Search

Summary Background Margetuximab, a novel, investigational, Fc-engineered, anti-HER2 monoclonal antibody, is designed to more effectively potentiate innate immunity than trastuzumab. We aimed to evaluate the safety, tolerability, and antitumour activity of margetuximab plus pembrolizumab (an anti-PD-1 monoclonal antibody) in previously treated patients with HER2-positive gastro-oesophageal adenocarcinoma. Methods CP-MGAH22–05 was a single-arm, open-label, phase 1b–2 dose-escalation and cohort expansion study done at 11 academic centres in the USA and Canada and 15 centres in southeast Asia (Korea, Taiwan, and Singapore) that enrolled men and women aged 18 years or older with histologically proven, unresectable, locally advanced or metastatic, HER2-positive, PD-L1-unselected gastro-oesophageal adenocarcinoma, with an Eastern Cooperative Oncology Group performance status of 0 or 1, who had progressed after at least one previous line of therapy with trastuzumab plus chemotherapy in the locally advanced unresectable or metastatic setting. In the dose-escalation phase, nine patients were treated: three received margetuximab 10 mg/kg intravenously plus pembrolizumab 200 mg intravenously every 3 weeks and six received the recommended phase 2 dose of margetuximab 15 mg/kg plus pembrolizumab 200 mg intravenously every 3 weeks. An additional 86 patients were enrolled in the phase 2 cohort expansion and received the recommended phase 2 dose. The primary endpoints were safety and tolerability, assessed in the safety population (patients who received at least one dose of either margetuximab or pembrolizumab) and the objective response rate as assessed by the investigator according to both Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1, in the response-evaluable population (patients with measurable disease at baseline and who received the recommended phase 2 dose of margetuximab and pembrolizumab). This trial is registered with ClinicalTrials.gov , NCT02689284 . Recruitment for the trial has completed and follow-up is ongoing. Findings Between Feb 11, 2016, and Oct 2, 2018, 95 patients were enrolled. Median follow-up was 19·9 months (IQR 10·7–23·1). The combination therapy showed acceptable safety and tolerability; there were no dose-limiting toxicities in the dose-escalation phase. The most common grade 3–4 treatment-related adverse events were anaemia (four [4%]) and infusion-related reactions (three [3%]). Serious treatment-related adverse events were reported in nine (9%) patients. No treatment-related deaths were reported. Objective responses were observed in 17 (18·48%; 95% CI 11·15–27·93) of 92 evaluable patients. Interpretation These findings serve as proof of concept of synergistic antitumour activity with the combination of an Fc-optimised anti-HER2 agent (margetuximab) along with anti-PD-1 checkpoint blockade (pembrolizumab). Funding MacroGenics.

Purpose: The oxaliplatin plus S-1 and cisplatin plus S-1 regimens are interchangeably used in the management of advanced gastric cancer. The previously reported G-intestinal (G1) and G-diffuse (G2) intrinsic gene expression signatures showed promise for stratifying patients according to their tumor sensitivity to oxaliplatin or cisplatin. Experimental Design: The proof-of-concept, multicenter, open-label phase II “3G” trial was done to prospectively evaluate the feasibility and efficacy of using genomic classifiers to tailor treatment in gastric cancer. Patients’ tumors were classified as “G1” or “G2” using a nearest-prediction template method, or “G3” (unclear assignment) when FDR ≥ 0.05. The first 30 patients in the “G1” cohort were assigned oxaliplatin plus S-1 (SOX) chemotherapy; thereafter, subsequently recruited “G1” patients were treated with cisplatin plus S-1 (SP) chemotherapy. “G2” patients and “G3” patients were treated with SP and SOX chemotherapy, respectively. Results: A total of 48, 21, and 12 patients, respectively, were given “G1,” “G2,” and “G3” genomic assignments. Median turnaround time was 7 days (IQR, 5–9). Response rates were 44.8%, 8.3%, 26.7%, and 55.6% for the “G1-SOX,” “G1-SP,” “G2,” “G3” cohorts, respectively; and was higher in G1 patients treated with SOX compared with SP (P = 0.033). Exploratory analyses using the genomic classifier of Lei and colleagues validated the utility of the metabolic signature as a biomarker for predicting benefit from chemotherapy (log-rank P = 0.004 for PFS), whereas the Asian Cancer Research Group classifier did not demonstrate any predictive value. Conclusions: This bench-to-bedside effort establishes a reasonable turnaround time for gene expression profiling and possible utility of genomic classifiers in gastric cancer treatment stratification. Clin Cancer Res; 24(21); 5272–81. ©2018 AACR.

