Search

Table S1: Summary of baseline characteristics for the current analysis cohort vs. the Remainder of HERA patients Table S2: Summary of Disease-free Survival by subgroup of patients Table S3: Summary of Overall Survival by subgroup of patients Table S4: Comparison of median follow-up time (based on reverse censoring for OS) Table S5: Predictive effect of p27 biomarker (Low/High) for DFS: Cox proportional hazards model without adjustment for other variables. Table S6: Predictive effect of p27 biomarker (Low/High) for DFS, adjusted for variables of clinical interest: Multivariate Cox proportional hazards model. Table S7: Predictive effect of p27 biomarker (Low/High) for DFS, adjusted for variables of clinical interest and the predictive effect of ER Local: Multivariate Cox proportional hazards model Table S8: Predictive effect of p27 biomarker (Low/High) for DFS, adjusted for variables of clinical interest and the predictive effect of ESR1: Multivariate Cox proportional hazards model Table S9: Predictive effect of cyclin D1 biomarker (continuous) for DFS: Multivariate Cox proportional hazards model without adjustment for other variables. Table S10: Predictive effect of cyclin D1 (continuous) biomarker for DFS, adjusted for variables of clinical interest: Multivariate Cox proportional hazards model. Table S11: Predictive effect of cyclin D1 biomarker (continuous) for DFS, adjusted for variables of clinical interest and the predictive effect of ER Local: Multivariate Cox proportional hazards model Table S12: Predictive effect of cyclin D1 biomarker (continuous) for DFS, adjusted for variables of clinical interest and the predictive effect of ESR1: Multivariate Cox proportional hazards model Figure S1. A: Histogram of TOPO2A biomarker Figure S1.B Histogram of Ki67 biomarker Figure S1.C: Histogram of Cyclin D1 biomarker Figure S1.D: Histogram of p27 biomarker Figure S2.A: Ki67 distribution by TOPO2A level Figure S2.B: Cyclin D1 distribution by TOPO2A level Figure S2.C: p27 distribution by TOPO2A level Figure S3.A: Ki67 distribution by Cyclin D1 level Figure S3.B: p27distribution by Cyclin D1 level

Introduction: Previous studies have shown that HER2+ breast cancer is biologically heterogenous and intrinsic subtypes can be identified (luminal A, luminal B, HER2-enriched [HER2-E] and basal-like). HER2-E predominates and is associated with higher response rates following anti-HER2-based chemotherapy or dual HER2 blockade (only with lapatinib and trastuzumab). We explored the ability of intrinsic subtypes to predict pCR in pts treated with anti-HER2 neoadjuvant therapy. Methods: KRISTINE (NCT02131064) is an open-label, phase 3 study of neoadjuvant trastuzumab emtansine + pertuzumab (T-DM1+P) vs docetaxel + carboplatin + trastuzumab + pertuzumab (TCH+P) in pts with HER2+ EBC. Treatment-naive pts with stage II–IIIC HER2+ EBC were randomized to receive 6 cycles of T-DM1+P or TCH+P and assessed for the primary endpoint, pCR (ypT0/is, ypN0). HER2 and hormone receptor (HR) status were centrally assessed. Gene expression (RNA) was assessed by a custom 800-gene codeset on the nCounter platform. Intrinsic subtypes were assessed with the research-based PAM50 classifier. Results: KRISTINE randomized 444 pts (data cutoff, Dec 3, 2015; TCH+P, n=221; T-DM1+P, n=223). PAM50 results were available for 354 pts (79.7% of the intent-to-treat [ITT] population). Baseline characteristics and efficacy in the PAM50 population were similar to that of the ITT population. The HER2-E subtype represented 54.8% of the samples (Table 1). Differences were observed by HR status. Almost all luminal tumors (131/132) were identified within HR+ disease. Of HR+ tumors, 32% were identified as HER2-E. In the TCH+P arm, the pCR rate was 72.1% for HER2-E vs 32.8% in the other subtypes combined (Table 2.) In the T-DM1+P arm, the pCR rate was 62.2% for HER2-E vs 26.9% in the other subtypes combined. No major differences were observed in pCR rates within the HER2-E subtype according to HR status. Further multivariable analyses assessing differences between treatment