Prognostic factors in ductal carcinoma in situ (DCIS)

Transcription

1 Prognostic factors in ductal carcinoma in situ (DCIS) S. Altintas, M. Huizing, W. Tjalma Ductal carcinoma in situ of the breast (DCIS) is a clinical entity which is discovered as microcalcifications on screening mammography, it rarely represents a palpable disease. Asymptomatic women with DCIS receive treatments that are similar to women with invasive breast cancer and therefore experience substantial psychological distress despite the fact that they have an excellent prognosis and normal life-expectancy. It is also true that, in spite of aggressive treatment approaches, some patients do recur. In analogue with invasive breast cancer, DCIS is a heterogeneous disease with different prognostic profiles. The high incidence of DCIS and variations in its treatment with different outcomes led to the introduction of the Van Nuys Prognostic index (VNPI) developed in 1996 by Silverstein. This index is a simple decision-making tool to improve or at least standardise DCIS care and had been incorporated in our daily practice since Data on that experience were analysed. We tried to obtain a better understanding of the molecular behaviour of DCIS laesions and looked for predictive and prognostic markers associated with disease-free survival (DFS). The next step was the use of micro-array analysis with the Genomic Grade Index (GGI), based on four proliferation genes, and the proliferation index Ki-67. These two indices, which are considered to be predictive for the behaviour of invasive breast cancer, were incorporated into the VNPI. Furthermore, we looked if the tumour microenvironment might play a crucial role in local relapse of DCIS and risk of subsequent invasive disease. (Belg J Med Oncol 2012;6: ) Introduction Ductal carcinoma in situ of the breast (DCIS), a precursor of invasive breast cancer, represents a spectrum of abnormal cells confined to the breast duct. Per definition, DCIS has not yet invaded beyond its intraductal origin and may never invade its neighbouring tissue. So in fact, it is a curable disease after local treatment. 1 With the advent of widespread screening for invasive breast cancer in the early to mid-1980 s, the detection and thus the incidence of DCIS have increased dramatically. 2 Treatment options are mastectomy and local excision with or without radiotherapy. Mastectomy and local excision with radiotherapy are both effective approaches with long-term survival (96-98%). Still, there is a higher local recurrence risk in patients treated with conservative treatment (12%, half of whom have invasive cancer). Randomised trials show that the addition of radiotherapy after local excision halves the risk for both invasive and non-invasive Authors: Ms. S. Altintas MD PhD, Ms. M. Huizing MD PhD, Department of Medical Oncology and Breast Unit, Antwerp University Hospital, Edegem, Belgium; W. Tjalma MD PhD, Department Gynaecological Oncology and Breast Unit, Antwerp University Hospital, Edegem, Belgium. Please send all correspondence to: Ms. S. Altintas MD PhD, Department Medical Oncology, Antwerp University Hospital, Wilrijkstraat 10, 260 Edegem, Belgium, tel: +32 (0) ; Sevilay.altintas@za.be. Conflict of interest: the authors have nothing to disclose and indicate no conflicts of interest. Keywords: DCIS, VNPI, prognostic markers. 164

2 Table 1. Van Nuys Prognostic Index Score 2003, University of California Score Tumour size (mm) Margins* (mm) <1 Pathological Classification non-high grade** no necrosis*** non-high grade** necrosis*** high grade** with/without necrosis*** Classification > <40 Distance of tumour to margin of biopsy specimen **Nuclear grade ***Comedo necrosis, not individual cells local recurrences, with equivalent survival. 3-6 Today, we are not able to identify accurately which patient with DCIS will progress to invasive breast cancer and how to prevent this progression. Identifying predictive and prognostic clinicopathological factors and biomarkers that are reflective of biology would better inform therapeutic decision-making. 7,-9 The natural history of DCIS is not fully elucidated. 10 Autopsy studies suggest that latent DCIS is relatively common, ranging from % in a series of women who died of causes other than breast cancer. 11,12 No more than one third of all in situ carcinomas will progress to invasive breast cancer. Given the lack of clarity about prognostic and predictive factors, and the heterogeneity of the disease, important therapeutic questions so far remained unanswered. Currently for every four diagnoses of invasive breast cancer, there is one diagnosis of DCIS. The risk of DCIS increases steadily from age 40-0 years. The risk increases much more slowly after the age of 0 years and plateaus after the age of 60 years and this is highly and consistently associated with the concurrent increase in rates of mammography screening. Risk factors for DCIS includes high mammographic density, family history, increasing age, menopausal oestrogen with progestin therapy, late age at menopause, nulliparity, late age at first birth and high postmenopausal body mass index. Other tumour-related risk factors are: grade (high, intermediate and low), presence or absence of necrosis, type (comedo, cribriform, papillary, micropapillary and solid) and size (the average size is 1-1. cm). The most aggressive form is called comedo-type with high-grade cellular and nuclear features. The Van Nuys Prognostic Index 1996 (VNPI), based upon tumour size, pathological grade and tumour margins, is a guideline for the treatment of ductal carcinoma in situ (DCIS). It was thought to strongly decrease overtreatment (Table 1). 7 Treatment modalities are mastectomy and tumourectomy with or without radiotherapy. Only tumourectomy as therapeutic intervention is rather rare and in most centres radiotherapy is applied after breast conservative surgery (Table 2). Table 2. Treatment of Choice based on VNPI 2003 Score Treatment 4,,6 tumourectomy 7,8,9 tumourectomy + radiotherapy 10,11,12 mastectomy In 2003, age was added to the VNPI as a fourth prognostic factor. We examined changes in treatment modality after applying the VNPI (1996 and 2003) retrospectively to 104 cases treated between 1996 and 2003 and investigated if the addition of age to the Index caused a shift in treatment. The influence of tumour size, pathological grade, tumour margins and age on DFS was calculated with the Kaplan-Meier method and the log-rank test. The median follow-up was 36 months. Twelve patients showed recurrence (11.%), of whom seven were invasive (8%). The introduction of age caused no significant shift in treatment modalities. Significant differences in DFS were seen between large ( 41 mm) and small ( 1 mm) tumours (p=0.0074), old (>60 years) and young (<40 years) patients (p=0.024) and Van Nuys Subgroup 2 and 3 (p=0.04). Tumour margins and pathological grade showed no significant difference in DFS. We concluded that the VNPI can be a useful tool in the treatment of DCIS. 16

