A phase II/III randomized study to compare the efficacy and safety of rigosertib plus gemcitabine versus gemcitabine alone in patients with previously untreated metastatic pancreatic cancer

BACKGROUND

Rigosertib (ON 01910.Na), a first-in-class Ras mimetic and small-molecule inhibitor of multiple signaling pathways including polo-like kinase 1 (PLK1) and phosphoinositide 3-kinase (PI3K), has shown efficacy in preclinical pancreatic cancer models. In this study, rigosertib was assessed in combination with gemcitabine in patients with treatment-naïve metastatic pancreatic adenocarcinoma.

MATERIALS AND METHODS

Patients with metastatic pancreatic adenocarcinoma were randomized in a 2:1 fashion to gemcitabine 1000 mg/m(2) weekly for 3 weeks of a 4-week cycle plus rigosertib 1800 mg/m(2) via 2-h continuous IV infusions given twice weekly for 3 weeks of a 4-week cycle (RIG + GEM) versus gemcitabine 1000 mg/m(2) weekly for 3 weeks in a 4-week cycle (GEM).

RESULTS

A total of 160 patients were enrolled globally and randomly assigned to RIG + GEM (106 patients) or GEM (54). The most common grade 3 or higher adverse events were neutropenia (8% in the RIG + GEM group versus 6% in the GEM group), hyponatremia (17% versus 4%), and anemia (8% versus 4%). The median overall survival was 6.1 months for RIG + GEM versus 6.4 months for GEM [hazard ratio (HR), 1.24; 95% confidence interval (CI) 0.85-1.81]. The median progression-free survival was 3.4 months for both groups (HR = 0.96; 95% CI 0.68-1.36). The partial response rate was 19% versus 13% for RIG + GEM versus GEM, respectively. Of 64 tumor samples sent for molecular analysis, 47 were adequate for multiplex genetic testing and 41 were positive for mutations. The majority of cases had KRAS gene mutations (40 cases). Other mutations detected included TP53 (13 cases) and PIK3CA (1 case). No correlation between mutational status and efficacy was detected.

CONCLUSIONS

The combination of RIG + GEM failed to demonstrate an improvement in survival or response compared with GEM in patients with metastatic pancreatic adenocarcinoma. Rigosertib showed a similar safety profile to that seen in previous trials using the IV formulation.

Phase Ib study of mapatumumab in combination with sorafenib in patients with advanced hepatocellular carcinoma (HCC) and chronic viral hepatitis

261

Background: Mapatumumab is a fully human agonist monoclonal antibody to the tumor necrosis factor-related apoptosis-inducing ligand receptor 1 (TRAIL-R1). Sorafenib targets Mcl-1, a key TRAIL resistance protein, and accordingly could enhance mapatumumab's pro-apoptotic activity. Based on this and on preclinical data on mapatumumab in HCC cell lines, the current dose-escalation study is evaluating mapatumumab in combination with sorafenib in patients with advanced HCC.

Methods: Eligible patients had advanced HCC, Child-Pugh A or Model for End-Stage Liver Disease score < 15, and were positive for hepatitis B surface antigen or hepatitis C antibody. Intravenous mapatumumab was administered at 3, 10 or 30 mg/kg every 21 days with sorafenib (400 mg twice daily) until disease progression or unacceptable toxicity. Dose-limiting toxicities (DLTs) included events considered at least possibly related to mapatumumab and/or its interaction with sorafenib. Tumor measurements were performed every 2 cycles. Dose escalation required at least 3 patients in a cohort to receive >= 50% of full-dose sorafenib in the first 2 cycles.

Results: To date, 19 patients have been enrolled in the 3 mg/kg (n=6), 10 mg/kg (n=9) and 30 mg/kg (n=4) cohorts and have received a median of 4 cycles (range 1 to 24 cycles); 4 patients have received >= 11 cycles. The maximum tolerated dose has not been reached. DLTs that prompted expansion of the 3 mg/kg and 10 mg/kg cohorts were elevations of amylase and/or lipase (1 at 3 mg/kg, 1 at 10 mg/kg). Other Grade 3-4 events considered at least possibly related to mapatumumab and/or its interaction with sorafenib included hepatic pain (1/17), thrombocytopenia (1/17), increased aspartate aminotransferase (1/17), increased lipase (1/17), and increased gamma-glutamyltransferase (1/17). Two patients have had a partial response and 4 patients have had stable disease lasting > 12 weeks, based on investigator assessment.

