Search

Purpose: To explore whether the hypomethylating agent 5-azacytidine (5-aza) restores platinum sensitivity. Experimental Design: Patients with advanced cancer relapsed/refractory to platinum compounds were treated with 5-aza 20-50 mg/m2/d IV (D 1-5) and oxaliplatin 15-30 mg/m2/d IV (D 2-5) (“3+3” design escalation phase followed by expansion phase; NCT01039155). PK studies were performed. Global DNA and gene-specific methylation changes (baseline to D12) in blood and tissue, tissue platinum level, and CTR1 (copper transporter involved in platinum uptake) expression were assessed (expansion phase). Results: Overall, 37 patients were treated (median age, 59 yrs; men, 49%; prior: oxaliplatin, 68%; carboplatin, 30%; cisplatin, 16%) (escalation, n=21; expansion, n=16). No DLT was noted at the maximum dose level tested (5-aza, 50 mg/m2/d; oxaliplatin 30 mg/m2/d; used in expansion phase). The most common adverse events were anemia (49% of pts.) and fatigue (32%). LINE-1 was measured (surrogate for global DNA methylation) in pre- and post-treatment blood samples (9 patients) and in tumor samples (7 patients). All blood samples showed reduction in global DNA methylation (baseline to D12) (median, -22%; range, [-32%, -12%], p The mean cytoplasmic CTR1 score was 217.5 and 177.5 on D1 and D12, respectively (mean difference, -40; 95% CI: -69.1,-10.9; p=.02). The respective mean nuclear CTR1 score was 67.5 and 42.5 (mean difference, -25; 95% CI: -55.5,+5.5; p=.08). Oxaliplatin (D2) and 5-aza (D1 and 5) mean systemic exposure based on plasma AUCall resulted in a dose-dependent trend (oxaliplatin 15 to 30 mg/m2, 704.3 to 1149.0 hr·ng/mL; 5-aza 20 to 50 mg/m2, 112.5 to 404.5 [D1] and 143.7 to 426.5 hr·ng/mL [D5]). When the same dose of oxaliplatin was used with increasing doses of 5-aza from 25-50 mg/m2, a reduction in mean oxaliplatin exposure was noted (AUCinf: 1919.4 to 1455.9 hr·ng/mL). No significant differences in other non-compartmental modeled parameters estimated were observed. The total tumor oxaliplatin level was measured in 7 patients. The pre-dose levels ranged from Conclusion: 5-aza combined with oxaliplatin was safe. Hypomethylation was inconsistent in tumor tissue, the CTR1 score decreased and the tumor oxaliplatin level increased. Citation Format: Apostolia M. Tsimberidou, Kirk Culotta, Ignacio Wistuba, Siqing Fu, Aung Naing, Gerald Falchook, Sarina Piha-Paul, Ralph Zinner, Jaime Rodriguez-Canales, Guangan He, Zahid H. Siddik, Jaroslav Jelinek, Woonbok Chung, Yang Ye, Rabih Said, Kenneth Hess, David J. Stewart, Razelle Kurzrock, Jean-Pierre Issa. Phase I study of 5-azacytidine and oxaliplatin in patients with advanced cancers relapsed or refractory to platinum compounds. [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 CT201. doi:10.1158/1538-7445.AM2014-CT201

Background: Increased c-Met activity has been implicated in numerous oncologic processes. However, its prevalence among different cancers, association with demographics and genetic aberrations, and implication in therapy resistance has not been fully described. Patients and Methods: Between May 2010 to November 2012, 3607 patients were referred to the investigational cancer therapeutics department at MD Anderson Cancer Center for consideration of a phase I protocol. Among these patients, 1115 underwent MET amplification testing via fluorescence in-situ hybridization and were included in this analysis. Copy numbers are expressed as a ratio to the centromeric gene CEP7 with amplification defined as MET/CEP7 ≥2 or >20 gene copies in >10% of nuclei. Patient demographic, histologic, and molecular characteristics were stratified based on MET amplification status. Among patients with MET amplification, 20 were treated on a phase I protocol and evaluated for therapy response. Responses were categorized by RECIST criteria with objective responses defined as stable disease lasting ≥6 months or any observed partial responses. Categorical variables were compared via Chi-Squared and Fisher Exact Tests, while continuous variables were compared via Student's T-Test. Overall survival from the date of first consultation at our phase I clinic were compared via the Log-Rank test. Results: Twenty-nine (2.6%) patients exhibited MET amplification (median 3.48; range 2.05-16.14). Of these, 2 exhibited MET/CEP7 ratios 20 MET copies among >10% of nuclei. The highest prevalence was in tumors of adrenal (2 of 13; 15%) and renal (4 of 28; 14%) origins. Stratifying by MET amplification status, no differences were observed in demographic characteristics including age, ethnicity, or gender (all P>0.10). Amplified patients exhibited more organ systems involved with metastatic disease (median 3 vs. 4, P=0.0005). MET amplification was correlated with higher incidence of BRAF mutation, high-grade histology, and PTEN loss (all P Conclusions: Although caution must be stressed when interpreting these results, MET amplification exhibits an association with higher metastatic burden, specific molecular characteristics (BRAF mutation, PTEN loss), and high grade histology. Patients with MET amplification in this cohort exhibited a high response rate to VEGFR inhibitors while c-Met targeted agents showed little activity. Citation Format: Chad Tang, Denis Jardim, Debora Gagliato, Gerald Falchook, Kenneth Hess, Siqing Fu, Jennifer Wheler, Ralph Zinner, Aung Naing, Apostolia Tsimberidou, Funda Meric-Bernstam, David Hong. Analysis of 1115 patients tested for MET amplification and therapy response in the MD Anderson phase I clinic. [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 4424. doi:10.1158/1538-7445.AM2014-4424

The identification of new therapeutic uses for existing agents has been proposed as a means to mitigate the escalating cost of drug development. A common approach to such repurposing involves screening libraries of agents for activities against cell lines. In silico methods using knowledge from the biomedical literature have been proposed to constrain the costs of screening by identifying agents that are likely to be effective a priori. However, results obtained with these methods are seldom evaluated empirically. Conversely, screening experiments have been criticized for their inability to reveal the biological basis of their results. In this paper, we evaluate the ability of a scalable literature-based approach, discovery-by-analogy, to identify a small number of active agents within a large library screened for activity against prostate cancer cells. The methods used permit retrieval of the knowledge used to infer their predictions, providing a plausible biological basis for predicted activity.

Background: Cell-free (cf) DNA from the plasma of cancer patients offers an easily obtainable, low-risk, inexpensive and repeatedly applicable source of biologic material for mutation analysis of druggable targets and monitoring molecular changes in tumor(s) during and after therapeutic interventions. Novel, fast, and accurate diagnostic systems are needed for further development of plasma cfDNA testing in personalized therapy. Methods: cfDNA from plasma samples of patients with advanced cancers who progressed on systemic therapy was purified and 100 ng was used for testing for V600 BRAF mutations using the prototype molecular diagnostics (IdyllaTM) fully integrated real-time PCR-based prototype platform (Biocartis, Mechelen, Belgium) with a quick turnaround time (< 90 minutes). The IdyllaTM platform and the BRAF V600 mutation prototype assay were used for research purposes only. Results were compared to mutation analysis of archival primary or metastatic tumor tissue from a CLIA-certified laboratory if available. Results: cfDNA was extracted from plasma samples of 81 patients with advanced cancers (colorectal, n=37; melanoma, n=16; non-small cell lung, n=10; breast, n=4, Erdheim-Chester histiocytosis, n=3; appendiceal, n=3; other cancers, n=8). BRAF mutations were detected in 31% (25/81) of plasma samples and in 42% (30/72) of available archival tumor samples, resulting in concordance for patients who had plasma and tissue tested (N=72) in 90% (65/72) of cases (kappa=0.80, 95% confidence interval 0.65- 0.94). In 6 of 7 discrepant cases identical plasma cfDNA samples were tested using an alternative cfDNA BRAF mutation PCR-based method (BEAMing, Sysmex Inostics, Baltimore, MD), which yielded 100% agreement. Longitudinally collected plasma samples were available in 3 patients (appendiceal, melanoma and papillary thyroid cancer) treated with BRAF targeting combinations and changes in the amount of BRAF-mutant cfDNA corresponded with changes in tumor markers or subsequent responses to therapy visualized via imaging. Conclusions: Detecting V600 BRAF mutations in cfDNA from plasma using the IdyllaTM platform is a fast and noninvasive alternative to mutation testing of tumor tissue with an acceptable level of concordance and sensitivity, and should be investigated further for testing and monitoring of BRAF mutation status in patients with cancer. Citation Format: Helen J. Huang, Bart Claes, Gerald S. Falchook, Veronica R. Holley, Aung Naing, Sarina A. Piha-Paul, Apostolia M. Tsimberidou, Ralph G. Zinner, Daniel D. Karp, Siqing Fu, Vivek Subbiah, David S. Hong, Jennifer J. Wheler, Vanda M. Stepanek, Goran Cabrilo, Rajyalakshmi Luthra, Agop Y. Bedikian, Bryan K. Kee, Cathy Eng, Michael J. Overman, Kevin B. Kim, Tania Ivens, Erwin Sablon, Geert Maertens, Razelle Kurzrock, Funda Meric-Bernstam, Filip Janku. BRAF mutation testing in cell-free DNA from plasma of patients with advanced cancers using a novel, rapid, automated molecular diagnostics prototype platform (IdyllaTM). [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 5584. doi:10.1158/1538-7445.AM2014-5584

Purpose: To evaluate whether PIK3CA mutations and/or PTEN loss predict better clinical outcomes to phase I trials of PI3K/AKT/mTOR inhibition-based therapies in patients with metastatic, recurrent or refractory endometrial cancers. Methods: Outcome analyses were conducted on 73 patients with advanced endometrial cancers who were seen in a designated Phase I clinic at MD Anderson Cancer Center from 1/1/2006 to 4/30/2013, and tested for PIK3CA mutations and/or PTEN loss (an immunoreactive score of 3 or less) in a CLIA-certified molecular diagnostic laboratory, in accordance with the IRB guidelines. All data were obtained from patients' electronic medical records. Results: All patients (n=73) had received at least one systemic therapy before the referral (median = 2). Approximately 24% of tested patients (n=13/55) harbored PIK3CA mutations. Of the 13 patients with PIK3CA mutations, 12 (92%) were treated in clinical trials including a PI3K/AKT/mTOR pathway inhibitor with or without chemotherapy (matched therapy). Matched therapy led to a total of 54% PR/SD≥6months (23% PR and 31% SD≥6months) and a median PFS of 7.9 months. The response rate and PFS are much lower in the 17 patients who did not have PIK3CA and PTEN tested; 17.6% SD≥6months and median PFS of 1.9 months (p=0.037and p=0.013, respectively). Of those who had PTEN tested (n=41), 21 patients (51%) had PTEN loss. All 21 patients with PTEN loss received matched therapies, which led to a total of 38% PR/SD≥6months (14% PR and 24% SD≥6months) and a median PFS of 5 months. That is not significantly different from therapy in patients who do not have PIK3CA and PTEN tested (p=0.167and p=0.076, respectively). Of 30 patients with PIK3CA mutations and/or PTEN loss (n=30), matched therapy led to a significantly longer median PFS of 4.8 months (p=0.03) but similar response rate (15% PR and 25% SD≥6months) to those in patients without PIK3CA and PTEN tests (p=0.114). Patients with unknown PIK3CA and PTEN status received significantly fewer regimens targeting at PIK3CA pathway compared to those with known status (p Conclusions: Matched therapy using PI3K/AKT/mTOR pathway inhibitor -based regimens led to higher clinical response rate including CR/PR/SD ≥ 6months and PFS in patients with PI3Kinase mutation but not PTEN loss, indicating that different biological mechanisms are involved in these molecular aberrations. Further evaluation in patients with advanced endometrial cancers may require dissect individual molecular aberration for personalized cancer therapy. Citation Format: Xiaochun Liu, Mingmou Hou, Jennifer Wheler, Aung Naing, David Hong, Apostolia Tsimberidou, Sarina Piha-Paul, Gerald Falchook, Ralph Zinner, Filip Janku, Ramondetta Lois, Kathleen Schmeler, Karen Lu, Razelle Kurzrock, Siqing Fu. Differential effects of PIK3CA mutations and PTEN loss in patients with advanced endometrial cancers on matched therapy. [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 2838. doi:10.1158/1538-7445.AM2014-2838