3 The next step was a search for molecular markers, besides clinicopathologic features, that could also play an important role in developing a local recurrence and thus predict outcome. Cases of DCIS diagnosed in two Belgian University Centres were included. Paraffin-embedded material and Hematoxylin- and Eosin-stained slides of DCIS cases were reviewed and tumour size, margin width, nuclear grade, and comedo necrosis were assessed. Molecular markers (oestrogen receptor, progesterone receptor, HER1-4, Ki67, and c-myc) were assayed immunohistochemically on 19 patients with a median age of 4 years (range 29-78). Applied treatment strategies were correlated with the prospective use of the VNPI score. The median time of follow-up was 4 months (range -23). Twenty-three patients developed a recurrence (14.%). The median time to recurrence was 46 months (range -23). Before the introduction of the VNPI, 37.% of the DCIS patients showed a recurrence while thereafter 6.7% recurred (p<0.00). Independent predictors of recurrence were tumour size, margin width, pathological class and age, but none of the studied molecular markers showed this. Overexpression of HER4 in the presence of HER3 was found to be associated with a better DFS (p<0.0). This study confirms the value of the VNPI score and questions the benefit of an aggressive approach in the low-risk DCIS laesions. In a third part of our research we looked for a genomic profiling in DCIS. As is the case in invasive breast cancer, molecular portraits and gene expression profiling, which allow individualized medical treatment, seem to be essential in classifying DCIS in distinct molecular subtypes. Since DCIS and invasive breast cancer seem to show concordant biological behaviour, incorporation of proliferation markers into the VNPI seemed to be attractive for prognostication and tailored treatment. A qrt-pcr genomic grade index, based on four proliferation genes (PCR-GGI) selected from the GGI microarray signature of Sotiriou (Genomic Grade Index, MapQuant DX Genomic Grade, Ipsogen) was applied to 88 representative DCIS tissue samples. We assessed the prognostic value of the VNPI after integration of those proliferation genes and the proliferation index Ki-67. DCIS samples were divided into three VNPI subgroups (low-risk [score 4-6], intermediate-risk [score 7-9] and high-risk [score 10-12]) based on nuclear grade ± necrosis, tumour size, margin width, and age. Nuclear grade was substituted by the genomic grade index (GGI) to generate the VNPI-GGI and combined with the Ki-67 to generate the VNPI-Ki67. A total of 88 cases (median age 4 years) was identified out of 168 DCIS patients. Median follow-up was more than five years. A statistical association was observed between a high VNPI score and a higher risk of recurrence (HR=7.72 [9% CI ], p=0.049). Ki-67 did not improve the prognostic value of VNPI (HR=6., [9% CI ], p=0.08). In contrast, the VNPI-GGI could identify more accurately high-risk DCIS patients with early relapses within years (HR=18.14 [9% CI ], p=0.01). As a last part of the study we aimed to look further for alterations in the tumour microenvironment. The association between the CD10 expression pattern and risk of relapse was studied. The local death of myoepithelial cells (MECs) is believed to be a pre-requisite to tumour invasion. We thus hypothesised that loss of CD10 expression, a MEC surface peptidase, would signify basement membrane disruption and confer increased risk of relapse in DCIS. CD10 expression was evaluated by quantitative RT-PCR and immunohistochemistry using paraffin embedded samples of normal breast tissue (n=11); of morphologically normal ducts associated with DCIS (n=10); and of DCIS without an invasive component (n=14). CD10 immunostaining was only observed in MECs in normal tissues and DCIS. Normal tissue showed high mrna expression levels of CD10, whereas DCIS showed a variable range. After a median follow-up of six years, DCIS with CD10 expression below the levels observed in normal tissue (71%) demonstrated a higher risk of local relapse (HR=1.88; [9CI: ], p=0.001) in univariate analysis. No relapse was observed in the patients expressing high CD10 mrna levels (29%) similar to the ones observed in normal tissue. In multivariate analysis including known prognostic factors, low CD10 mrna expression remained significant (HR=2.2; [9%CI: ], p=0.008) as did the recently revised VNPI score (HR:2.03; [9%CI: ], p=0.006). We could conclude that the decrease of CD10 expression in MECs is associated with a higher risk of relapse in DCIS; 166

