Anderson, S. Keith

Abstract Background: Tumor-infiltrating lymphocytes (TILs) at diagnosis are reported to be prognostic in triple-negative breast cancer (BC). Analysis of a small subset of 209 HER2+ patients (pts) with 49 events concluded that higher levels of S-TILs are associated with increased trastuzumab benefit (Loi, 2014). Here we report the largest study to date evaluating S-TILs and their prognostic and predictive association with clinical outcome in N9831 pts treated with either chemotherapy or chemotherapy plus trastuzumab. Methods: Samples assessed were from primary tumors of pts on N9831 arm A (standard AC→T chemotherapy) and arm C (concurrent chemotherapy with trastuzumab) (Perez, 2011). S-TILs were evaluated on H&E whole tumor slides by a single pathologist with ∼10% of cases read by two pathologists in tandem. The percent of stromal lymphocytic infiltrates (S-TILs) was quantitated in deciles; ≥60% S-TILs was used for the categorical cutoff (Denkert, 2010). The association between S-TILs, treatment (tx) and recurrence-free survival (RFS) was studied and the interaction between S-TILs, trastuzumab benefit and RFS was calculated. Results: 489 pts from arm A (chemo) and 456 pts from arm C (chemo with trastuzumab) were assessed and were similar to pts in the overall trial; all had RFS information and a median follow-up of 4.4yr. Tumors from 54% of pts in arms A and C were HR+; 14% were node-negative. Tumors with high S-TILs were more likely to be hormone receptor-negative (p< 0.0001). In multivariable analyses including nodal status, hormone receptor status, tx arm, tumor size, tumor grade, and age, ≥60% S-TILs was significantly associated with RFS (HR 0.20; 95%CI 0.064–0.65, p=0.007) in arm A but not in arm C (HR 1.1; 95%CI 0.42–2.8, p=0.87); the interaction term of arm and ≥60% S-TILs was significant (p=0.042). Semi-continuous deciles were associated with RFS in arm A (p<0.0002) but not in C (p=0.37). Hormone receptor status was an independent prognostic factor in arm A (HR 0.61; 95%CI 0.41-0.93, p=0.02) but not in C (HR 0.79; 95%CI 0.44-1.41, p=0.42). In arm A the 10yr Kaplan-Meier estimates for RFS were 90.9% and 64.5% for high S-TILs and low S-TILs pts, respectively (HR 0.23; 95%CI 0.073–0.73, p=0.013). In arm C the 10yr Kaplan-Meier estimates for RFS were 80.0% and 80.1% for high S-TILs and low S-TILs pts, respectively (HR 1.26; 95%CI 0.5–3.2, p=0.63). Arm A (N=489)Arm C (N=456)VariableHR (95% CI)p-valueHR (95% CI)p-valueNodal statusLymph node(-)1.00 (ref)<0.00011.00 (ref)0.0131-3+0.58 (0.25, 1.35) 2.99 (0.40, 22.70) 4-91.46 (0.64, 3.35) 3.29 (0.42, 25.80) 10+2.42 (1.04, 5.64) 8.22 (1.06, 63.60) HR statusnegative1.00 (ref)0.01981.00 (ref)0.42positive0.61 (0.41, 0.93) 0.79 (0.44, 1.41) S-TILs status<60%1.00 (ref)0.0071.00 (ref)0.87≥60%0.2 (0.06, 0.65) 1.08 (0.42, 2.79) Tumor gradeGrade 1 or 21.00 (ref)0.501.00 (ref)0.51Grade 31.18 (0.73, 1.91) 1.24 (0.65, 2.39) Tumor size1.08 (0.98, 1.19)0.121.00 (0.93, 1.07)0.92Age0.99 (0.97, 1.00)0.131 (0.98, 1.03)0.87 Conclusions: In exploratory analyses from this subset HER2+ population from N9831, S-TILs were associated with RFS in patients treated with chemotherapy alone, and were not shown to be associated with RFS in patients treated with chemotherapy plus trastuzumab. Citation Format: Edith A Perez, Karla V Ballman, S Keith Anderson, E Aubrey Thompson, Sunil S Badve, Helen Bailey, Frederick L Baehner. Stromal tumor-infiltrating lymphocytes(S-TILs): In the alliance N9831 trial S-TILs are associated with chemotherapy benefit but not associated with trastuzumab benefit [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 S1-06.