The purpose of this study is to test aerosol delivery of Interkeukin-2 (IL-2) in patients > 12 yrs. with lung metastases to establish feasibility and safety in the pediatric population and determine the maximum tolerated dose (MTD) for potential combination therapies in patients with Osteosarcoma (OS) lung metastases. Organ-specific delivery of agents by the aerosol route offers many advantages for the treatment of OS since it targets the organ where metastases are and provides significantly less toxic effects. OS survival has remained at 65-70% for the last 20 years. Our pre-clinical data using a human OS mouse model demonstrated therapeutic efficacy of aerosol IL-2 in OS lung metastases (p = 0.03) and increase apoptosis measured by TUNEL (p= 0.003). In addition, aerosol IL-2 increased local immune cell proliferation as demonstrated by an increase in CD4+ T cells and NK cells with no evidence of T regulatory cells. Most recently, IL-2 was shown to induce autophagy, an evolutionary, conserved, intracellular self-defense mechanism, in CD4+ T cells and NK cells and contributes to growth factor withdrawal cell death. Inhibition of autophagy augments immune cell function allowing a better anti-tumor effect. Immunohistochemistry staining of OS lung tumors from mice treated with aerosol IL-2 showed an increase in Beclin expression as evidence of autophagy induction. We initiated a Phase I/II clinical trial of aerosol IL-2 to include patients > 12 years with lung metastases. The primary objective is to determine feasibility and safety of aerosol IL-2. The secondary objective is to determine local effects of aerosol IL-2 by analyzing pre and post-treatment surgical samples for evidence of increase NK cell number, function and autophagy induction. Patients received aerosolized IL-2 for 3 consecutive weeks (21 day cycle) for 2 cycles with 1-week rest between cycles providing no drug limiting toxicities (DLTs) occur. The first treatment is delivered in the hospital. Subsequent treatments are given at home providing patients had been educated. From dose levels 1-4 only 1-patient/dose level was enrolled. From dose level 5 on, 3 patients will be enrolled. Once MTD is reached an additional 14 patients will be treated. 34 to 56 patients may be enrolled. Clinical response will be assessed by chest CT and determined using modified Response Evaluation Criteria in Solid Tumor (RECIST) after 2nd cycle. Results: 5 patients have been enrolled, ages 18, 35, 66, 20 and 15; all male, 3 with Ewing's Sarcoma and 2 with Osteosarcoma. No side effects had been detected and serum levels of IL-2 were only detected in the fourth patient but were very low. Conclusion: Aerosol IL-2 seems so far feasibly and safe. Completion of the Phase I will provide information on the maximum tolerated dose (MTD) to use on the Phase II where only patients with OS lung metastases will be enrolled. Culmination of this trial will provide the basis for future combination therapy trials. Citation Format: Nancy Beatriz Gordon, Najat Daw Bitar, Peter Anderson, Sergei Guma, Hsuan-Chu Chien, Lacey M. McQuinn, Joshua P. Hein, Ralph Zinner, Eugenie S. Kleinerman, Aung Naing. Aerosol interleukin-2 for the treatment of patients ≥ 12 years old with osteosarcoma lung disease: A phase I/II study. [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 CT406. doi:10.1158/1538-7445.AM2014-CT406

Background: Detection and monitoring of oncogenic mutations in cell-free urinary DNA opens the possibility of a new paradigm for a truly non-invasive method of individualized care for metastatic cancer patients, which would enable the quantitation of mutational tumor load and respective concordance to therapeutic responsiveness followed by detection of emerging genomic alterations underlying acquired resistance. Methods: Cell-free DNA was isolated from single and/or multiple sequential urine samples from patients with advanced cancers or Erdheim-Chester disease and BRAF V600E mutation in the tumor tissue from a CLIA-certified laboratory, who progressed on systemic therapy. Assays for quantitative assessment of BRAF V600E mutation in cell-free urinary DNA were developed using digital droplet PCR methodology (RainDance, MA) with enrichment of mutation-containing DNA fragments by pre-amplification of the BRAF gene. Detection limits were established as wild-type (0.107%). Results: Cell-free DNA was extracted from urine of 33 patients with diverse advanced cancers (melanoma, n=11; colorectal cancer, n=8; non-small cell lung cancer, n=4; papillary thyroid carcinoma, n=4; other, n=4) or Erdheim-Chester disease (n=1) with BRAF V600E mutation in the tumor tissue. Of these 33 patients, 25 (76%) had the same mutation in urinary cell-free DNA (mutant, n=14; low-mutant, n=11). In addition, 17 patients had longitudinal analysis of percentage of BRAF V600E mutation to wild-type in sequentially collected urine samples and the dynamics of BRAF V600E mutation in the urine correlated with response to therapy in 13 (76%) of them. Conclusion: Our data suggest that detecting BRAF V600E mutation in cell-free DNA from urine can offer a noninvasive alternative to mutation testing of tumor tissue with acceptable concordance and should be investigated further for testing and monitoring of mutation status in patients with cancer and Erdheim-Chester disease. Citation Format: Filip Janku, Gerald S. Falchook, Sarina A. Piha-Paul, Aung Naing, Veronica R. Holley, David S. Hong, Vivek Subbiah, Daniel D. Karp, Ralph G. Zinner, Siqing Fu, Jennifer J. Wheler, Funda Meric-Bernstam, Vanda M. Stepanek, Rajyalakshmi Luthra, Lorieta Leppin, Latifa Hassaine, Karena Kosco, Jason C. Poole, Cecile Rose T. Vibat, Mark G. Erlander. Droplet digital PCR detection and longitudinal monitoring of BRAF mutations in cell-free urinary DNA of patients with metastatic cancers or Erdheim-Chester disease. [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 LB-170. doi:10.1158/1538-7445.AM2014-LB-170

Background: Cell-free (cf) DNA in the plasma of cancer patients offers an easily obtainable, low-risk, inexpensive, and repeatedly available source of biologic material for mutation analysis and monitoring of molecular changes throughout cancer therapy. Methods: DNA in plasma from patients with advanced cancers who progressed on systemic therapy was tested for BRAF V600 and KRAS G12 and G13 mutations using the ICE COLD-PCR platform. ICE COLD-PCR, “Improved and Complete Enrichment COamplification at Lower Denaturation,” selectively amplifies mutant DNA by exploiting differences in denaturation temperatures between mutant DNA duplexes and normal “wild-type” DNA duplexes. KRAS Exon 2 and BRAF Exon 15 ICE COLD-PCR was performed on plasma samples. Amplicons were analyzed using Sanger sequencing and results were compared to the mutation status of archival primary or metastatic tumor tissue as determined in a CLIA-certified laboratory during routine clinical care. Results: Plasma samples from 77 patients with advanced cancers and known tumor tissue BRAF and/or KRAS mutation status (colorectal cancer, n=38; melanoma, n=17; non-small cell lung cancer, n=7; other cancers, n=15) were obtained before treatment and, if possible, sequentially during therapy and tested for BRAF (42), KRAS (34) or BRAF and KRAS (1) mutations in cfDNA. BRAF mutations were detected in 93% (40/43) of archival tumor samples compared to 70% (30/43) of plasma cfDNA samples (agreement 77%). In addition, 20 patients treated with systemic therapy had serial plasma samples collected and the change in relative abundance of BRAF-mutant compared to wild-type cfDNA corresponded with the clinical course of 15 patients and was discrepant for 1 patient; in 5 patients no BRAF mutated cfDNA was detected at any time point. KRAS mutations were detected in 83% (29/35) of archival tumor samples compared to 74% (26/35) of plasma cfDNA samples (agreement 80%). In addition, 12 patients treated with systemic therapy had serial plasma collected and the change in relative abundance of KRAS-mutant compared to wild-type cfDNA corresponded with clinical course in 10 patients; in 2 patients no KRAS mutated cfDNA was detected at any time point. Conclusions: Detection of BRAF and KRAS mutations in cfDNA can provide a fast and noninvasive alternative to mutation testing in tumor tissue with a potential to be used for monitoring response to cancer therapy. Citation Format: Filip Janku, Ben Legendre, Katherine Richardson, Gerald S. Falchook, Aung Naing, Veronica R. Holley, Siqing Fu, David S. Hong, Sarina A. Piha-Paul, Jennifer J. Wheler, Ralph G. Zinner, Vivek Subbiah, Apostolia M. Tsimberidou, Daniel D. Karp, Vanda M. Stepanek, Goran Cabrilo, Rajyalakshmi Luthra, Funda Meric-Bernstam, Agop Y. Bedikian, Bryan K. Kee, Cathy Eng, Michael J. Overman, Kevin B. Kim, Amy Kruempel, Jaclyn Pope, Courtney Cubrich, Grant Wu, Marcia Lewis, Razelle Kurzrock. BRAF and KRAS mutation testing in plasma cell-free DNA with ICE COLD-PCR in patients with advanced cancers. [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 5607. doi:10.1158/1538-7445.AM2014-5607

Background: Carcinoma of unknown primary (CUP) represents a heterogeneous group of cancers without an identifiable primary disease site where staining for a vast array of immunohistochemical markers is undertaken to identify tumor lineage. Given rapid advances in characterizing cancer genomes, we sought to describe the molecular profiles of patients with advanced CUP treated with novel targeted therapies on phase I clinical trials to identify pathways of interest to drive future therapeutic strategies. Methods: Of patients with advanced malignancies seen in the Phase I clinic, we identified 16 CUP patients on whom adequate tissue was available to perform next-generation sequencing (NGS) from FFPE sections using a targeted NGS assay in a CLIA laboratory (Foundation One, MA). Over 2000 exons of 186 cancer-related genes plus over 30 introns from 14 genes often rearranged in cancer were fully sequenced for point mutations, insertions/deletions, copy number alterations (CNAs) and select gene fusions. Results: Of 16 tumor specimen, 14 demonstrated genomic alterations with a median of 4 aberrations within each specimen (range 0 - 10, total of 69 alterations); two patients did not demonstrate any tested changes. All patients with detected mutations possessed an exclusive unique molecular profile of aberrations with no two patients demonstrating an identical panel of mutations. A total of 69 alterations were identified of which 18 (26%) led to aberrant cell signaling within the PI3K/AKT/MTOR (including 4 with PIK3CA mutations) or the MAPK signaling cascades, primarily KRAS mutations in 3 patients. Also noted were 13 (19%) alterations impairing cell cycle arrest most commonly affecting cyclin dependent kinases, 10 (14%) impacting epigenetic regulation and DNA methylation, 10 (14%) affecting known tumor suppressor genes including TP53 in 8 patients, and 10 (14%) unique alternations in transcriptional regulators. Conclusion: With thorough molecular profiling, we can identify the unique fingerprint of potential actionable aberrations in a subset of advanced CUP patients. Herein we identified activation of the PI3K/AKT/MTOR and RAS/RAF/MEK signaling pathways combined with impaired cell cycle arrest in 8 of 16 (50%) tested patients, suggesting that a combination of PI3K and MAPK cascade inhibitors in combination with a cell cycle inhibitor may be of therapeutic value to these patients. Further matching patients to these actionable aberrations is underway, thereby translating these therapeutic targets to the bedside with the development of novel targeted therapies. Note: This abstract was not presented at the meeting. Citation Format: Ishwaria M. Subbiah, Gauri Varadhachary, Apostolia M. Tsimberidou, Jennifer J. Wheler, Vivek Subbiah, Filip Janku, Sinchita Roy Chowdhury, Ralph Zinner, Funda Meric-Bernstam, David S. Hong. One size does not fit all: Fingerprinting advanced carcinoma of unknown primary through comprehensive profiling identifies aberrant activation of the PI3K and MAPK signaling cascades in concert with impaired cell cycle arrest. [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 4700. doi:10.1158/1538-7445.AM2014-4700

e22082 Background: Biomarkers associated with efficacy of HER2-directed treatments are lacking. We evaluated HER2 ECD levels and HER2 SNPs in patients treated on a clinical trial of combination tra...

