3A1 and CDH13. The methylation levels of ID4, CCND2, and CDH13 increased significantly from normal breast tissue to CCL and to DCIS/invasive carcinoma. RASSF1, SCGB3A1 and SFRP5 had significant higher methylation levels in CCL compared to normal breast tissue, but showed no significant differences between CCL, DCIS and invasive carcinoma. Also, no difference was found between CCLs with and without atypia, or CCLs with or without synchronous cancer. In conclusion, promoter hypermethylation for several established tumour suppressor genes is already present in CCLs, underlining that promoter hypermethylation is an early event in breast carcinogenesis. Atypia in CCL or the presence of synchronous more advanced lesions does not seem to be accompanied by higher methylation levels.

Columnar cell lesions on breast needle biopsies: is surgical excision necessary? A systematic review

OBJECTIVE

This systematic review was conducted to provide treatment recommendations for patients with a diagnosis of columnar cell lesions (CCLs) in a breast core needle biopsy (CNB).

BACKGROUND

CCLs are putative breast cancer precursors and are often associated with (in situ) carcinoma in excision specimens. Although several studies reported on the progression risk and underestimation rate of a CNB diagnosis of CCL, there is no consensus regarding optimal clinical management in this context.

METHODS

We searched MEDLINE, Embase, and Cochrane databases from 1990 to October 2010 for studies on patients with a CNB diagnosis of CCL without atypia, CCL with atypia and atypical ductal hyperplasia associated with CCL followed by surgical excision or clinical follow up.

RESULTS

Of 1759 selected articles, 24 were included in this review. The pooled underestimation risks for (in situ) carcinoma were as follow: CCL without atypia 1.5% (95% confidence interval [CI] 0.6%-4%), CCL with atypia 9% (95% CI: 5%-14%), and atypical ductal hyperplasia associated with CCL 20% (95% CI: 13%-28%), based on the whole groups of patients with a CNB. Studies including CCLs with long-term clinical follow-up showed a trend toward a limited elevated breast cancer risk.

CONCLUSIONS

On the basis of the (in situ) carcinoma underestimation rates of patients with a CNB diagnosis of CCL with atypia and atypical ductal hyperplasia associated with CCL, surgical excision should be considered. For CCL without atypia, more studies with a long-term follow-up are required, but so far, surgical excision biopsy does not seem to be necessary.

Clinicopathologic characteristics and prognosis for molecular subtypes in low-grade breast carcinoma: comparison with grade one invasive ductal carcinoma-not otherwise specified

The aim of this study was to investigate the clinicopathologic characteristics and prognosis value for molecular subtypes of low-grade breast carcinoma (LGBC) compared with grade one invasive ductal carcinoma-not otherwise specified (G1-IDC-NOS). A retrospective review of 688 LGBC and 1 037 G1-IDC-NOS patients was classified into four different molecular subtypes based on the IHC-based definitions for ER, PR, and c-erbB-2. In LGBC, lymph node metastasis, the percentage of III/IV TNM stages, the expression of Ki-67 and p53 in luminal A subtype were lower than in other subtypes (P<0.01). In addition, the variations of Ki-67 and p53 expression were observed in different subtypes of G1-IDC-NOS (P<0.01). Compared with G1-IDC-NOS, LGBC has higher proportion in the ER positive, PR positive, HER-2 negative, luminal A subtype, Ki-67 negative, and lymph nodes negative group (P<0.01). Furthermore, the overall survival of luminal A and luminal B is higher than triple-negative and HER-2/neu subtype both in LGBC and G1-IDC-NOS in 262 LGBC and 330 G1-IDC-NOS patients with proper follow-up. The classification of molecular subtype together with clinicopathologic factors can significantly improve the traditional prognosticators in predicting outcome for LGBC and G1-IDC-NOS. And it may contribute to guide the treatment for LGBC and G1-IDC-NOS in the future.

