M.C. Falcou

Extensive pure ductal carcinoma in situ of the breast: Identification of predictors of associated infiltrating carcinoma and lymph node metastasis before immediate reconstructive surgery

AIM To identify predictors for infiltrating carcinoma and lymph node involvement, before immediate breast reconstructive surgery, in patients with an initial diagnosis of extensive pure ductal carcinoma in situ of the breast (DCIS). PATIENTS AND METHODS Between January 2000 and December 2009, 241 patients with pure extensive DCIS in preoperative biopsy had underwent mastectomy. Axillary staging (sentinel node and/or axillary dissection) was performed in 92% (n = 221) of patients. Patients with micro-invasive lesions at initial diagnosis, recurrence or contralateral breast cancer were excluded. RESULTS Respectively 14% and 21% of patients had a final diagnosis of micro-invasive carcinoma (MIC) and invasive ductal carcinoma (IDC). Univariate analysis showed that the following variables at diagnosis were significantly correlated with the presence of either MIC or IDC in the mastectomy specimen: palpable tumor (p = 0.002), high grade DCIS (p = 0.002) and detection of an opacity by mammography (p = 0.019). Axillary lymph node (ALN) involvement was reported in 9% of patients. Univariate analysis suggested that a body mass index higher than 25 (p = 0.007), a palpable tumor (p = 0.012) and the detection of an opacity by mammography (p = 0.044) were associated with an increased rate of ALN involvement. CONCLUSION Skin-sparing mastectomy and immediate breast reconstruction (IBRS) has become increasingly popular, especially for patients with extended DCIS of the breast. This study confirmed that extended DCIS is associated with a substantial risk of finding MIC or IDC on the surgical specimen but also ALN involvement. Adjuvant systemic treatment and/or radiotherapy could be indicated for some of these patients after the surgery. Patients should be informed of the rate of 1) complications associated to IBRS that will potentially delay the introduction of systemic or local therapy 2) complications associated to radiotherapy after IBRS.

Validation over time of a nomogram including HER2 status to predict the sentinel node positivity in early breast carcinoma

BACKGROUND The molecular subtypes of breast cancer have different axillary status. A nomogram including the interaction covariate between estrogen receptor (ER) and HER2 has been recently published (Reyal et al. PLOS One, May 2011) and allows to identify the patients with a high risk of positive sentinel lymph node (SLN). The purpose of our study was to validate this model on an independent population. METHODS We studied 755 consecutive patients treated at Institut Curie for operable breast cancer with sentinel node biopsies in 2009. The multivariate model, including age, tumor size, lymphovascular invasion and interaction covariate between ER and HER2 status, was used to calculate the theoretical risk of positive sentinel lymph node (SLN) for all patients. The performance of the model on our population was then evaluated in terms of discrimination (area under the curve AUC) and of calibration (Hosmer-Lemeshow HL test). RESULTS our population was significantly different from the training population for the following variables: median tumor size in mm, lymphovascular invasion, positive ER and age. The nomogram showed similar results in our population than in the training population in terms of discrimination (AUC=0.72 [0.68-0.76] versus 0.73 [0.7-0.75] and calibration (HL p=0.4 versus p=0.35). CONCLUSIONS Despite significant differences between the two populations concerning variables which are part of the nomogram, the model was validated in our population. This nomogram is robust over time to predict the likelihood of positive SLN according to molecular subtypes defined by surrogate markers ER and HER2 determined by immunohistochemistry in clinical practice.

Distant Metastasis Free Survival (DMFS) in Breast Cancer Patients with Micrometastases (pN1mi) in the Sentinel Lymph Node (SLN): Results in 582 Positive-SLN Patients in a Single Institution.