2563 Background: Besides its selective activity to induce autophagy and caspase-independent apoptosis in argininosuccinate synthetase (ASS)-deficient cancer cells, ADI-PEG20 (pegylated arginine dei...

TPS2639 Background: Preclinical models have shown that concurrent inhibition of the mammalian target of rapamycin (mTOR) pathway signaling cooperates with epidermal growth factor (EGFR)/vascular en...

e13521 Background: Genomic profiling is increasingly used in the management of patients with cancer. We used “precision medicine” with genomic profiling to treat patients referred to our Phase I Clinical Trials Program. Methods: Genomic analysis of patients referred from 9/07 to 10/13 was performed in a CLIA-certified lab. Patients with tumor alterations were treated with matched targeted therapy (MTT), when available. Results: Of 3,745 patients, 2,089 (55.8%) had ≥1 alteration (actionable, 1,740): KRAS 20%; PTEN 16%; PI3K 11%; BRAF 10%; NRAS 8%; EGFR 4%; and CKIT 2%; 1,187 patients were treated (median age, 58 yrs; women, 61%): MTT: 673 (56.7%); non-matched therapy (NMT): 514 (43.3%). Results are shown in Table. The clinical benefit (CB) rates in patients with PI3K/AKT/mTOR pathway alterations treated with MTT by tumor type were: breast 40% (18/45); gynecologic 33% (20/60); head and neck 24% (6/25); colorectal 3% (1/32); and other 33% (18/54) (p=.006). The CB rates in patients with RAS/RAF/MEK pathway al...

10595 Background: Therapeutic options are limited for advanced sarcomas. Search for molecular alterations could lead to identification of novel therapies for refractory sarcoma patients. Rarity pos...

e22154 Background: Patients with advanced carcinoma of unknown primary (CUP) have limited effective therapeutic options given the challenges of phenotypic and genotypic diversity. To identify futur...

e22139 Background: Cell-free (cf) DNA from the plasma of cancer patients offers an easily obtainable source of biologic material for analysis and monitoring of molecular aberrations. Novel technolo...

3516 Background: BRAF V600 mutations, present in 5-10% of patients (pts) with metastatic colorectal cancer (mCRC), are considered poor prognostic markers, and

2576 Background: VEGF inhibition-mediated tumor hypoxia promotes invasive growth by transcriptional activation of hypoxia-induced factors mediated by tumor hypoxia. Accordingly, we conducted this p...

3607 Background: This retrospective study aims to investigate the activity of retreatment with anti-EGFR-based therapy in order to explore the concept of clonal evolution by evaluating the impact o...

2606 Background: Activation of the PI3K/AKT/mTOR pathway has a key role in breast cancer development and therapeutic resistance. Next generation sequencing (NGS) has expanded our understanding of t...

PURPOSE:FBXW7 is a tumor suppressor gene responsible for the degradation of several proto-oncogenes. Preclinical data suggest that FBXW7 mutations sensitize cells to mTOR inhibitors. Clinicopathologic characteristics of cancer patients with FBXW7 mutations and their responses to mTOR inhibitors remain unknown. METHODS:Using multiplex gene panels we evaluated how the FBXW7 mutation affected the cancer phenotype of patients referred to a phase I clinic starting January 2012. Whenever possible patients positive for FBXW7 mutation were treated with regimens containing an mTOR inhibitors and their outcomes were reviewed. RESULTS:FBXW7 mutations were detected in 17 of 418 patients (4.0%). Among tumor types with more than 10 patients tested, FBXW7 mutations occurred in colorectal cancer (7/49; 14.3%), squamous cell cancer of head and neck (2/18; 11.1%), liver (1/13; 7.7%), and ovarian cancers (1/40; 2.5%). No one clinical, pathological or demographic feature was characteristic of the FBXW7-mutated patient population. The mutation occurred in isolation in only 2/17 (12%) patients, and KRAS was frequently found as a concomitant mutation, especially in patients with colorectal cancer (6/7; 86%). Ten patients were treated on a protocol containing an mTOR inhibitor, with a median time to treatment failure of 2.8 months (range, 1.3-6.8). One patient with liver cancer (fibrolamellar subtype) continues to have a prolonged stable disease for 6.8+ months. CONCLUSION:In patients with advanced cancers, somatic mutations in FBXW7 usually occur with other simultaneous molecular aberrations, which can contribute to limited therapeutic efficacy of mTOR inhibitors.

407 Background: c-MET is a tyrosine kinase receptor involved in cancer growth, migration, development of metastasis and angiogenesis. Deregulation of this gene has been associated with poor prognosis and resistance to treatment in prostate, bladder and renal cell cancer (RCC). Methods: Patients with GU malignancies referred to the phase I clinic were evaluated for the presence of c-MET mutation and amplification. Their outcomes on protocols containing c-METinhibitors were also analyzed. Results: c-MET amplification was detected in 7 out of 97 (7.2%) patients (4/27 renal, 1/18 urothelial and 2/12 adrenocortical carcinoma), while c-MET mutation/variant in 3 out of 53 (5.7%) patients (2/20 renal and 1/15 prostate cancer). No demographic characteristics were associated with either c-MET abnormalities, but patients tested positive for either mutation or amplification had a higher number of metastatic sites (median number of sites of 4 versus 3 for wild type patients). Median overall survival from phase I consult was 4.4 and 10.6 months for patients with and without a c-MET alteration, respectively. Twenty-nine (25%) patients were treated under a protocol containing a c-MET inhibitor. Overall 6 (21%) patients had a partial response (all prostate and RCC) and 10 (34%) had stable disease as best response. Median time to tumor progression was 2.3 months (0.4-19.7) for all treated patients. No responses were detected in patients with a c-MET genetic abnormality or treated with a specific c-MET inhibitor as a single agent. Conclusions: c-MET mutation and amplification are present in different GU malignancies conferring a worse prognosis for patients referred to the phase I program. We observed activity of c-MET inhibitors for patients without c-MET genetic abnormalities and when c-MET inhibition was combined with other targets/drugs.

The PTEN/MMAC1/TEP (PTEN) tumor suppressor gene at 10q23.3 is mutated in multiple types of sporadic tumors including breast cancers and also in the germline of patients with the Cowden's breast cancer predisposition syndrome. The PTEN gene encodes a multifunctional phosphatase capable of dephosphorylating the same sites in membrane phosphatidylinositols phosphorylated by phosphatidylinositol 3'-kinase (PI3K). We demonstrate herein that loss of PTEN function in breast cancer cells results in an increase in basal levels of phosphorylation of multiple components of the P13K signaling cascade as well as an increase in duration of ligand-induced signaling through the P13K cascade. These alterations are reversed by wild-type but not phosphatase inactive PTEN. In the presence of high concentrations of serum, enforced expression of PTEN induces a predominant G1 arrest consistent with the capacity of PTEN to evoke increases in the expression of the p27Kip1 cyclin dependent kinase inhibitor. In the presence of low concentrations of serum, enforced PTEN expression results in a marked increase in cellular apoptosis, a finding which is consistent with the capacity of PTEN to alter the phosphorylation, and presumably function, of the AKT, BAD, p70S6 kinase and GSK3 alpha apoptosis regulators. Under anchorage-independent conditions, PTEN also induces anoikis, a form of apoptosis that occurs when cells are dissociated from the extracellular matrix, which is enhanced in conjunction with low serum culture conditions. Together, these data suggest that PTEN effects on the PI3K signaling cascade are influenced by the cell stimulatory context, and that depending on the exposure to growth factors and other exogenous stimuli such as integrin ligation, PTEN can induce cell cycle arrest, apoptosis or anoikis in breast cancer cells.

Background: Next generation sequencing (NGS) has revealed significant heterogeneity among patients’ tumors. Molecular aberrations may be both prognostic and predictive. Methods: We retrospectively analyzed 57 patients with advanced or metastatic breast cancer, with NGS profiling of their tumors, treated on Phase I trials at MD Anderson Cancer Center and UCSD-Moores Cancer Center. The objectives of this study were to: 1) characterize molecular aberrations and, 2) correlate molecular aberrations with response to targeted therapies. Results: Fifty-seven patients with advanced or metastatic breast cancer with NGS profiling of their tumors were included. Fifty-six of 57 patients (98%) with NGS analysis of their tumors demonstrated at least one molecular aberration. A total of 216 somatic aberrations in 71 different genes were identified including variants in the same gene. There were 99 mutations, 97 amplifications, 11 deletions, 5 splices, 2 truncations, 1 fusion and 1 rearrangement. The most commonly mutated genes were: TP53 (26 patients, 46%); PIK3CA (18 patients, 32%); PIK3R1, GATA3 (5 patients, 9% each); BRCA2, CDH1, and NF1 (3 patients, 5% each). The most common amplifications occurred at: CCND1, MYC (12 patients, 21% each); HER2 (official gene name ERBB2), MCL1 (7 patients, 12% each); and FGFR1 (6 patients, 11%). In addition, PTEN deletion and CDKN2 deletion were found in 6 (11%) and 3 (5%) patients, respectively. Among those genes with the highest number of gene variants included: TP53 (26 variants); PIK3CA (7 variants); GATA (5 variants); PIK3R1 (4 variants); and BRCA 2 (4 variants). No two patients demonstrated the same spectrum of molecular aberrations. Forty-four of 57 patients were evaluable for response on targeted therapies excluding single-agent hormone therapy (13 patients did not receive targeted therapy). Thirty-seven patients received therapies that were either directly or indirectly matched to NGS aberrations resulting in 2 patients with complete responses (CR), 5 with partial response (PR), 10 with stable disease ≥ 6 months (SD) and, 20 with progressive disease (PD). Seven patients who did receive matched therapy demonstrated SD (3 patients) and PD (4 patients) 3 SD (Fisher's exact test, p=0.88). Conclusion: Unique profiles of genomic aberrations in patients with breast cancer were demonstrated. Further evaluation NGS in larger patient cohorts is warranted including studies that prospectively match aberrations to targeted therapies. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A212. Citation Format: Jennifer J. Wheler, Barbara A. Parker, Jack Lee, Roman Yelensky, Stacy Moulder, Apostolia M. Tsimberidou, Filip Janku, Johnique T. Atkins, Ralph Zinner, Richard Q. Schwab, Maria Schwaederle, Vivek Subbiah, Siqing Fu, Philip J. Stephens, Razelle Kurzrock. Next generation sequencing (NGS) in 57 patients with advanced or metastatic breast cancer: Identification of unique genomic profiles and correlation with response. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A212.

