Abstract P6-09-14: Prognostic significance of a modified residual disease in breast and nodes (mRDBN) algorithm after neoadjuvant therapy for breast cancer

Background: Patients achieving pathologic complete response to neoadjuvant chemotherapy have excellent disease free and overall survival. For patients with residual disease, the residual disease in breast and lymph node (RDBN) method provides useful prognostic information. RDBN is calculated as follows: 0.2*tumor size (in cm)+lymph node status (0-3) + tumor grade (1-3). pCR, low, intermediate and high risk of recurrence categories correspond to RDBN index of 0, 0.1 to 2.9, 3 to <4.4, and ≥ 4.4, respectively. We hypothesized that the prognostic accuracy of RDBN may be improved by also taking into account the residual tumor cellularity.

Methods: Retrospective review of 614 consecutive patients who underwent neoadjuvant therapy for breast cancer was performed. At our institution, tumor size/volume reduction in the breast is determined using the equation:

Estimated % tumor size reduction = [(pre-therapy clinical size – “revised” pathology tumor size)/pre-therapy clinical size]*100.

“Revised” pathology tumor size is calculated by multiplying the largest dimension of the gross tumor bed by the invasive tumor cellularity of the tumor bed (in comparison to the pre-therapy core biopsy sample). For example, if a 3 cm tumor bed has only 50% cellularity for invasive cancer (in comparison to pre-therapy core biopsy), the revised tumor size is 1.5 cm. Hence, we were able to use the “revised tumor size” for calculating the modified RDBN index (mRDBN). We also used gross tumor bed size for gross RDBN (gRDBN) to compare with mRDBN. mRDBN and gRDBN could be calculated on 459 of the 514 cases. Chi-Square statistical analysis was performed.

Results: Mean follow up was 33.1 months (median 31, range 4-70).

The results are shown in Table 1 & 2.

Table 1. Overall Recurrence and MortalityRDBN Score Category Overall Recurrence Mortality  nn (%)RR95% CI; pn (%)RR95% CI; pmRDBN (n=459)High5829 (50.0)19.63[7.22, 53.40]; p=<0.000118 (31.0)16.24[4.97, 53.10]; p=<0.0001Intermed16433 (20.1)7.9[2.86, 21.78]; p=<0.000115 (9.1)4.79[1.41, 16.21]; p=0.006Low803 (3.8)1.47[0.34, 6.42]; p=0.694 (5.0)2.62[0.60, 11.41]; p=0.23pCR1574 (2.5)REF 3 (1.9)REF gRDBN (n=459)High8131 (38.3)15.02[5.49, 41.09]; p=<0.000119 (23.5)12.28[3.74, 40.26]; p=<0.0001Intermed14932 (21.5)8.43[3.05, 23.26]; p=<0.000116 (10.7)5.62[1.67, 18.89]; p=0.003Low722 (2.8)1.09[0.20, 5.82]; p=12 (2.8)1.45[0.25, 8.51]; p=1pCR1574 (2.5)REF 3 (1.9)REF 

Table 2; Reclassification of gRDBN categoriesgRDBN mRDBN ReclassificationClassificationnLow (%)Intermed (%)High (%)Low7272 (100)0 (0)0 (0)Intermed1498 (5.4)140 (93.9)1 (0.7)High810 (0)24 (29.6)57 (70.4)

Conclusions: Both mRDBN and gRDBN provide prognostic information; however, separation of categories is improved with mRDBN (Table 1). mRDBN reclassified 30% of the high risk-gRDBN patients into intermediate risk category with a recurrence rate of 20%, leaving the 'true' high risk subgroup with a revised recurrence rate of 50% (Table 2). RDBN index also identified a group of low risk patients who have prognosis similar to patients with pCR.

Citation Format: Farrugia DJ, Landmann A, McAuliffe PF, Diego EJ, Johnson R, Bonaventura M, Soran A, Dabbs DJ, Clark B, Lembersky BC, Puhalla SL, Brufsky A, Jankowitz R, Davidson NE, Ahrendt GM, Bhargava R. Prognostic significance of a modified residual disease in breast and nodes (mRDBN) algorithm after neoadjuvant therapy for breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-09-14.

Abstract P3-05-14: A neoadjuvant window trial of endocrine response in women with invasive lobular carcinoma

Background:

Patients with invasive lobular carcinoma (ILC) would be expected to have favorable outcomes compared to patients with invasive ductal carcinoma (IDC) given that ILC is more often hormone receptor-positive (HR+), human epidermal growth factor receptor 2 (HER2)-negative, of lower grade, and displays decreased proliferation markers. Based on our preclinical studies showing differential hormone response in HR+ ILC vs. IDC and on recent studies suggesting differences in endocrine treatment response between patients with ILC vs. IDC, we designed a biomarker-driven, neoadjuvant window trial for newly diagnosed women with HR+, HER2-negative ILC. We hypothesize that Ki67 will be reduced by 85% in the fulvestrant arm compared with 60% and 75% reduction in the tamoxifen and anastrozole arms, respectively, and that Ki67 reduction will correlate with alterations in expression of ER and ER-regulated genes. Differential Ki67 effect will serve as a surrogate for outcome of patients with ILC on endocrine therapy.