Abstract Recent studies have shown that associations between common genetic variation and gene expression in breast tumors provide insight into the functional mechanisms underlying breast cancer risk loci. However, the relationship between genetic variation, gene expression, and risk of triple negative (TN) breast cancer remains largely unexplored. We performed a genome-wide expression quantitative trait locus (eQTL) analysis of TN breast cancer using 668 formalin-fixed paraffin embedded TN tumors from the Triple Negative Breast Cancer Consortium (TNBCC) to identify novel genes relevant to TN breast cancer risk and to further explore the biology underlying known TN risk loci. Cis-eQTLs were defined as correlations between single nucleotide polymorphisms (SNP) and expression of genes within 1Mb, and trans-eQTLs were defined as the remainder of all SNP-gene correlations. In a genome-wide analysis, we identified 68,012 cis-eQTLs, representing correlations between 42,193 SNPs and 2,092 genes that were significant at a 10% false discovery rate (FDR). We also identified 70 trans-eQTLs across two unique genes at the same FDR threshold. Five cis-eQTLs involved SNPs previously associated with TN breast cancer risk (p<5.0x10-5) in 3,700 TN cases and 4,700 controls from the TNBCC. Three of these, located on chromosomes 5p (cis-eQTL p=4.8x10-34; risk p=1.1x10-7), 7p (cis-eQTL p=6.1x10-10; risk p=1.3x10-7), and 17q (cis-eQTL p=1.8x10-15; risk p=2.9x10-7), have previously been reported in public databases of normal tissue. The remaining two TN breast cancer cis-eQTLs, located on chromosomes 8q (cis-eQTL p=6.3x10-6, risk p=8.3x10-6) and 14q (cis-eQTL p=3.1x10-6, risk p=1.2x10-5), have not been reported as eQTLs in other tissues. In addition, 12 of 25 known TN breast cancer risk loci (PEX14, MDM4, 2q31.1, ESR1, 11q13.1, 11q24.3, 12p13.1, PTHLH, NTN4, 12q24, 19p13.1, MLK1), contained at least one significant cis-eQTL. In summary, we have identified five cis-eQTLs that may identify novel risk loci for TN breast cancer. Citation Format: Kristen S. Purrington, Drakoulis Yannoukakos, Jane Carpenter, Heli Nevanlinna, Angela Cox, Gianluca Severi, Christine Ambrosone, Amanda Ewart Toland, Andrew K. Godwin, Hiltrud Brauch, Peter A. Fasching, Penelope Miron, Jenny Chang-Claude, Nicholas G. Martin, Grant W. Montgomery, Vessela Kristensen, Hoda Anton-Culver, Paul Goodfellow, Janet E. Olson, Hugues Sicotte, Naresh Prodduturi, Daniel W. Visscher, Jeanette E. Eckel-Passow, S. Keith Anderson, Seth Slettedahl, Curtis Olswold, Xianshu Wang, V. Shane Pankratz, Susan Slager, Wei Zheng, Arto Mannermaa, Ute Hamann, Diana M. Eccles, Celine M. Vachon, Fergus J. Couch. Expression quantitative trait locus analysis of triple negative breast cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3266. doi:10.1158/1538-7445.AM2014-3266

