Outcome prioritization and preferences among older adults with cancer starting chemotherapy in a randomized clinical trial

INTRODUCTION

Older adults with cancer facing competing treatments must prioritize between various outcomes. This study assessed health outcome prioritization among older adults with cancer starting chemotherapy.

METHODS

Secondary analysis of a randomized trial addressing vulnerabilities in older adults with cancer. Patients completed three validated outcome prioritization tools: 1) Health Outcomes Tool: prioritizes outcomes (survival, independence, symptoms) using a visual analog scale; 2) Now vs. Later Tool: rates the importance of quality of life at three times-today versus 1 or 5 years in the future; and 3) Attitude Scale: rates agreement with outcome-related statements. The authors measured the proportion of patients prioritizing various outcomes and evaluated their characteristics.

RESULTS

A total of 219 patients (median [range] age 71 [65-88], 68% with metastatic disease) were included. On the Health Outcomes Tool, 60.7% prioritized survival over other outcomes. Having localized disease was associated with choosing survival as top priority. On the Now vs. Later Tool, 50% gave equal importance to current versus future quality of life. On the Attitude Scale, 53.4% disagreed with the statement "the most important thing to me is living as long as I can, no matter what my quality of life is"; and 82.2% agreed with the statement "it is more important to me to maintain my thinking ability than to live as long as possible".

CONCLUSION

Although survival was the top priority for most participants, some older individuals with cancer prioritize other outcomes, such as cognition and function. Clinicians should elicit patient-defined priorities and include them in decision-making.

RX‐3117, an oral hypomethylating agent to treat advanced solid tumors: Interim results from an ongoing phase 2a study in advanced urothelial cancer (aUC)

501

Background: RX-3117 is an oral small molecule hypomethylating agent, cyclopentyl pyrimidyl nucleoside that is activated by uridine cytidine kinase 2. RX-3117 shows efficacy in various xenograft models, including those of gemcitabine resistant bladder cancer. Preliminary data from Stage 2 of a Phase 2a clinical study of RX‐3117 as a single agent in subjects with aUC is described below. Methods: This Phase 2a study (2‐stage design, NCT02030067) evaluates the efficacy of RX‐3117 in eligible patients (aged ≥ 18 years) with refractory aUC. Primary objectives include safety and efficacy of the recommended Phase 2 dose and schedule identified in the Phase 1 portion of the study. Patients received 700 mg of oral RX‐3117 daily for either 3 weeks with 1 week off in each 4-week cycle or 4 continuous weeks. The primary endpoint is a ≥ 20% rate of progression free survival benefit (i.e., proportion of patients with stable disease for at least 4 months) and/or a 10% of evaluable patients with a partial response or better. Results: As of October 2017, 17 patients (12 males, 5 females) with aUC were treated with RX‐3117. The median age was 66 years, ECOG performance status was 0 to 1 and 53% received ≥ 3 prior therapies. Metastatic disease sites included lung, liver, lymph nodes, and mediastinum. Four patients achieved stable disease for 4 cycles of RX‐3117 treatment; one patient received treatment for 168 days and another patient for 301 days. One patient showed tumor shrinkage as measured by RECIST (‐15.5%) after 4 cycles of RX-3117; another patient showed a 19% tumor reduction after 1 cycle. The most frequent related adverse events were G1 diarrhea (13%), fatigue (13%), nausea (10%), G1/G2 anemia (10%), vomiting (10%) and G3 thrombocytopenia (10%). Conclusions: RX‐3117 is safe and well tolerable and shows preliminary evidence of anti-tumor activity. The study continues to enroll patients with aUC in Stage 2. Clinical trial information: NCT02030067.

A phase 1 study of p53MVA vaccine in combination with pembrolizumab

206

Background: Mutations in the p53 gene are present in a majority of solid tumors, resulting in the accumulation of oncogenic and potentially immunogenic p53 protein within tumor cells. We have developed a genetically engineered Modified Vaccinia Ankara (MVA) virus to express wild type p53 transgene (p53MVA) as an immunotherapeutic strategy. We observed robust p53-specific CD8+ T cell responses which were further enhanced by anti-PD-1 treatment. This phase 1 study evaluates the safety and tolerability of the p53MVA vaccine in combination with pembrolizumab. Methods: Patients with non-small cell lung, head and neck squamous cell (HNSCC), hepatocellular (HCC), renal cell, melanoma, bladder, soft tissue sarcoma, triple-negative breast (TNBC), pancreatic (PDAC) and MSI high colorectal cancer failing or intolerant to standard treatment were eligible. Confirmation of p53 involvement by immunohistochemistry or mutational analysis was required for eligibility. A 3-at-risk rolling design was employed and patients received 5.6 x 108 pfu p53MVA for 3 doses in combination with 200 mg pembrolizumab for 7 doses every 3 weeks. Blood draws for immunological monitoring were obtained pre-treatment and up to week 19. Results: Two patients (TNBC and HNSCC) showed clinical benefit associated with durable p53-specific CD8+ T cell responses with gene expression pathway scores of T cell function and associated immune response categories elevated for more than 6 months. The TNBC patient had complete regression of cutaneous metastases and stable disease for > 6 months. Two patients (TNBC and PDAC) were on study for 35 and 9 weeks respectively; however, p53-specific T cells were not generated. Four patients (HCC, PDAC and 2 with TNBC) had rapidly progressive disease. Conclusions: Targeting p53 antigen associated with cancer using a viral vaccine p53MVA in combination with PD-1 blockade represents a novel immunotherapeutic approach capable of stimulating systemic immune responses and associated clinical benefit. Clinical trial information: NCT02432963.

Multiomic molecular comparison of primary versus metastatic pancreatic tumors

213

Background: Pancreatic cancer metastasizes very early, as evidenced by the fact that > 70% of patients with operable disease ultimately develop metastases. Thus, it is likely that the molecular characteristics of primary pancreatic tumors are similar to metastatic lesions. We compared the frequency of genetic alterations and protein expression from primary vs. metastatic pancreatic tumors, and from metastases from different sites. By focusing on actionable genetic and proteomic information, we sought to explore whether targeted therapies could be tailored to patients at metastatic progression based on primary surgical material. Methods: Next generation DNA sequencing (NGS) data of 208 genes and a limited set of protein markers were analyzed from pancreatic tumors of 431 patients enrolled in the Know Your Tumor initiative. Of the 208 genes tested, mutations in 70 were considered potentially actionable based on preclinical and clinical evidence. We compared 146 primary pancreatic tumors against 285 metastatic lesions, and examined subgroups for liver vs. lung vs. other metastatic lesions. Molecular alterations were compared between independent groups for each gene/protein using Fisher’s exact test. Significance was assessed using a false discovery rate adjusted q-value threshold of 0.05. Results: No differences in the specific mutation or expression pattern were observed between primary vs. metastatic lesions, nor across the site of metastasis after correcting for multiple hypotheses. Even the proportion of actionable alterations (including mutations in the homologous recombination DNA repair pathway) was similar across subgroups. Conclusions: Comparison of the muli-omic profile of primary vs. metastatic pancreatic adenocarcinoma reveals that the molecular architecture is very similar, and that actionable alterations are identified at the same frequency. This is unlike the data observed from other solid tumors, (e.g. colon and breast cancer), in which substantial molecular discordance and heterogeneity exists between primary tumors and metastatic sites, but is consistent with the belief that primary pancreatic cancers metastasize early and thus are molecularly indistinguishable from metastatic lesions.