arms and treatment benefit across subtypes is ongoing. Table 1. Intrinsic subtypesn (%)ITT (n=354)HR- (n=143)*HR+ (%) (n=200)*HER2-E194 (54.8)123 (86.0)64 (32.0)Luminal A60 (16.9)1 (0.7)57 (28.5)Luminal B74 (20.9)074 (37.0)Basal-like26 (7.3)19 (13.3)5 (2.5)*Central HR status unknown for n=11 Table 2. pCR by intrinsic subtype TCH+PT-DM1+P npCR, n (%)npCR, n (%)pCR difference (95% CI)HER2-E10475 (72.1)9056 (62.2)-9.89 (-23.11, 3.32)HER2-E and HR+ *3724 (64.9)2715 (55.6)-9.31 (-33.56, 14.94)HER2-E and HR- *6448 (75.0)5937(62.7)-12.29 (-28.56, 3.98)Other subtypes combined6722 (32.8)9325 (26.9)-5.9 (-20.36, 8.45)Luminal A254 (16.0)3510 (28.6)12.57 (-8.18, 33.32)Luminal B3211 (34.4)4212 (28.6)-5.80 (-27.19, 15.58)Basal-like107 (70.0)163 (18.8)-51.25 (-85.49, -17.01)*Central HR status unknown for n=7 Conclusions: In this analysis from the KRISTINE study, HER2-E was the most common intrinsic subtype and was associated with the highest pCR rate with both regimens. Results are consistent with previous findings. The luminal A and B subtypes were well associated with HR+ status. A sizeable subgroup of the HER2-E subtype was HR+ (32%), and pCR rates within the HER2-E subtype seemed independent of HR status. Citation Format: Prat A, Slamon D, Hurvitz SA, Press MF, Lewis Phillips G, Lopez Valverde V, Kiermaier A, Helms H-J, Martin M, de Haas SL. Association of intrinsic subtypes with pathological complete response (pCR) in the KRISTINE neoadjuvant phase 3 clinical trial in HER2-positive early breast cancer (EBC) [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr PD3-06.

Background: The primary analysis of the randomized, double-blind, placebo-controlled APHINITY trial, published in 2017, including 4804 patients (pts) with HER2-positive, early breast cancer with 45.4 months' median follow-up, demonstrated that adjuvant pertuzumab (P) added to trastuzumab and chemotherapy, statistically significantly improved invasive disease-free survival (IDFS) compared with placebo (Pla) added to trastuzumab and chemotherapy overall and for pts with node-positive (N+) disease. In 2019, updated descriptive analyses of IDFS with 74.1 months' median follow-up, demonstrated sustained benefit of adding P both overall (HR, 0.76; 95% CI, 0.64-0.91), and for N+ disease (HR, 0.72; 95% CI, 0.59-0.87), while confidence intervals remained wide for the node-negative (N-) cohort (HR, 1.02; 95% CI, 0.69-1.53). There is great interest to explore how these significant overall results translate into absolute treatment benefits for different patient subpopulations. Methods: Subpopulation Treatment Effect Pattern Plot (STEPP) is an exploratory, graphical method that plots estimates of treatment effect for overlapping patient subpopulations defined by a covariate of interest. Four continuous covariates of interest are considered for defining subpopulations in this report: i) a clinical composite risk score (see below), ii) TILs percentage, iii) HER2 FISH copy number, and iv) a clinical-biological composite risk score combining the previous three factors. Pts with lowest values for the covariate comprise the extreme left STEPP subpopulation, and pts with highest values comprise the extreme right subpopulation. The clinical composite risk score for IDFS based on the overall cohort was calculated using a Cox regression model including the following prespecified clinical characteristics: number of positive nodes, tumor size, age, and centrally-reviewed hormone receptor status. Composite risk scores were scaled between 0 and 100 with higher scores reflecting higher risk of an IDFS event. An example of low clinical risk factors would be T1N0 and aged 40-64; while high risk would be T3N2 or higher and ages Results: Table of 6-year IDFS percents (%) from Aphinity STEPPs, Overall and for N- and N+ cohorts. For each analysis, results are shown for the two subpopulations at either extreme of the STEPP (i.e. lowest and highest risk or values) as well as the intermediate STEPP subpopulation. Conclusions: Based on the two extreme and one intermediate subpopulations of