Background: Actionable mutations confer a survival advantage to cancer cells and can provide therapeutic targets. Cell-free (cf) DNA in the plasma of cancer patients offers an easily obtainable, low-risk, and inexpensive source of biologic material for mutation analysis. Methods: cfDNA in plasma samples from patients with advanced cancers who progressed on systemic therapy was tested for 21 mutations in BRAF (codon 600), EGFR (exons 19-21), KRAS (codons 12, 13), and PIK3CA (exons 9, 20) using the BEAMing method. Results were compared to mutation analysis results from archival primary or metastatic tumor tissue from a CLIA-certified laboratory. Results: cfDNA was extracted from 155 patients with advanced cancers. BRAF mutations were detected in 27% (33/122) plasma samples and in 30% (37/122) of archival tumor samples resulting in agreement in 88.5% cases (kappa=0.72, 95% confidence interval [CI] 0.58- 0.86). EGFR mutations were detected in 8% (6/77) plasma samples and in 6% (5/77) tumor samples resulting in agreement in 98.7% cases (kappa=0.90, 95% CI 0.71- 1.09). KRAS mutations were detected in 50% (55/110) of plasma samples and in 53% (58/110) of tumor samples, resulting in agreement in 82.7% cases (kappa=0.66, 95% CI 0.51- 0.80). PIK3CA mutations were detected in 20% (19/94) of plasma samples and in 14% (13/94) of tumor samples, resulting in concordance in 91.5% cases (kappa=0.70, 95% CI 0.51- 0.89). Of interest, 2 of 9 patients with BRAF mutations (both melanoma) in the tumor, but not in cfDNA had plasma collection at the time of progression on prior BRAF targeting therapy. Also, a patient with V600E BRAF-mutated melanoma, who previously progressed on a MEK inhibitor, displayed a G12R KRAS mutation in cfDNA and a patient with V600K BRAF-mutated melanoma with additional E542K and H1047R PIK3CA mutations in cfDNA was ultimately refractory to BRAF inhibitor treatment. Also, a V600K BRAF mutation was found in the cfDNA of a patient with colorectal cancer and a G12V KRAS mutation. In addition, 2 patients with EGFR-mutated lung adenocarcinoma (exon 19 and L858R), who progressed on erlotinib, had T790M EGFR (n=2) and E545K PIK3CA (n=1) mutations in cfDNA, plausibly explaining secondary resistance to erlotinib. A patient with Erdheim-Chester disease and a V600E BRAF mutation also had a T790M EGFR mutation in cfDNA. Finally, a patient with H1047R PIK3CA mutant ovarian cancer showed an additional G12C KRAS mutation in cfDNA and experienced a partial response followed by early progression on a PI3K inhibitor. Patient characteristics, mutation types and detailed discrepancy analysis will be presented. Conclusions: Detecting actionable mutations in cfDNA is a noninvasive alternative to mutation testing in tumor tissue with an acceptable level of concordance, and should be investigated further for selecting appropriate targeted therapies and monitoring mutation profiles. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B26. Citation Format: Filip Janku, Philipp Angenendt, Aung Naing, Gerald S. Falchook, Apostolia M. Tsimberidou, Veronica R. Holley, Siqing Fu, David S. Hong, Jennifer J. Wheler, Sarina A. Piha-Paul, Ralph G. Zinner, Daniel D. Karp, Vivek Subbiah, Kevin B. Kim, Funda Meric-Bernstam, Frank Diehl, Razelle Kurzrock. Actionable mutations in cell-free DNA in plasma of patients with advanced cancers referred for experimental targeted therapies. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B26.

The purpose of this study is to determine safety of aerosol IL-2 in patients ≥12 yrs. with lung metastases to establish the maximum tolerated dose (MTD) to use for combination therapy with infused natural killer (NK) cells in patients with osteosarcoma (OS) lung metastases. OS survival has remained at 65-70% for the last 20 years. One recent therapeutic approach has been to use adoptive cell-therapy to non-specifically augment immune function in vivo. However, therapeutic efficacy of immune cells depends upon proliferation and persistence in the tumor area. Organ-specific delivery of interleukin-2 (IL-2) offers that benefit. We initiated a Phase I/II clinical trial of aerosol IL-2 to include patients ≥12 years with lung metastases. The primary objective is to determine feasibility and safety of aerosol IL-2. Patients only received aerosolized IL-2 for 3 consecutive weeks (21 day cycle) for 2 cycles with 1-week rest between cycles providing no drug limiting toxicities (DLTs) occur. Each patient is educated on treatment administration the first time in the hospital. Further treatments are administered at home and remote spirometry and pulse oximetry are recorded before each treatment. Data is downloaded on a web portal and captured into the patient's medical record. We anticipated the maximum tolerated dose (MTD) to be at dose level 5. From dose levels 1-4 only 1 patient/dose level was enrolled. From dose level 5 on, 3 patients will be enrolled and the 3 x 3 model will be followed. Once MTD is reached an additional 14 patients will be treated at the MTD. 34 to 56 patients may be enrolled. Clinical response will be assessed by chest CT and determined using modified Response Evaluation Criteria in Solid Tumor (RECIST) after 2nd cycle. So far, 4 patients have been enrolled, ages 18, 35, 66, and 20; all male, 3 with Ewing's Sarcoma and 1 with osteosarcoma. Side effects were grade 1 fatigue (n= 1), grade 1 cough (n= 2), and grade 1 wheezing (n=1). None of these adverse events have been attributed to the IL-2 therapy since symptoms resolved without intervention while on therapy. Serum levels of IL-2 in the first three patients were undetectable and they had progressive disease. The fourth patient is on treatment. Supporting this clinical study is pre-clinical data from our lab. Immunocompetent mice treated with aerosol IL-2 had significant increase in the number of local NK cells in the lung when compared with aerosol phosphate buffer saline (PBS) group. Using a human OS mouse model we demonstrated therapeutic efficacy of aerosol IL-2 in OS lung metastases when compared with aerosol PBS group (p = 0.03) and there was significant increase in apoptosis measured by TUNEL (p= 0.003). In addition, aerosol IL-2 significantly increased proliferation of local infused NK cells in the lungs when compared with aerosol PBS group (p= 0.03 and p=0.007 at 24 and 72 hours respectively) with no proliferation demonstrated in other organs including bone marrow. In conclusion, clinical evaluation of aerosol IL-2 therapy will potentially benefit patients with OS who succumb almost always due to lung disease and allow future combination therapy with infused NK cells. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C64. Citation Format: Nancy B. Gordon, Peter Anderson, Sergei R. Guma, Hsuan-Chu Chien, Lacey M. McQuinn, Joshua P. Hein, Ralph G. Zinner, Eugenie S. Kleinerman, Aung Naing. Clinical evaluation of aerosol IL-2 in patients with lung metastases to test feasibility and safety for future combination therapy using infused natural killer cells. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C64.

Background: Cell-free (cf) DNA in the plasma of cancer patients offers an easily obtainable, low-risk, and inexpensive and repeatedly applicable source of biologic material for mutation analysis for druggable targets and monitoring of the molecular changes of tumor during and after therapeutic interventions. Novel, fast, and accurate diagnostic systems are needed for further implementation of cfDNA testing in personalized therapy. Methods: cfDNA in plasma samples from patients with advanced cancers who progressed on systemic therapy was purified and 100 ng was used for testing for V600 BRAF mutations using the prototype molecular diagnostics (MDx) fully integrated real-time PCR-based prototype platform (Biocartis, Mechelen, Belgium) with a quick turnaround time (< 90 minutes). Both the MDx prototype platform and the BRAF V600 mutation prototype assay were used for research purposes only. Results were compared to mutation analysis of archival primary or metastatic tumor tissue from a CLIA-certified laboratory. Results: cfDNA was extracted from plasma samples of 52 patients with advanced cancers (colorectal, n=23; melanoma, n=14; non-small cell lung cancer, n=6; ovarian, n=2; other cancers, n=7). BRAF mutations were detected in 35% (18/52) of plasma samples and in 48% (25/52) of archival tumor samples, resulting in agreement in 86.5% cases (kappa=0.73, 95% confidence interval 0.55- 0.91). In 7 cases with BRAF mutation in the tumor and wild-type BRAF in cfDNA, cfDNA mutation analysis using an alternative PCR-based method (BEAMing) was requested and results will be presented at the meeting along with detailed patient characteristics, mutation types and treatment outcomes. Conclusions: Detecting V600 BRAF mutations in cfDNA from plasma with the MDx prototype platform is a fast and noninvasive alternative to mutation testing of tumor tissue with an acceptable level of concordance and sensitivity, and should be investigated further for testing and monitoring of BRAF mutation status in patients with cancer. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A202. Citation Format: Helen J. Huang, Bart Claes, Gerald S. Falchook, Apostolia M. Tsimberidou, Veronica R. Holley, Vivek Subbiah, Ralph G. Zinner, Aung Naing, Daniel D. Karp, Sarina A. Piha-Paul, Jennifer J. Wheler, Siqing Fu, David S. Hong, Vanda M. Stepanek, Rajyalakshmi Luthra, Funda Meric-Bernstam, Tania Ivens, Erwin Sablon, Geert Maertens, Bryan K. Kee, Razelle Kurzrock, Stefan Scherer, Filip Janku. BRAF mutation testing in cell-free DNA from plasma of patients with advanced cancers using a novel, rapid, automated molecular diagnostics prototype platform. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A202.

Background: P is a small molecule tyrosine kinase inhibitor that selectively inhibits VEGFR1-3, PDGFR-α, PDGFR-β, c-kit, and FGFR 1-3 and is approved for renal cancer and STS. T is a potent, highly selective, allosteric inhibitor of MEK1/2 approved for BRAF-mutated advanced melanoma. Increased signaling of growth factor receptor tyrosine kinases have been shown to be important to cell survival in multiple tumor types, and their inhibition has been effective therapeutically in STS, DTC, and ChCa. In addition, increased RAS/RAF/MEK/ERK signaling plays a critical role in the development of many different types of cancers, including ChCA and DTC. We hypothesized that combination therapy with P and T would be tolerable and effective for this range of tumor types. Methods: We designed a phase I, open label, dose escalation trial with P+T open to patients with advanced solid tumors. The study was a standard 3+3 design with an expansion cohort of 25 pts each for advanced DTC, STS, and ChCA for correlative endpoints and PK studies. Pts were treated with T held constant at 1 mg QD and P at 400 mg QD (DL1), 600 mg QD (DL2)and 800 mg QD (DL3); T was then escalated to 1.5 mg QD (DL4) and 2 mg QD (DL5). Eligibility included adequate end organ function, PS 0-1, and PD within 6 months in the expansion cohort of DTC. The primary objective was to determine the safety and tolerability of P+T and to find the MTD. Secondary objectives include assessing preliminary efficacy and exploration of PK/PD endpoints with tumor biopsies. Results: We report the results of the dose escalation component of the study. 26 pts were accrued to 5 dose levels and treated for a median of 3 cycles (range Conclusions: P+T is a tolerable combination with some preliminary signs of efficacy. Accrual to expansion cohorts in DTC, STS, and ChCA for further exploration of efficacy and correlative endpoints is ongoing. This study was approved and funded by the National Comprehensive Cancer Network (NCCN) from general research support provided by GlaxoSmithKline. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B279. Citation Format: Nilofer Azad, Douglas Ball, Steven Sherman, Michelle Rudek, Gerald Falchook, Barry Nelkin, Filip Janku, Nicholas Papadopoulos, Ashley O'Conner, Shabina Ahmed, Breelyn Wilky, Susan Markus, Jing Gong, David Cosgrove, Marianna Zahurak, Ralph Zinner, Razelle Kurzrock. A phase I study determining the safety and tolerability of combination therapy with Pazopanib (P), a VEGFR/PDGFR/Raf inhibitor, and GSK1120212 (Trametinib: T), a MEK inhibitor, in advanced solid tumors with expansion cohorts in advanced differentiated thyroid cancer (DTC), cholangiocarcinoma (ChCA), and soft tissue sarcoma (STS). [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B279.