Background:The risk of developing distant metastases (DM) in pN1mi and isolated tumor cells (pN0i+) patients remains under question. Does occult axillary node metastases is an additional factor for using an adjuvant systemic therapy (AST) in early breast cancer ?Patients and Methods:Among 2695 patients operated on from 2000 to 2006 for SLN, 582 patients had a positive SNB: 307 were pN1, 154 pN1mi and 121 pN0i+ (6th AJCC-classification). All patients underwent an immediate or delayed Axillary Lymph Node Dissection (ALND). We report the results for DMFS [median follow-up of 56 months (2-105)].We used Kaplan-Meier method and Cox regression for multivariate analysis.Results:ALND were positive in pN1, pN1mi and pN0i+ patients respectively in 127 (41,3%), 20 (13%) and 14 (11.6%) of these patients. On univariate and multivariate analysis, positive ALND, mitotic index, pathologic tumor size were significantly related to the DMFS; on multivariate analysis, the type of axillary lymph node metastases was an additional significative factor. There was not relationship between pN0i+ and the development of DM. Surprisingly, patients with pN1mi had a 2.8 higher risk for developing DM than pN1 patients. pN1 patients receive more AST than pN1mi (75% and 22%), however AST was not prognostic (p=0.49).Conclusion:In our series, patients with pN1mi were associated with a worse prognosis related to DMFS compared to pN1. Use of AST and/or biological primary tumor characteristics could explain this paradoxical result. Longer follow-up and larger series are needed to determine the prognostic significance of axillary occult metastases. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 308.

Abstract P5-07-05: Development of individual adjuvant chemotherapy benefit estimating program for breast cancer patients management

Background To predict the individual benefit of adjuvant chemotherapy (CT) is challenging especially in estrogen receptor tumors HER2 negative tumors. Existing tools are based on data from randomized trials or on genomic tests with benefit calculated for outmoded CT. The aim of this study was to develop an exportable model to predict from real life data the invidual benefit of adjuvant anthracyclines/taxanes based CT. Patients and methods Women with estrogen receptor-positive HER2-negative tumors without metastasis at the time of the diagnosis treated at the Institut Curie - Centre Rene Huguenin (St Cloud, France) between 2000 and 2008 were included. Clinical characteristics, pathological data and information concerning the treatments and outcomes had been prospectively registered. We divided the study population into 2 groups: patients who did not receive adjuvant CT and patients who did. Multivariate survival analysis and prediction were performed according to the Cox model (proportional hazard model). Non-informative variables in the Cox model were excluded from the final model. The individual benefit of CT was calculated by comparing predicted distant disease-free survival without chemotherapy (using a model constructed in patients treated without adjuvant CT) and modeled distant disease-free survival with CT. Benefit of chemotherapy was validated by cross validation and compared to Adjuvant online predictions. Results Data from 3385 women were available: 2137 treated without adjuvant chemotherapy and 1248 with. The models to predict survival without and with CT were based on tumor size, number of metastatic nodes, grade and KI67 (all patients had ER-positive-HER2 negative tumors). The discrimination and the calibration of the models were excellent: C-index were 0.81 and 0.76 for patients treated without and with CT. In this population, the mean 10-year benefit provided by CT was 6.2% (median: 1.5%, min: -4%, max: 35%). The accuracy of the model was internally validated and over-performed predictions of adjuvant online because Ki67 and HER2 status are included in our model (p Conclusion We constructed a tool to evaluate the benefit of adjuvant anthracycline / taxane at an individual level. This tool is based on a model that outperform currently availble methods. Citation Format: Rouzier R, Stevens D, Bonneau C, Falcou M-C, Hequet D, Pierga J-Y, Lerebours F. Development of individual adjuvant chemotherapy benefit estimating program for breast cancer patients management. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P5-07-05.

Abstract P3-07-63: Ki67 cut-off point to predict the benefit of adjuvant chemotherapy in ER+ HER2- breast cancer patients

BACKGROUND: The benefit of chemotherapy for patients with estrogen receptor(ER)+ HER2- breast cancers is an ongoing question. The evaluation of the tumor9s proliferation by Ki67 may guide the indication but no consensual predictive cut-off has been set yet. MATERIALS AND METHODS: This study included women with a first ER+HER2- invasive breast cancer treated by primary surgery between 2003 and 2008. Data was collected prospectively. Ki67 cut-off was sequentially set each 1% from 5% to 30%. For each threshold, the interaction between Ki67 and adjuvant chemotherapy was integrated in a multivariate Cox model to determine when it became statistically significant in predicting distant-disease free survival (DDFS). Using different Ki67 cut-offs, we also compared DDFS of patients who had or not an indication of chemotherapy, depending on whether they actually received it or not. RESULTS: Among the 3221 breast cancers, median Ki67 was 10%, with a mean of 15% (S.D = 14). Ki67 was an independent prognosis factor whether the cut-off was set to 14% or to 20% (p CONCLUSION: In ER+ HER2- breast cancers, a Ki67 level of expression >= 20% was predictive of benefit from adjuvant chemotherapy. A threshold at 14% was not as discriminant. Based on this large cohort study, we recommend the use of a cut-off at 20% to decide whether patientes with ER positive tumors should receive chemotherapy or not/. Citation Format: Rouzier R, Rossi L, Lerebours F, Savignoni A, Falcou M-C, Bonneau C, Pierga J-Y, Guinebretiere J-M. Ki67 cut-off point to predict the benefit of adjuvant chemotherapy in ER+ HER2- breast cancer patients. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-07-63.