Conclusions: Mapatumumab was well-tolerated at doses up to 30 mg/kg in combination with sorafenib in patients with HCC and viral hepatitis. A randomized phase II study of this combination in patients with advanced HCC is planned.

[Table: see text]

Oncologic outcomes following radical resection with intraoperative electron radiation therapy in patients with recurrent colorectal cancer

508

Background: Of the 40,000 rectal cancer patients treated annually, up to 15% will experience a local recurrence (LR). Resection only series for LR report 30–60% positive margin rates. For this reason our institution utilizes intra-operative electron radiation therapy (IOERT) in patients undergoing resection of locally advanced rectal cancer recurrences (LA-RRC). The purpose of this study was to evaluate the oncologic outcomes of patients treated for a LA-RRC with resection and IOERT

Methods: Under IRB approval a prospective database of patients with LA-RRC undergoing resection and IOERT from April 2001 to January 2010 was analyzed. Demographic, clinical, and pathologic variables were collected. The Kaplan-Meier method estimated 5-year overall survival (OS) and Fisher's Exact test's explored associations for developing a LR.

Results: Fifty patients (32 male, mean age 58 ± 11) underwent resection and IOERT (median dose 1500 cGy, energy 9 MeV). Primary procedures included pelvic exenteration (n = 17), APR (n = 10), LAR (n = 7), isolated pelvic/retroperitoneal tumor (n = 7), para-aortic lymphadenectomy (n = 7), and partial colectomy (n = 2). 56% (n = 28) required multi-visceral resections including partial/total cystectomy (n = 14), small bowel resection (n = 10), vaginectomy (n = 6), hysterectomy (n = 3), oophorectomy (n = 3), nephrectomy (n = 2), and other (n = 4). 34 pelvic sidewall resections and 12 partial sacrectomies were required. Margins were R0 in 37 (74%) and R1 in 13 (26%). Median LOS was 10 days and there were no peri-operative deaths. Five-year OS was 32% with a 19-month follow-up of survivors. 32% (n = 16) developed a pelvic LR, of which 2 were isolated and 14 were in the setting of distant metastasis. No difference was noted in OS (p = 0.5) or LR (p = 0.3) in patients with R0 vs. R1 resections. Initial T (p = 0.1), N (p = 0.2), and stage (p = 0.7) were not found to be associated with LR.

Conclusions: This study reveals that resection combined with IOERT provides good local control and reasonable long-term survival in patients with LA-RRC. The similar OS and LR following an R0 and R1 resections could be interpreted as a measure of the beneficial effect of IOERT on microscopic disease.

No significant financial relationships to disclose.

The role of mTOR and hypoxia pathway activation in promoting tumorigenesis of intrahepatic cholangiocarcinoma

218

Background: Cholangiocarcinoma (CC) is an aggressive malignancy that often presents in advanced stages with poor response rates to conventional chemotherapy. Recent studies of CC have uncovered several oncogenes and tumor suppressors implicated in the transition from biliary hyperplasia to malignancy. We have developed a mouse model of intrahepatic CC that results from deletion of the phosphatase and tensin homolog (PTEN) gene and the von Hippel Lindau (VHL) gene, a tumor suppressor that regulates hypoxia inducible factor (HIF) expression. Immunohistochemical (IHC) analysis of the liver tumors revealed strong staining of p-AKT and HIF-1α. This study aims to determine the expression of the same downstream targets p-AKT and HIF1α in a dataset of human CC tumors.

Methods: Archived samples from 33 CC patients were sectioned and stained via IHC for p-AKT and HIF1α expression. Samples were evaluated using a proportion score (PS) for neoplastic cells on a scale of 0 to 5, and an intensity score (IS) expressed on a scale of 0 to 3. Normal liver samples were used as controls.