the STEPP analyses shown in the table, the intermediate clinical composite risk subpopulation and the highest TILs percentages had the largest absolute improvements in 6-year IDFS percents for P compared with Pla. Table of 6-year IDFS percents (%) from Aphinity STEPPs, Overall and for N- and N+ cohorts.6-year IDFS %Overall (N=4804)Node-Negative (N=1799)Node-Positive (N=3005)PPlaΔ±SEPPlaΔ±SEPPlaΔ±SEOverall Average Results90.687.82.8±0.995.094.90.1±1.187.983.44.5±1.2Clinical composite riskLowest risk (0 - 21)95.396.2-0.9±1.396.196.5-0.4±1.5---Intermediate (39 - 63)92.687.35.3±1.995.091.04.0±3.093.686.76.9±2.3Highest risk (81 - 100)80.575.84.7±2.8---79.475.44.0±3.2TILs percentageLowest values (0-9)90.487.82.6±2.094.795.2-0.5±2.087.282.64.6±2.7Intermediate (13-21)89.487.71.7±2.194.294.10.1±2.285.484.80.6±2.7Highest values (≥31)95.689.36.3±1.798.194.93.2±1.792.384.97.4±2.4HER2 copy numberLowest values (1-8)87.186.40.7±2.292.794.8-2.1±2.284.182.12.0±2.8Intermediate (9.5-11)91.889.02.8±1.994.996.1-1.3±1.990.783.37.4±2.5Highest values (13-32)90.588.91.6±2.096.095.10.9±2.087.785.32.4±2.6Clinical-biological composite riskLowest risk (0-21)96.796.40.3±1.298.295.72.5±1.6---Intermediate (40-60)93.489.53.9±1.991.792.7-1.0±2.594.288.85.4±2.2Highest risk (79-100)80.175.94.2±2.7---79.575.24.3±3.2 Citation Format: Richard D. Gelber, Xin Victoria Wang, Bernard F. Cole, David Cameron, Fatima Cardoso, Vivianne Tjan-Heijnen, Ian Krop, Sherene Loi, Roberto Salgado, Astrid Kiermaier, Elizabeth Frank, Debora Fumagalli, Carmela Caballero, Evandro de Azambuja, Marion Procter, Emma Clark, Eleonora Restuccia, Sarah Heeson, Jose Bines, Sibylle Loibl, Martine Piccart-Gebhardt. 6-year absolute invasive disease-free survival (IDFS) benefit of adding adjuvant pertuzumab to trastuzumab and chemotherapy for patients with early HER2-positive breast cancer: A STEPP analysis of the APHINITY (BIG 4-11) trial [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS10-01.

Introduction KRISTINE (NCT02131064) is an open-label, phase 3 study of neoadjuvant trastuzumab emtansine + pertuzumab (T-DM1+P) vs docetaxel + carboplatin + trastuzumab + pertuzumab (TCH+P) in pts with HER2+ EBC (all pts, Table). In exploratory analyses we assessed whether HER2 expression and gene amplification levels and PIK3CA mutation were associated with pathologic complete response (pCR) in KRISTINE. Methods Treatment-naive pts with stage II–IIIC HER2+ EBC were randomized to receive 6 cycles of T-DM1+P or TCH+P and assessed for the primary endpoint, pCR (ypT0/is, ypN0). HER2 status was confirmed centrally as immunohistochemistry (IHC) status of IHC3+ and/or HER2/CEP17 gene ratio ≥2 by in situ hybridization (ISH) before study entry. Baseline HER2 mRNA, PIK3CA mutation, and hormone receptor (HR) status were evaluated. Rates of pCR were compared across biomarker subgroups, including: HER2 mRNA, HER2 IHC staining percentage, HER2 gene ratio, and PIK3CA mutation. Incidence of overlap between biomarkers was assessed. All analyses were descriptive. Results KRISTINE randomized 444 pts (data cutoff, Dec 3, 2015; TCH+P=221; T-DM1+P=223). The biomarker population was representative of the ITT population. Biomarker values and distribution were balanced across treatment arms. Lower pCR rates were seen in both treatment arms across biomarker subgroups with lower vs higher HER2 expression levels (Table). PIK3CA mutation was associated with numerically lower pCR with T-DM1+P but not TCH+P. Evaluation of 15 pts who progressed during T-DM1+P neoadjuvant therapy suggests that nonresponse to therapy in this arm may be associated with low HER2 levels (mRNA, IHC, and ISH). PIK3CA mutation rate was higher in tumors with ≤median HER2 mRNA (38%) vs >median (16%), and in focal (53%) and variable (41%) HER2 expression vs homogeneous (24%). A numerically higher proportion of HR positivity was seen in the HER2 IHC2+ (81%) subgroup vs IHC3+ (55%), and in the focal (83%) and variable (62%) HER2 subgroups vs homogeneous (53%). Table TCH+PT-DM1+P BiomarkernpCR (%)npCR (%)Difference in pCR rates (%)95% CIAll