Trial Design: This multicenter study (NCT02206984) will enroll 150 women with HR+ and HER2-negative ILC. A mandatory research breast tumor biopsy will be performed at baseline. Fifty patients will be randomized to each of three open-label treatment arms for 21 days: fulvestrant (two 250 mg IM injections on both day 1 and day 14), anastrozole (1mg orally daily), or tamoxifen (20 mg orally daily). Biomarkers of response will be assessed on baseline and post-treatment tumor tissue. Patients will proceed to definitive surgery on day 21 after study drug exposure, or they will undergo a second research breast core biopsy if further neoadjuvant treatment is planned.

Eligibility Criteria: Eligible patients include postmenopausal women with newly diagnosed, HR+, HER2-negative ILC (excluding pleomorphic subtype) measuring ≥ 1cm, with adequate organ function, ECOG PS ≥ 2, and agreeable to baseline research breast tumor biopsy.

Specific Aims: The primary endpoint is percent change from baseline to post-treatment Ki67 values in ILC tissue after 21 days of endocrine treatment. Comparisons across study arms will be made using a general linear model adjusting for institutional effect, with 80% power estimated for pairwise comparisons of log2(% staining) between treatment arms, allowing for 10% attrition. Secondary endpoints include post-therapy Ki67, and change in ER and PR protein expression by IHC. Finally, planned correlative studies include evaluation of gene expression, epigenetic markers, and DNA sequence variants in ILC tissues in an effort to identify biomarkers of endocrine response and putative drivers of endocrine resistance in ILC.

Target Accrual: This study will be open to enrollment by August 2015 at the University of Pittsburgh. Additional sites will be opened through the Translational Breast Cancer Research Consortium (TBCRC). We anticipate an accrual rate of 8 patients per month.

(Funding from Susan G. Komen® and AstraZeneca).

Citation Format: Jankowitz RC, McAuliffe PF, Sikora MJ, Butler L, Ahrendt G, Johnson R, Diego E, Bonaventura M, Puhalla S, Lembersky B, Clark B, Brufsky A, Kurland BF, Davidson NE, Dabbs DJ, Oesterreich S. A neoadjuvant window trial of endocrine response in women with invasive lobular carcinoma. [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 P3-05-14.

Abstract P1-01-15: Breast cancer biomarkers: Global tumor biomarker quality assurance study of intratumor heterogeneity

Introduction

Biomarkers (BM) are often performed on core biopsies(CB) of breast tumors for treatment purposes, yet comprehensive information on BM heterogeneity is lacking. This quality assurance study is to survey BM intra-tumor heterogeneity (ITH) by comparing BM from CB and the examination of entire breast tumors from surgical excisions (SE). This interim report supplies data on 45/100 collected cases to date.

Methods:

Whole tumor cases, 3.0cm or less were collected in years 2013-2014, excluding neoadjuvant cases and were examined with CB by two pathologists. All tissues were processed using 2013 ASCO/CAP guidelines. Immunohistochemistry (IHC) for ER/PR/Her2/Ki67 was performed on breast core biopsies and each block of a tumor surgical excision (1-13 blocks per case). IHC was performed using the Ultra Ventana-Roche device with FDA IVD clones ER (SP1), PR(1E2) Her2(4B5) Ki67(30-9). Whole slides were semi-quantitated for ER/PR using HScore (0-300), Her2 was interpreted with 2013 ASCO/CAP guidelines and Ki67 proliferation index (PI) as estimated whole slide percent of tumor cell nuclear stain. Mean HScores and PI were calculated for all BM on SE, and CB deviations from the SE mean were calculated for ER/PR/Ki67. ITH of BM was assessed by comparing results among all tissue blocks for each case. Her2 results were descriptive. Eight cases lacked CB Ki67 PI.

Results

There were no cases in which categorical hormone receptors results (positive/negative) differed between CB and SE.

2/45 (4%) cases were Her2 positive, and 13/45(28%) were equivocal by IHC with one case amplified by in situ hybridization.

Comparing ER between CB/SE , 3/45 differed in HScores by 30% or more. 10/45 SE ER had ITH of 25% or more among different tissue blocks. 14/45 cases had CB/SE PR difference in HScore of 30% or more. 18/45 cases had SE PR with ITH HScore of 25% or more between tissue blocks.

Ki67 PI differed between CB/SE by more than 25% in 18/37 cases, with ITH among tissue blocks of more than 25% in 23/37 cases. The variability of CB-SE results and ITH among tissue blocks are summarized in table 1.

Table 1 Summary of ResultsCB-SE ERITH ERCB-SE PRITH PRCB- SE PIITH PI>30%>25%>30%>25%>25%>25%3/4510/45 (22%)14/45 (31%)18/45 (40%)18/37(49%)23/37(62%)CB-SE: core biopsy vs surgical excision; ER PR as H Score

Conclusions

In this interim report of 45/100 cases, (1) ITH of 25% or more occurred for ER content in 22 % of cases (2) similar ITH exists for PR content in 40% of cases. (3) Ki67 PI tissue block ITH was expressed with variation of 25% or more in 23/37 cases.

(4) While this degree of variability of BM may not affect treatment decisions, the tissues show sufficient BM variability that should stimulate concern for accuracy of results of prognostic gene expression profiles and genomic testing of breast tumors.

Citation Format: Dabbs DJ, Clark B. Breast cancer biomarkers: Global tumor biomarker quality assurance study of intratumor heterogeneity. [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 P1-01-15.