Purpose To develop a genomic signature that predicts benefit from trastuzumab in human epidermal growth factor receptor 2–positive breast cancer. Patients and Methods DASL technology was used to quantify mRNA in samples from 1,282 patients enrolled onto the Combination Chemotherapy With or Without Trastuzumab in Treating Women With Breast Cancer (North Central Cancer Treatment Group N9831 [NCCTG-N9831]) adjuvant trastuzumab trial. Cox proportional hazard ratios (HRs), adjusted for significant clinicopathologic risk factors, were used to determine the association of each gene with relapse-free survival (RFS) for 433 patients who received chemotherapy alone (arm A) and 849 patients who received chemotherapy plus trastuzumab (arms B and C). Network and pathway analyses were used to identify key biologic processes linked to RFS. The signature was built by using a voting scheme. Results Network and functional ontology analyses suggested that increased RFS was linked to a subset of immune function genes. A voting scheme model was used to define immune gene enrichment based on the expression of any nine or more of 14 immune function genes at or above the 0.40 quantile for the population. This model was used to identify immune gene–enriched tumors in arm A and arms B and C. Immune gene enrichment was linked to increased RFS in arms B and C (HR, 0.35; 95% CI, 0.22 to 0.55; P < .001), whereas arm B and C patients who did not exhibit immune gene enrichment did not benefit from trastuzumab (HR, 0.89; 95% CI, 0.62 to 1.28; P = .53). Enriched immune function gene expression as defined by our predictive signature was not associated with increased RFS in arm A (HR, 0.90; 95% CI, 0.60 to 1.37; P = .64). Conclusion Increased expression of a subset of immune function genes may provide a means of predicting benefit from adjuvant trastuzumab.

Abstract Purpose: We hypothesized that vertical blockade of VEGF signaling by combining bevacizumab with sorafenib in patients with recurrent glioblastoma would result in a synergistic therapeutic effect. We also investigated whether VEGF, VEGFR2 and hypoxia-inducible factor-1α single-nucleotide polymorphisms (SNP), circulating biomarkers of angiogenesis, and MRI markers such as apparent diffusion coefficient (ADC) are correlated with treatment efficacy and/or toxicity. Experimental Design: Patients received bevacizumab (5 mg/kg every 2 weeks) with sorafenib (200 mg twice a day, weekly, days 1–5; group A). Due to toxicity, the starting sorafenib dose was subsequently modified to 200 mg every day (group B). Results: Fifty-four patients were enrolled: 19 patients in group A and 35 in group B. Objective response rate was 18.5% with median duration of 6.7 months (range 0.5–24.1 months). Six-month progression-free survival (PFS6) was 20.4% (11/54), and median overall survival (OS) was 5.6 months [95% confidence interval (CI), 4.7–8.2]; outcome was similar between the two dose groups. We identified SNPs in the VEGF and VEGFR2 promoter regions, which were associated with PFS6 (P &lt; 0.022). Among molecular markers of angiogenesis, a higher log2 baseline level of stromal cell–derived factor-1 was associated with PFS6 success (P = 0.04). Circulating endothelial cells decreased during treatment with subsequent increase at disease progression (P = 0.022). Imaging analysis showed a trend associating ADC-L with poor outcome. Conclusions: The bevacizumab/sorafenib combination did not improve outcome of patients with recurrent glioblastoma versus historic bevacizumab-treated controls. Biologic markers of response and resistance to bevacizumab in gliomas were identified which merit prospective validation. Clin Cancer Res; 19(17); 4816–23. ©2013 AACR.