Meeting abstracts To circumvent significant toxicities associated with high dose systemic IL-2 treatment, organ-specific delivery such as inhalational IL-2 has been studied. Our pre-clinical data demonstrated therapeutic efficacy of aerosol IL-2 in a human OS mouse model with no toxicity. The

Background: Cell-free (cf) DNA and circulating tumor cells (CTC) in the blood of cancer patients offer an easily obtainable, low-risk, inexpensive, and repeatedly available source of biologic material for mutation analysis and monitoring of molecular changes throughout cancer therapy. Methods: DNAs from plasma and CTC from patients with advanced cancers who progressed on systemic therapy were tested for BRAF V600 and KRAS G12/G13 mutations using the ICE-COLD-PCR platform. ICE COLD-PCR, "Improved and Complete Enrichment COamplification at Lower Denaturation” selectively amplifies mutant DNA by exploiting differences in denaturation temperatures between mutant DNA duplexes and normal “wild-type” DNA duplexes. KRAS Exon 2 and BRAF Exon 15 ICE COLD-PCR was performed on plasma samples, and from matched CTCs collected using ScreenCell® MB kits (ScreenCell, Sarcelles, France). Amplicons were analyzed by Sanger sequencing methods and results were compared to the mutation status of the archival primary or metastatic tumor tissue as determined in a CLIA-certified lab. Results: Blood samples from 59 patients with advanced cancers (colorectal cancer, n=32; melanoma, n=12; non-small cell lung cancer, n=7; other cancers, n=8), were obtained before treatment and, if possible, sequentially during therapy and tested for BRAF (30), KRAS (29) or BRAF and KRAS (1) mutations. BRAF mutations were detected in 97% (30/31) of archival tumor samples compared to 65% (20/31) of cfDNA samples (agreement 68%) and to 3% (1/31) of CTC samples (agreement 6%). KRAS mutations were detected in 90% (26/29) of archival tumor samples compared to 86% (25/29) of cfDNA samples (agreement 83%) and to 10% (3/29) of CTC samples (agreement 21%); however, CTCs had different KRAS mutation subtypes than those in tumor tissue. Of interest, in 3 patients with serial blood collection for cfDNA BRAF mutations, 2 (melanoma, Erdheim-Chester disease) had BRAF V600E cfDNA mutations at baseline, which disappeared during the response to subsequent therapy. Another patient (melanoma) did not have a BRAF V600E cfDNA mutation immediately after being taken off BRAF inhibitor therapy due to intolerance, but a BRAF V600E cfDNA mutation emerged when the patient was treated with non-BRAF targeting therapy. In 2 patients with serial blood collections for cfDNA KRAS mutations, 2 (colorectal cancer) did not have KRAS cfDNA mutations at baseline, but KRAS cfDNA mutations (G13D, G12D, respectively) emerged following disease progression and subsequent therapy. Conclusions: Detection of BRAF and KRAS mutations in cfDNA can provide a fast and noninvasive alternative to mutation testing in tumor tissue. Although these mutations can be also detected in CTC, the level of concordance with tumor tissue results is low. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C203. Citation Format: Filip Janku, Ben Legendre, Katherine Richardson, Gerald S. Falchook, Aung Naing, Veronica R. Holley, David S. Hong, Ralph G. Zinner, Siqing Fu, Apostolia M. Tsimberidou, Vivek Subbiah, Daniel D. Karp, Sarina A. Piha-Paul, Jennifer J. Wheler, Vanda M. Stepanek, Goran Cabrilo, Rajyalakshmi Luthra, Funda Meric-Bernstam, Amy Kruempel, Jaclyn Pope, Courtney Cubrich, Grant Wu, Yanggu Shi, Marcia Lewis, Razelle Kurzrock. BRAF and KRAS mutation testing in cell-free DNA and circulating tumor cells from blood of patients with metastatic cancers. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C203.

Background: Detection and monitoring of oncogenic mutations in cell-free urinary DNA opens the possibility of a new paradigm for a truly non-invasive method of individualized care for metastatic cancer patients, enabling the quantitation of mutational tumor load and respective concordance to therapeutic responsiveness followed by detection of emerging genomic alterations underlying acquired resistance. Methods: Cell-free DNA was isolated from single and/or multiple sequential urine samples from patients with advanced cancers and BRAF V600E, KRAS G12D or G12V mutations in the tumor tissue from a CLIA-certified laboratory, who progressed on systemic therapy. Assays for quantitative assessment of BRAF V600E, KRAS G12D and G12V mutations in cell-free urinary DNA were developed using droplet digital PCR methodology (RainDance, MA) with enrichment of mutation-containing DNA fragments by pre-amplification of BRAF and KRAS genes. Mutation sensitivity of at least 0.03% was achieved by spike-in experiments of input DNA from cell-lines containing BRAF and KRAS mutations. Healthy controls (N=6) yielded baseline signals that were ∼10-fold less than observed for 0.03% sensitivity. Results: Cell-free DNA was extracted from urine of 25 patients with diverse advanced cancers (colorectal cancer, n=8; melanoma, n=7; non-small cell lung cancer, n=6; papillary thyroid carcinoma, n=2; appendiceal carcinoma, n=1; and glioblastoma, n=1) with BRAF V600E (N=18), KRAS G12D (N=5) and KRAS G12V (N=2) in the tumor tissue. Of 18 patients with BRAF V600E mutations in the tumor, 17 (94%) had the same mutation in urinary cell-free DNA. In addition, all 5 (100%) patients with KRAS mutations (G12D, n=5; G12V, n=2) in the tumor tissue DNA had these same mutations in urinary cell-free DNA.A total of 5 patients with BRAF V600E mutations had longitudinal analysis of percentage of cell-free urinary DNA BRAF V600E mutation to wild-type in sequentially collected urine samples. Although the numbers are small the detected amount of BRAF mutant copies are in agreement with a clinical course. Conclusion: Our preliminary data suggest that detecting BRAF V600E, KRAS G12D, and G12V mutations in cell-free DNA from urine can offer a noninvasive alternative to mutation testing of tumor tissue with excellent concordance, and should be investigated further for testing and monitoring of mutation status in patients with cancer. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B175. Citation Format: Filip Janku, Gerald S. Falchook, Sarina A. Piha-Paul, Aung Naing, Apostolia M. Tsimberidou, Veronica R. Holley, Daniel D. Karp, Ralph G. Zinner, Siqing Fu, Jennifer J. Wheler, David S. Hong, Funda Meric-Bernstam, Vanda M. Stepanek, Rayjalakshmi Luthra, Lorieta Leppin, Latifa Hassaine, Karena Kosco, Jason C. Poole, Mark G. Erlander. Detection and monitoring of BRAF and KRAS mutations in cell-free urinary DNA of metastatic cancer patients by droplet digital PCR. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B175.

Background: Novel, fast, and accurate diagnostic systems are needed for further implementation of personalized therapy. Mutations in the BRAF gene can provide actionable targets for cancer therapy in melanoma and other tumor types. Methods: The molecular diagnostics (MDx) prototype platform (Biocartis, Mechelen, Belgium) is a fully integrated real-time PCR-based system with high sensitivity (1%) and quick turnaround time (< 90 minutes), which requires no sample preparation and Results: Seventy-nine pts (melanoma, n=34; colorectal, n=20; papillary thyroid, n=6; ovarian, n=4; other cancers, n=15) with available tissue and CLIA laboratory BRAF results were identified (BRAF V600 mutation, n=49; wild-type BRAF, n=30). Of the 70 pts for whom the same tissue block was used for MDx and CLIA, BRAF results (V600 mutation yes/no) were concordant in 67 of them (96%; kappa 0.91; 95% CI 0.81-1.01; discrepancies: V600E mutation from CLIA only in prostate cancer, V600E mutation from CLIA only in melanoma, V600E mutation from MDx only in colon cancer). In addition, BRAF results by MDx were discrepant with CLIA in one case (melanoma) in the mutation subtype (V600K vs. V600E). In all 79 pts (including 9 patients with different blocks tested) MDx and CLIA BRAF results (V600 mutation yes/no) were concordant in 75 (95%; kappa 0.89; 95% CI 0.79-1.00) of them (additional discrepancy: V600K mutation from CLIA only in colon cancer). Of 49 pts with BRAF mutations detected by MDx, 30 were treated on protocols (on the basis of the CLIA lab results) with BRAF/MEK inhibitors and 8 (27%) had a partial response. Of interest, 2 of 3 pts with BRAF mutations from CLIA, but not MDx, received a BRAF/MEK inhibitor and did not respond. Detailed patient characteristics, mutation types and discrepancy analysis will be presented. Conclusions: The BRAF V600 mutation MDx prototype platform is a fast (turn-around time about 1.5 hours) and simple ( Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C198. Citation Format: Helen J. Huang, Bart Claes, Gerald S. Falchook, Benoit Devogelaere, Aung Naing, Siqing Fu, Sarina Piha-Paul, David S. Hong, Veronica R. Holley, Apostolia M. Tsimberidou, Vanda M. Stepanek, Kevin B. Kim, Laura S. Angelo, Vivek Subbiah, Jennifer J. Wheler, Ralph G. Zinner, Daniel D. Karp, Rajyalakshmi Luthra, Erwin Sablon, Geert Maertens, Funda Meric-Bernstam, Razelle Kurzrock, Stefan Scherer, Filip Janku. BRAF mutation testing of archival tumor samples with a novel, rapid, fully-automated molecular diagnostics prototype platform. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C198.