A pilot study of vinorelbine safety and pharmacokinetics (PKs) in patients (pts) with varying degrees of liver dysfunction (LD)

432

Background: Prospective data to support dosing guidelines for intravenous (IV) vinorelbine in pts with LD are limited. One study recommends standard 30 mg/m2 weekly for total bilirubin ≤2 mg/dL and a 50% dose reduction for bilirubin > 2 mg/dL. Another study recommends no dose modifications for bilirubin up to 3X the upper limit of normal (ULN). This phase I study describes the safety and PKs of vinorelbine in pts with varying degrees of hepatic impairment. Methods: Pts with treatment-refractory solid tumors received weekly IV 30 mg/m2 vinorelbine for normal liver function, 20 mg/m2 for mild LD defined as bilirubin < 1.5 mg/dL but aspartate aminotransferase (AST)/alanine aminotransferase (ALT) 1.5-2.5X ULN or alkaline phosphatase (ALK) 1.5-3X ULN, 15 mg/m2 or 30 mg/m2 for moderate LD defined as bilirubin 1.5-3.0 mg/dL and/or AST/ALT > 2.5 or ALK > 3X ULN, and 7.5 mg/m2 or 20 mg/m2 for severe LD defined as bilirubin > 3.0 mg/dL. Vinorelbine PKs were evaluated to describe its relationship with liver function. Results: 47 patients were enrolled (61.7% gastrointestinal and hepatobiliary cancers) and a total of 248 treatment-related adverse events (AEs) occurred. All-grade myelosuppression was the most common AE overall (41.1%). Out of 71 grade ≥3 AEs, 17 (23.9%) were grade 4 in severity with the majority (15/17) being myelosuppression. Rates of grade 4 hematologic AEs in the normal liver function and the 7.5 mg/m2 severe LD group were low ( < 10%). 4/17 grade 4 hematologic AEs were observed in the mild LD group while the remainder and majority occurred in the moderate-severe LD groups. Vinorelbine PKs were not correlated with the degree of LD, however free drug levels or levels of the active metabolite 4-O-deacetylvinorelbine were not measured. Conclusions: Weekly vinorelbine at 30 mg/m2 and 7.5 mg/m2 appears safe with normal liver function and severe LD, respectively. High rates of grade 4 myelosuppression with 15-30 mg/m2 vinorelbine in moderate-severe LD raise concerns for its safety in this population. Vinorelbine PKs are not affected by LD. However, it is possible that levels of the active metabolite could be higher in pts with LD if its elimination is impacted by LD to a greater extent than the parent drug. Clinical trial information: NCT00540982.

RX 3117: Activity of an oral antimetabolite nucleoside in subjects with pancreatic cancer–Preliminary results of stage II of the phase Ia/IIb study

396

Background: RX-3117 is an oral small-molecule antimetabolite, cyclopentyl pyrimidyl nucleoside that is activated by uridine cytidine kinase 2. RX-3117 has shown efficacy in xenograft models of gemcitabine resistant pancreatic, bladder and colorectal cancer. Data from stage 2 of the Phase 1b/2a clinical study of RX3117 as a single agent in subjects with metastatic pancreatic cancer is described below. Methods: Stage 2 of the Phase 1b/2a study (NCT02030067) is designed to evaluate safety, tolerability and efficacy following treatment with 700 mg administered orally once-daily for 5 consecutive days with 2 days off per week for 3 weeks with 1 week off in each 4 week cycle. Eligible subjects (aged ≥ 18 years) had relapsed/refractory metastatic pancreatic cancer. The primary endpoint is a ≥ 20% rate of progression free survival (PFS) benefit (i.e., proportion of subjects with stable disease for at least 4 months) and/or a 10% of evaluable subjects with a partial response rate or better. Results: As of Sep 2017, 44 subjects have been enrolled (22 females, 22 males). The median age was 68 years, ECOG performance statuses were 0 (13 subjects) and 1 (31 subjects) and 6 subjects had received 4 or more prior therapies. One subject had an unconfirmed partial response and 21 subjects met the primary endpoint of stable disease with a duration of 30-224 days. The most frequent adverse events were mild to moderate anemia (19%), mild to moderate fatigue (15%), mild to moderate diarrhea (11%), and severe anemia (9%). Conclusions: This ongoing trial shows an early efficacy signal where RX-3117 is active against advanced pancreatic cancer. The study continues to enroll subjects with advanced pancreatic cancer into stage 2. A phase 2 study with nab-paclitaxel in first-line patients with advanced pancreatic cancer has been started. Clinical trial information: NCT02030067.

Activity of RX-3117, an oral antimetabolite nucleoside, in subjects with metastatic bladder cancer resistant to gemcitabine: Preliminary results of a phase Ia/IIb study

4544

Background: RX-3117 is an oral small molecule antimetabolite, cyclopentyl pyrimidyl nucleoside that is activated by uridine cytidine kinase 2. RX-3117 has shown efficacy in xenograft models of gemcitabine resistant pancreatic, bladder and colorectal cancer. Preliminary data from an analysis of a phase 1b/2a clinical study of RX3117 in metastatic bladder cancer is described. Methods: This phase 1b/2a study (NCT02030067) was designed to evaluate safety, tolerability and efficacy following treatment with 700 mg administered orally once-daily for 5 consecutive days with 2 days off per week for 3 weeks with 1 week off in each 4 week cycle in a 2-stage design. Eligible subjects (aged ≥ 18 years) were those with relapsed/refractory metastatic bladder cancer with any number of prior therapies. Prior therapy with platinum-based chemotherapy was required. The primary endpoint was to assess the efficacy and safety of RX-3117 in metastatic bladder cancer, with secondary aims of evaluating PFS and CBR. Results: With 9 subjects enrolled, median age was 66 years, ECOG PS was 0-1. All subjects had received gemcitabine/cisplatin in the perioperative or metastatic setting, and 4 subjects had received 3 or more prior therapies. The most frequent related adverse events were anemia, mild-moderate fatigue, vomiting and diarrhea. No dose limiting toxicities were observed. PFS and CBR will be presented at the meeting, as 5 subjects continue to receive therapy at the time of this submission. One subject continues on treatment at 139 days with persistent stable disease. Molecular profiling of his bladder tumor showed alterations in ARID1A, FBXW7, FGFR3, NF1, and TERT. The patient previously responded to an FGFR3 inhibitor but progressed after 9 months, with ctDNA assessments showing incurrence of TP53 alteration. Clinical benefit with RX-3117 was achieved in spite of incurrence of this alteration. Conclusions: RX-3117 demonstrated an excellent safety profile, and prolonged stable disease was seen in 1 subject who failed prior cisplatin/gemcitabine and FGFR3 inhibition. Activity persisted despite development a putative resistance alteration detected by ctDNA. Clinical trial information: NCT02030067.