Abstract Purpose: Surrogate biomarkers are lacking for predicting or monitoring response to anti-angiogenesis therapies. We investigated whether circulating biomarkers (CBMs) of angiogenesis (i.e., bFGF, SDF-1α, HGF, sKIT, Ang-2, and PlGF) correlated with treatment efficacy in a pooled analysis of metastatic colorectal cancer (mCRC) and recurrent glioblastoma multiforme (rGBM) patients (pts) treated with bevacizumab/sorafenib (BEV/SOR). Experimental procedures: Peripheral blood samples were obtained for CBM analyses from pts enrolled in the North Central Cancer Treatment Group (NCCTG) phase II studies N054C (mCRC, n=75, Grothey, ASCO 2010) and N0776 (rGBM, n=54, Galanis, ASCO 2010). Blood collection time points were: baseline (BL), cycle 1 day 3 (C1D3), prior to treatment cycles 2, 3, 5, 7, 9, 11, 13, and after the pt went off study. Plasma CBM levels were determined by commercially-available ELISAs (R&D Systems) and performed according to manufacturer's instructions. For each CBM, a linear regression model that adjusted for study was used to evaluate the correlation with either progression-free survival (PFS) success/failure or progression within 14 days of the last blood draw for the following measures: absolute BL levels, log2 BL levels; C1D3 log2-fold change (FC) from BL, prior to cycle 2 log2-FC from BL, last draw log2-FC from BL, and last draw log2-FC from prior draw. Results summary: There appears to be no significant difference in the PFS or overall survival (OS) between the pts on these two studies; therefore, we pooled the patients from both studies for a combined analyses. We found no difference in absolute or log2 levels for any of the CBMs during the course of BEV/SOR treatment. In these pooled analyses, we found increased absolute and increased log2 BL SDF-1α levels were significantly correlated with PFS success (p=0.023 and p=0.02, respectively). After correcting for multiple comparisons (method of Benjamini & Hochberg) the p-values &lt;0.05 had a false-discovery-rate of 41.4%. None of the other CBM measures were significantly correlated with either PFS success/failure or progression within 14 days of the last blood draw (all p&gt;0.10). Conclusions: This study provides preliminary evidence that SDF-1α may be a useful CBM for determining treatment efficacy in these pts treated with BEV/SOR. Further prospective validation of SDF-1α for predicting pt progression with this treatment combination in advanced disease (e.g., mCRC and rGBM) appears warranted. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5562. doi:1538-7445.AM2012-5562

Abstract The androgen receptor (AR) is the principal target for treatment of non–organ-confined prostate cancer (PCa). Androgen deprivation therapies (ADT) directed against the AR ligand–binding domain do not fully inhibit androgen-dependent signaling critical for PCa progression. Thus, information that could direct the development of more effective ADTs is desired. Systems and bioinformatics approaches suggest that considerable variation exists in the mechanisms by which AR regulates expression of effector genes, pointing to a role for secondary transcription factors. A combination of microarray and in silico analyses led us to identify a 158-gene signature that relies on AR along with the transcription factor SRF (serum response factor), representing less than 6% of androgen-dependent genes. This AR-SRF signature is sufficient to distinguish microdissected benign and malignant prostate samples, and it correlates with the presence of aggressive disease and poor outcome. The AR-SRF signature described here associates more strongly with biochemical failure than other AR target gene signatures of similar size. Furthermore, it is enriched in malignant versus benign prostate tissues, compared with other signatures. To our knowledge, this profile represents the first demonstration of a distinct mechanism of androgen action with clinical relevance in PCa, offering a possible rationale to develop novel and more effective forms of ADT. Cancer Res; 71(5); 1978–88. ©2011 AACR.

Purpose Findings from the human epidermal growth factor receptor 2 (HER2) –positive National Surgical Adjuvant Breast and Bowel Project (NSABP) B31 trial suggested that MYC/HER2 coamplification (> 5.0 copies/nucleus) was associated with additional benefit from adjuvant trastuzumab in patients with early-stage breast cancer. To further explore this relationship, we investigated associations between MYC amplification and disease-free survival (DFS) in a similar adjuvant trastuzumab HER2-positive breast cancer trial—North Central Cancer Treatment Group (NCCTG) N9831. Patients and Methods This analysis included 799 patients randomly assigned to receive chemotherapy alone or with concurrent trastuzumab on N9831. Fluorescence in situ hybridization (FISH) was performed by using a dual-probe mixture for MYC and centromere 8 (MYC:CEP8) on tissue microarrays. MYC amplification was prespecified as MYC:CEP8 ratio > 2.2 or average MYC copies/nucleus > 5.0. Exploratory variables included polysomy 8. Results In comparing DFS (median follow-up, 4.0 years) between treatments, patients with MYC:CEP8 ratio ≤ 2.2 (n = 618; 77%) and > 2.2 (n = 181; 23%) had hazard ratios (HRs) of 0.46 (P < .001) and 0.67 (P = .33), respectively (interaction P = .38). Patients with MYC copies/nucleus ≤ 5.0 (n = 534; 67%) and > 5.0 (n = 265; 33%) had HRs of 0.52 (P = .002) and 0.48 (P = .02), respectively (interaction P = .94). Patients with MYC:CEP8 ratio < 1.3 with normal chromosome 8 copy number (n = 141; 18%) and ≥ 1.3 or < 1.3 with polysomy 8 (n = 658; 82%) had HRs of 0.66 (P = .28) and 0.44 (P < .001), respectively (interaction P = .23). Patients with MYC copies/nucleus < 2.5 (n = 130; 16%) and ≥ 2.5 (n = 669; 84%) had HRs of 1.07 (P = .87) and 0.42 (P < .001), respectively (interaction P = .05). Conclusion We did not confirm the B31 association between MYC amplification and additional trastuzumab benefit. Exploratory analyses revealed potential associations between alternative MYC/chromosome 8 copy number alterations and differential benefit of adjuvant trastuzumab.