Results: Using a combined score of at least 5 to determine positive expression, 18% of the 33 evaluated samples displayed positive p-AKT expression and 61% displayed positive HIF1α expression. 15% of cases displayed concomitant expression of both p-AKT and HIF1α, while 39% of evaluated samples had expression of neither protein.

Conclusions: Delineating the molecular pathways that lead to CC formation is crucial in developing novel therapeutics for a disease in drastic need of more effective therapies. Our group has developed a mouse model of intrahepatic CC that suggests a synergistic role of PTEN and VHL in the tumorigenesis of CC with upregulation of p-AKT and HIF1α. A modest percentage of a cohort of human samples demonstrates a similar expression pattern, potentially defining a distinct molecular subtype of this cancer with implications for tumor behavior and response to treatment. Future studies will correlate these molecularly-defined subtypes to the clinicopathologic characteristics of the tumors with the ultimate aim of establishing biomarkers of disease prognosis as well as potential predictive indicators of treatment activity.

No significant financial relationships to disclose.

Cost-effectiveness of adjuvant therapy for treatment of stage II colon cancer in patients with diabetes

421

Background: The decision to treat stage II colon cancer patients with adjuvant chemotherapy involves assessment of life expectancy, risk for recurrent disease, and the potential benefit and likelihood of adverse effects from treatment. This is a challenging task, particularly for patients with pre-existing chronic illnesses such as diabetes, which may increase side effects and potentially lessen response to chemotherapy.

Methods: We use the Archimedes Model to analyze cost effectiveness of adjuvant therapy in stage II colon cancer patients with pre-existing diabetes. The Archimedes Model is a large-scale, integrated mathematical model of human physiology, diseases, and healthcare systems, including pathways relating to diabetes, cardiovascular disease, and cancers of the breast, lung, and colon. The colon cancer model is built using the SEER, ACCENT databases and meta-analyses of clinical trials. Costs relating to colon cancer, diabetes and its complications are based on Medicare reimbursement rates. We simulate a trial in which stage II colon cancer patients are subjected to two treatment strategies: no treatment vs. adjuvant chemotherapy by FOLFOX regimen. We report incremental cost-effectiveness ratio (ICER), measured by cost per quality-adjusted life year (QALY) gained, of adjuvant therapy compared with no treatment.

Results: Cost effectiveness is strongly dependent on a patient's tumor profile, age and duration of diabetes. For instance, adjuvant therapy saves ∼0.2 QALYs per person in stage IIA patients who are 75 and older and have been diagnosed with diabetes >10 years, at an ICER of >$150,000/QALY gained. In contrast, it saves 1.1 QALYs per person in stage IIB patients aged 60-65, recently diagnosed with diabetes, at an ICER of <$30,000/QALY gained. Results are sensitive to assumptions on efficacy and side effects of chemotherapy in diabetic patients, as well as cost of adjuvant therapy.

Conclusions: The current study suggests that the decision to proceed with adjuvant chemotherapy requires careful assessment of severity of diabetes in stage II colon cancer patients with pre-existing diabetes.

No significant financial relationships to disclose.

Phase I study of the combination of everolimus (RAD001) with 5FU/LV in patients with refractory solid malignancies

512

Background: mTOR is a controller of cellular growth and other processes that is activated by several oncogenic pathways in cancer. Everolimus is an oral inhibitor of mTOR activation. The mTOR inhibitor, temsirolimus, when combined with high dose 5FU, was limited by significant mucosal toxicity. Purpose: Phase I study to determine the maximum tolerated doses (MTD) and the safety of everolimus combined with 5FU/LV in patients with refractory solid tumors.