pts22155.722344.4-11.3-20.5, -2.0HER2 IHC3+19460.819549.7-11.1-20.9, -1.3IHC2+2520.0287.1-12.9-31.2, 5.5HER2 IHC 2+/3+ fraction* Focal (median10763.610851.9-11.7-24.8, 1.4HER2/CEP17 gene ratio ≥2 to Conclusions None of the evaluated biomarker subgroups showed superior pCR rates for T-DM1+P vs TCH+P. Lower HER2 levels based on mRNA and protein expression and gene amplification were associated with numerically lower pCR in both treatment arms. PIK3CA mutation results differ from prior observations. Unfavorable biomarkers for pCR, like low HER2, PIK3CA mutation, and HR positivity, showed a tendency for co-expression, which should be considered when planning future trials in HER2+ BC. Citation Format: de Haas SL, Hurvitz SA, Martin M, Kiermaier A, Lewis Phillips G, Xu J, Helms H-J, Slamon D, Press MF. Biomarker analysis from the neoadjuvant KRISTINE study in HER2-positive early breast cancer (EBC) [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 P6-07-09.

Background The use of trastuzumab (H)-based therapy for HER2−positive BC has significantly altered outcomes. Yet up to 64% of metastatic patients (pts) fail to respond (Robert JCO 2006) and most pts will progress within 24–42 months (m) following an initial response for metastatic disease. Preclinical data suggest several in vitro resistance mechanisms but confirmatory in vivo data are lacking, preventing optimal personalized care. This ongoing proof-of-concept study examines serial bx in pts with HER2−positive metastatic BC in order to assess BM profiles across multiple lines of treatment. Methods: Key eligibility criteria include: centrally confirmed HER2 status, minimum of 1 disease site considered suitable for serial bx, normal coagulation profile and cardiac function, prior adjuvant/neoadjuvant taxane and H completed ≥12 m and ≥6 m respectively, ECOG ≤2. Pts receive q3wk H with clinician choice of taxane (docetaxel 75–100 mg/m2 q21, paclitaxel 80 mg/m2 weekly or 175 mg/m2 q21 [TH]). At the time of progressive disease, pts receive capecitabine (X) and H. Two core bx and 1 optional fine needle aspirate are performed at the following times: baseline (after 3 weeks [w] of TH), at 6 w, at time of first progression prior to XH, and at the time of progression on XH. Tumor assessments are performed at 6 w, then 9 w intervals, with cardiac assessment every 6 m. Primary endpoint aims to explore and potentially define BM signatures that could alter during HER2−targeted therapy and predict for decreased or increased sensitivity to H-based treatment. Secondary endpoints include bx safety, ORR and TTP. At time of baseline bx, pts have the option to complete a pt satisfaction questionnaire. Results: Between August 2009 and June 2011, 58 pts were screened; 29 enrolled. The other 29 pts were screen failures, with 3 pts (5%) specifically declining entry due to the requirement for serial bx. Median age is 53 y (range 37–86), with 9 and 20 pts having recurrent or de novo metastatic disease. Eight pts received adjuvant systemic treatment, including H in 4 pts. Baseline bx performed in the breast, n (%): 15 (52), bone: 6 (21), liver: 4 (14), and lymph nodes: 4 (14); 14 and 27 pts underwent bx at 3 and 6 w respectively. After a median of 34 w of treatment 11 pts progressed on TH and 7 pts underwent planned bx. There were 12 SAEs, none related to bx. Most AEs were grade 1/2 (95%) with 3% grade 3 and 1 death due to intercurrent illness. From a total of 43 bx, 12 (27%) grade 1/2 AEs (hematoma 2, transient hypotension 2, mild pain 8) were reported, with resolution in all instances. Fifteen pts consented to the pt satisfaction substudy. Conclusions: Preliminary results of the first reported serial bx study in metastatic BC demonstrated 95% pt acceptance of this approach. Evaluation of BM profiles will be conducted following planned recruitment of 50 pts. To date, the feasibility and safety of obtaining serial bx in metastatic BC is supported by the current safety profile and patient uptake. Updated recruitment data will be presented. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-22-01.