Validation dans le temps d’un nomogramme prédictif de positivité du ganglion sentinelle axillaire en fonction des sous-types moléculaires de cancer du sein

Les sous-types moleculaires de cancer du sein ont des statuts axillaires differents. Un nouveau nomogramme incluant l’interaction entre les recepteurs aux oestrogenes (RO) et le statut HER2 a recemment ete publie et permet d’identifier, avant la chirurgie, les patientes ayant un haut risque d’atteinte du ganglion sentinelle axillaire (GS) [1]. L’objectif de notre etude etait de valider ce modele sur une population independante.

PD02-08: Validation over Time of a Nomogram Predicting the Sentinel Node Positivity in Early Breast Carcinoma According to the Molecular Subtypes Classification.

Background: The molecular subtypes of breast cancer have different axillary status. A new nomogram including the interaction covariate between estrogen receptor (ER) and HER2 status has been recently published ( Reyal et al. PLOSone, may 2011 ) and allows to identify before surgery the patients with a high risk of positive sentinel lymph node (SLN). The purpose of our study was to validate this model on an independent population. Patients and methods: We studied 755 consecutive patients treated for operable breast cancer with sentinel node biopsies in 2009, from the Institut Curie breast cancer prospective database. Baseline characteristics were compared between our population and the population used to build the model, using Chi-square test for categorical variables and Kruskal-Wallis test for continuous variables. The multivariate model, including age, tumor size, lymphovascular invasion and interaction covariate between ER and HER2 status, was used to calculate the theoretical risk of positive sentinel lymph node (SLN) for all patients. The performance of the model on our population was then evaluated in terms of discrimination (area under the curve AUC) and of calibration (Hosmer-Lemeshow HL test). Results: Characteristics of our population were significantly different from the training population for the following variables: tumor size (median 12mm [1-60] versus 13mm [1-100] p=0.005), lymphovascular invasion (18.6% versus 23.7% p=0.006), positive ER (91.4% versus 87% p=0.002) and age as followed: 56.7% of patients ≤ 60 versus 63.1%, 17.5% of patients between 60 and 65 versus 14.1% and 25.8% of patients above 65 versus 22.8% p=0.01. The nomogram showed similar results in our population than in the training population in terms of discrimination (AUC= 0.72 [0.68−0.76] versus 0.73 [0.7−0.75] and calibration (HL p= 0.4 versus p=0.35). Conclusions: Despite significant differences between the two populations concerning variables which are part of the nomogram, the model was validated in our population. Our study shows that this nomogram is efficient and robust over time to predict the likelihood of positive SLN according to molecular subtypes defined by surrogate markers ER and HER2 determined by immunohistochemistry in clinical practice. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD02-08.

P3-07-17: Optimal Number of Sentinel Lymph Nodes That Could Avoid Completion of Axillary Lymph Node Dissection in Operable Breast Cancer.

Background Axillary lymph node dissection (ALND) has been the standard of care for breast cancer patients with sentinel lymph node (SLN) metastasis. It is now under discussion since the publications of ACOZOG Z0011 and NSABP-32. The 1–3% rate of axillary recurrence is very low in these studies, whereas the rate of additional nodal metastasis after the completion of ALND varies from 15% to 35%. The purpose of this study was to determine the optimal number of nodes to be removed in order to obtain an axilla free of disease after initial surgery. Patients and methods : 4928 consecutive patients treated for breast cancer ≤ 2 cm with primary breast surgery with SNB between 2000 and 2009 were reviewed from the Institut Curie breast cancer prospective database. 1165 patients with a positive SLN (macrometastasis, micrometastasis and isolated tumor cells) underwent ALND. The proportion of patients with positive ALND was analysed according to the total number of sentinel lymph nodes biopsied. Results were compared with Chi-square test for qualitative variables and with Kruskal-Wallis for quantitative variables. Results : Among the 1165 patients, 308 (26.4%) had a positive ALND. Among them, 81.1% of patients had a macrometastasis in the SLN versus 45.6% in the 857 patients with negative ALND (p Conclusions : We showed that patients with positive SLN have a very low risk ( Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-07-17.

