The prognostic role of HER2 expression in ductal breast carcinoma in situ (DCIS); a population-based cohort study

All patients diagnosed with a primary DCIS between 1986 and 2004 in two Swedish counties (Uppland and Västmanland) were included in this DCIS cohort (n = 458). The majority underwent breast-conserving surgery (79 %), and clear surgical margins were obtained in 88 % among all cases irrespective of surgical intervention. All cases were histopathologically re-examined. Tumour tissue biopsies of 1.0 mm in duplicate from paraffin blocks of samples taken as part of standard care were used to construct tissue microarrays (TMA). Follow up was complete up to December 15^th, 2013. This study was approved by the Ethics Committee at Uppsala University, Sweden (Dnr 2005:118) and no informed consent was needed.

Scoring and IHC protocols are previously described in detail [21, 25]. SISH was performed on the automated instrument, Ventana Benchmark (Ventana Medical Systems, Tucson, AZ), as per the manufacturer’s protocols for the INFORM HER2 DNA probe and chromosome 17 probes. Testing for the HER2 gene and chromosome 17 was performed on sequential sections. Both probes are labelled with dinitrophenol and denaturation occurred on the instrument with enzyme digestion in protease 3 for eight minutes. The detection system used a multimer labelled with goat anti-rabbit antibody horseradish peroxidase as the linking step. Visualization occurred with the sequential addition of silver acetate as the source of ionic silver, hydroquinone, and hydrogen peroxide to give black metallic silver precipitate at the probe site. Counterstaining was performed with hematoxylin II on the instrument. The time taken for the complete run was 6.5 h. Both HER2 and chromosome 17-detection were performed on the same slide run. Gene amplification was assessed using the American Society of Clinical Oncology/College of American Pathologists guideline and Australian HER2 Advisory Board criteria for single HER2 probe testing (diploid, 1 to 2.5 copies/nucleus; polysemy >2.5 to 4 copies/nucleus; equivocal, >4 to 6 copies/nucleus; low-level amplification, >6 to 10 copies/ nucleus; and high-level amplification >10 copies/nucleus) and for dual HER2/CHR17 probe testing (non-amplified ratio <1.8; equivocal ratio, 1.8 to 2.2; gene amplification, >2.2) [26]. The HER2 status was predominantly relying on the SISH data. For those cases in which SISH failed, HER2 status was based on the IHC data and cases were considered HER2 positive if the IHC score was 3+, using the HercepTest©. As for hormone receptors (estrogen receptor alpha (ER) and progesterone receptor (PR)), tumours with 10 % or more nuclei stained were considered ER and PR positive, respectively. Staining intensity was not taken into consideration.

Baseline characteristics among women with different HER2 status were compared by Chi-square test for categorical variables or analysis of variance for continuous variables. Survival analyses were performed using Kaplan-Meier curves, including the Log-Rank test. Cox proportional hazards regression models were used to generate hazard ratios (HRs) with 95 % confidence intervals (CIs), with adjustment for radiotherapy, age at diagnosis (continuous), tumour size (two categories: ≤ 25 mm; >25 mm or multifocal), and ER status. Stratification analyses were performed by mode of detection, type of surgery and ER in different multivariate models. All statistical tests were two-sided, and p-values less than 0.05 were considered significant. Data were analysed using the SPSS Statistics, version 19 (IBM, Chicago, IL, USA) and SAS 9.3 (Cary, NC, USA).

Primary endpoints were Ipsilateral Breast cancer Events (IBE) and Invasive Breast Cancer Recurrences (IBCR). IBEs were divided in new in situ events and invasive IBEs. As for IBCR, an event was defined as an invasive IBE, a regional recurrence, a contralateral invasive breast cancer or distant metastasis. IBCR free survival was calculated based on the time from primary surgery to the date of any invasive breast cancer event, date of death or date of last follow-up as by December 15^th 2013. For an event to be classified as IBE or IBCR, a minimum of 3 months had to pass after the primary surgery.

Baseline characteristics by HER2 status are presented in Table 1. The mean age at diagnosis was 58.6 years, ranging from 30 to 90 years of age. A total of 202 women were diagnosed in Västerås, whereas the remaining 256 patients were diagnosed in Uppsala. In most cases, the DCIS was detected by mammographic screening (75.5 %). All patients except one underwent surgery with the majority of patients operated with breast conserving surgery (BCS) (78.6 %). Surgery was followed by postoperative radiotherapy (RT) in 35.2 % of all cases, although primarily for patients operated with BCS of whom 44.0 % received RT (Table 1). None of the patients received hormonal therapy or chemotherapy according to Swedish clinical guidelines.