The lack of efficacious therapies have hindered the progress in improving the survival of HCC patients worldwide. As a disease that associates with the underlying liver injuries caused by various etiologies and remains asymptomatic during most of its decades long progression, HCC presents challenges to accurately depict its biological landscape and determine suitable therapeutic intervention. An approach to tackle this issue is to define significant biological variants that impact patient survival in each disease stage, then attempt to chart paths of potential therapies. To identify significant biological determinants associated with the overall survival (OS) of early stage, non-metastatic HCC patients, we constructed focused Nanostring panels that are represented by 1,164 genes of validated components and transcriptional outputs of multiple signaling pathways, cellular machineries, and tumor immunity. An unbiased analysis of patient survival and gene transcript levels revealed four distinct groups of genes that are significantly prognostic. The first two (Group 1 & 2) are associated with poorer prognosis when expression is high (hazard ratio > 2), whereas the remaining two (Group 3 & 4) correspond to significantly better OS with elevated transcript levels (hazard ratio Searching for druggable targets associated with poorer prognosis, we identified both EZH2 and CHEK1 as candidates from Group 2. In vitro treatment of three HCC cell lines with EPZ-6438 (EZH2 inhibitor) or GDC-0425 (CHEK1 inhibitor) resulted in significant modulation of H3K27me3 or pCHEK1 levels (s345) respectively. The growth of these cell lines was also dose-dependently suppressed by each inhibitor in colony formation assays . Furthermore, the combination of both inhibitors was able to achieve substantial cell growth inhibition at lower concentrations, suggesting a combinatorial effect of targeting both EZH2 and CHEK1 in HCC cell lines. Xenograft studies are ongoing to assess in vivo efficacy of this combination. In conclusion, utilizing well constructed Nanostring gene panels, we identified significant association of cell cycle machinery and adaptive tumor immunity with HCC patient survival. Our findings not only provided potential rationale for expand testing of cancer immunotherapies to non-metastatic, early stage HCC, but also revealed the potential utility of combining EZH2 and CHEK1 inhibitors to treat HCC. Citation Format: Pei-Zhen Miao, Jeff Cheng, Kwame Okrah, Bonnie Liu, Charlie Sun, Grace Cao, Tony Guo, Shijing Fu, Marie-Claire Wagle, Elizabeth Blackwood, David Shames, Garret Hampton, Lukas Amler, Astrid Kiermaier, Yevgeniya Fridlyand, Mark Lackner, Shih-Min Huang, Jian Zhou, Zhi Dai. Analysis of non-metastatic HCC patient tumors revealed the significance of cell cycle regulation and tumor immunity in association with overall survival and identified clinically relevant druggable targets [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 1750. doi:10.1158/1538-7445.AM2017-1750

Background: In HER2+ early breast cancer (eBC) pCR rates after standard neoadjuvant chemo- + anti-HER2 therapy differ according to hormone-receptor (HR) status. Molecular analysis reveals HER2+/HR+ BC as a distinct entity within HER2+ BC. The ADAPT HER2+/HR+ phase II trial aims to identify early responders to endocrine + anti-HER2 therapy. Methods: The trial completed recruitment in January 2015 (n=376). Patients (pts.) were randomized to 12 weeks of neoadjuvant therapy: A:T-DM1 (3.6 mg/kg q3w) vs. B:T-DM1 with endocrine therapy (ET) (pre-: tamoxifen; postmenopausal: aromatase inhibitor) vs C:trastuzumab q3w+ET. After surgery, standard chemotherapy at investigators' discretion and completion of 1y trastuzumab were recommended. Trial tests pCR (yPN0 and ypT0/is) in after T-DM1 or T-DM1+ET compared to T+ET. Biomarkers are measured at baseline and after 3 weeks. Results: Pre-planned interim analysis (n=130) aimed to identify an early-response biomarker (e.g. Ki-67 drop) and to validate trial assumptions. Median age was 49 years; 55% were pre-menopausal; 40% had cT1 tumors, 51% cT2; 68% had cN0, 27% cN1; 75% had G3. Median baseline Ki67 was 30%. In all arms, more than 95% received all 4 therapy cycles. 