A phase I study to assess the safety, pharmacokinetics (PK), pharmacodynamics (PD), and anti-tumor activity of oral COH29, a novel ribonucleotide reductase (RNR) inhibitor in adult patients (pts) with advanced solid tumors

TPS2600

Background: Human RNR catalyzes the rate-limiting step in the formation of deoxyribonucleotide triphosphates (dNTPs) necessary for DNA repair and replication. Rapidly dividing tumor cells are especially sensitive to RNR inhibition due to elevated dNTP requirements. Overexpression of the RNR RRM2 subunit is also associated with neoplasia, metastasis, and poor prognosis. COH29 is an aromatically substituted thiazole compound that is a novel small molecule inhibitor of RNR activity, and exhibits unique mechanisms and target specificity that overcomes the weaknesses of other small molecule RNR inhibitors. Preclinically, it is more potent than hydroxyurea and gemcitabine, and is not associated with iron chelating-related toxicities such as hypoxia. Cell lines deficient in BRCA1 also exhibit greater sensitivity to COH29 than BRCA1 wildtype cell lines, implicating inhibition of DNA repair mechanisms in line with PARP inhibitors. Methods: In this Phase I, single site, dose escalation, safety study pts will receive oral COH29 twice a day for 21 days of a 28-day cycle. Eligible pts are age ≥ 18 years, ECOG ≤ 2, able to take oral medication, have adequate organ and marrow function, and diagnosed with any solid tumor refractory to standard therapies. Dose escalation will be pursued utilizing a Simon’s accelerated titration design, which allows skipping of dose levels (dose doubling) during the accelerated dose-finding phase. Primary objectives are to determine the maximum tolerated dose of COH29, toxicities per CTCAEv4, and PKs. Secondary objectives include assessment of objective response per RECIST 1.1 every 2 cycles. PD assessment includes measurement of plasma CK18 levels to determine degree of cellular apoptosis, evaluation of dNTP pool levels in peripheral blood mononuclear cells (PBMCs) to evaluate RNR inhibition, as well as measurement of PAR expression in PBMCs to assess PARP inhibition. Quantitation of tumor RRM2 expression using dual-color immunohistochemistry will be explored as a predictive biomarker of anti-tumor response to COH29. Clinical trial information: NCT02112565.

Risk factors for hospitalizations (HOS) among older adults with gastrointestinal (GI) cancers receiving chemotherapy

e21523

Background: Older adults undergoing chemotherapy for GI cancers are at increased risk of HOS due to treatment related toxicity; however, there are limited data regarding which individuals are at greatest risk. We therefore sought to identify risk factors for HOS among older adults with GI cancers receiving chemotherapy. Methods: We performed a secondary analysis of patients age ≥ 65 years with GI cancer who participated in either of 2 prospective studies used to develop (n = 500) and validate (n = 250) a geriatric assessment (GA) based chemotherapy toxicity score for older adults with cancer. The incidence of HOS within 30 days post treatment was determined. The following patient characteristics were captured pre-chemotherapy: demographics, cancer type, stage, laboratory values, chemotherapy type, and GA measures (functional status, comorbidity, psychological state, cognitive function, nutritional status, and social support). Univariate and multivariate logistic regressions were used to estimate the odds ratio (OR) to identify risk factors. Results: A total of 199 adults age 65+ (median 73; range 65-94) with GI cancers (colorectal 43%, gastric/esophageal 25%, pancreas/biliary 32%; Stage I-III 42%, stage IV 58%) receiving chemotherapy (67% poly-chemotherapy) were included. 5-FU chemotherapy was administered alone or in combination in 126 (63%) patients. Sixty five (33%) patients had ≥1 HOS (1 HOS: 55, 2 HOS: 9, 3 HOS: 1). In univariate analysis, hospitalized patients were more likely to be female (p = 0.02), have stage IV disease (p = 0.03), have a diagnosis of non-colorectal GI cancer (p = 0.04), have poly-pharmacy (≥ 5 medications, p < 0.01), decreased hearing (p = 0.05), cardiac comorbidity (p < 0.01), and low serum albumin (p = 0.05). On multivariate analyses, patients who were female (OR = 2.06, 95% CI: 1.05-4.06), with cardiac comorbidity (OR = 3.73, 95% CI: 1.78-7.83), or a diagnosis of stage IV non-colorectal GI cancer (OR = 3.75, 95% CI: 1.50-9.39) were more likely to be hospitalized. Conclusions: HOS during chemotherapy treatment are common among older adults with GI cancers. Female sex, cardiac comorbidity, and a diagnosis of stage IV non-colorectal GI cancer are risk factors for HOS.

Combining pembrolizumab and palliative radiotherapy in gastroesophageal cancer to enhance anti-tumor T-cell response and augment the abscopal effect

TPS220

Background: Pembrolizumab has demonstrated encouraging preliminary evidence of activity in small cohorts of gastroesophageal cancer patients (pts) when selected for tumor PD-L1 overexpression. Strategies should be further developed to enhance the potency of PD-1 inhibitors, even in pts whose tumors are not considered to have detectable PD-L1 expression. The abscopal effect is a rarely observed though well annotated clinical phenomenon in cancer radiotherapy (RT), in which non-irradiated metastatic lesions are noted to regress after radiation to the primary tumor site. In preclinical models, this phenomenon has been attributed to immune-mediated effects, while RT has been observed to increase tumor PD-L1 expression. This lends credence to ongoing efforts to combine RT and immune checkpoint inhibition. We present the study design for a phase II trial combining pembrolizumab and palliative RT in metastatic gastroesophageal squamous cell and adenocarcinoma pts. Methods: A target accrual of 14 pts will be assigned to palliative RT of 30 Gy over 10 fractions for symptoms from their primary tumor or a single target metastatic site in combination with fixed dose pembrolizumab 200 mg every 3 weeks. Symptomatic brain metastases will be excluded. Pembrolizumab will be continued until disease progression, intolerable toxicity, or 35 administrations (~2 years). Pts will be required to undergo sequential biopsies of a non-target metastatic site (inclusive of malignant ascites) with timepoints occurring pre-treatment and ~21 days after completion of RT. Biomarkers of immune response and overall response rate (RECIST1.1, irRECIST) will be a composite primary endpoint; progression-free and overall survival, and toxicities will be secondary endpoints. Quantitative spatial image analyses of biopsy samples will be performed to examine changes in tumor cell-stromal interactions inclusive of CD4, CD8, Treg, and MDSC populations as well as tumor PD-L1/PD-1 expression levels. The results will assist in determining mechanisms of response and resistance to combination RT and PD-1 targeting strategies. Clinical trial information: NCT02830594.

Molecular biomarkers as predictors of patient survival in pancreatic adenocarcinoma (PDA): An analysis of the Know Your Tumor initiative (KYT)

278

Background: Recent studies have expanded our knowledge of the genomic landscape of PDA. While critical and in some cases, potentially actionable alterations are being identified, limited outcomes data have thus far made it difficult to validate the relevance of these observations. Methods: The Pancreatic Cancer Action Network (PanCAN) and Perthera have facilitated commercial tumor molecular profiling for over 400 PDA pts since 2014 through KYT, and have developed a database of molecular and clinical information useful for data mining of biomarker-survival correlations. The survival significance of biomarkers was assessed using standard statistical methodology including Kaplan-Meier analysis and Cox proportional hazard models. Results: Linked molecular and outcomes data were available for 360 pts, of which 173 had treatment (tx) information available. Pathogenic mutations from targeted NGS, protein expression from IHC, and protein phosphorylation from RPPA were screened for correlations with overall survival (OS) and progression-free survival (PFS) independent of tx received. As shown in the table, mutations in 3 genes were associated with a better OS; while mutations in 8 genes were associated with poorer OS. Only two mutations were correlated with PFS in 1st or 2nd-line tx ( BRCA2 and KDM6A, worse PFS). Positive expression of 7 proteins, including TS, TOP1, and RRM1, were associated with reduced OS but were not correlated with PFS. High levels of phospho-ribosomal protein S6 were associated with both poor OS (HR=10.3, p=0.001) and poor PFS (HR=9.6, p=0.006). Conclusions: Multiple biomarkers had significant correlations with OS in PDA, while fewer were correlated with PFS. Growth of this registry database will further validate tx-specific predictive biomarkers for use in pts with multi-omic profiling data. [Table: see text]

Integration of palliative care for patients with solid tumors on phase I clinical trials

