Occult breast cancer in breast reduction specimens

Introduction

In Denmark, the risk of developing invasive breast cancer before the age of 75 is 11% [1]. Screening for breast cancer is offered routinely from age 50-69 [2]. Patients undergoing breast reduction are in risk of having occult invasive breast cancer or ductal carcinoma in situ at the time of surgery. Detection of such malignancies can be made pre- or postoperatively. In order to detect malignancies preoperatively the Danish guidelines suggest that women 40 + years of age ought to have a clinical mammography prior to breast reduction surgery [3].

Postoperative diagnosis can be verified at the pathologic examination. Two different principles in this exist: (1) consistent microscopic examination of all specimens or (2) macroscopic examination, only followed by microscopy in case of focal abnormalities.

This study aimed to evaluate if macroscopic examination of breast reduction specimens is a sufficient procedure for detection of cancer. Second, we intended to map the methods for specimen examination in Denmark.

Method

Two sub-studies were undertaken:

1. A study regarding quality of specimen examination method.

   Here we quantified the subsequent occurrence of breast cancer in women who had earlier had breast reduction surgery. Furthermore, every cancer case was thoroughly evaluated in order to decide if the cancer might have been present and overlooked at the time of breast reduction surgery.

2. A mapping of current specimen examination method used in Departments of Pathology in Denmark.

Study 1 – quality of specimen examination method

Study population and data processing

The Danish National Pathology Register comprises information on all pathoanatomical analysis of specimens performed by Danish pathology departments. Information regarding all tissue specimens is registered with the international coding system “Systematized Nomenclature of Medicine” (SNOMED) [4]. Using this register it was possible to identify 1120 women who have had breast reduction surgery at Odense University Hospital from January 2005 until December 2010. Of these 363 fulfilled the inclusion criteria; female sex, bilateral breast reduction surgery and no previous breast cancer.

All specimens from the included 363 women were examined at the Pathology department at Odense University Hospital. At this department macroscopic examination is manual slicing of the tissue and examination of the tissue slices by palpation and vision. Microscopic examination is only performed when suspicious areas with focal abnormalities are found.

Information on possible diagnoses of invasive breast cancer and ductal carcinoma in situ registered after breast reduction surgery were ascertained from the Danish National Pathology Register by searching for relevant SNOMED codes.

The follow-up period was calculated from the day of breast reduction surgery until a registered diagnosis of breast cancer, death or September 1. 2015 occurred. Histological details for women diagnosed with invasive breast cancer or ductal carcinoma in situ can be seen in Table 1.

Table 1. Flowchart of selection of the study population.

* Specimens from patients were registered and examined at the Department of Clinical Pathology (DCP), Odense University Hospital (OUH). Patients were identified in Patobank using the following codes: T04030, T04020, T04800, T04010 and T04000 from January 1. 2005 to December 31. 2010. Exclusion by the following codes: P306Y3, P306X4, P30990, P32940 and P31060.

** Unilateral removal of a lipoma in two patients, a fibroadenoma in one patient, an abscess in two patients, a fistula in one patient and a nonosseous extramedullary plasmacytoma in one patient.

*** except malignant melanoma.

Table 2. Characteristics of the 5 cases of invasive breast cancer (IBC) and ductal carcinoma in situ (DCIS) in the cohort.

Patient (age at surgery)	Follow-up(years)	Diagnosis	Tumour characteristics	Pathology examinationmethod
1 (38,5)	0	Multifocal invasive ductal carcinoma, DCIS AL metastasis: 1/13	19 mm diameter (29 mm including DCIS areas) Grade II. ER pos, PR neg, HER2 normal	Macro and Micro
2 (43,5)	3,6	Invasive ductal carcinoma. AL metastasis: 1/17	63  mm diameter Grade II. ER pos, HER2 normal.	Macro and Micro
3 (38,8)	4,9	Invasive ductal carcinoma with satellites, DCIS. AL metastasis: 0/15	30 mm diameter Grade III. ER pos, HER2 normal, Ki67-PI >80%	Macro
4 (48,4)	5,2	Invasive ductal carcinoma, DCIS AL metastasis: 9/15	35 mm diameter Grade III. ER neg, HER2 normal Ki67-PI = 70%	Macro and Micro
5 (44,3)	7,0	DCIS. Sentinel node: 0/2	Area >60 mm diameter Van Nuys group 2	Macro

AL: axillary lymph node; ER: estrogen receptor; PR: progesterone receptor; HER2: HER2 receptor expression; Ki67-PI: Ki67 proliferation index.

Data on expected numbers of invasive breast cancers was obtained from NORDCAN. NORDCAN comprise information on incidence, mortality, prevalence and survival statistics on more than 50 major cancers in the Nordic countries [1]. No information on ductal carcinoma in situ is retrievable from this database. In order to calculate the expected number of breast cancers in our population, data for age specific incidence rates from 2006 to 2014 from NORDCAN was used.

Statistics

Statistical calculations were made using Stata 14. Statistical significance was defined as a two-sided p-value <.05. Risk ratio for the examination procedures was calculated using Fishers exact test.