16 SAEs were reported in 13 pts (A:7; B:6; C:3); all CTC grade 1 (1), 2 (11) or 3 (4); all pts completely recovered without sequelae. Overall pCR rate was 30.8%: T-DM1: 40.5%, T-DM1+ET: 45.8%, T+ET: 6.7%. The difference between either arm T-DM1 arm vs. T+ET was significant (p Conclusions: The WSG-ADAPT HER2+/HR+ phase II trial is internationally the first large prospective randomized phase II trial specifically conducted within this distinct subtype. Interim analysis demonstrated for the first time clinically meaningful pCR rates (>40%) after short therapy (12 weeks) of T-DM1 ± ET without systemic chemotherapy in HER2+/HR+ eBC. Final efficacy and safety data will be presented at the meeting together with results of the correlative science program. Citation Format: Harbeck N, Gluz O, Christgen M, Braun M, Kuemmel S, Schumacher C, Potenberg J, Kraemer S, Kleine-Tebbe A, Augustin D, Aktas B, Forstbauer H, Tio J, Liedtke C, Kates RE, Wuerstlein R, de Haas SL, Kiermaier A, Kreipe HH, Nitz U. Final analysis of WSG-ADAPT HER2+/HR+ phase II trial: Efficacy, safety, and predictive markers for 12-weeks of neoadjuvant TDM1 with or without endocrine therapy versus trastuzumab+endocrine therapy in HER2-positive hormone-receptor-positive early breast cancer. [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 S5-03.

As cancer cells die, they release small pieces of DNA into the bloodstream. These molecules of DNA are called circulating tumor DNA (ctDNA). ctDNA carry somatic aberrations that are found in the originating tumor tissue and can serve as excellent biomarkers. They can be extremely valuable as tools for disease monitoring, and samples for predictive and prognostic biomarkers. However, limited quality and quantity of formalin-fixed paraffin-embedded (FFPE) tissues and circulating cell-free DNA (cfDNA) samples available from patients pose a big challenge to performing DNA quantification, mutation profiling and copy number variation analysis. We explored the feasibility of using Droplet Digital PCR (ddPCR) technology to address this challenge. Using ddPCR, we studied the ability to perform high-sensitivity mutation analysis and accurate quantification of allele frequencies, and compared this approach to alternate methods such as next-generation sequencing (NGS). For mutation profiling, we examined a total of 22 mutation detection ddPCR assays across 7 oncogenes - KRAS, PIK3CA, EGFR, BRAF, NRAS, cKIT and AKT1. The assay sensitivity for the majority of mutation detection assays was found to be approximately 0.03%. We have identified/reconfirmed various mutations using matching FFPE/plasma samples from patients in various cancer types. Moreover, we have adapted the previously published HER2 CNV ddPCR assay to detect HER2 copy number variation in cfDNA. We will present data showing validation of this platform using commercially available DNA control, archival FFPE tissue and cfDNA samples and demonstrate improved sensitivity and specificity. The platform is being routinely used in our labs to analyze samples from multiple clinical trials. These efforts enable the development of a non-invasive method to overcome existing challenges to provide molecular understanding of patient's tumor evolution, and aid in the development of personalized therapies for cancer patients. Citation Format: Nga Wan Rachel Tam, Teiko Sumiyoshi, Rajesh Patel, Sundari Sarma, Astrid Kiermaier, Rajiv Raja. Ultrasensitive detection of genomic alterations in cell-free DNA by Droplet Digital PCR. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2409. doi:10.1158/1538-7445.AM2015-2409

Background: The anti-HER2 humanized monoclonal antibodies trastuzumab (T) and pertuzumab (P) have complementary mechanisms of action. CLEOPATRA showed that P+T+docetaxel (D) significantly improved progression-free survival (PFS) compared with placebo (Pla)+T+D in the first-line treatment of HER2-positive MBC (HR = 0.62, 95% CI 0.51–0.75; p < 0.001), with a similar safety profile. A panel of biomarkers (BMs) was assessed in tumor tissue and in serum samples to explore their potential predictive and/or prognostic value. Methods: Samples were collected at baseline prior to the first dose of study drug (N = 808; 58–99% of which were assessable for BMs due to tissue availability/technical reasons). HER2, HER3, IGF1R, PTEN, and pAKT were assessed by immunohistochemistry and a modified H-score was derived for each. HER1, HER2, HER3, AREG, and betacellulin tumor mRNA levels were assessed by quantitative