138

Background: Cancer patients receiving Phase I clinical trials are a population with advanced disease, high symptom burden, and with multiple QOL concerns including use of advance directives. Methods: An NCI funded R01 is currently in progress (2014-2019) as a randomized clinical trial to test a palliative care intervention (PCI) in this population. The PCI includes comprehensive patient assessment, goals of care communication, interdisciplinary care planning and patient teaching. Aims and hypotheses test the impact of the PCI on symptoms, QOL, resource use, spirituality and distress. Outcomes for the study (N = 400) will be conducted at the conclusion of the RCT. This paper reports preliminary baseline data of the first 100 subjects accrued. Results: Subjects mean age was 59 years and 59% were female, similar to prior trials. Forty eight percent (48%) were ethnic minorities, higher than prior trials (3% non-white, Finlay E, 2014; 9%, Parsons JA, et al. PLoS One 2012) with colon (22%) and lung (21%) cancers as dominant. Patients had a mean of 2 comorbidities (range 0-8) and 40% of the patients are > 65 years of age. The most common symptoms reported by PRO-CTCAE (1 = least to 5 = most concern) were sexuality (4.2), fatigue (2.9), pain (2.6) and anxiety (2.5). Psychological Distress Thermometer, (0 - least to 10 = most distress) mean score was 4. FACIT spiritual concerns (1 = least to 5 = most concerning) identified greatest concerns of illness strengthening faith (2.0), strength from faith (2.0) and sense of harmony (2.6). There was limited use of supportive care services, PC consultation, or advanced directives. Conclusions: The population of Phase I trial patients is an important group for palliative care integration with major unmet needs in symptom management and advance care planning.

Multi-Omic profiling (MoP) for patients (pts) with pancreatic cancer (PDA): Initial results of the Know Your Tumor (KYT) initiative

282

Background: Emerging insights into the molecular biology of tumors and recent developments in MoP have led to the identification of targets that can be used for selecting treatment (tx) regimens. Multiple testing platforms are available, though most studies to date have used only next generation DNA sequencing (NGS). Methods: The Pancreatic Cancer Action Network (PanCAN) and Perthera have initiated an IRB-approved registry trial wherein we facilitate commercial MoP on tumor tissue from PDA patients. MoP includes NGS, immunohistochemistry (IHC), and phosphoproteomics (PHO). The results are reviewed by a team of PDA specialists in the context of the pt’s tx history. Tx options are prioritized based on the actionable molecular abnormalities. Pts are being followed longitudinally to assess physician acceptance of the tx options and track survival outcomes. Results: From 06/2014 to 09/2015, Perthera reports were delivered for 117 pts. 44% of the analyses were for second-line tx and 43% for ≥ third-line. Tumor based NGS and IHC were available for 75% and 90% of pts, respectively, and research use only PHO data was available for 20 pts. Actionable findings, defined based on a high response rate in pts with an identified molecular abnormality (in any cancer type), or based on a mechanism/pathway-defined implication of response to tx were identified in 43% of pts, primarily based on NGS. Incorporation of IHC refined and expanded chemotherapy tx options in all pts. PHO revealed pathway activation (e.g. mTOR, JAK-STAT, MET, RET, or EGFR) in 16/20 samples. Interestingly, only 45% of KRAS mutant samples had PHO-defined activation of the MEK-ERK pathway. Conclusions: MoP resulted in actionable findings in 43% of PDA pts using NGS and IHC. Incorporation of PHO data could have a significant impact on MoP-based therapy options. Analysis of the MoP in the context of patient history by a PDA specialist provided a service to treating oncologists in the community that was frequently incorporated into their treatment decisions. Survival outcomes will be presented. As additional testing platforms and results become available, the breadth and confidence of actionable markers is expected to increase.

A phase I study of MEK162 and FOLFOX in chemotherapy-resistant metastatic colorectal cancer

679

Background: Pre-clinical studies have shown that the combination of MEK inhibitors and 5-FU improves antitumor activity and that MEK inhibition overcomes both 5FU and platinum resistance. This phase I study was conducted to determine the maximum tolerated dose (MTD) of the combination MEK162 and FOLFOX. Methods: Patients (pts) with metastatic colon or rectal cancer who progressed or failed prior 5FU, irinotecan, oxaliplatin and anti-EGFR therapy (in cases of RAS wild type tumors) received twice daily MEK162 in combination with every-2-week FOLFOX. Two dose levels of MEK162 (30mg and 45mg) were investigated in a standard 3 + 3 escalation design in combination with standard doses of FOLFOX without bolus 5-FU. Dose limiting toxicity (DLT) was defined as any treatment-related grade (G) 3 or 4 non-hematological toxicity (with the exception of G3 diarrhea or vomiting < 48 hrs) or G 4 neutropenia or thrombocytopenia within the first 2 cycles (4 weeks) of the treatment. Limited pharmacokinetic (PK) analysis of 5FU, oxaliplatin and MEK162 was performed at the MTD level. Results: 16 pts were enrolled (median age (range) 53 yrs (49-78); 11 men; ECOG 0/1 in 9/7 patients). No DLT was noted on the study. The MTD of MEK162 was 45 mg PO BID. An additional 6 pts (for a total of 12) were enrolled at the MTD for PK analysis and none of them developed DLT defining toxicities. A median of 8 cycles (range 1-19) was administered. Treatment-related ≥ grade 3 toxicities included anaphylaxis due to oxaliplatin (n = 1), CPK elevation (n = 2), neutropenia (n = 1), peripheral neuropathy (n = 3), thrombocytopenia (n = 1), retinal vascular disorder (n = 1), and acneiform rash (n = 1). 10 pts had SD at 2 months (m) by radiographic assessment, 5 of whom with stabilizations of > 5 months (5-10 months). There were no significant differences in the PKs of 5FU or oxaliplatin when administered with or without MEK162. Conclusions: The combination of MEK162 and FOLFOX has a manageable toxicity profile and promising antitumor activity in heavily pretreated metastatic colorectal cancer patients. Clinical trial information: NCT02041481.

The Know Your Tumor (KYT) initiative: A national program of multi-omic molecular profiling (MoP) for patients (Pts) with pancreatic cancer (PDA)

279

Background: There has been a significant increase in the utilization of MoP to facilitate the selection of therapies for pts with advanced cancers. However, coordinated efforts to perform uniform MoP and collect clinical outcomes data have only successfully been executed in limited academic settings. In an effort to “democratize” the availability of MoP for PDA patients in a broad spectrum of clinical settings, The Pancreatic Cancer Action Network (PanCAN) and Perthera initiated the KYT initiative, an academic, industry, and advocacy group collaboration. Methods: PDA pts self-referred to the Patient Central call center at PanCAN. Pts interested in enrolling in an IRB-approved registry trial were then referred to Perthera to facilitate consent, tissue acquisition, and commercial MoP of the pt’s tumor. Test results were reviewed by a team of PDA specialists in the context of the pt’s treatment (tx) history. A report detailing tx options was delivered to the pt’s oncologist. Pts are being actively followed longitudinally to assess physician acceptance of the tx options and to track outcomes. Results: From 06/2014 to 09/2015, 382 pts were enrolled into KYT. Of these, 273 pts from 192 physicians (community and academic) in 38 states were enrolled. Reasons for non-enrollment were: pts no longer interested (67%); pts too ill or not appropriate for KYT (17%); the pt’s physician would not sanction MoP (16%). To date, 162 pt tumor samples have been obtained, and 117 reports delivered. 75% of samples had adequate tissue for genomic profiling (NGS), with 87% of samples harboring a KRAS mutation. Actionable findings (i.e. linked to a specific tx option) identified by NGS included mutations in BRCA2 (5%), PALB2 (1%), ATM (4%); BRAF (2%), PIK3C/PIK3R (7%), STK11 (5%), amplification of ERBB2 (3%), FGFR (2%), PDGFR (2%); and RET fusions (2%). Of the 117 profiles, 43% revealed actionable findings; 58% presented off-label therapy options; and 48% led towards high priority clinical trials. Conclusions: MoP of pts with PDA is feasible irrespective of their geographical location or access to an academic center. Updated information on tx selection and patient outcomes will be provided.