263 Background: The PRONOUNCE trial was a phase III, randomized, open-label U.S. study comparing pemetrexed (PemC) with bevacizumab (PCB) in the treatment of advanced chemo-naive nonsquamous non-small cell lung cancer (NS-NSCLC). The primary endpoint of improved PFS without grade 4 toxicity for PemC over PCB was not met. There were no differences in PFS or OS. This analysis was conducted to estimate differences in costs of treatment and related patient care between the groups to inform evidence-based treatment decision making. Methods: A cost-minimization analysis was performed using data from the PRONOUNCE trial and publically available U.S. unit costs. Drug costs modeled included induction, maintenance, and subsequent therapies. Other estimated medical costs were drug administration, concomitant medications, and treatment of grade 3/4 toxicities. A probabilistic sensitivity analysis (PSA) was conducted to model joint uncertainty associated with model parameters. Results: The average cost of treatment with PemC was $4,689 less than PCB ($30,334 vs $35,024). Drug acquisition and administration costs were less in all modeled settings (induction, maintenance, and subsequent lines) for PemC. Costs associated with concomitant medications and grade 3/4 toxicities were similar ($13,286 vs $13,738). PemC was cost saving in 99% of 10,000 PSA iterations. Conclusions: Based on the PRONOUNCE trial and model results, PemC may present a similarly effective but less costly treatment option for patients in the United States compared with PCB. While treatment decision making should consider the overall safety profile and patient preferences, this study provides estimates of the costs of treatment of these two therapeutic options. The analysis was based on clinical trial data; therefore, the results may not be generalizable to all patients. [Table: see text]

Purpose: Histone deacytylase inhibitors are promising targets for therapeutic interventions intended to reverse aberrant epigenetic states associated with cancer. Valproic acid (VPA) is a strong histone deacetylase inhibitor. Lenalidomide (CC-5013, Revlimid) is an immunomodulatory drug with an enhanced potency and pharmacological profile for the regulation of cellular immunity and cytokine response such as TNFa and IL-12. VPA can be additive or synergistic with lenalidomide in inducing cell apoptosis. Objectives of this Phase I study were to evaluate the tolerability and efficacy of VPA and lenalidomide. Experimental Design: Patients in sequential cohorts ("3+3" design) received escalating doses of VPA with lenalidomide. Prophylactic anticoagulation was not required for this protocol. VPA was administered daily for 7 days with 7 days off; and lenalidomide was administered daily for 28 days. At MTD, 10 additional patients with adenocystic carcinoma were enrolled for efficacy. Results: Twenty six patients with advanced cancer (14 M/12 F), median age = 56 (38-70), median number of prior therapies = 2 (0-12) were enrolled. MTD was reached at VPA 30 mg/ kg for 7 days on with 7 days off, and lenalidomide 25 mg daily. Most common toxicities included fatigue, rash, neutropenia and thrombocytopenia. One patient had deep venous thrombosis and two others had pulmonary embolism. DLTs included grade III confusion (n=3), grade III somnolence (n=1) and grade III gait disturbance (n=1). Even though 2/12 patients from dose expansion had DLT during the first cycle of VPA 30 mg /kg for 7 days on and 7 days off with lenalidomide 25 mg PO QD, majority of patients required dose reduction of VPA to 5- 20 mg /kg during the subsequent cycles due to fatigue and drowsiness. Seven patients with adenocystic carcinoma had stable disease over 4 cycles. Of non-adnocystic carcinoma patients, one patient with melanoma and another patient with parathyroid carcinoma had stable disease for 6 cycles and 8 cycles respectively. Conclusions: Lenalidomide 25 mg PO QD combined with VPA (5 -20 mg /kg for 7 days on 7 days off) was well tolerated. No significant tumor reductions were noticed in patients with adenocystic carcinoma. Stable disease was noted in patients in adenocystic carcinoma, melanoma and parathyroid carcinoma. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C69. Citation Format: Mehmet A. Bilen, Siqing Fu, David S. Hong, Jennifer J. Wheler, Maen Abdelrahim, Gerald S. Falchook, Apostlia M. Tsimberidou, Filip Janku, Sarina Piha-Paul, Ralph Zinner, Razelle Kurzrock, Aung Naing. Phase I trial of lenalidomide and valproic acid in patients with advanced cancer. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C69.

Purpose: To evaluate whether PIK3CA mutations and/or PTEN loss predict better clinical outcomes to phase I trials of PI3K/AKT/mTOR inhibition-based therapies in patients with metastatic, recurrent or refractory cervical cancers. Methods: Outcome analyses were conducted on 36 consecutive patients with advanced cervical cancers who were initially seen in a designated Phase I clinic at MD Anderson Cancer Center from 1/1/2006 to 6/30/2012, and tested for PIK3CA mutations and/or PTEN loss in a CLIA-certified molecular diagnostic laboratory, in accordance with the IRB guidelines. All data were obtained from patients’ electronic medical records. Results: All patients (n=36) had received at least one systemic therapy before the referral. Approximately 50% of tested patients (n=18/36) harbored PIK3CA mutations (n=9/34, 26% including seven E545K, one E542K, and one E545K/D549H in squamous cell carcinoma [7/19, 37%], adenocarcinoma [1/12, 8%] and small cell carcinoma [1/1, 100%]) and/or PTEN loss (n=12, 63%). Six patients were removed from further analyses because they did not enroll to a phase I study (n=2) or had only one negative result while the other test was not done (n=4). At their initial phase I therapy, the patients with PIK3CA mutations and/or PTEN loss (n=18) achieved four PR and seven SD ≥ 6 months (61% CR/PR/SD≥6months) with a median progression-free survival (PFS) of 6.7 months, compared to one CR and one SD ≥ 6 months (17% CR/PR/SD≥6months, p=0.026) with a median PFS of 2.5 months (p=0.001) in patients without PIK3CA mutations and/or PTEN loss (n=12). Analyses in patients receiving mTOR inhibitor-based phase I regimens showed that patients with PIK3CA mutations and/or PTEN loss (n=10) achieved three PR and four SD ≥ 6 months (70% CR/PR/SD≥6months) with a median PFS of 6.6 months, compared to one SD ≥ 6 months (17% CR/PR/SD≥6months, p=0.119) with a median PFS of 3.2 months (p=0.014) in patients without PIK3CA mutations and/or PTEN loss who received mTOR inhibitor-based phase I regimens (n=6); or one PR and three SD ≥ 6 months (50% CR/PR/SD≥6months, p=0.63) with a median PFS of 5.3 months (p=0.86) in patients with PIK3CA mutations and/or PTEN loss who did not receive mTOR inhibitor-based phase I regimens (n=8). Conclusions: The most frequent PIK3CA mutation in patients with advanced cervical cancers was E545K. Matched therapy using mTOR inhibitor-based regimens led to dramatic clinical responses, and a significantly longer PFS in patients with PIK3CA mutations and/or PTEN loss. Thus, targeting aberrant PI3K pathway is warranted for further evaluation in patients with advanced cervical cancers. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B191. Citation Format: Ming-Mo Hou, Xiaochun Liu, Sarina A. Piha-Paul, Filip Janku, Jennifer Wheler, Aung Naing, Diane Bodurka, Kathleen Schmeler, Karen Lu, Ralph Zinner, David Hong, Funda Meric-Bernstam, Razelle Kurzrock, Siqing Fu. Target-matched therapies in patients with advanced cervical cancers harboring PIK3CA mutations and/or PTEN loss. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B191.

LBA8003 Background: PemC and PCB are regimens used for first-line treatment of advanced NS-NSCLC. The primary objective was to compare progression-free survival without Grade 4 toxicity (G4PFS) between two vs three drug regimen arms. Methods: Patients ≥18 years, Stage IV NS NSCLC, AJCC (v7.0), and ECOG PS 0/1 were enrolled. Patients were randomized (1:1); received 4 cycles of induction (PemC: Pem, 500 mg/m2 and C, AUC = 6; PCB: P, 200 mg/m2, C, AUC = 6, and B, 15 mg/kg) followed by Pem (PemC Arm) or B (PCB Arm) maintenance therapy in the absence of progressive disease or discontinuation. Secondary endpoints were PFS, overall survival (OS), overall response rate (ORR), and disease control rate (DCR). The study was powered for G4PFS; assuming hazard ratio (HR) of 0.75; there was 80% power to detect superiority of PemC over PCB with a 2-sided type I error of 0.10. Efficacy data were analyzed by intent-to-treat principle using the log-rank test for time-to-event variables, and an exact test for ORR and DCR. Safety data were evaluated using CTCAE v3 for patients who received ≥1 dose of study treatment. Results: Patients were randomized to PemC (N = 182) or PCB (N = 179). Baseline factors were balanced between arms: median age 66/66 years; % female 42/42; % PS=0, 47/47; % stage IV M1a 29/30; for PemC vs PCB, median G4PFS (months) was 3.91/2.86 (HR = 0.85, 90% CI 0.7, 1.04, p = 0.176); PFS and OS had HR = 1.06 (95% CI 0.84, 1.35), p = 0.610, and HR = 1.07 (95% CI 0.83, 1.36), p = 0.616, respectively. The ORR (%) 23.6/ 27.4 and DCR (%) 59.9/57.0 were for PemC vs PCB, respectively. Significantly more drug-related grade 3/4 anemia (18.7% vs 5.4%), and thrombocytopenia (24.0% vs 9.6%) were seen on PemC; significantly more grade 3/4 neutropenia (48.8% vs 24.6%) and grade 1/2 alopecia (28.3 % vs 8.2%) were seen on PCB. Conclusions: PemC was not superior to PCB in G4PFS; no difference in PFS or OS was observed for the two- vs three-drug regimens. There were no unexpected toxicities; the toxicity profiles demonstrated distinctions by arm, and both regimens demonstrated tolerability. Clinical trial information: NCT00948675.

e13575 Background: Mucositis is a well-known side effect of mTOR inhibitor (temsirolimus) treatment of cancer patients, limiting the use of mTOR inhibitors. We investigated whether combination treatment of temsirolimus with another agent that was not known to cause mucositis would result in a higher or lower incidence of mucositis. We further investigated if there was an association between the severity of mucositis and tumor response to the temsirolimus-based combination treatment. Methods: This retrospective data review was approved by IRB and conducted in the department of Investigational Cancer Therapeutics at MD Anderson Cancer Center. We reviewed the electronic medical records of 87 patients who were enrolled in temsirolimus-based combination phase I trials (n=3), including physician notes from the date they enrolled on the study to the date they came off treatments. We first examined the incidence of mucositis during the course of study and its severity. We assessed patient response to the study treatment with severity of the mucositis. Results: The overall incidence of mucositis in the 3 temsirolimus-based combination trials was 46%, similar to the published data for the single agent of 41%. The grades of mucositis were: for grade 1 (n=14), grade 2 (n=17) and grade 3 (n=9). No grade 4 mucositis was observed. Neither gender nor ethnicity altered the incidence of mucositis. The median onset time was 14 days after the start of the studies. Of 87 patients, 20 responded to the study treatments with at least stable disease for longer than 6 months or a partial or complete response. The incidence of the mucositis was not associated with the response to temsirolimus-based treatment. Severity of mucositis did not correlate with response to the temsirolimus-based regimen. (grades 1/2 vs. ≥ grade 3; p=0.512). Conclusions: The incidence of mucositis did not increase significantly in patients who received a temsirolimus-based regimen compared to single-agent temsirolimus. Severity of mucositis did not correlate with tumor response to the temsirolimus-based regimen. Future study is warranted to optimize management for mTOR inhibitor-induced mucositis.

1051 Background: Matching molecular alterations in patients with breast cancer to targeted therapy may increase response rates. Methods: We analyzed 22 sequential breast cancer patients with next generation sequencing (NGS) profiling of their tumors profiling (FoundationOne) treated on Phase I trials at the Clinical Center for Targeted Therapy at MD Anderson Cancer Center. Objectives included: (1) characterize molecular alterations including mutations, amplifications and deletions, (2) evaluate associations between molecular alterations and response to therapy. Results: Twenty-two breast cancer patients, all female with median age 56 years were included. Twenty-one of 22 of patients (95%) analyzed demonstrated at least one molecular alteration. Twenty-one of 22 patients were evaluable for response to therapy (1 patient had not yet reached restaging). Sixteen patients were treated on Phase I trials with targeted therapy that was either directly or indirectly associated with a molecular alteration. Seven of these 16 patients (44%) achieved stable disease (SD) ≥6 months/partial response (PR)/complete response (CR) including 3 patients with PR and 1 patient with CR. Examples of molecular alterations include: mutations in PIK3CA (8 patients), PIK3R1 (2 patients), PTEN (1 patient), cMET (1 patient), NF1 (one patient); amplifications in CCND1 (8 patients), FGFR1 (4 patients), MYC (3 patients), MCL1 (3 patients), IRS2 (2 patients) and CCNE1(1 patient). Conclusions: A majority of patients with advanced or metastatic breast cancer whose tumors underwent NGS profiling demonstrated molecular alterations. Of the 9 patients who had either SD≥6 months/PR/CR on Phase I treatment, 7 (78%) had molecular alterations, either directly or indirectly associated with therapy. Further investigation of larger cohorts of patients with NGS is ongoing.