reverse transcription polymerase chain reaction (qRT-PCR) and c-myc by fluorescence in situ hybridization. PCR-based methods were used for mutational analyses of 8 hotspots within PIK3CA on tumor DNA, and for assessments of FCγR polymorphisms in DNA extracted from whole blood. Serum HER2 extracellular domain, AREG, EGF, and TGFα were detected by ELISA. Exploratory subgroup analyses of PFS by BM level were performed for each BM. Median values were used as cutoffs for high vs low levels, except for PTEN, PIK3CA mutation, and c-myc amplification, where cutoffs were applied following a biologic rationale. Univariate Cox regression analyses assessed the prognostic value on a background of HER2-targeted treatment (relationship of each BM to clinical outcome, both arms pooled) and predictive value (association of BMs with treatment benefit) of each BM. Analyses were exploratory; therefore no adjustment for multiplicity was made. Results: Significant prognostic variables identified were: HER2 protein (p = 0.05), HER2 mRNA (p = 0.008), HER3 mRNA (p = 0.03), and PI3KCA mutation (p = 0.0001). The HR was Summary: The BM analysis from CLEOPATRA is consistent with results from previous phase II studies: HER2 remains the only marker suitable for use when selecting patients for anti-HER2 therapy. The current analyses did not identify any additional predictive markers that would allow refinement of the HER2-positive target population treated with P+T+D. The lack of a HER2 treatment-naive control group may partly explain the lack of a signal. The PI3K mutational status may identify a poor prognostic group within the HER2-positive population that could be appropriate for clinical trials of HER2-targeted molecules in combination with PI3K pathway-targeted agents. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr S5-1.

Background: NeoSphere showed that P with H and docetaxel (T) had a significantly higher pCR rate (45.8%) compared with TH (29%) or other P combinations (HP, TP) (Gianni, SABCS 2010). HER2+/ER+ and HER2+/ER− BC have different patterns of gene expression, and treatment response seems to be driven by different biologic pathways (Bianchini, ASCO 2011). This effect was also seen in NeoSphere. A broad panel of biomarkers (BMs) was assessed in tumor specimens and in sera to characterize the molecular profile of the ITT population and to explore BMs with different treatments and in different subsets of pts. Methods: Core biopsies and sera from 387/417 pts collected at baseline were available for BM analyses. HER2, all its truncated forms, HER3, IGF1R, PTEN, and pAKT were assessed by IHC and a modified H-score was derived per each marker. HER1, HER2, HER3, amphiregulin (AREG) and betacellulin mRNA expression in tumor tissue was assessed by qRT-PCR. C-myc was assessed by FISH. Mutational analyses of 8 hot spots within PIK3CA on tumor DNA was done via TaqMan-PCR. ELISA was used to detect HER2 extracellular domain (ECD), AREG, EGF, and TGFα in serum. Treatment interaction test was performed to test for association between BMs and pCR. In the analysis, median values were used as cutoffs, except for PTEN, PIK3CA mutation, and c-myc amplification, where cutoffs were applied following a biologic rationale. Results: In the overall population, high HER2 membrane protein expression measured by modified H-score was associated with a significantly higher pCR rate in the THP arm (p=0.001), with a large benefit from combination of P with TH (OR=3.74, CI 80%, 2.21−6.34; p=0.0009). The cutoff derived from NeoSphere has no immediate clinical applicability. A decreased likelihood of pCR was detected for pts with PI3K mutations in all 4 arms in a preliminary analysis limited to about 70% of cases. A final analysis will be presented. A significantly different distribution of BM expression in the ER+ vs ER− subgroups was found for 10 of 16 BMs (FDR Summary: The level of HER2 membrane protein expression was associated with increased benefit from combination of P with TH. The PI3K mutational status appears to have a relevant role in defining the likelihood of response with all treatments. The different treatment effects according to ER status are reflected in a different molecular BM profile in ER+ vs ER− tumors, with the IGF1R membrane score being correlated with ER− tumors. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr S5-1.