RAS mutational status and CEA production at initial presentation in metastatic colorectal cancer

542

Background: Serial CEA testing is recommended in the surveillance of patients with resected stage II-IV colorectal cancer. However, the sensitivity of CEA in identifying metastatic disease has not been evaluated in the settings of RAS mutant (MT) and RAS/BRAF wild-type tumors (WT). In order to evaluate the impact of RAS mutational status on CEA production, we retrospectively evaluated a single-institute metastatic colorectal cancer (mCRC) population. Methods: We retrospectively reviewed, in a single center, all cases with mCRC with known RAS mutational status based on next generation sequencing (ONCO44 and ONCO48). These assays identify clinically relevant mutations in BRAF, KRAS, and NRAS. Additional eligibility criteria included the availability of CEA levels and imaging studies at first diagnosis of mCRC. Patient demographics, primary tumor location and sites of metastatic disease at 1st diagnosis were captured. CEA levels were stratified as normal or elevated based on a cut point of 5ng/ml. Results: 139 mCRC patients satisfied the eligibility criteria (75 RAS-MT, 59 RAS/BRAF-WT, and 5 BRAF-MT). BRAF-MT patients were excluded from the analysis due to their small sample size. Patients with RAS/BRAF-WT tumors were more likely to present with metastatic disease to the liver, but this did not reach statistical significance (p = 0.056). There was no difference in the incidence of normal CEA at presentation in RAS-MT (30%) and RAS/BRAF-WT (28%) cohorts. CEA production was dependent on the pattern of metastatic disease. Elevated CEA was associated with the presence of liver metastases versus no metastases among RAS-MT (92% vs 47% p <.0001) and RAS/BRAF-WT patients (82% vs 50% p = 0.0101). RAS status did not impact the likelihood of CEA production within the hepatic metastases and non-hepatic metastases groups. Conclusions: RAS status does not appear to influence CEA production in patients with mCRC. CEA elevations are highly associated with liver metastases and are less prevalent in patients without hepatic involvement. These findings confirm the limited predictive value of CEA for non-hepatic recurrence, irrespective of RAS status.

Pilot trial of CRLX101 in patients (pts) with advanced, chemotherapy-refractory gastroesophageal cancer (GEC)

44

Background: Camptothecin (CPT) derivatives such as irinotecan have activity in 2nd-line therapy in advanced GEC with reported response rates of 0-15%. CRLX101 is an investigational nanoparticle-drug conjugate (NDC) with a CPT payload. Preclinical evidence indicates preferential uptake in tumors, and animal GEC xenograft models demonstrate superiority of CRLX101 over irinotecan. A pilot trial was conducted at recommended phase 2 dosing (RP2D) to assess preferential uptake of CRLX101 in tumor vs. adjacent normal tissue in endoscopically accessible tumors in patients with chemotherapy-refractory GEC. Data demonstrating preferential tumor uptake of CRLX101 has been presented separately and here we report on the clinical outcomes of patients enrolled. Methods: All pts initiated CRLX101 dosed intravenously at RP2D (15 mg/m2) on days 1 and 15 of a 28-day cycle until disease progression or intolerant toxicity. While detection of preferential CRLX101 tumor uptake was the primary endpoint, with 10 pts enrolled a secondary analysis could be performed with the study having 90% power to detect ≥ 1 responder if the true response rate is ≥ 21%. Responses were assessed using RECIST 1.1. Results: Between Dec. 2012 and Dec. 2014, 10 patients with chemotherapy-refractory (median 2 prior lines of therapy, range 1-4) GEC and adenocarcinoma histology were enrolled and evaluable for response and toxicity. The median time-to-progression was 1.9 mo (range 0.6-8.7 mo). Best response was seen in 1 pt with stable disease (SD) for 8 cycles. Only ≥ grade 3 toxicities related to CRLX101 occurred in a single patient with grade 3 anemia and chest pain who was able to resume therapy without any further toxicity after CRLX101 was reduced to 12 mg/m2. Conclusions: CRLX101 demonstrated minimal activity with SD as best response in this heavily pretreated population. Future efforts with CRLX101 in advanced GEC should focus on combination strategies. Its favorable toxicity profile and evidence of preferential tumor uptake support further clinical research of combining CRLX101 with other targeted therapies such as anti-angiogenesis (ramucirumab) and/or immune checkpoint inhibitors. Clinical trial information: NCT01612546 Clinical trial information: NCT01612546.

Effect of baseline carbohydrate antigen 19-9 (CA19-9) level on overall survival (OS) in NAPOLI-1 trial: A phase III study of MM-398 (nal-IRI), with or without 5-fluorouracil and leucovorin (5-FU/LV), versus 5-FU/LV in metastatic pancreatic cancer (mPAC) previously treated with gemcitabine-based therapy

425

Background: CA19-9 has been shown to correlate with response to therapy and OS in patients with mPAC. NAPOLI-1, a randomized phase 3 study evaluated nal-IRI, a nanoliposomal formulation of irinotecan, with or without 5-FU/LV vs 5-FU/LV in patients with mPAC previously treated with gemcitabine-based therapy. Nal-IRI+5-FU/LV significantly improved OS (primary endpoint) vs 5-FU/LV (6.1 mo vs 4.2 mo; unstratified hazard ratio [HR] = 0.67; P = 0.012). CA19-9 response (≥50% decline from baseline) was superior with nal-IRI+5FU/LV compared with 5-FU/LV (29% vs 9%; P=0.0006). Nal-IRI alone did not show a statistical improvement in survival. Methods: Patients with a recorded baseline CA19-9 measurement were divided into quartiles to evaluate the treatment effect pattern of CA19-9 from nal-IRI+5-FU/LV and 5-FU/LV arms. Quartile ranges were based on 404 available CA19-9 values from randomized patients (N=417). Unstratified Cox proportional hazards regression was used to estimate HRs and corresponding 95% CIs. Effect of baseline CA19-9 on time to response, progression-free survival, and response will be presented. Results: Of patients randomized to receive nal-IRI+5-FU/LV (n = 117) or 5-FU/LV enrolled contemporaneously (n = 119), 218 received study drug and had a baseline CA19-9 measurement. Results show a greater treatment effect on OS with higher CA19-9 level relative to 5-FU/LV. Conclusions: In patients with mPAC previously treated with gemcitabine-based therapy, nal-IRI+5-FU/LV significantly improved OS supported by progression free survival and objective response rate. The CA19-9 serum level can provide important information with regards to overall survival. Clinical trial information: NCT01494506. [Table: see text]

Preliminary observations of blood-based (BB) molecular testing in a subset of patients with pancreatic cancer (PDA) participating in the Know Your Tumor (KYT) initiative