The mean follow up time was 183.5 months (range 3–329 months). A total of 105 IBEs were identified of which 54 were a new in situ and 51 an invasive IBE in the entire cohort of 458 patients. Among women treated with BCS (N = 324), recurrences were detected among 95 women comprising 49 in situ and 46 invasive cases. Eleven of the in situ IBEs were followed by a subsequent invasive IBE. One hundred and six IBCR were identified. The first invasive event was an invasive IBE in 50 cases, regional axillary metastasis in four, loco-regional together with distant metastases or distant metastasis only in twelve cases and a contralateral invasive cancer in 40 cases. In total, 32 patients developed generalized disease; fifteen subsequently to an invasive IBE, eight after an invasive contralateral cancer and nine had no reported prior local or regional recurrence. Eighteen women died from breast cancer and another 114 had died from other causes.

Of all 458 women, 420 could be HER2 classified using available SISH or IHC data; 132 were classified as HER2 positive (31.4 %) and 288 as HER2 negative. Of 344 cases with available SISH data, 118 (34.3 %) were HER2 positive and of 408 cases with IHC data 103 (25.2 %) were classified as HER2 positive (i.e. 3+). In comparison, SISH and IHC data showed concordance in 296 of 332 (89.2 %) cases with available data for both SISH and IHC. Twenty-nine of those 332 cases were classified as HER2 2+ by IHC, 21 of those 29 (72.4 %) were SISH positive and eight (27.6 %) were SISH negative. In 38 cases, data on HER2 status was missing all together. HER2 status was significantly related to tumor size, nuclear grade (NG) and hormone receptor status, and the HER2 positive tumours tended to be detected by screening more often than HER2 negative tumors. HER2 status was not associated with age at diagnosis, type of surgery or RT (Table 1). In analyses assessing risk of recurrences according to established patient-and tumor characteristics, clinically detected DCIS were more prone to recur locally (HR 1.78 (1.04–3.07), larger DCIS lesions were correlated to a borderline significant increased risk of in situ IBEs but not invasive IBEs (HR 1.88 (0.97–3.62) and 0.64 (0.30–1.34)), respectively. Other assessed factors were not associated with recurrences (Table 2).

Among the 420 women with available HER2 status, a total of 102 women experienced recurrences during follow-up. HER2 positivity in the primary DCIS was not a risk factor for IBE in women undergoing BCS (Log-Rank P = 0.40, HR 1.20 (95 % CI, 0.78–1.85)), (Fig. 1a and Table 3). Interestingly, divided by type of IBE, HER2 positivity showed a borderline statistically significant increased risk of in situ IBEs (Log-Rank P = 0.09, HR 1.63 (95 % CI, 0.92–2.89), and no association with risk of invasive IBEs (Log-Rank p = 0.48, HR 0.78 (95 % CI, 0.40–1.55)), (Fig. 1b–c and Table 3). Results from the multivariate analyses did not differ substantially from the univariate analyses (Table 3).

The risk of IBCRs was statistically significantly lower subsequent to a HER2 positive DCIS compared to a HER2 negative primary DCIS (Log-Rank P = 0.03, HR 0.60 (95 % CI, 0.38–0.94)) (Fig. 2 and Table 3). Remarkably, the curves in the Kaplan-Meier plot did not separate until after almost 10 years (Fig. 2). In the multivariate analyses, HRs after adjustments were very similar to the crude analyses (Table 3).

The survival analyses were stratified for ER-status, and for patients with an ER negative DCIS, HER2 positivity predicted a significantly lower risk of IBCR (Log-Rank, P = 0.003), (Fig. 3a), which was not the case for ER positive DCIS patients (Log-Rank, P = 0.76), (Fig. 3b). In analyses restricted to women who undergoing BCS, ER status remained an effect modifier for the association between HER2 status and IBCR.

The prognostic value of age was studied in survival analysis using age as at dichotomous variable of ≤ 50 years (N = 141, 31 %) or > 50 years (N = 317, 69 %). Age was not proven to be a prognostic indicator of neither invasive nor in situ recurrence (data not shown). However, age added substantial information to the prognostic value of HER2 status. For presumably premenopausal women aged 50 or below, HER2 expression in the DCIS revealed no prognostic information (Fig. 3c). In contrast, for women who were older than 50 years of age at diagnosis, HER2 positivity was associated with a markedly improved recurrence-free survival for invasive disease (Log-Rank, P = 0.002), (Fig. 3d). The age-stratified analyses for in situ IBEs did not reveal any significant results.