Object Second-generation antiepileptic drugs (AEDs) are increasingly used in the care of patients with glioma. There is little data on how this practice compares with the use of traditional AEDs in this population. This noninferiority analysis compares seizure outcomes and side effects in patients with glioma treated with phenytoin and levetiracetam monotherapy. Methods The authors retrospectively reviewed the records of 500 consecutive patients with glioma who were treated in clinical trials from 2001 to 2008 at 3 Mayo Clinic campuses. To be eligible for the study, these patients had to have had at least 1 clinical seizure and to have undergone follow-up for at least 6 months. Seizure outcomes, defined by the occurrence of a second seizure, time to second seizure, and seizure frequency, along with AED side effects, were compared between cohorts treated with phenytoin or levetiracetam Results Seventy-six patients were identified, 25 treated with phenytoin and 51 with levetiracetam. Sixty-four percent of the patients had a Grade 4 astrocytoma. There was no difference in seizure outcome between the phenytoin and levetiracetam groups when comparing time to second seizure (p = 0.584), second seizure rates (p = 0.561), and average seizures per month (p = 0.776). When adjusting for age, sex, type of seizure, type of glioma, and dosage using univariate and multivariate models, there were no differences between the treatment groups and none of these covariates were statistically significant for explaining the second seizure rates between treatment groups (all p values > 0.05). The incidence of side effects in the levetiracetam group was 6% versus 20% in the phenytoin group (p = 0.106). Additionally, 36% of the patients in the phenytoin group had dose adjustments unrelated to breakthrough seizures compared with only 10% in the levetiracetam group (p = 0.010) Conclusions In this study, patients with glioma treated with levetiracetam and phenytoin had similar seizure control. Patients treated with levetiracetam experienced fewer side effects and required fewer nonseizure-related dose adjustments than patients treated with phenytoin. Levetiracetam is a safe, effective, and preferred alternative for seizure management in patients with glioma.

Abstract Polymorphisms in genes critical to cell cycle control are outstanding candidates for association with ovarian cancer risk; numerous genes have been interrogated by multiple research groups using differing tagging single-nucleotide polymorphism (SNP) sets. To maximize information gleaned from existing genotype data, we conducted a combined analysis of five independent studies of invasive epithelial ovarian cancer. Up to 2,120 cases and 3,382 controls were genotyped in the course of two collaborations at a variety of SNPs in 11 cell cycle genes (CDKN2C, CDKN1A, CCND3, CCND1, CCND2, CDKN1B, CDK2, CDK4, RB1, CDKN2D, and CCNE1) and one gene region (CDKN2A-CDKN2B). Because of the semi-overlapping nature of the 123 assayed tagging SNPs, we performed multiple imputation based on fastPHASE using data from White non-Hispanic study participants and participants in the international HapMap Consortium and National Institute of Environmental Health Sciences SNPs Program. Logistic regression assuming a log-additive model was done on combined and imputed data. We observed strengthened signals in imputation-based analyses at several SNPs, particularly CDKN2A-CDKN2B rs3731239; CCND1 rs602652, rs3212879, rs649392, and rs3212891; CDK2 rs2069391, rs2069414, and rs17528736; and CCNE1 rs3218036. These results exemplify the utility of imputation in candidate gene studies and lend evidence to a role of cell cycle genes in ovarian cancer etiology, suggest a reduced set of SNPs to target in additional cases and controls. (Cancer Epidemiol Biomarkers Prev 2009;18(3):935–44)