Joseph Michael Anderson

Multigene Expression Assay and Benefit of Radiotherapy After Breast Conservation in Ductal Carcinoma in Situ.

Background Most women with ductal carcinoma in situ (DCIS) will receive breast-conserving surgery (BCS) and radiation (RT). RT can be omitted for women at low risk of local recurrence (LR). The Oncotype DX DCIS score (DS) predicts LR risk after BCS alone. This study assesses the impact of RT and DS on LR risk. Methods Population-based cohort analysis of individuals with DCIS treated by BCS ± RT from 1994-2003. Treatment and outcomes were determined by linkage and chart review. We used a propensity score-adjusted multivariable model to examine associations between DS and LR and evaluate the impact of RT. All statistical tests were two-sided. Results The cohort includes 571 individuals treated by BCS alone, 689 cases treated with BCS + RT. Median follow-up was 9.4 years. On multivariable analysis, factors associated with LR include RT, age at diagnosis, tumor size, and multifocality. Adjusting for these factors, the DS risk group was statistically significantly associated with LR risk (hazard ratio high/intermediate = 1.75, 95% confidence interval = 1.28 to 2.41, P < .001). Women with a low-risk DS treated by BCS alone had an LR risk of 10.6% at 10 years and a small benefit from RT, while those with a high DS had a higher risk of LR (25.4%) after BCS alone and greater benefit from RT. A subgroup of patients with favorable clinicopathological features had a high-risk DS; these patients had a higher than expected risk of LR after BCS alone and a greater benefit with RT. Conclusions The DS molecular assay improves risk stratification and estimates of RT benefit in individuals with DCIS treated with breast-conserving therapy.Abstract Background: Most women with ductal carcinoma in situ (DCIS) will receive breast-conserving surgery (BCS) and radiation (RT). RT can be omitted for women at low risk of local recurrence (LR). The Oncotype DX DCIS score (DS) predicts LR risk after BCS alone. This study assesses the impact of RT and DS on LR risk. Methods: Population-based cohort analysis of individuals with DCIS treated by BCS ± RT from 1994–2003. Treatment and outcomes were determined by linkage and chart review. We used a propensity score–adjusted multivariable model to examine associations between DS and LR and evaluate the impact of RT. All statistical tests were two-sided. Results: The cohort includes 571 individuals treated by BCS alone, 689 cases treated with BCS + RT. Median follow-up was 9.4 years. On multivariable analysis, factors associated with LR include RT, age at diagnosis, tumor size, and multifocality. Adjusting for these factors, the DS risk group was statistically significantly associated with LR risk (hazard ratio high/intermediate = 1.75, 95% confidence interval = 1.28 to 2.41, P < .001). Women with a low-risk DS treated by BCS alone had an LR risk of 10.6% at 10 years and a small benefit from RT, while those with a high DS had a higher risk of LR (25.4%) after BCS alone and greater benefit from RT. A subgroup of patients with favorable clinicopathological features had a high-risk DS; these patients had a higher than expected risk of LR after BCS alone and a greater benefit with RT. Conclusions: The DS molecular assay improves risk stratification and estimates of RT benefit in individuals with DCIS treated with breast-conserving therapy.

Molecular Characterization of Breast Cancer Core Biopsy Specimens by Gene Expression Analysis Using Standardized Quantitative RT-PCR.