268

Background: Obtaining an adequate tissue sample for molecular profiling to guide therapy selection for patients with advanced cancers can be clinically difficult, and/or patients may not want to undergo a biopsy. There has been a growing interest in the use of BB tests, including cell free DNA (cfDNA) and exosome/circulating tumor cell based-analyses as surrogates for tumor tissue (TT) testing. Validation of BB tests in PDA is possible because > 90% of PDAs harbor KRAS mutations – thus providing a reliable “internal control.” Methods: The Pancreatic Cancer Action Network (PanCAN) and Perthera initiated an IRB-approved registry trial for patients with pancreatic cancer wherein we facilitated commercially available, CLIA certified multi-Omic profiling including next generation DNA sequencing (NGS), immunohistochemistry, and phosphoproteomics on patient tumor samples. In a subset of these patients, we incorporated BB testing. Results: A KRAS mutation has been identified in 87% of KYT patients based on TT NGS in general. As of this report, 17 BB test results (cfDNA NGS) were available. In 8 patients we were able to compare the cfDNA NGS directly with TT NGS. BB testing identified a KRAS mutation in 1/8 compared to 8/8 from the tumor tissue. Of the 9/17 patients who did not have corresponding TT NGS, a KRAS mutation was found on BB testing in 3/9 samples. 4/17 BB cases overall revealed KRAS mutations, but when cases were filtered for patients with extensive metastatic disease and progressive disease at the time of blood sampling, the KRAS mutation rate increased to 2/5 overall. Actionable findings (i.e. linked to a specific therapeutic option) were identified in 6/17 cases, of which 3/17 had both an actionable mutation and a KRAS mutation. Conclusions: Although we are aware of the limitations of this study, we recommend that for patients who have biopsiable disease, a TT test should still be the gold standard for molecular profiling. For those without biopsiable disease, the KYT program presents an opportunity to determine the parameters that influence the probability of obtaining reliable molecular profiling information from a BB sample.

Gene-mediated cytotoxic immunotherapy as adjuvant to surgery or chemoradiation for pancreatic adenocarcinoma

BACKGROUND

While surgical resection of pancreatic adenocarcinoma provides the only chance of cure, long-term survival remains poor. Immunotherapy may improve outcomes, especially as adjuvant to local therapies. Gene-mediated cytotoxic immunotherapy (GMCI) generates a systemic anti-tumor response through local delivery of an adenoviral vector expressing the HSV-tk gene (aglatimagene besadenovec, AdV-tk) followed by anti-herpetic prodrug. GMCI has demonstrated synergy with standard of care (SOC) in other tumor types. This is the first application in pancreatic cancer.

METHODS

Four dose levels (3 × 10(10) to 1 × 10(12) vector particles) were evaluated as adjuvant to surgery for resectable disease (Arm A) or to 5-FU chemoradiation for locally advanced disease (Arm B). Each patient received two cycles of AdV-tk + prodrug.

RESULTS

Twenty-four patients completed therapy, 12 per arm, with no dose-limiting toxicities. All Arm A patients were explored, eight were resected, one was locally advanced and three had distant metastases. CD8(+) T cell infiltration increased an average of 22-fold (range sixfold to 75-fold) compared with baseline (p = 0.0021). PD-L1 expression increased in 5/7 samples analyzed. One node-positive resected patient is alive >66 months without recurrence. Arm B RECIST response rate was 25 % with a median OS of 12 months and 1-year survival of 50 %. Patient-reported quality of life showed no evidence of deterioration.

CONCLUSIONS

AdV-tk can be safely combined with pancreatic cancer SOC without added toxicity. Response and survival compare favorably to expected outcomes and immune activity increased. These results support further evaluation of GMCI with more modern chemoradiation and surgery as well as PD-1/PD-L1 inhibitors in pancreatic cancer.

Phase I study of the halichondrin B analogue eribulin mesylate in combination with cisplatin in advanced solid tumors

BACKGROUND

Eribulin mesylate is a synthetic macrocyclic ketone analogue of Halichondrin B that has demonstrated high antitumor activity in preclinical and clinical settings. This phase I study aimed to determine the maximum tolerated dose (MTD), dose-limiting toxicities (DLTs), and pharmacokinetics in combination with cisplatin (CP) in patients with advanced solid tumours.

METHODS

Thirty-six patients with advanced solid tumours received eribulin mesylate 0.7-1.4 mg m(-2) and CP 60-75 mg m(-2). Eribulin mesylate was administered on days 1, 8, and 15 in combination with CP day 1 every 28-day cycle. The protocol was amended after dose level 4 (eribulin mesylate 1.4 mg m(-2), CP 60 mg m(-2)) when it was not feasible to administer eribulin mesylate on day 15 because of neutropenia; the treatment schedule was changed to eribulin mesylate on days 1 and 8 and CP on day 1 every 21 days.

RESULTS

On the 28-day schedule, three patients had DLT during the first cycle: grade (G) 4 febrile neutropenia (1.0 mg m(-2), 60 mg m(-2)); G 3 anorexia/fatigue/hypokalemia (1.2 mg m(-2), 60 mg m(-2)); and G 3 stomatitis/nausea/vomiting/fatigue (1.4 mg m(-2), 60 mg m(-2)). On the 21-day schedule, three patients had DLT during the first cycle: G 3 hypokalemia/hyponatremia (1.4 mg m(-2), 60 mg m(-2)); G 4 mucositis (1.4 mg m(-2), 60 mg m(-2)); and G 3 hypokalemia (1.2 mg m(-2), 75 mg m(-2)). The MTD and recommended phase II dose was determined as eribulin mesylate 1.2 mg m(-2) (days 1, 8) and CP 75 mg m(-2) (day 1), on a 21-day cycle. Two patients had unconfirmed partial responses (PR) (pancreatic and breast cancers) and two had PR (oesophageal and bladder cancers).

CONCLUSIONS

On the 21-day cycle, eribulin mesylate 1.2 mg m(-2), administered on days 1 and 8, in combination with CP 75 mg m(-2), administered on day 1 is well tolerated and showed preliminary anticancer activity.

A phase I study of an MVA vaccine targeting p53 in cancer

3089

Background: Despite chemotherapy, survival for most patients with metastatic gastrointestinal cancer is <2 years. These dismal statistics reflect lack of effective therapy. We are conducting a first-in-human trial of modified vaccinia Ankara (MVA), an attenuated viral vaccine that targets wild-type (wt) p53. This approach applies to the majority of p53-involved tumors, since the bulk of the p53 sequence is identical, except for individual point mutations. Our phase I study evaluates the safety and tolerability of the vaccine. Methods: Patients with metastatic colon, gastric and pancreas cancer that failed standard treatment were eligible for vaccination if >10% of the cells stained strongly positive for p53. A standard 3+3 design was employed and patients were accrued to either the 108 pfu or 5.6 x 108pfu dose levels. A total of 3 injections were given, each spaced apart by 3 weeks and patients were evaluated for dose limiting toxicities. Results: Three patients were accrued to the 108 pfu dose level; 2 patients with colon cancer and 1 patient with pancreatic cancer. There were no dose limiting toxicities and the injection was well tolerated and only local irritation at the injection site was seen. The dose was escalated to 5.6 x 108 pfu; 3 patients were accrued, 1 with colon cancer and 2 with pancreatic cancer. There were no dose limiting toxicities observed in the 3 patients. Grade 2 toxicities that were at least possibly related to the vaccine include injection site reaction and fatigue seen in one patient each. One patient with pancreatic cancer had stable disease on CT after completion of the 3 injections. This dose level is being expanded to accrue an additional 6 patients to evaluate safety and immunogenicity using cytokine flow cytometry to characterize p53-specific CD4+ and CD8+ T cell response. Of the cell-surface markers employed, PD1 showed extraordinary strong expression prior to vaccine injection. All patients had a humoral response to the MVA backbone. Conclusions: Our MVA p53 vaccine is well tolerated with minimal grade 1-2 toxicities. The highest dose tested is 5.6 x 108 pfu and additional patients are being accrued to evaluate immunogenicity. Since many cancers have mutant p53, this is an attractive target and may potentially be used with many other cancers. Clinical trial information: NCT01191684.