Abstract P5-14-18: Extended Pure Ductal Carcinoma In Situ of the Breast on Preoperative Biopsies. Prognostic Factors for Infiltrating Carcinoma and Lymph Node Involvement

Introduction: The aim of our study was to identify predictive factors of infiltrating carcinoma and lymph node involvement in patients with an initial diagnosis of extended pure ductal carcinoma in situ (DCIS) of the breast. Material and Methods: 241 patients diagnosed with extended pure ductal carcinoma in situ (DCIS) underwent treatment at the Institut Curie (2000-2009) consisting of mastectomy with or without immediate breast reconstruction (IMR). Axillary staging (sentinel node and/or standard procedure) was performed in 92% of patients. Patients with micro-invasive lesions at diagnosis, recurrence or contralateral breast cancer were excluded. Differences between groups were analysed by Chi-square or Fisher Exact tests for categorical variables and Student9s t-test for continuous variables. Survival analyses were performed using KaplanMeier, with comparisons using the logrank test and hazard ratios estimated using the Cox proportional hazard model. P-values were considered significant when below 0.05. Results: Respectively 15% and 20%of patients had a final diagnosis of micro-invasive (MIC) and invasive ductal carcinoma (IDC). The median sizes of the DCIS and IDC were respectively 40mm [0-95] and 6mm [2-50] according to final histological assessment. Univariate analysis showed that the following variables at diagnosis were significantly correlated to the presence of either MIC or IDC in the mastectomy specimen; palpable tumor (p=0.02), high grade DCIS (p=0.02), detection of an opacity on mammography (p=0.01). Axillary lymph node involvement was reported in 9% of patients. In univariate analysis a BMI>25 (p=0.007), a palpable tumor (p=0.01), the detection of an opacity (p=0.04) were associated with an increase rate of lymph node involvement. A IMR was performed in 69% of patients. These patients were younger (P 25 (p=0.06), a palpable tumor (p=0.0004), an opacity (p=0.01) and extended microcalcifications (p=0.02) were associated with a higher rate of loco-regional recurrence. Immediate breast reconstruction was not a significant risk factor for loco-regional recurrence (p=0.31). Conclusion: Extended pure ductal carcinoma of the breast on preoperative biopsies is associated with a substantial risk of finding not only micro-invasive or invasive carcinoma on the mastectomy specimen but also axillary lymph node involvement. Some risk factors have been identified and should be used to exclude patients from immediate reconstruction surgery due to an increased risk of getting adjuvant systemic treatment and radiotherapy. Immediate breast reconstructive surgery was not associated with an increased risk of loco regional recurrence in our series. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P5-14-18.

Do isolated cells (pN0i+) in the sentinel lymph node change the post-operative management in breast cancer?.

Abstract #206 Background: immunohistochemical (IHC) analysis of the sentinel lymph node (SLN) allows detection of occult metastases not routinely diagnosed by conventional techniques. There is, however, no consensus concerning the post-operative management of those patients with IHC-positive (pN0i+) nodes: should one re-operate, change the medical treatment or alter the irradiation fields?
 Patients and methods: 2692 patients with early invasive breast cancer underwent conservative treatment with SLN biopsy between 2000 and 2006. SLN were evaluated with frozen section followed by serial-section HES and IHC if HES showed no tumour cells. Lymph node staging followed the accepted pTNM classification: pN0, pN0i+ (≤ 0.2mm, IHC+), pNmi (0.2-2mm) and pN1a (> 2mm). In 1506 patients with T1pN0 tumours : 143 were pN0i+, that is 10%. We compared the post-operative management of pN0 patients, who had no completion axillary dissection (CAD), to those pN0i+ who did. All positive SLNs underwent CAD according to our institutional protocol.
 Results: 15 of 143 (10.5%) pN0i+ patients showed metastases in their CAD; a single node in 10 cases, 2-3 in 4 and > 3 in one patient. Univariate analysis showed chemo- and hormono-therapy to be more frequently administered in pN0i+ (24.5% vs. 77.6%) compared to pN0 (9.1% vs. 55.8%) patients; p Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 206.