Methods: Using a standard 3+3 design, starting doses were everolimus 15mg weekly, and q2 weekly 5FU 400mg/m2 bolus and 1800mg/m2 infusion (over 46 hours) and LV 400mg/m2. Daily dosing of everolimus was instituted during level 3 as data emerged about improved target inhibition and adverse event profile with daily dosing. Dose escalation to the maximum planned levels was achieved: everolimus 10mg daily, 5FU 400mg/m2 bolus and 2400mg/m2 infusion, and LV 400mg/m2. Dose limiting toxicities (DLT) were assessed in cycle 1 (4 weeks), defined as any grade 3/4 non-hematologic toxicity (except grade 3 skin rash, nausea and diarrhea), or complicated grade 3-4 heme toxicity. Tumors were measured every 8 weeks.

Results: From 03/2008 and 11/2009, 21 patients were treated. Median age 58 (range 35-77), male/female 13/8, PS 0/1 10/11. One DLT was seen at dose level 1 (grade 3 hypersensitivity/angioedema); two at dose level 6 (grade 3 diarrhea, grade 3 hypophosphatemia). As such, dose level 5 is the MTD. Dose reductions were required in 4 patients for mouth sores (2), nasal sores (1), and grade 3 transaminitis (1). Responses are as follows: 11 stable disease and 8 progressive disease; 2 patients were not evaluable.

Conclusions: MTD is everolimus 5mg daily, with q2 weekly 5FU 400mg/m2 bolus and 2400mg/m2 infusion and LV 400mg/m2. We are currently expanding the study to evaluate the addition of oxaliplatin and panitumumab to the 5-FU/LV and everolimus base.

[Table: see text]

[Table: see text]

A phase II study of AZD6244 in advanced or metastatic hepatocellular carcinoma

e15574

Background: HCC is a deadly disease with few effective treatments. It is characterized by frequent MEK/ERK activation in the absence of RAS or RAF mutation. AZD6244 is a specific inhibitor of MEK that can be administered orally. Methods: Multi-center, single-arm study with a two-stage design utilizing RR as primary endpoint. Patients had advanced or metastatic HCC not amenable to surgical therapy or who failed locoregional therapy. Pts with Childs-Pugh A and compensated B were eligible. No prior systemic therapy was allowed. A pre-treatment biopsy was performed for p-ERK analysis. AZD6244 was administered orally at a dose of 100 mg bid. Results: 19 patients were enrolled, completing the first planned stage. One patient withdrew during cycle 1 with grade 2 AEs and is not evaluable. One patient is too early for evaluation. Another was removed from study for toxicity prior to evaluation. Two patients exhibited clinical progression and were removed from study without imaging (both assigned response of PD). Of 16 evaluable patients, PR + CR was 0%, SD 37.5%, PD 62.5%. Toxicities were generally low grade and included nausea (75%), transaminitis (63%), Alkaline phosphatase (50%), vomiting (50%), fatigue (44%), diarrhea (44%) and skin rash (44%). Grade 3–4 toxicity included transaminitis (38%), hyperbilirubinema (25%), and hypokalemia (13%). Median TTP was 8 weeks (95% CI: 6.6, 11.1). Conclusions: The study has been stopped at the interim analysis due to lack of radiographic response. Time to progression is not suggestive of significant prolonged stable disease. AZD6244 is well tolerated, but appears to have minimal activity in advanced HCC. Analysis of ERK phosphorylation status in biopsy samples is ongoing.

No significant financial relationships to disclose.

Diabetes mellitus affects response to neoadjuvant chemoradiotherapy in the management of rectal cancer

INTRODUCTION

Although diabetic patients with rectal cancer have poorer outcomes than their nondiabetic counterparts, few studies have looked at diabetics' response to therapy as an explanation for this disparity. This study compares the neoadjuvant chemoradiotherapy (CRT) response in diabetic and nondiabetic patients with locally advanced rectal cancers.

METHODS

This is a single-institution, retrospective review of rectal cancer patients who received CRT followed by resection from 1995 to 2006. Pretreatment tumor-node-metastasis (TNM) staging was determined using endorectal ultrasound, computed tomography (CT) scan, and magnetic resonance imaging (MRI); post-treatment staging was determined by pathological review.