2611 Background: Pathway crosstalk and emergence of drug resistance have limited the activity of targeted therapies, including anti-angiogenic agents and mTORC1 inhibitors. However, this vulnerability may be addressed by rationally combining targeted therapies to improve outcomes. We sought to determine the MTD/recommended phase II dose and define the pharmacokinetics (PK) of the combination of sorafenib (S), bevacizumab (B), and temsirolimus (T) in patients with advanced solid tumors. Methods: S was given once or twice daily, B was given intravenously every three weeks, and T was given intravenously weekly. Doses were escalated in a stair-step dose escalation with 6 planned dose levels (DL) in a standard 3+3 design. Responses were defined using RECIST 1.1. Results: To date, 51 patients have been enrolled. Three patients withdrew consent after only one dose. One dose-limiting toxicity (DLT) occurred at DL 5 (grade 4 fatigue, pain) and one DLT occurred at DL 6 (G3 headache). However, 5/6 patients required dose reduction on DL6 (G2 nausea, fatigue, hand/foot syndrome, rash). Thus, DL5 (B 10mg/kg, T 20mg, S 200mg twice daily) is being explored in dose expansion (N=10) and tumor expansion cohorts as the recommended phase II dose. The most common adverse events (>10%) were fatigue (69%; G3/4 6%), nausea (38%, G3/4 2%), pain (35%, G3/4 10%), thrombocytopenia (33%, G3/4 8%), mucositis (33%, 0%), constipation (29%, G3/4 0%), hand/foot syndrome (23%, G3/4 0%), rash (19%, G3/4 0%), hypertension (17%, 0%), neuropathy (17%, G3/4 0%), diarrhea (15%, G3/4 6%), vomiting (15%, G3/4 2%), headache (12%, G3/4 2%), and hypertriglyceridemia (10%, G3/4 2%). Of 34 patients evaluable for response, 4 had PR including ovarian (-57%, +6m), gastric (-46%, +5m), colorectal (-32%, +5 m), and thyroid (-31%, +6m) cancer. Eight patients had SD for greater than 4 months including ovarian (0%, +7m), colorectal (-5%, +8m), endometrial (-26%, +5m; -16%, +4m; -15%, +6m), leiomyosarcoma (-5%, +6m), squamous lung cancer (-6%, +6m), and triple negative breast (-15%, +4m) cancer. Conclusions: The combination of S, B, and T is well tolerated and demonstrates preliminary evidence of tumor activity in a variety of solid tumors. PK studies at the recommended phase II dose are ongoing. Clinical trial information: NCT01187199.

2604 Background: Lenalidomide (immunomodulatory and antiangiogenic drug) and temsirolimus (mTOR inhibitor) are potentially synergistic (Raje N, Blood 2004;104(13):4188). We conducted a phase I study of the combination in patients with advanced cancers. Methods: During the escalation phase, lenalidomide (PO days 1-21) and temsirolimus (IV, Q weekly) were given at the following respective doses: level 1 (10 mg, 15mg); level 2 (10 mg, 20 mg); level 3 (20 mg, 20 mg); and level 4 (20 mg, 25 mg) (1 cycle = 28 days). A “3+3” study design was used. The maximum tolerated dose, dose-limiting toxicity, and response were assessed. Results: Forty- three patients were treated (median age: 58 (21-80) years; male/female: 26:17). The most common diagnoses were colorectal cancer (N=5), sarcoma (N=5) and adenoid cystic carcinoma (N=4). Overall, 121 cycles (median: 2) were administered. No dose limiting toxicities were observed. The maximum tested dose (level 4) was used for expansion. Grade 3-4 toxicity (all reversible) occurred in 19 (44%) patients: neutropenia (N=12), thrombocytopenia (N=6), and infection (N=1). Of 43 patients, 31 (72%) received anticoagulation prophylaxis. There were no thrombotic events. Of 36 patients who were evaluable for response, PR was noted in 1 (2.7%), and stable disease (SD) was noted in 17 (47%) patients (SD ≥6 months: 17%). Tumors with SD≥6 months were soft tissue sarcoma, 2/5 (40%), adenoid cystic carcinoma, 1/4 (25%), parotid ca, 1/2 (50%), adrenocortical ca 1/3 (33%), and neuroendocrine ca, 1/4 (25%) (Table). The median progression-free survival was 2.2 months (95% CI, 1.5-2.9) and overall survival was 7.8 months (95% CI, 5.1-10.6). Conclusions: Lenalidomide and temsirolimus combination was well tolerated and had antitumor activity in selected participants with advanced cancer. Clinical trial information: NCT01183663. [Table: see text]

LBA8003 The full, final text of this abstract will be available at abstract.asco.org at 7:30 AM (EDT) on Monday, June 3, 2013, and in the Annual Meeting Proceedings online supplement to the June 20, 2013, issue of Journal of Clinical Oncology. Onsite at the Meeting, this abstract will be printed in the Monday edition of ASCO Daily News.

e22086 Background: The prognostic significance of KRAS mutations varies by mutation type. We compared the outcomes of patients with different KRASmutations treated in phase I clinical trials. Methods: Patients with advanced cancer and KRAS mutation who were treated in the phase I program from 9/08 to 12/12 were analyzed. The presence and type of KRAS mutation in the patients’ tumor samples were assessed in a CLIA-certified laboratory using PCR, sequenom analysis or next-generation sequencing. We compared clinical outcomes according to type of mutation, tumor, and therapy. Results: KRAS mutations were identified in 365 patients (codon 12, n=278; codon 13, n = 49; other codon, n = 38). The median age was 59 yrs (range, 18-87); and 46% men. Of 365 patients, 260 had KRAS mutation alone; 105 patients had 130 additional aberrations: PI3KCA, n=45 (39%); p53, n=29 (25.2%); PTEN, n=22 (19%); other, n=34 (29%). Of 260 patients, 159 were treated and evaluable for response. Outcomes in patients with KRAS mutation alone or with KRAS and additional molecular aberrations are shown in the Table. Conclusions: Keeping in mind that the number of patients was small, our results demonstrate heterogeneity in outcomes of patients with KRAS mutations. Combination studies of MEK inhibitors with other drugs may have improved antitumor activity. [Table: see text]

11086 Background: Mutations in the BRAF gene provide actionable targets for cancer therapy in melanoma and other tumor types. Novel, fast, and accurate diagnostic systems are needed for further implementation of personalized therapy. Methods: The molecular diagnostics (MDx) prototype platform (Biocartis, Mechelen, Belgium) is a fully integrated real-time PCR-based system with high sensitivity (1% mutant in wild-type [wt] background) and fast turnaround time (< 90 minutes), which requires no sample preparation and

e13534 Background: Patients with metastatic cervical cancers have a poor prognosis and few effective therapies available. Thus, access to a phase I trial may provide a viable option. Methods: Outcomes of 54 consecutive patients with metastatic cervical cancers seen in a phase I clinic at MD Anderson Cancer Center from 1/1/2006 to 12/31/2012 were reviewed in accordance with the MD Anderson IRB guidelines. Results: The 54 patients included Caucasians (n=34, 63%), Hispanics (n=9, 17%), African Americans (n=7, 13%), and others (n=4, 7%). Pathological diagnoses were squamous cell carcinoma (n=33, 60%), adenocarcinoma (n=12, 22%), adenosquamous (n=3, 6%), poorly differentiated carcinoma (n=2, 4%), small cell carcinoma (n=2, 4%), clear cell carcinoma (n=1, 2%), and melanoma (n=1, 2%). Initial staging was as follows: I (n=16, 30%), II (n=11, 20%), III (n=11, 20%), and IV (n=16, 30%). Prior treatment included surgery (n=23, 43%), radiotherapy (n=41, 76%), and systemic chemotherapy (n=54, 100%) including one regimen (n=30, 55%), two (n=15, 28%) and more (n=9, 17%). At their initial visit to the phase I clinic, the median age was 47 years. The majority of patients had ECOG PS 0-1 (n=45, 83%), and MDACC prognostic scores 0-1 (n=48, 89%). Almost all the 54 referred patients were enrolled into a phase I trial (n=52, 96%). Initial phase I trials these patients had received led to one CR, five PR, and 12 SD ≥ 6 months (CR/PR/SD≥6 months=35%). Of the patients who progressed on their first protocols, 11 patients (21%) were enrolled into a second protocol, leading to 36% of PR/SD≥6 months (n=4). Among those enrolled into a third (n=4) and a fourth (n=1) study, no CR/PR/SD≥6 months was achieved. The median overall survival in these 54 referred patients was 10.6 months. Among 36 patients who underwent molecular marker studies, 15 were found to harbor PIK3CA mutations and/or PTEN loss (42%) and achieved 60% of CR/PR/SD≥6 months (3 PR and 6 SD≥6 months). Conclusions: Aberrant PI3K pathway was frequently identified in advanced cervical cancers, and the high rate of CR/PR/SD≥6 months was observed in these patients treated under a phase I trial, suggesting that phase I trials, especially mutation matched regimens, may provide potential benefits to these patients.

Background: Gene amplifications in breast cancer may indicate poor prognosis disease, particularly for patients with multiple amplifications. Endocrine response and resistance in breast cancer maybe revealed to PI3K/AKT/mTOR pathway activation. Patients and Methods: We analyzed tissue with next generation sequencing (NGS) from 3 patients with estrogen receptor (ER)-positive breast cancer treated on a dose escalation study with anastrozole, an aromatase inhibitor, and everolimus, an mTOR inhibitor. Genomic libraries were captured for 3230 exons in 182 cancer related genes plus 37 introns from 14 genes, often rearranged in cancer (Foundation Medicine, Cambridge, MA). Results: Twelve of 50 evaluable patients with ER-positive breast and gynecologic tumors treated with anastrozole (1mg PO daily) and everolimus (either 5mg PO daily or 10mg PO daily) had SD ≥ 6 months/PR/CR. NGS analysis on 3 of the 12 patients tumors demonstrated multiple amplifications, including: CCND1, CCNE1, FGFR1, MYC, IRS2 and MCL1. One patient had a complete response (CR) and two patients achieved partial responses (PR) with an 80% decrease in disease in one patient and a 38% decrease in disease in the other. All 3 patients had ER-positive breast cancer and previously experienced progression of disease while receiving hormone therapy. These amplifications seen have all been linked to hormone therapy resistance in ER-positive breast cancers. Two of these patients also had direct alterations in the PI3K/AKT/mTOR pathway: PIK3CA and PIK3R1 mutations in one patient and PTEN loss in another. Conclusions: Molecular profiling of 3 patients with ER-positive breast cancer with PR/CR on treatment with anastrozole/everolimus revealed multiple genetic alterations. Among these alterations were amplifications that may be associated with resistance to hormone therapies. Further exploration of combination hormonal therapies and correlative molecular profiling is warranted. Citation Format: Ralph Zinner, Johnique T. Atkins, Filip Janku, Stacy Moulder, Phil Stephens, Roman Yelensky, Robert Wolff, Razelle Kurzrock, Jennifer J. Wheler. Next generation sequencing demonstrates multiple gene amplifications and mutations in 3 patients with estrogen receptor-positive breast cancer with responses to treatment with combination aromatase and PI3K/AKT/mTOR pathway inhibition. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1209. doi:10.1158/1538-7445.AM2013-1209