Background: Core biopsies (CBx) are increasingly the initial diagnostic procedure of choice for breast masses. Clinical studies of the 21 gene assay, which was optimized for small amounts of fixed paraffin-embedded tumor tissue, have been performed on core biopsies in the neoadjuvant setting (Gianni et al, JCO 2005 and Chang et al, Br Cancer Res Treat 2008). We report here the results of pathology review and quantitative RT-PCR analysis of a large number of CBx in clinical practice.Methods: Cases included all Oncotype DX tests reported by the Genomic Health laboratory from July 15, 2005 through May 31, 2009. Pathology review of H and E stained sections, manual microdissection, RNA quantification, and quantitative RT-PCR analysis for each gene were performed according to the pre-specified eligibility criteria and methods used in the NSABP and Kaiser Permanente clinical validation studies. Descriptive analysis was performed on de-identified data with IRB approval.Results: There were 103,863 submissions to the clinical laboratory, of which 11,757 (11.3%) were identified on pathology review as CBx. The initial submission success rate was 94.3% on all cases, and was slightly lower in the CBx (91.6%) than in surgical resections (SRx) (95.7%). The most common reason for initial failure was insufficient tumor or no tumor found in 5.6% of CBx and 3.0% of SRx. Insufficient RNA ( 97%. As expected, the rate of manual microdissection to remove non-tumor elements was higher in SRx than CBx (33.0 vs 3.9% of cases). In addition, the total yield of RNA, on average, was greater in SRx than CBx (4.2 µg vs 2.5 µg). The average Recurrence Score (RS) in CBx and SRx was similar - RS mean (±SD) = 18.9 (±12.4) in CBx and 19.7 (±11.3) in SRx. A wide range of RS was observed in CBx. The proportion of tumors with RS Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6021.

Abstract S5-04: A large prospectively-designed study of the DCIS score: Predicting recurrence risk after local excision for ductal carcinoma in situ patients with and without irradiation

Background: DCIS patients need better tools to align the aggressiveness of treatment with the aggressiveness of their disease. The DCIS Score (DS) was validated as a predictor of ipsilateral breast recurrence (IBR; DCIS or invasive) in 327 E5194 patients treated by breast-conserving surgery (BCS) without radiation (RT) (Solin,2013). This Ontario population based DCIS study of 3335 women with DCIS from 1994 to 2003 (Rakovitch,2013) was conducted to test the DCIS Score as a predictor of recurrence risk in patients treated with BCS alone and in patients treated with BCS+RT. Methods: REMARK guidelines were followed. Breast pathologists centrally reviewed all HE 718 received BCS without RT (N=571 with CM) and 846 received BCS+RT (N=689 with CM). Median follow-up was 9.4 years. Among 1260 pts with CM, 100 pts treated with BCS alone had an IBR (DCIS, N=44; invasive, N=57); 86 pts treated with BCS+RT had an IBR (DCIS, N=32; invasive, N=55). In the primary analysis, among 571 patients treated by BCS alone with CM the continuous DS was significantly associated with IBR in ER+ patients (HR 2.26; 95%CI 1.41,3.59; P=0.001) and in all patients (HR 2.15; 95%CI 1.43,3.22; P= Conclusions: DCIS Score quantifies recurrence risk for DCIS patients treated by BCS with or without RT. Integrating the DCIS Score with established risk factors, such as multifocality, age, and tumor size, can help identify DCIS patients treated with BCS alone with low 10 year risk ( Citation Format: Eileen Rakovitch, Sharon Nofech-Mozes, Wedad Hanna, Frederick L Baehner, Refik Saskin, Steven M Butler, Alan Tuck, Sandip Sengupta, Leela Elavathil, Prashant A Jani, Michel Bonin, Martin C Chang, Elzbieta Slodkowska, Joseph M Anderson, Farid Jamshidian, Diana B Cherbavaz, Steven Shak, Lawerence Paszat. A large prospectively-designed study of the DCIS score: Predicting recurrence risk after local excision for ductal carcinoma in situ patients with and without irradiation [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr S5-04.

Quantitative gene expression by RT-PCR in classic and variant lobular carcinoma in ER+ breast cancer.