PHI-55: (NCI#7427): A phase I study of halichondrin B analog (E7389) in combination with cisplatin (CDDP) in advanced solid tumors: A CCC, NCI/CTEP-sponsored trial (grant U01 CA 062505)

2564

Background: E7389 is a structurally simplified synthetic macrocyclic ketone analog of the marine sponge natural product Halichondrin B, which inhibits microtubule dynamics via a novel mechanism characterized by suppression of microtubule growth, lack of effect on microtubule depolymerization, and sequestration of tubulin into nonfunctional aggregates. Methods: The goals of this trial were to determine the DLT, the MTD, and PK of E7389 (administered on days 1, 8 and 15 every 28 days) in combination with CDDP (administered on day 1 every 28 days) in patients (pts) with advanced solid tumors. The protocol was amended after dose level 4 (E7389 1.4 mg/m2, CDDP 60 mg/m2) when it was not feasible to administer E7389 on day 15 due to neutropenia; the treatment schedule was changed to E7389 days 1 and 8 and CDDP day 1 every 21 days. Eligibility criteria included normal organ function and < 2 prior chemotherapy regimens. Results: To date, 36 pts have been treated (E7389 0.7-1.4 mg/m2 and CDDP 60-75 mg/m2). Median age 61 years; 19 males; the most common tumor types were lung (8), pancreatic (5), head and neck (6). 36% ECOG 0, 56% ECOG 1, 8% ECOG 2; Median number of cycles was 3 (1 – 8). There were 3 pts with DLT’s on the 28-day cycle: gr 4 febrile neutropenia (1.0/60); gr 3 anorexia/fatigue/hypokalemia (1.2/60); and gr 3 stomatitis/fatigue (1.4/60). There were 3 pts with DLTs treated on the 21-day schedule: gr 3 hypokalemia/hyponatremia (1.4/60); gr 4 mucositis (1.4/60); and gr 3 hypokalemia (1.2/75). With 2 DLTs out of 6 pts at E7389 1.4 mg/m2 and CDDP 60 mg/m2, E7389 was reduced, CDDP was escalated, and the MTD was determined to be E7389 1.2 mg/m2 and CDDP 75 mg/m2(1 patient out of 6 with a DLT). At the MTD, protocol defined dose modifications or delays were required in 2 of the 6 patients by cycle 2. Notably, all DLTs were observed in patients exposed to at least 2 prior lines of chemotherapy. Two pts had an unconfirmed PR (pancreatic, breast) and 2 had a PR (esophagus, transitional bladder). Conclusions: On a 21 day schedule,E7389 in combination with CDDP appears well tolerated and showed preliminary activity. The MTD was determined to be E7389 1.2 mg/m2 and CDDP 75 mg/m2. Clinical trial information: 00400829.

SWOG S1115: Randomized phase II clinical trial of selumetinib (AZD6244; ARRY 142886) hydrogen sulfate (NSC-748727) and MK-2206 (NSC-749607) versus mFOLFOX in patients withmetastatic pancreatic cancer after prior chemotherapy

TPS4145

Background: Pancreatic cancer remains a deadly disease and despite advances in chemotherapy treatment, survival for most patients is still less than one year. Over 80% of pancreatic cancers are KRAS mutant which activates the PI3K/AKT pathway and signals downstream to mTOR leading to cell growth, proliferation and survival. Recent data has shown that blocking both the PI3K/AKT and MEK pathways simultaneously is effective in KRASmutant tumors. Our trial is a novel, molecular targeted treatment approach for patients with metastatic pancreatic cancer that has the potential to establish a new treatment paradigm. Methods: S1115 was activated in SWOG in August 2012 and is currently IRB approved at 130 institutions within SWOG and the Clinical Trials Support Unit (CTSU). Patients (performance status 0 or 1) with metastatic pancreatic cancer failing standard gemcitabine chemotherapy are randomized to MK-2206 135 mg orally weekly plus selumetinib 100 mg orally daily or mFOLFOX IV every 2 weeks. Eligibility criteria allow metastatic patients who have progressed within 6 months of receiving adjuvant gemcitabine. Patients receiving prior 5-fluorouracil (excluding radiation-sensitizing doses), capecitabine, oxaliplatin, MEK or PI3K/AKT inhibitors are not eligible. Stratification factors include duration of prior systemic therapy and presence of liver metastases. The primary endpoint of this study is overall survival (OS) in patients treated with the combination of MK-2206 and selumetinib compared with those treated with mFOLFOX. Based on previous studies, median OS in the control group is approximately 6 months. Assuming a one-sided type 1 error of 10%, 80% power, approximately 2 years of accrual and 1.5 years of follow-up, 120 eligible patients will be accrued to detect an improvement in median survival from 6 to 9 months (corresponding to a 1.5 hazard ratio). Prospective tumoral tissue collection will be undertaken. ClinicalTrials.gov Identifier: NCT01658943. Support: NCI grants CA32102 & CA38926 Clinical trial information: NCT01658943.

Phase I study of ursodeoxycholic acid in combination with 5-fluorouracil, leucovorin, oxaliplatin, and bevacizumab for metastatic colorectal cancer

e14182

Background: Significant treatment advances have doubled the median survival in patients with metastatic colorectal cancer (mCRC). An emerging body of literature links cancer with aberrant metabolism. These findings suggest that treatment directed against CRC metabolism may add to improved survival outcomes. The bile acid, ursodeoxycholic acid (UDCA), lowers cholesterol and improves insulin sensitivity when used clinically. Moreover, UDCA protects against intestinal polyp development in animal and human studies. UDCA effects may be mediated through activation of FXR, since bile acids are the endogenous ligands for this nuclear receptor. FXR is a tumor suppressor in mouse models of colon cancer and a prognostic biomarker in patients with CRC. Activation of FXR with bile acids in mouse models suppresses intestinal tumorigenesis. We propose to determine the dose and safety of UDCA when added to standard therapy for mCRC. Methods: Trial Design: UDCA doses will be escalated using a 3+3 design which will stop at the MTD or at two levels above an Active Dose (AD), whichever is the lowest. The AD is defined as the dose that will activate FXR. After a 7 day run-in period of UDCA only, cytotoxic combination chemotherapy and antiangiogenic agent for mCRC will be given along with the UDCA. The use of a run-in period allows for metabolic studies to identify serum biomarkers of FXR activation and to determine an AD. Treatment or intervention planned: UDCA is given during a run-in period of 7 days. At the end of the run-in period, standard doses of 5-fluorouracil, leucovorin, oxaliplatin, and bevacizumab are added. Patients are treated until excessive toxicity, progression of disease, or surgical resection. Major eligibility criteria: Patients with mCRC are eligible. Patients with greater than 10% body weight loss within 6 months of entering the protocol, a diagnosis of diabetes and active diabetes treatment are excluded from the study. Results: Current enrollment: Cohorts 1, 2, 3 have been completed without DLT. Enrollment to Cohort 4 began in 12/2011. Conclusions: Continue dose escalation until MTD or 2 levels above AD.Clinical trial registry number: NCT00873275.

Phase I trials of PS-341 (bortezomib, B) in combination with topotecan (T) in advanced solid tumors: A potential pharmacokinetic (PK) interaction

3108

Background: Preclinical data has demonstrated that B when combined with camptothecins enhances apoptotic effects which may be independent of NF-κB status and dependent on the sequence of exposure to the proteasome inhibitor. Other data suggests PK interactions of B with some chemotherapeutic agents. It is hypothesized that B will act synergistically with T and enhance the apoptotic effects of T. Conflicting data exists regarding schedule dependent synergistic cytotoxicity of the combination. This study examined both sequences of administration and determined the PK of each sequence. Methods: A standard 3 + 3 schedule was used in both sequences (S). The dose of B was fixed with escalating doses of T. S1: B 1.3 mg/m2 on days 1, 4, 8, and 11. T 2 to 3.5 mg/m2 on days 1 and 8 was given 6 h before B. S2: B 1.3 mg/m2 on days 1, 4, 8, 11 followed by T 2 to 3 mg/m2 given 6 h after B. Cycle length was q28 days. PK sampling for both sequences was performed. Results: 54 pts (S1: 27, S2: 27) received a median of 2 cycles (range 1-7) of therapy. Pts were heavily pre-treated. For S1, the maximum tolerated dose (MTD) of B and T were 1.3 and 3 mg/m2 respectively. For S2, the MTD of B and T were 1.3 and 2.5 mg/m2 respectively. Dose-limiting toxicities (2 pt in each sequence) were gr 3 thrombocytopenia in S1, and gr 4 thrombocytopenia in S2. Stable disease was seen in 6/27 pts in S1, and 6/27 in S2. PK results for S2 show no significant difference between the day 1 and day 8 PKs. However, the mean AUC (454±150 mg/Lxhr) and Cmax (123±51 mg/L) at the MTD of 2.5 mg/m2 reached on this trial is similar to those reported with single agent doses of 4 mg/m2 given over 30 minutes weekly x 3. As a result, the estimated topotecan clearance (11.3±5 L/hr) when given following bortezomib is approximately one half of the reported clearance for the single agent. (Curtis et al, J Clin Pharmacol, 50:268, 2010). Conclusions: The tolerability and toxicity of both sequences was similar. MTD differs with S2 having a lower tolerated dose of T. PK results suggest a possible drug-drug interaction with decreased clearance of T with S2. PK results for S1 are pending. Supported by CCSG CA62505.

Phase I pharmacokinetic study of dasatinib (BMS-354825) in patients with advanced malignancies and varying levels of liver dysfunction: S0711, a SWOG early therapeutics committee study

3078

Background: Dasatinib (D) is a first in class Src kinase inhibitor, and inhibits BCR-Abl, c-Kit, PDGFR-beta, EPHA2 and Src family kinases including Src, Lck, Yes, Fyn at nanomolar concentrations. Initially FDA approved for use in imatinib resistant CML. It is a small molecule targeted therapy hepatically metabolized primarily by CYP3A4. We conducted a phase I study to determine maximum tolerated dose (MTD) and pharmacokinetics (PK) of D in patients (pts) with liver dysfunction (LD). Methods: Pts with advanced solid tumors or lymphoma, Zubrod ≤2, no baseline ascites or pleural effusions, adequate renal and bone marrow function, received PO D daily. Cycles q28 days. Pts stratified into 4 LD groups: normal, mild, moderate, severe, using Child-Pugh classification (CPC). Data also collected for NCI ODWG Organ Dysfunction Working GroupCriteria. D dose was escalated in sequential cohorts of pts within each LD category. Blood analysis for D concentrations were determined during cycle 1 using a validated LC-MC/MS assay. Study objectives included characterizing safety, tolerability, PK, identifying the MTD and obtaining preliminary evidence of efficacy. Results: 54 registered pts, 51 pts received 51 cycles of D at doses starting at 100 mg in mild LD (50-140 mg). Median age 60, male 55%, Zubrod 1 70%. CRC 27%. Groups: normal-17, mild-20, moderate-13, severe-1 pt(s). Related AEs include fatigue 35%, diarrhea 27%, anemia 27%, nausea 25%, vomiting 21%, lymphopenia 13%, rash 13%, pleural effusion 8%. 1 DLT of increased CK in a pt in moderate LD with past history of similar episode with previous sorafenib. Previous linear PK disposition, and no accumulation. No apparent PK differences between normal and mild groups. Cycle 1 Day 1 D 140 mg Normal Group: Cmax 129 ng/mL. Mild Group 140mg: Cmax 157 ng/mL. Prolonged disease stabilization (≥4 cycles) in 6 pts, 3 CRC (4,5,8); 1 Pancreas, HCC, Bladder (4,5,6). Conclusions: Recommended dose for Dasatinib given PO QD for pts with mild, moderate, or severe LD using clinical criteria with CPC and no baseline ascites, are 140 mg, 70 mg, insufficient pts, respectively. Dose adjustment not necessary in pts with mild LD.

Phase I trial of gene-mediated cytotoxic immunotherapy combined with resection for pancreatic adenocarcinoma

241

Background: While surgical resection of pancreatic adenocarcinoma provides the only chance of cure, long-term survival is still poor. Immunotherapy approaches may improve outcomes. Gene Mediated Cytotoxic Immunotherapy (GMCI) generates a systemic anti-tumor vaccine effect through intra-tumoral delivery of an adenoviral vector expressing the HSV-thymidine kinase gene (AdV-tk) followed by anti-herpetic prodrug administration. This is the first application of GMCI in pancreatic cancer.

Methods: This study evaluated 4 dose levels of AdV-tk (3x1010 to 1x1012 vector particles) injected into pancreatic tumors via EUS 2 weeks before resection. Patients then underwent attempt at resection. If resection was undertaken, AdV-tk was injected into the resection bed. If resection was not possible, AdV-tk was injected into the primary tumor. The prodrug, Valacyclovir, was given for 14 days after each injection. Postoperative therapy was not protocol-driven.

Results: The study accrued 14 patients with 12 completing therapy: 3 at each of the 4 planned dose levels. One patient died of an unrelated myocardial infarction 2 days after initial injection and one patient dropped out mid-treatment after metastases were found at surgery. Median age was 67 years (range 40-81). Of 12 patients explored, 4 were not resected due to distant metastases (N=3) or locally advanced disease (N=1). Three patients had Grade 3 possibly-related adverse events: 2 abdominal pain and one dehydration with renal insufficiency. There were no dose limiting toxicities and no grade 4 clinical adverse events. Grade 3-4 laboratory abnormalities were AST/ALT, bilirubin, alkaline phosphatase and lipase, all in patients with obstructive jaundice. Post-operative complications included 2 patients who developed abscesses requiring drainage. Six of 12 patients are alive 5-34 months after start of treatment including 5 resected patients and one unresected.

Conclusions: AdV-tk can be safely injected into potentially resectable pancreatic tumors prior to planned resection. Early results are encouraging and justify further evaluation in a Phase 2 study.

Prenatal buprenorphine exposure and sexually dimorphic nonreproductive behaviors in rats

This set of experiments examined the effects of prenatal buprenorphine (BUP) exposure on three measures of sexual differentiation in rats. Pregnant female rats were divided into four treatment groups: 0.6 mg/kg BUP, 0.3 mg/kg BUP, a pair-fed control (PFC), and an untreated control (UTC). Drugs were injected starting on gestation day (GD) 6 and continuing through GD 20 with a 48-h interval between drug administrations. Three variables were examined in the offspring: anogenital (AG) distance on postnatal day (PND) 1, spontaneous parental behavior on PNDs 23-28, and saccharin consumption on PNDs 42-55. Whereas prenatal BUP exposure had no effect on AG distance, spontaneous parental behavior was impaired in the 0.6-mg/kg-exposed offspring on two measures: pup-retrieval latencies and pup-directed behaviors. Furthermore, although both control groups and the 0.3-mg/kg-exposed offspring showed the expected sex difference in consumption of a 0.25% saccharin solution, this difference was not displayed by the 0.6-mg/kg-exposed offspring. These findings suggest that exposure to relatively high doses of buprenorphine during development may have long-term effects on behavior.