Digital mammography: more microcalcifications, more columnar cell lesions without atypia

The incidence of columnar cell lesions in breast core needle biopsies since full-field digital mammography in comparison with screen-filmed mammography was analyzed. As tiny microcalcifications characterize columnar cell lesions at mammography, we hypothesized that more columnar cell lesions are diagnosed since full-field digital mammography due to its higher sensitivity for microcalcifications. In all, 3437 breast core needle biopsies performed in three hospitals and resulting from in total 55 159 mammographies were revised: 1424 taken in the screen-filmed mammography and 2013 in the full-field digital mammography period. Between the screen-filmed mammography and full-field digital mammography periods, we compared the proportion of mammographies that led to core needle biopsies, the mammographic indication for core needle biopsies (density, microcalcifications, or both) and the proportion of columnar cell lesions with or without atypia. The columnar cell lesions were graded according to Schnitt, and we included atypical ductal hyperplasia arising in the context of columnar cell lesions. Proportions were compared using χ(2) tests and prevalence ratios were adjusted for age and hospital. We found that more core needle biopsies per mammogram were taken in the full-field digital mammography period (7.6%) compared with the screen-filmed mammography period (5.0%, P<0.0001). Microcalcifications were more often diagnosed with full-field digital mammography than with screen-filmed mammography (adjusted prevalence ratio: 1.14, confidence interval 95%: 1.01-1.28). Core needle biopsies from the full-field digital mammography era showed more columnar cell lesions (10.8%) than those from the screen-filmed mammography era (4.9%; adjusted prevalence ratio: 1.93, confidence interval 95%: 1.48-2.51), particularly due to more columnar cell lesions without atypia (8.2% respectively 2.8%) while the proportion of columnar cell lesions with atypia remained nearly constant (2.0 vs 2.6%). In conclusion, since the implementation of full-field digital mammography, more microcalcifications are seen at mammography, more often resulting in core needle biopsies, which especially yields more columnar cell lesions without atypia.

Progression risk of columnar cell lesions of the breast diagnosed in core needle biopsies

Columnar cell lesions (CCLs) of the breast are recognized as putative precursor lesions of invasive carcinoma, but their management remains controversial. We therefore conducted a retrospective study on 311 CCLs, diagnosed in 4,164 14-gauge core needle biopsies (CNB): 221 CCLs without atypia (CCL), 69 with atypia (CCL-A), and 21 atypical ductal hyperplasias originating in CCL (ADH-CCL). Two groups were identified: "immediate treatment" group undergoing excision within four months after the CNB diagnosis of CCL (N = 52) and the "wait-and-see" group followed up to 8 years (median 3.5 years, N = 259). In 7 of 31 women (22.5%, 1 CCL, 4 CCL-A, 2 ADH-CCL) who underwent immediate surgical excision and were initially biopsied for microcalcifications, ductal carcinoma in situ (DCIS) was present and in 2/31 women (6.5%, 1 CCL, 1 CCL-A) invasive carcinoma. In 2/21 excisions (9.5%, 1 CCL, 1 CCL-A) initially biopsied for a density, DCIS was present and invasive carcinoma in 5/21 excisions (23.8%, 2 CCL, 3 CCL-A). In the wait-and-see group, 9/259 women (3.5%) developed invasive carcinoma, 6 ipsi, and 3 contralaterally. Progression risks of CCL-A and ADH-CCL were 18% and 22%,versus 2% for CCL without atypia (p < 0.001). In conclusion, CCL-A or ADH-CCL in a CNB were associated with a high risk of DCIS/invasive carcinoma in immediate surgical excision biopsies. The 8-years progression risks for CCL-A and ADH-CCL were around 20%. This illustrates that an atypical CCL in a CNB may signal the presence of concurrent lesions or development of advanced lesions in future and may justify ("mini") surgical excision.

Invasive tubular carcinoma of the breast frequently is clonally related to flat epithelial atypia and low-grade ductal carcinoma in situ

Low-grade precursor lesions, such flat epithelial atypia (FEA), low-grade ductal carcinoma in situ (lg-DCIS), and lobular neoplasia (LN) often coexist with invasive tubular carcinomas (TCs) of the breast. To evaluate a possible clonal relationship, we have examined a series of 27 TC and the surrounding putative precursor lesions using loss of heterozygosity analysis and mitochondrial DNA sequencing. In these lesions (22 FEA, 10 lg-DCIS, 3 LN), loss of heterozygosity was most frequently observed on the long arm of chromosome 16 as well as at chromosome 8p21, 3p14, 1p36 and 11q14 with a high degree of homology of allelic losses between FEA, lg-DCIS and tubular carcinomas. In the adjacent invasive tubular carcinomas, mitochondrial DNA sequencing revealed identical mutation patterns in 50% of the lg-DCIS and in 12 of 21 (57%) informative cases of FEA. No direct association was seen between TC and LN or columnar cell lesions without nuclear atypia. Our data indicate, that in the majority of cases lg-DCIS and FEA are directly related to tubular breast cancer with a possible precursor role.