11 Background: Classic lobular carcinoma is characterized by a distinctive morphology, loss of E-cadherin commonly due to mutation or deletion of CDH1 on chromosome 16q, and a variable clinical course. Variants (pleomorphic, solid and alveolar) with distinct morphologies and potential differences in outcome have been described (Rosen 2009). Herein we provide an 8-year update of the patterns of quantitative gene expression as measured by the 21 gene Oncotype DX assay observed between ductal NOS (DC) and classic and variant lobular carcinomas. METHODS All tumors analyzed in the Genomic Health laboratory from 6/1/04-5/31/12 were included. Central path used WHO criteria for classification of classic lobular (CL), solid and alveolar lobular (SAL), and pleomorphic lobular (PL) carcinomas. Quantitative expression of 16 cancer related genes was measured on a scale from 2 to 15 (relative to reference genes) where a 1 unit increment is associated with an ~2-fold change in expression. Descriptive statistics for RS & individual genes [ER, PR, invasion gene group (IGG) and proliferation gene group (PGG)] among the subtypes were obtained. Comparisons of means among the subtypes were adjusted to control the overall error rate under any complete or partial null hypothesis. RESULTS DC accounted for 81.8% of 286,726 cases, CL 7.3%, SAL 0.4% and PL 0.4%. For all types a continuous range of RS was observed. DC had the greatest percentage of high risk RS followed by SAL, PL and CL. DC had the highest mean RS and PL and CL had the lowest RS. SAL had the highest mean ER expression and CL and PL had the lowest ER expression. These results may reflect a submission bias and are not population based. The proportion with ER+/PR- phenotype was slightly different among the subtypes: SAL (24.0%) and PL (19.4%) had a higher incidence compared to DC (14.1%) and CL (15.2%). SAL had the highest PGG expression; CL had the lowest. DC had the highest IGG; CL had the lowest. CONCLUSIONS CL and the lobular carcinoma variants are characterized by differential patterns of gene expression. Outlier cases are not infrequent within each of the special subtypes in this large observational cohort. The variation in gene expression, noted by histologic subtype, will be presented in detail.

Quantitative gene expression by RT-PCR in the tubular, cribriform, mucinous, and papillary histologic subtypes of invasive breast cancer.

10 Background: ER+ special histologic breast cancer subtypes are prognostically significant. Here we report the special histologic subtypes of ER+ breast carcinoma and associated patterns of observed gene expression as measured by the 21 gene OncotypeDX assay. METHODS All tumors from 6/1/04-5/31/12 were included in analyses. Central path used WHO criteria. Ductal NOS (DC), tubular (TC), cribriform (CC), mucinous (MC) and papillary (PC) carcinomas were included. Quantitative expression of 16 cancer related genes was measured on a scale from 2 to 15 (relative to reference genes) where a 1 unit increment is associated with ~2-fold change in expression. Recurrence Score (RS) was calculated as published. Descriptive stats for the RS and individual genes [ER, PR, invasion gene group (IGG) and proliferation gene group (PGG)] were obtained. Comparisons of means were adjusted to control the overall error rate under any complete or partial null hypothesis. RESULTS DC accounted for 95.1%, TC 0.7% , CC 0.3%, MC 3.1% and PC 0.7% of 246,555 cases. For all types a wide continuous range of RS was noted. DC had the highest mean RS, followed in decreasing order by MC, TC, CC and PC. PC had the highest ER and PC and CC the highest PR. TC had the lowest ER. ER was not different between CC and MC but PR was. The proportion with ER+/PR- phenotype was different among the subtypes: TC (8.1%) and CC (7.1%) had the lowest incidence whereas MC (12.9%) and PC (10.3%) were more similar to DC (14.1%). TC had lowest PGG expression. MC and PC had lower IGG expression compared to other subtypes. CONCLUSIONS The special subtypes had a wide continuous range of RS: DC had the highest mean RS, followed in decreasing order by MC, TC, CC and PC. PC had the highest ER and PC and CC the highest PR. TC had the lowest ER (may reflect submission bias). ER was not different between CC and MC but PR was. The proportion with ER+/PR- phenotype was different among the subtypes: TC and CC had the lowest incidence, MC and PC were more similar to DC. TC had the lowest PGG expression. MC and PC had lower IGG expression. [Table: see text].

P1-07-11: Consistency and Control in Clinical Assay Technology over Time: The Oncotype DX Recurrence Score and Assessment of Single Gene Expression Levels.