4 Key Messages for daily practice 1 Recognise DCIS as a highly curable disease entity with an excellent long-term disease-free survival and overall survival, especially the low risk group. 2 One-size-fits-all theory cannot be applied to DCIS patients since all current therapies have their own short- and long-term side-effects. 3 Mastectomy cures DCIS for up to 98%. 4 Radiotherapy has an important role in a subgroup of patients. Hormonal therapy is not standard of care in DCIS. 6 VNPI is a useful clinical tool in defining subgroups and can tailor treatment options. 7 Molecular characteristics, Genomic Profile and the microenvironment play an important role in the transition from in situ to invasive. this knowledge has the potential to improve DCIS management. Conclusion DCIS is currently a growing health issue due to the extended mammographic screening and the routine use of mammography in symptomatic patients. Based on molecular and genetic studies, DCIS has been recognised as the precursor laesion of invasive breast cancer and adequate risk stratification may be important to tailor treatment. The Van Nuys Prognostc Index (VNPI) 2003 is a reliable clinical tool that includes four different predictors (age, grade, size and margin) providing a statistical risk of local failure at twelve years. The most useful aspect of the VNPI is its ability to define a low risk subset whose likelihood of local recurrence is so low that radiotherapy provides no significant additional benefit for local control and will even cause more harm. In this thesis, we address the practical use of this tool to identify a low risk subset which exhibited a local recurrence rate of 7.1% at 4 months. This was even lower than the incidence of breast cancer in our Flemish area. Overall, the VNPI can be a useful tool for communicating treatment decisions in both the breast unit and with the individual patient. Further refinement of this risk assessment tool lies in the development of molecular markers, gene expression profiling and studies of the micro- environment. Several markers described in this thesis can be used in combination with the VNPI to identify those patients who are at risk for developing recurrence and/or invasive breast cancer. In the future, further translational research, including molecular markers, genomic grading and CD10, should be incorporated in randomised phase III trials. Till these trials are conducted we strongly recommend the use of the VNPI in daily practice. This index is a simple decision-making tool to improve or at least standardize DCIS care. References 1. Wellings SR, Jensen HM: On the origin and progression of ductal carcinoma in the human breast. J. Natl Cancer Inst 1973;0: Ernster VL, Ballard-Barbash R, Barlow WE et al. Detection of ductal carcinoma in situ in women undergoing screening mammography. J Natl Cancer Inst 2002;94: Meijnen P, Peterse JL, Bogaerts J et al. Breast-conserving treatment with or without radiotherapy in ductal carcinoma-in-situ: ten-year results of European Organisation for Research and Treatment of Cancer randomised phase III trial a study by the EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. J Clin Oncol. 2006;24: Fisher B, Dignam J, Wolmark N et al. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet. 1999;33: Houghton J, George WD, Cuzick J et al. Radiotherapy and tamoxifen in women with completely excised ductal carcinoma in situ of the breast in the UK, Australia, and New Zealand: randomised controlled trial. Lancet. 2003;362:

5 6. Emdin SO, Granstrand B, Ringberg A et al. SweDCIS: Radiotherapy after sector resection for ductal carcinoma in situ of the breast. Results of a randomised trial in a population offered mammography screening. Acta Oncol. 2006;4: Silverstein MJ. The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast. Am J Surg 2003;186: Holland R, Peterse J, Mills R, et al (1994). Ductal carcinoma in situ proposal for a new classification. Semin Diagn Pathol 11: Cornfield DB, Palazzo JP, Schwartz GF et al. The prognostic significance of multiple morphologic features and biologic markers in ductal carcinoma in situ of the breast: a study of a large cohort of patients treated with surgery alone. Cancer 2004;100: Sanders ME, Schuyler PA, Dupont WD et al (200). The natural history of low grade ductal carcinoma in situ of the breast in women treated by biopsy only revealed over 30 years of long-term follow-up. Cancer 103: Nielsen M, Thomsen JL, Primdahl S et al: Breast cancer and atypia among young and middle-aged women: a study of 110 medicolegal autopsies. Br J Cancer 1987;6: Bhatal PS, Brown RW, Lesueur GC et al: Frequency of benign and malignant breast lesions in 207 consecutive autopsies in Australian women. Br J Cancer 198;1: