Introduction: Pancreatic cancer (PC) is a leading cause of cancer-related deaths globally, with a dismal all-stage 5-year survival rate of 8%. Systemic chemotherapy can prolong survival and palliate symptoms in patients with metastatic PC (MPC). The phase III MPACT trial demonstrated superiority of first-line nab -paclitaxel plus gemcitabine vs gemcitabine monotherapy across all endpoints, including the primary endpoint of overall survival (OS) in patients with MPC. Promising results have been observed with regimens containing nab -paclitaxel and 5-fluorouracil (5-FU); however, toxicity is a major concern with high-dose intermittent therapy. The goal of this study is to determine whether metronomic therapy (continuous, low-dose treatment) with FABLOx can reduce the associated toxicities without affecting efficacy. Methods: Patients (aged 18 - 65 years) with previously untreated, histologically or cytologically confirmed MPC and an ECOG performance status of 0 or 1 were eligible. Patients with preexisting peripheral neuropathy grade > 1 were excluded. Phase I of the study assessed potential dose-limiting toxicities (DLTs) to determine the recommended phase II dose (RP2D); phase II was designed to evaluate efficacy and further assess safety of the RP2D. During the dose-determining phase, a minimum of 6 patients were to be enrolled in each consecutive dosing cohort, and 5-FU, nab -paclitaxel, and oxaliplatin doses were to be de-escalated to the next lower dose if ≥ 2 of 6 patients experienced a DLT in cycle 1 (to a maximum of dose level −2). Patients received 5-FU 180 mg/m 2/day on days 1 to 14; nab -paclitaxel 75 mg/m 2 on days 1, 8, and 15; bevacizumab 5 mg/kg on days 1 and 15; leucovorin 20 mg/m 2 on days 1, 8, and 15; and oxaliplatin 40 mg/m 2 on days 1, 8, and 15 of each 28-day cycle in cohort I (the starting dose level). The primary endpoint of phase I was incidence of DLTs, and the secondary endpoint was safety. Patients were treated until disease progression, unacceptable toxicity, withdrawal of consent, physician decision, or death. Results: Two DLTs were observed in 1 patient (grade 3 anemia requiring a transfusion and grade 3 mucositis unresponsive to medical treatment within 4 days of onset); cohort I was expanded from 6 to 12 patients. The median age was 57.5 (range, 41 - 64) years; 3 women and 9 men were enrolled, and 3 patients continue to be followed up for survival analysis. Patients received a median of 7.0 treatment cycles of nab -paclitaxel, bevacizumab, leucovorin, and 5-FU and 5.5 cycles of oxaliplatin. The evaluation of grade ≥ 3 treatment-emergent adverse events is pending. The objective response rate was 42% (5 partial responses); the median PFS and OS were 5.6 and 11.5 months, respectively. Conclusion: Results from phase I of this study demonstrate that metronomic FABLOx treatment is feasible for patients with MPC. Preliminary data suggest promising antitumor activity with the regimen. NCT02620800.

ABSTRACT Aim: In a retrospective study of 18 pts with unresectable (UR) or borderline resectable (BR) LAPC, neoadjuvant therapy with FOLFIRINOX with or without subsequent chemoradiation (CCRT) resulted in an R0 resection rate (RR) of 44% (Hosein et al, BMC Cancer 2012). The reported 1-year progression-free survival (PFS) was 83 % and the 1-year overall survival (OS) was 100 %. Toxicity profile was tolerable. In order to confirm these preliminary results, we analyzed a large cohort of pts treated in a similar fashion with mature follow-up. Methods: Between 2008 and 2013, 51 treatment-naive pts with LAPC were treated with first-line FOLFIRINOX with neoadjuvant intent. Pts were categorized as BR or UR using the NCCN criteria. Pts received FOLFIRINOX chemotherapy (at the full dose as described in the ACCORD-11 trial) until maximum response or tolerability, and then underwent surgery if their imaging suggested resectability. Pts then received CCRT if they were still UR or BR after FOLFIRINOX. The end points of this retrospective analysis were OS, PFS, R0 RR and toxicity profile. Results: A total of 429 cycles were given with a median of 8 (range 2-29); 27 (53%) went on to receive CCRT. After a median follow-up of 17 mo (range 2-56), the Kaplan-Meier median OS was 35 mo (95% CI 26-45), the 3-yr OS rate was 42% and the median PFS was 14 mo (95% CI 11 – 16). By imaging criteria, 13 (26%) were converted to resectability and 10 (4 BR and 6 UR) of these had successful R0 resections. Pts who had R0 resections had a significantly longer survival than pts who did not (3-yr OS rate 67% vs 21%, log rank p = 0.042). Grade 1&2/3&4 chemotherapy-related toxicities were neutropenia (39%/20%), neutropenic fever (0%/12%), thrombocytopenia (53%/16%), anemia (63%/10%), fatigue (76%/6%), nausea (57%/4%) vomiting (22%/4%), neuropathy (53%/4%) and diarrhea (37%/10%). Conclusions: FOLFIRINOX followed by chemoradiotherapy is feasible as neoadjuvant therapy in patients with unresectable LAPC. Although the resection rate was only 20%, the median OS of almost 3 years is appreciably longer than historical survival rates for this population. Prospective controlled trials testing this algorithm in LAPC are ongoing. Disclosure: All authors have declared no conflicts of interest.

4620 Background: In a phase I study, a single dose of 90Y-labeled anti-mucin humanized antibody, hPAM4 (90Y-hPAM4), led to several transient reductions or stabilization of lesions in advanced pancreatic cancer, with bone marrow toxicity limiting the maximum tolerated dose to 20 mCi/m2. Preclinical studies showed gemcitabine enhanced radioimmunotherapy, so a phase Ib study was undertaken to evaluate repeated treatment cycles of 90Y-hPAM4 plus gemcitabine. Methods: Patients (pts) with previously untreated, locally advanced or metastatic, pancreatic cancer were treated in 4-week cycles (200 mg/m2 gemcitabine once-weekly; 111In-hPAM4 the 1st wk for imaging, biodistribution, and dosimetry; 90Y-hPAM4 once-weekly the last 3 wks), which could be repeated in the absence of progression or unacceptable toxicity. The 90Y-dose was escalated by patient cohort following a 3+3 design, with tumor responses assessed by CT and FDG/PET imaging, and by CA19.9 serum levels. Results: Eight pts (3F/5M, 56–72 years old, 7 with metastatic disease) have now been treated at the first 2 dose levels (6.5 and 9.0 mCi/m2 90Y-hPAM4 x 3) with hematologic toxicity all transient Grade 1–2 (NCI CTC v3). 111In-hPAM4 imaging showed normal biodistribution, evidence of tumor targeting and acceptable dosimetry estimates to normal organs per treatment cycle. Two pts had tumor responses to initial treatment with significant decreases in FDG metabolic activity on PET imaging, regression of lesion sizes on CT, and CA19.9 decreases. Both pts continue in excellent performance status now at 9 and 11 months after study entry, after receiving a total of 3 and 4 treatment cycles, respectively, without additional toxicity. A 3rd pt with a stable response by PET and CT 4 weeks after initial treatment and decreases in CA19.9 levels is now undergoing a 2nd treatment cycle. Four other pts had early progression of disease by or before post-treatment week-4 evaluation, and the remaining pt is still being evaluated. Conclusions: Dose escalation is continuing after fractionated radioimmunotherapy with 90Y-hPAM4 plus low-dose gemcitabine demonstrated therapeutic activity at the first two 90Y dose levels, with minimal hematologic toxicity, even after 4 treatment cycles. [Table: see text]

Objective: Gastrointestinal stromal tumors (GISTs) have been recognized as the most common mesenchymal tumor of the GI tract. New effective chemotherapies have been defined for the treatment of GIST but have not been validated by phase III trials. We sought to determine both current population-based incidence and if improved outcomes noted in both individual centers and clinical trials have also been observed in a large prospective cancer registry. Materials and Methods: The Surveillance, Epidemiology and End Results (SEER) database thirteen-center cumulative tumor registry (April 2005 release) was queried from 1992-2002 to determine incidence and associated outcomes for patients diagnosed with GIST. Confirmation of incidence trends were confirmed using the incident Florida Cancer Data System, which is the largest non-SEER registry in the US (FCDS, 2005 release). Results: A twentyfivefold age-adjusted increase in incidence of GIST from 0.028 per 100000 in 1992 to 0.688 per 100000 in 2002 was observed. This increase is mostly due to reclassifying smooth-muscle tumors as GISTs but also represents a fifty percent increase in population and age-adjusted gastrointestinal mesenchymal tumor diagnosis since 1992. Despite rising GIST incidence rates, there was a marked improvement in survival since 2000 coinciding with the introduction of the tyrosine kinase inhibitor imatinib into clinical practice. Conclusions: The diagnosis of GIST has dramatically increased since 1992. Survivals have dramatically improved since 2000. Proper diagnosis is critical to identify those patients who will benefit from adjuvant imatinab chemotherapy. This is the first report to demonstrate a population-based marked survival benefit for patients with GIST coincident with the introduction of imatinib.

17061 Background: Platinum-based doublets are used as treatment for advanced or metastatic non-small cell lung cancer (NSCLC), but chemotherapy must be tailored to decrease side effects. Oxaliplatin is more potent than cisplatin, requiring fewer DNA adducts to provide equivalent cytotoxicity in vitro studies. Oxaliplatin was active as a single agent and in combination with vinorelbine, paclitaxel, and gemcitabinein phase II studies of patients with NSCLC. A phase II study was conducted to evaluate the efficacy and safety of oxaliplatin combined with docetaxel for NSCLC. Methods: Patients with stage-IIIB or -IV, chemotherapy-naive NSCLC received docetaxel 70 mg/m2, oxaliplatin 130 mg/m2, and pegfilgrastim 6 mg every 21 days for up to 6 cycles. The primary endpoint was overall response rate (ORR); secondary endpoints were progression-free and overall survival (PFS and OS), and safety. Results: Twenty-nine patients were treated; 15 (51.7%) were women, 25 (76%) were white, 17 (58.6%) were hispanic, 21 (72.4%) had adenocarcinomas, 24 (83%) had a PS ECOG 1, 93% had stage-IV disease and 28% had brain metastases. There were 10 partial responses in 27 evaluable patients for an ORR of 37% (90% confidence interval [CI], 21.7%–54.7%). Median PFS for 29 treated patients was 4.6 months (95% CI, 2.6–6.5 months); 12-month PFS was 14.8% (95% CI, 3.4%– 34.0%). Median OS was 10.9 months (95% CI, 8.9–16.8 months); 12-month OS was 40% (95% CI, 18.5%–60.8%) and 18-month OS was 16% (95% CI, 1.4%–45.7%). There were no unusual or unexpected adverse events. The most common grade-3 and -4 toxicities were anemia (14% of patients) and hyperglycemia (10%). There were only 2 reports of neutropenia; both were grade 1 or 2. Conclusions: These phase II findings suggest that the combination of oxaliplatin and docetaxel is active and well tolerated, and offers a feasible treatment alternative for patients with advanced or metastatic NSCLC. [Table: see text]

661 Background: The incorporation of G into DNA enhances cleavage complexes in vitro when combined with a topo I inhibitor. Topo I poisons require enzyme interaction with DNA to exert activity. Methods: Two stage accrual design, primary endpoint: response (RR) using RECIST criteria. Inclusion criteria: male and female patients (pts) with MBC, prior anthracycline therapy, measurable disease, ECOG PS of ≤ 2, adequate organ function, and ≤ 3 prior chemotherapy regimens for MBC. 51 eligible pts received therapy with G at 1000mg/m2 and I at 100mg/m2 on days 1 and 8 of a 21-day cycle. Optional tumor biopsies were obtained in 9 pts (18%) prior to therapy to determine localization of topo I using immunofluorescence. PK: Irinotecan: A validated limited sampling strategy was used. Gemcitabine: Serial blood samples were collected over 24 hrs following the first dose. Intracellular nucleotides were quantitated in PBMCs. Results: 45 pts have been evaluated with a RR of 27% (CR=0, PR=12; 95% CI 13–37%). 4 pts had SD for ≥6 months for a clinical benefit rate (PR+SD) of 36%. 3 pts received < 1 cycle of therapy before protocol withdrawal and were not evaluable for RR. RR for the final 3 patients will be available at the time of presentation. 7/9 tissue biopsies were assessable for topo I with results listed below. PK and toxicity data will be available at presentation.Conclusion: GI is active in MBC. Topo I localization can be measured in MBC. In this limited data set, the two lowest nuclear to cytoplasmic (N/C) ratios were associated with lack of response to irinotecan. Further validation is needed. [Table: see text] [Table: see text]

17088 Background: Irinotecan (I) has an active role and potential as a radiosensitizing agent in patients with NSCLC. SWOG 9504 established the concept of “consolidation” chemotherapy with 3 cycles of docetaxel (D) after chemo/radiation (CRXT). We evaluated the efficacy and safety of administering weekly doses of carboplatin/irinotecan (CI) concomitantly with radiation therapy followed by D chemotherapy for patients (pts) with stages IIIA/B NSCLC. Methods: We have enrolled 23 pts, treatment included: C: (AUC = 2) and I: 30 mg/m2, weekly concomitant with radiation therapy, and D: 75 mg/m2 every 3 weeks for 3 times after CRXT was finished. The daily administered dose of radiation was 1.8 Gy, 5 days a week for 5 weeks, 25 fractions, (45 Gy) to the primary tumor and mediastinum (primary planning target volume: PPTV. After 45 Gy, the primary tumor and involved nodal metastasis (secondary planning target volume: SPTV) was boosted at 2 Gy per day to 18 Gy in 9 fractions. The total dose given was 63 Gy in 35 fractions over seven weeks. Evaluation of response has been done with RECIST criteria. Results: Median age is 55 years (range: 42–78), 18 (78%) of the pts are female, 17 (74%) are white, 13 (57%) are Hispanic, and 14 (65%) had an ECOG 1. Most common histologies are poorly differentiated and squamous cell carcinomas: 14 pts (61%), and half of the pts are stage IIIB. 111 weeks of CI and 39 cycles of D have been administered and we have documented 22 grade 3/4 adverse events (AE) among them: 4 pts pneumonia (17%), 3 pts radiation pneumonitis (13%), 2 pts: dehydration o neutropenia o dyspnea (9%) and 1 pt with diarrhea, nausea and vomiting (4.5%). No grade 4 neutropenia, esophagitis or diarrhea was reported. 12 severe AE were reported including: 2 pts with pneumonia (9%), 2 pts with dehydration (9%), and 2 pts with radiation pneumonitis (9%) requiring hospitalization. The rest of SAE were unrelated to therapy. Data for response is available in 18 pts. A partial response was seen in 10 pts (56%), and stable disease in 6 pts (33%). 2 pts are ineligible for response. Median survival (and PFS) has not been reached and it will be presented in the meeting. Conclusions: CI administered with radiation therapy and followed by D is safe and efficacious in the treatment of stage III NSCLC. [Table: see text]

Background: The Trucut Biopsy needle (TCB) has recently been introduced for use in EUS guided biopsy. Although a few small studies have compared EUS FNA with EUS TCB, the accuracy of FNA alone vs. the combination of FNA + TCB has not been evaluated. Aim: To evaluate the safety and accuracy of standard FNA (22-gauge needle) with TCB (19 gauge needle) and FNA + TCB combined. Methods: From October 2003 to November 2004, all patients who were referred for EUS guided biopsy and had lesions 20 mm or greater accessible through the esophagus or stomach were included. Cytology and histological specimens from the FNA and TCB were each reviewed by a different cytopathologist who were blinded to the results of the other technique. No on-site pathology assistance was available for either FNA or TCB. Results: A total of 19 patients were included (11F, 8M), with a mean age of 66 years. A total of 7 patients (37%) had previous biopsy procedures that failed to provide the correct diagnosis. The following sites were sampled: pancreatic mass (3), sub-epithelial gastric mass (3), mediastinal mass (10) and lung (3). More FNA then TCB passes were perfomed with a mean of 3.7 (range: 2-6) for FNA and 2.7 (range: 1-4) for TCB (p= 0.002). In 11 patients (58%), tissue was acquired with the first TCB pass. Fourteen patients were found to have malignant disease and five benign disorders. Overall diagnostic accuracy of FNA, TCB and FNA + TCB were 74%, 90% and 100% respectively. Although the accuracy of the combined techniques was superior to standard FNA alone, there was no statistical difference (p= 0.7). One patient had chest pain and fever three days after mediastinal mass biopsy which resolved with oral antibiotics. Conclusion: In carefully selected patients, both FNA and TCB can be safely performed. Although TCB had a similar accuracy requiring fewer needle passes when compared with FNA, the combination of TCB and FNA had an accuracy of 100% in our series. We conclude that TCB should always be considered for tissue acquisition in addition to FNA in lesions accessible through the esophagus or stomach. This technique may be particularly important for centers where a cytopathologist is not present to evaluate adequacy of FNA samples.

7268 Background: Combining Oxaliplatin and Docetaxel (DOCOX) is a biologically rational and feasible option. Oxaliplatin cross-links DNA strands and has shown more activity in vitro than cisplatin. Docetaxel is a mitotic spindle inhibitor, approved as first-line therapy for metastatic (or advanced) non-small cell lung cancer (NSCLC). Methods: We designed a phase II trial to evaluate the efficacy and safety of DOCOX for first-line treatment of patients (pts) with stage IIIB/IV NSCLC, ECOG PS 0–1. We have enrolled 24 pts out of 30 initially planned. Treatment included: docetaxel 70 mg/m2 followed by oxaliplatin 130 mg/m2 on day 1, and pegfilgastrim 6 mg subcutaneous on day 2, for a maximum of six cycles if there was evidence of response. Results: Median age is 59 years (range 39–70), 12 of the patients are female, 18 are white, 12 are Hispanic. Nineteen pts had an ECOG PS 1. The most common histology is adenocarcinoma (18 pts) and most of them are stage IV: 21/24 (88%). Seven of the 24 pts (29%) presented with brain metastasis and received palliative radiation before enrollment. Data for response is available in 16 patients. A partial response was seen in 6 pts (37%, 90% CI:18–61%), and stable disease in 7 (44%, 90% CI:23–67%). Three pts (19%, 90% CI:5–42%) had disease progression. Four pts are ineligible for response. Seventy-nine cycles of chemotherapy have been administered and we have documented 7 severe (grade 3 and 4) adverse events, 2 of which might be related to this combination: grade 3 diarrhea (docetaxel/oxaliplatin), and grade 3 hypersensitivity (docetaxel). Other grade 3 toxicity reported included dyspnea and edema. No grade 3 or 4 hematologic toxicity was seen. Conclusion: The DOCOX combination is safe and efficacious in the treatment of NSCLC. Final results will be presented in the meeting. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Aventis, Sanofi Aventis, Sanofi Aventis, Sanofi

Purpose: The purpose of this phase I clinical trial was to determine the maximum-tolerated dose and toxicity of CP-609,754 in patients with solid tumors refractory to standard therapies, to determine the cellular effects of CP-609,754 on its molecular target (farnesyltransferase), and to determine the recommended phase II dose (RP2D) of this agent.Experimental Design: Consenting patients with adequate bone marrow, liver, and renal function were enrolled with an accelerated dose strategy with single-patient parallel cohorts in whom the drug was given orally either once or twice daily. Once a dose-limiting toxicity was encountered or two patients developed Common Toxicity Criteria 2: grade 2 toxicities, a modified Fibonacci sequence was initiated. Blood samples were collected during cycle 1 for pharmacokinetic and pharmacodynamic analyses.Results: A total of 68 cycles of CP-609,754 was administered to 21 patients enrolled in this study. The dose escalation was from 20 mg once daily to 640 mg twice per day, and at the highest dose level, one of six patients developed a dose-limiting toxicity of grade 3 neuropathy. The drug was otherwise well tolerated, and the maximum-tolerated dose was not reached because of the large number of tablets that would have been required for additional dose escalation. Pharmacokinetic analyses showed a proportional increase in exposure with dose, rapid oral absorption, and a half-life of similar to3 hours. Pharmacodynamic results predict a 95% maximal inhibition of peripheral blood mononuclear cell farnesyltransferase activity 2 hours postdose, on average, with a dose of 400 mg twice per day of CP-609,754.Conclusions: On the basis of the safety findings and the pharmacokinetic and pharmacodynamic analyses, the RP2D of CP-609,754 is greater than or equal to640 mg twice per day.