Background: ASCO® and CAP have recently highlighted the importance of consistency and control in clinical assay technology. To date the Onco type DX ® Recurrence Score ® (RS) has been ordered to assist in individualized treatment decision making in over 200,000 estrogen receptor positive, early stage breast cancer patients. The assay quantifies gene expression using RT-PCR from fixed paraffin embedded tissue (FPET) and employs a large number of controls and calibrators to enhance precision and reproducibility. Over 6 years of data on the assay presents an opportunity to explore consistency over time in the RS and quantitative single gene levels (ER, PR, HER2). Methods: All tumors successfully analyzed in the Genomic Health Laboratory from 1/1/05-3/31/11 were included. Descriptive statistics for the RS, the average reference gene expression level, and expression levels for quantitative single genes were obtained for each calendar year. This was done for the entire data set as well as for subgroups defined by histological tumor type. The associations by year between HER2 and GRB7, ER and HER2, and ER and PR expression levels were explored including scatterplots and summary statistics. Results: There were a total of 207,691 breast cancer cases, and the number in each calendar year increased over time as shown in the Table. In general the median age increased slightly over time, as did the proportion of patients in the low RS Risk Group. There was no systematic change over time in the average reference gene expression level, or in the expression levels for the individual genes ER, PR or HER2 during the period from 2008 to 2011 when individual gene testing was provided as part of the Oncotype DX assay report. The relationships between HER2 and GRB7, ER and HER2, and ER and PR remained consistent over time. Conclusions: Active monitoring of the Oncotype DX assay as mandated by ASCO/CAP shows a high degree of consistency in results for both the multigene Recurrence Score and the quantitative single gene results. These results and the approaches used for monitoring consistency are relevant to other institutions in their efforts to maintain and improve assay quality control. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-07-11.

Abstract P3-10-15: Quantitative Gene Expression by RT-PCR in Classic and Variant Forms of Lobular Carcinoma in Estrogen Receptor Positive Invasive Breast Cancer

Background: Classic lobular carcinoma is characterized by a distinctive morphology, loss of E-cadherin commonly due to mutation or deletion of CDH1 on chromosome 16q, and a variable clinical course (Pestalozzi J Clin Oncol. 2008 26:3006). Variants (pleomorphic, solid and alveolar) with distinct morphologies and potential differences in outcome have been described (Rosen 2009). Herein we explore the patterns of quantitative gene expression as measured by the 21 gene Oncotype DX® assay observed between ductal NOS (DC) and classic and variant lobular carcinomas. Material and Methods: All tumors analyzed in the Genomic Health laboratory from 6/1/04-3/31/10 were included. Central path used WHO criteria for classification of classic lobular (CL), solid & alveolar lobular (SAL), and pleomorphic lobular (PL) carcinomas. Quantitative expression of 16 cancer related genes was measured on a scale from 0 to 15 (relative to reference genes) where a 1 unit increment is associated with an ∼2-fold change in expression. Descriptive statistics for RS & individual genes [ER, PR, invasion gene group (IGG) & proliferation gene group (PGG)] among the different subtypes were obtained. Comparisons of means among the subtypes were adjusted to control the overall error rate under any complete or partial null hypothesis. Results: DC accounted for 90.4% of 133,234 cases, CL 8.2%, SAL 0.7% and PL 0.8%. For all types a continuous range of RS was observed. DC had the greatest percentage of high risk RS followed by SAL, PL and CL. DC had the highest RS and CL had the lowest RS. SAL had the highest ER expression and CL had the lowest expression, but these results may refect a submission bias and are not population based. The proportion with the ER+/PR-phenotype was slightly different among the subtypes: SAL (24.2%) and PL (19.5%) had a higher incidence compared to DC (15%) and CL (15.6%). SAL had the highest PGG expression; CL had the lowest. DC had the highest IGG; CL had the lowest. Discussion: Classic lobular carcinoma and the lobular carcinoma variants are characterized by differential patterns of gene expression. Outlier cases are not infrequent within each of the special subtypes in this large observational cohort. The variation in gene expression, noted by histologic subtype, will be presented in detail. RS Distributions in Ductal NOS, Lobular and Lobular Variants Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-10-15.