RESULTS

110 patients were included; seventeen had diabetes and 93 were nondiabetics. Pretreatment staging was similar in both groups. Sixteen of the diabetics (94%) completed CRT compared to 92% (86/93) of the nondiabetics. Tumor downstaging rates were similar in the two groups (53% in diabetics, 52% in nondiabetics). Nondiabetic patients had a higher rate of nodal downstaging although not statistically significant (67% versus 27%, P = 0.80). While none of the diabetics patients achieved a pathologic complete response (pCR), 23% (21/93) of the nondiabetics did (P = 0.039). Local progression rates were higher in the diabetic group (24% versus 5%, P = 0.046).

CONCLUSION

Our study shows that neoadjuvant chemoradiotherapy in rectal cancer is less effective in diabetic patients than in nondiabetics. While minimal differences are found in the rate of downstaging, the rate of achieving a complete pathologic response was significantly higher in nondiabetic patients, and in fact was not seen in any of our diabetic patients. This may explain the poorer outcomes seen in diabetic patients with rectal cancer.

Unusually high rates of hypersensitivity to cetuximab in the middle-southern United States: Association with atopic phenotype

9051

Purpose: To confirm the anecdotal observation that pts in North Carolina (NC) and Tennessee (TN) treated with cetuximab (C-mab) experience hypersensitivity reactions (HSR) at a much higher rate than that reported nationally and internationally (= 3%). Methods: Data from pts treated with C-mab on clinical trials (CTs, n = 88) at 3 research sites in TN and NC were analyzed for grade 3 or 4 HSR. Additional information was obtained from medical records under IRB approval at the University of North Carolina (UNC) for 90 pts (including 35 pts on CTs) to determine whether history of other significant allergy was a risk factor for HSR. Prior allergy was defined as history of H1 antagonist use, or notation in the medical record of significant drug allergy (excluding narcotic allergies), bee sting allergy, eczema, allergic reactive airways disease, or food allergy. Pearson's or Fisher's exact tests were used for comparison of proportions. Results: Data for 88 pts on CTs and an additional 55 pts treated outside of CTs was included in this analysis. Pts had a variety of tumor types, the most frequent being colorectal and lung For the CT group (n = 88), the overall rate of grade 3 to 4 HSR was 21.6%, significantly higher than the rate noted in any large published trial (p < 0.0001 for comparison to data from Cunningham et al). All HSRs occurred during the first dose. There was no apparent protection from HSR afforded by premedication with dexamethasone in either the CT cohort or the overall UNC cohort. Upon examination of medical records of 90 pts treated at UNC (14.4% gr 3/4 HSR overall), there was a strong relationship between prior allergy history and chance of HSR; 10/28 pts with any HSR vs. 7/62 without allergy history (p = 0.006), and grade 3/4 HSR occurred in 8/28 versus 5/62, respectively (p = 0.01). Conclusions: At the sites in neighboring Southern US states studied, HSR was far more common than reported in national studies. History of prior allergy is a strong predictor of HSR, although is not specific enough to be used as the sole means of excluding pts from therapy. Further investigation of more specific predictors of HSR in the US Middle South is warranted, and pts being treated with C-mab in the region should be observed particularly closely during their first infusion.

No significant financial relationships to disclose.

Detection of shared MHC-restricted human melanoma antigens after vaccinia virus-mediated transduction of genes coding for HLA

To detect shared human melanoma Ag that are recognized by HLA-A2 restricted, melanoma-specific CTL derived from tumor infiltrating lymphocytes, we have developed a convenient method to insert and express foreign HLA genes capable of presenting Ag on target cell lines. Seventeen melanoma cell lines and 11 nonmelanoma cell lines were infected with recombinant vaccinia virus containing the HLA-A2.1 gene. Infection by the vaccinia virus resulted in expression of functional HLA-A2 molecules on the cell surface of virtually 100% of infected cells within a 3.5-h period. The results showed that 11 of 17 (65%) naturally HLA-A2- melanoma cell lines were specifically lysed by the HLA-A2-restricted, melanoma-specific TIL after infection with the vaccinia-HLA-A2.1 virus. None of the nine human nonmelanoma cell lines tested (three colon cancer, four breast cancer, or two immortalized non-tumor cell lines) or two murine melanoma cell lines were lysed by the HLA-A2-restricted TIL after vaccinia-HLA-A2.1 infection. Coinfection of the vaccinia virus containing the beta 2-microglobulin gene with the vaccinia-HLA-A2.1 virus increased the surface expression of HLA-A2 and subsequent lysis by melanoma-specific tumor infiltrating lymphocytes. With this new method we could extend previous findings demonstrating that shared melanoma Ag recognized by HLA-A2-restricted tumor infiltrating lymphocytes exist among melanoma cells from different patients regardless of HLA type. These Ag represent excellent candidates for the development of vaccines to induce T cell responses for the immunotherapy of patients with melanoma.