Background: BAY 94-9343 (BAY, Mesothelin-ADC) is a novel ADC consisting of a fully human monoclonal antibody directed against mesothelin, conjugated to a maytansine derivative DM4. Mesothelin is an internalizing antigen overexpressed by a number of solid tumors, including a majority of mesotheliomas and pancreatic and ovarian adenocarcinomas. BAY demonstrates mesothelin-specific antitumor activity in relevant xenografts. Objectives of this study were evaluation of safety, PK, immunogenicity, biomarkers, antitumor activity, and determination of Maximum Tolerated Dose (MTD) of BAY in patients (pts) with advanced solid tumors. Methods: BAY doses ranged from 0.15 to 5.5 mg/kg IV every 21 days. PK samples were collected in Cycles (C) 1 and 3; biomarker and immunogenicity samples were also collected. Tumor assessments were planned every 6 weeks. Results: 32 pts (16 mesothelioma, 9 pancreatic, 3 ovarian and 4 other) received BAY. No dose limiting toxicities (DLT) have been observed up to 5.5 mg/kg. One pt had possibly study drug related SAE of Grade 3 chest pain in C2. There were no other related Grade 3/4 AEs. One pt had a Grade 2 infusion reaction and one pt had possibly related Grade 2 pericardial effusion (PE). Two other pts had PE not related to BAY but to underlying cancer. The most common possibly related Grade 1/2 AEs were fatigue (10 pts), nausea (7), anorexia (4), vomiting (4), and weakness (3). PK of ADC and Total Antibody increased in a generally dose proportional manner; half-life was 4 to 5 days. DM4 Cmax increased in a dose proportional manner and DM4 metabolite exposure increased in a more than dose proportional manner; half-lives were approx. 2 and 6 days, respectively. Anti-BAY antibodies were observed in 3 pts at dose levels ≤ 2.4 mg/kg which was associated with decreased ADC exposure in C3. Exploratory analysis of cytokeratins (CK18 M30 and M65) and serum mesothelin as biomarkers will be discussed. One mesothelioma pt in 5.5 mg/kg cohort has confirmed partial response and is currently ongoing. Durable stable disease has been observed in 2 mesothelioma pts (6 and 10 cycles, respectively). Conclusions: BAY has been tolerated at doses of up to 5.5 mg/kg and dose escalation continues. There are no prohibitive findings with regards to PK and antidrug antibodies at higher dose levels. Initial signals of clinical activity are seen. Updated results will be presented. Citation Format: Johanna Bendell, George Blumenschein, Ralph Zinner, David Hong, Suzanne Jones, Jeffrey Infante, Howard Burris, Prabhu Rajagopalan, Martin Kornacker, David Henderson, Andrea Kelly, Raffit Hassan. First-in-human phase I dose escalation study of a novel anti-mesothelin antibody drug conjugate (ADC), BAY 94-9343, in patients with advanced solid tumors. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr LB-291. doi:10.1158/1538-7445.AM2013-LB-291

Background Combining aromatase inhibitors with PI3K/AKT/mTOR inhibitors is synergistic and may overcome hormone resistance. Patients and Methods We evaluated anastrozole and everolimus in patients with metastatic estrogen receptor (ER) and/or progesterone receptor (PR)-positive breast and gynecologic tumors. Endpoints were safety, antitumor activity and molecular correlates. Results Fifty-five patients were enrolled. Full doses of anastrozole (1 mg PO daily) and everolimus (10 mg PO daily) were well tolerated. Fifty patients were evaluable for response with a median age of 57 (range: 24-82) and a median number of prior therapies of 3 (range: 0-13). Twelve of 50 patients (24%) achieved stable disease (SD) ≥ 6 months/partial response (PR)/complete response (CR), including 9 of 32 patients (28%) with breast cancer, 2 of 9 patients (22%) with ovarian cancer and 1 of 6 patients (17%) with uterine cancer. Six of 8 patients (75%) with SD ≥ 6 months/PR/CR with molecular testing demonstrated at least one alteration in the PI3K/AKT/mTOR pathway including mutations in PIK3CA (n=3) and AKT1 (n=1) or PTEN loss (n=3). All 3 patients who had next generation sequencing demonstrated additional alterations: amplifications in CCNE1, IRS2, MCL1, CCND1, FGFR1 and MYC and a rearrangement in PRKDC. Six of 22 patients (27%) with molecular alterations in the PI3K/AKT/mTOR pathway achieved SD ≥ 6 months/PR/CR. Conclusion Combination anastrozole and everolimus is well tolerated at full approved doses and active in heavily pretreated patients with ER and/or PR-positive breast, ovarian and endometrial cancers. Patients with SD ≥ 6 months/PR/CR frequently had alterations in the PI3K/AKT/mTOR pathway. Citation Format: Jennifer J. Wheler, Stacey L. Moulder, Aung Naing, Filip Janku, Sarina A. Piha-Paul, Gerald S. Falchook, Ralph Zinner, Apostolia M. Tsimberidou, Siqing Fu, David S. Hong, Johnique T. Atkins, Maureen Cronin, Roman Yelensky, Phil Stephens. A dose-escalation study of anastrozole and everolimus in patients with advanced gynecologic and breast malignancies: tolerance, biological activity, and molecular alterations in the PI3K/AKT/mTOR pathway. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3514. doi:10.1158/1538-7445.AM2013-3514

Background The cell surface receptor tyrosine kinase c-Met and its ligand, the hepatocyte growth factor (HGF), are implicated in tumor cell migration, invasion, survival and proliferation. EMD 1214063 is a novel potent and highly selective reversible, ATP-competitive small molecule c-Met inhibitor. Methods This is a phase I, first in human (FIH) clinical trial with escalating doses of EMD 121406(NCT 01014936). The primary objective is to determine the MTD. Secondary objectives include evaluation of safety, pharmacokinetics, anti-tumor effect and pharmacodynamics (Pd). Eligible patients had advanced solid tumors not amenable to standard therapies. Following a classical 3 + 3 dose escalation scheme, successive cohorts of patients were treated with once daily oral EMD 1214063 according to two 21-day-cycle schedules, either days 1-14 followed by a 7-day rest(regimen 1, R1), or continuous three times weekly administration (regimen 2, R2). Pd markers were evaluated in paired tumor biopsies. Results As of 30 March 2011, a total of 41 patients had been enrolled, 21 in R1 and 20 in R2. The dose was escalated from 30 mg/day to 115 mg/day in R2 and to 230 mg/day in R1. One DLT was reported in R1 at 115 mg/day, an asymptomatic, grade 4 lipase and G3 amylase elevation. No other DLTs or treatment-related SAEs were observed. The remaining treatment-related AEs of grade 2 or higher included nausea (n = 1), vomiting (n = 1), anorexia (n = 1), diarrhea (n = 1), and fatigue (n = 1) in R1, and neutropenia (n = 1) in R2. 37 patients (90%) had no drug-related toxicity greater than grade 1. At the dose levels investigated, median Cmax and AUC values increased with dose. Immunohistochemical analysis of a patient with pre- and on-treatment biopsies showed a decrease in phospho-c-Met staining intensity under treatment. Preliminary anti-tumor activity has been observed, including an unconfirmed PR in one patient and stable disease lasting for at least 4 months in 5 patients. One patient with sarcomatoid bladder carcinoma and multiple MET copies due to polysomy of chromosome 7 achieved SD for 16+ months. Conclusions The MTD has not yet been reached and dose escalation of EMD 1214063 continues. Updated results of this FIH study will be presented.

Background Adjuvant chemotherapy provides a small, but significant, survival advantage for patients (pts) with completely resected stage II and III non-small-cell lung cancer (NSCLC). Prior data suggest that benefit from platinum-based adjuvant therapy is limited to pts with low levels of ERCC1 (E1) expression. RRM1 (R1) is the molecular target and key efficacy determinant for gemcitabine. We conducted a prospective clinical trial (SWOG-0720, NCT00792701) to determine if treatment selection based on in situ tumoral levels of E1 and R1 is feasible in a mixed environment of academic and community practices. Methods Eligibility was confined to pts with a complete resection of stage I disease and a tumor diameter of 2 cm or larger. Biomarker levels were determined at the protein level using automated quantitative analysis (AQUA, range of expression 1 - 255), and classified as high or low based on a priori cut-off values (E1 >65.0; R1 >40.0). Treatment consisted of four three-weekly cycles of cisplatin/gemcitabine (80 mg/m2 on day 1 and 1 g/m2 on days 1 & 8) for those with low E1 and/or R1 levels. Pts with high levels of both markers were observed. Feasibility was defined as treatment assignment within 12 weeks from surgery in at least 75% of eligible pts. Results 85 pts accrued between March 2009 and April 2011. Two pts were ineligible because of inadequate mediastinal lymph node sampling. E1 and R1 levels were successfully determined in all 83 eligible pts, and 72 pts (87%) were successfully assigned within the allotted time period. 64 pts (77%) were assigned to chemotherapy and 19 (23%) to observation. E1 levels ranged from 4.3 – 211.2 (median 44.7), and R1 levels ranged from 2.5 – 234.4 (median 39.3). Pts demographics: median age 64 y, age range 41-84 y, ♂/♀ = 48/52%, adeno/squam/other = 59/28/13, stage IA/IB = 38/62%, PS 0/1 = 52/48%. Conclusions Adjuvant treatment assignment is feasible in pts with completely resected NSCLC in the cooperative group environment, and treatment assignment to chemotherapy vs observation was not significantly different from prior experience. Support: NCI grants U10-CA014028, U10-CA032102, U10-CA038926, U10-CA046368, U10-CA046441, U10-CA105409, P30-CA022453, R01-CA129343 Disclosure All authors have declared no conflicts of interest.

10607 Background: p53 mutation is one of the most common mutations in human tumors and plays an important role in apoptosis, genomic stability and inhibition of angiogenesis. Methods: We retrospectively reviewed 121 consecutive patients (pts) with solid tumors referred to the Clinical Center for Targeted Therapy (Phase I program) who were tested for p53 mutation, and compared clinical characteristics and outcomes in p53 positive vs. wild-type (wt) pts. Results: Of the 121 tested pts, 65 (53.7%) were p53 mutation-positive. The presence of p53 mutation was not associated with sex and race. Tumors with p53 mutation metastasized more often to the liver (44 pts (67.7%) with mutated p53 vs. 26 pts (46.4%) with wt p53, p=0.0264); there was no difference in metastasis to the bones, lungs, brain, soft tissues and adrenals. p53 mutation also showed a trend toward an association with PTEN-loss (13 out of 43 (30%) with p53 mutated vs. 4 out of 37 (11%) with wt p53, p=0.053). Among pts with the p53 mutation, the best median progression-free survival (PFS) prior to phase I trial entry was 10.97 months (range 0.85–29.04 months) when the treatment regimen included bevacizumab and 4.37 months (range 0.92–14.98 months) when the treatment did not include bevacizumab (P=0.0046). Among patients with wt p53, there was no significant difference in PFS between bavacizumab-containing regimens and regimens without bevacizumab. The median OS from diagnosis was 7.7 years (95% CI 6.32–9.84 years) vs. 11.8 years (95% CI 10.37 years, not attained) for p53 mutant vs. wild-type disease (p = 0.03). Univariate analysis for OS from diagnosis showed that mutated p53, Hispanic race (vs. Caucasians) and the shorter time from diagnosis to metastasis were associated with a worse prognosis. Conclusions: Tumors with p53 mutation have a more aggressive clinical behavior, metastasize more to the liver and may have a higher rate of PTEN loss compared to tumors with wt p53; In addition, anti-angiogenic agents may be of therapeutic value in p53-mutated pts.