Study 2 – mapping of specimen examination method in Denmark

Information on clinical practice regarding examination method of tissue from breast reduction surgery was obtained through e-mails to all 12 Departments of Pathology in Denmark.

Results

Study 1 – quality of specimen examination method

For 84.3% (n = 306) of the women, only macroscopic examination was found to be relevant. Further microscopic examination was performed in the remaining 15.7% (n = 57) of the women. In two of the 306 and three of the 57 women breast cancer was found. One case was identified during the surgery and was excluded. The others were identified during the follow-up period.

Risk ratio for being diagnosed with breast cancer in the follow-up period showed a significantly increased risk for women whose tissue was initially examined microscopically. Risk ratio was 0.12 (95%CI 0.03–0.55), p = .0292.

Prevalence of invasive breast cancer was 0.3% (n = 4) and the incidence rate was 1.1% per 2868 person years. One patient was diagnosed solely with ductal carcinoma in situ. Incidence rate of ductal carcinoma in situ was 0.3% (n = 1) per 2868 person years. Mean follow-up was 7.9 years (95%CI 7.7–8.1).

The expected number of invasive breast cancers in the population at 8 years of follow-up was 3.47. No statistical significant difference between observed and expected incidence rates was found. Rate ratio was 1.17 (p = .734, 95%CI .28–7.04). The average age of the women were 38.8 years (95%CI 37.4–40.1).

We found no statistical significant difference in incidence rate of breast cancers based on the examination method (p = .1147).

Study 2 – mapping of specimen examination method in Denmark

The response rate was 83%, as 10 out of 12 departments answered the questionnaire. One of these did not received breast reduction specimens and was excluded. Six departments systematically conducted a microscopic examination of their specimens. The remaining three performed a microscopic examination only when considered relevant after the macroscopic examination.

Discussion

Our data suggest that the risk of missing a cancer at the examination of breast reduction specimens by macroscopic examination is low. The prevalence of invasive breast cancer in our study was in line with previous reported prevalence ranging from 0% to 0.99% [5–14]. Overlooked breast cancer is therefore very rare in breast reduction specimens. Invasive breast cancer is even rarer in studies that, like ours, subtracted data for women with prior breast cancer [5,7,8,10,11,14].

A former Danish autopsy study of 110 younger to middle aged women found a much higher prevalence of invasive breast cancer (2%) and ductal carcinoma in situ (14%). A total of 275 tissue samples were here thoroughly microscopically examined suggesting a correlation between the number of tissue samples and probability of finding histologic abnormalities on a microscopic level.

Many international studies recommend consistent microscopic examination based on findings of high abnormality prevalence in breast reduction specimens alone [6,10,11,14]. Detection of other abnormalities than invasive breast cancer or ductal carcinoma in situ do not have the same clinical consequences in Denmark, therefore these recommendations cannot be applied uncritically. Lobular carcinoma in situ for example is still regarded an incidental finding, although increasing the subsequent risk of carcinoma [15]. Comparison with international studies and application of their recommendations are therefore difficult.

Women with benign breast abnormalities, typically found in breast reduction specimens, generally have more precursor prone tissue that can lead to breast cancer [6,14]. Second, invasive breast cancer found at the time of operation immediately gets histologically diagnosed by microscopic analysis. Excluding the one case in our study that was diagnosed at the time of the operation from the statistical analysis results in no difference between the incidence rates of breast cancers based on the examination method.

We can’t exclude that a small breast cancer has been removed at the breast reduction without diagnosing it in the subsequent pathology examination. A further limitation is the relatively few numbers of women with invasive breast cancer and ductal carcinoma in situ in this study. Several large international retrospective studies have found fewer observed invasive breast cancer than expected in cohorts of women who underwent breast reduction surgery [13,16,17].

Identified cases

Evaluation of each identified breast cancer case is shown in Table 2. The timespan between the operations and the succeeding cancers, combined with tumor characteristics and clinic, makes it unlikely that any of the breast cancers were overlooked at the initial pathology examination (Supplementary material).

Evaluating the examination method

The examination procedure chosen by pathology departments should represent a method not only oncological safe but also one that makes economic sense.

The pathologists in this study found macroscopic abnormalities in 16% of the specimens, which is high when compared to Cook et al. [12]. This could reflect that macroscopic examination is performed more thoroughly when subsequent microscopic examination is not standard procedure. Despite this, breast tissue initially not considered suspicious of occult cancer by macroscopic examination can contain abnormalities [7,12,18].

Conclusion

Breast carcinoma is rare in specimens from breast reduction surgery. This study examined the quality of a time-efficient method where microscopic examination was only done when macroscopically suspect areas were found. We did not find evidence of undiagnosed breast cancers at the time of surgery. Therefore, the method seems appropriate. The method is, however, dependent on a careful macroscopic examination and we encourage the results to be replicated in other populations so national guidelines for macroscopic examination can be established.

Trial registration/ethics

This study was approved by the Danish Data Protection Agency and Danish Patient Safety Authority.

Disclosure statement

The authors report no conflicts of interest