Detection of shared MHC-restricted human melanoma antigens after vaccinia virus-mediated transduction of genes coding for HLA

To detect shared human melanoma Ag that are recognized by HLA-A2 restricted, melanoma-specific CTL derived from tumor infiltrating lymphocytes, we have developed a convenient method to insert and express foreign HLA genes capable of presenting Ag on target cell lines. Seventeen melanoma cell lines and 11 nonmelanoma cell lines were infected with recombinant vaccinia virus containing the HLA-A2.1 gene. Infection by the vaccinia virus resulted in expression of functional HLA-A2 molecules on the cell surface of virtually 100% of infected cells within a 3.5-h period. The results showed that 11 of 17 (65%) naturally HLA-A2- melanoma cell lines were specifically lysed by the HLA-A2-restricted, melanoma-specific TIL after infection with the vaccinia-HLA-A2.1 virus. None of the nine human nonmelanoma cell lines tested (three colon cancer, four breast cancer, or two immortalized non-tumor cell lines) or two murine melanoma cell lines were lysed by the HLA-A2-restricted TIL after vaccinia-HLA-A2.1 infection. Coinfection of the vaccinia virus containing the beta 2-microglobulin gene with the vaccinia-HLA-A2.1 virus increased the surface expression of HLA-A2 and subsequent lysis by melanoma-specific tumor infiltrating lymphocytes. With this new method we could extend previous findings demonstrating that shared melanoma Ag recognized by HLA-A2-restricted tumor infiltrating lymphocytes exist among melanoma cells from different patients regardless of HLA type. These Ag represent excellent candidates for the development of vaccines to induce T cell responses for the immunotherapy of patients with melanoma.

T-cell recognition of human melanoma antigens

The adoptive transfer of tumor-infiltrating lymphocytes (TILs) with interleukin-2 (IL-2) has antitumor activity in some patients with metastatic melanoma. We have analyzed molecular mechanisms of TIL recognition of human melanoma. Some cultured TILs specifically lysed autologous and some allogeneic melanomas sharing a variety of class I major histocompatibility complex (MHC) molecules. HLA-A2-restricted melanoma-specific TILs lysed many HLA-A2+ melanoma cell lines from different patients but failed to lyse HLA-A2- melanoma and HLA-A2+ nonmelanoma cell lines. However, these TILs were capable of lysing many naturally HLA-A2- melanomas after introduction of the HLA-A2.1 gene by vaccinia virus. These results indicate that shared melanoma antigens (Ag) are expressed in melanomas regardless of their human leukocyte antigen types. In order to identify these shared melanoma Ags, we have tested some known proteins expressed in melanoma. Expression of tyrosinase or HMB45 Ag correlated with lysis of TILs. We are also attempting to isolate antigenic peptides by high performance liquid chromatography separation and genes encoding melanoma Ag by cDNA expression cloning. The T-cell component of the antimelanoma response was also analyzed by determining the genetic structure of the T-cell receptor (TCR) used by melanoma TILs. However, we did not observe common TCR variable region usage by different melanoma TILs. We could establish melanoma cell clones and lines resistant to TIL lysis due to the absence of or defects in the expression of Ag, MHC, or beta 2-microglobulin molecules. These data indicate multiple mechanisms for melanoma escape from T-cell immunosurveillance.(ABSTRACT TRUNCATED AT 250 WORDS)