Aim The survival benefit of using a non-cross resistant second-line chemotherapy in the third-line setting in metastatic gastroesophageal cancer is unproven. We evaluated the utility of third-line chemotherapy in patients treated at a single institution. Methods Between 2010 and 2014, efficacy and toxicity data of patients who received three or more lines of systemic therapies for metastatic gastroesophageal adenocarcinoma at the National Cancer Centre Singapore was retrospectively analyzed. Results Thirty-two (6%) patients received three or more lines of chemotherapy. The median age and ECOG performance status were 59 years (36-82) and 1 (0-2), respectively. Majority of patients (88%) had tumor located in the stomach and 13 patients (41%) had diffuse histology or poorly cohesive or signet ring cells. Four (12%) patients had HER2-positive disease. Prior therapy was platinum (100%), fluoropyrimidine (97%), taxane (63%), irinotecan (28%), anthracycline (13%) and ramucirumab (3%). Third-line therapy consisted of 24 (75%) monotherapy, 6 (19%) doublet, 1 (3%) triplet chemotherapy and 1 (3%) clinical trial. Monotherapy irinotecan (44%) was most common, followed by docetaxel (19%) and paclitaxel (9%). Of 22 patients evaluable for response, there was 1 (5%) partial response, 9 (41%) stable disease. Median overall survival was 18.3 weeks (4.3-65.1). Of 30 patients evaluable for toxicities, 17 (57%) experienced at least one grade 3 or 4 toxicities. Conclusion The benefit of using non-cross resistant second-line regimens as third-line chemotherapy was small with moderate toxicity. Newer agents such as nivolumab or TAS-102 or clinical trial may be preferred.

Aim To evaluate the safety and efficacy of the combination therapy of fluorouracil, leucovorin and irinotecan (FOLFIRI) and aflibercept in Asian patients with metastatic colorectal cancer (mCRC), who had progressed after oxaliplatin-based chemotherapy. Methods This is a retrospective analysis of 19 mCRC patients who received FOLFIRI and aflibercept (4 mg/kg intravenously) every 2 weeks via a Named Patient Program (supported by Sanofi Aventis) in Singapore. Treatment was administered until disease progression or unacceptable toxicities. Kaplan–Meier method was used to estimate progression-free survival (PFS) and overall survival (OS). Efficacy and toxicities were summarized using descriptive statistics. Statistical analysis was performed using STATA 12.0 software. Results The majority (84%) of the patients were of chinese ethnicity. The median age was 59 years, with 63.2% of the patients having an Eastern Cooperative Oncology Group status of 1. Four patients (21.1%) achieved partial response and 8 patients (42.1%) achieved stable disease. After a median follow-up of 9.6 months [95% confidence interval (CI), 2.2–13.1 months], the median OS was 11.6 months (95% CI, 6.1 to not-estimable), and median PFS was 4.1 months (95% CI, 2.2–5.9). Majority of the toxicities were grade 1–2, and include leucopenia (84.2%), anemia (73.7%), liver enzyme elevation (68.4%) and fatigue (68.4%). The most frequently reported grade 3 toxicities were neutropenia and neutropenic complications (both 15.8%). All adverse events resolved with supportive management. Conclusion The clinical benefit and safety profile of the combination of FOLFIRI/aflibercept in Asian patients with mCRC are consistent with that of Western population. FOLFIRI/aflibercept may be an appropriate therapeutic option in Asian patients with mCRC previously treated with an oxaliplatin-based regimen.

730 Background: Tumoral KRAS mutations (KRAS mt) are detected in ~80% of APC and associated with a negative prognosis. Digital droplet PCR (ddPCR) has high sensitivity for circulating KRASmtbut narrower gene coverage. NGS using hybrid-capture methods has reported ctDNA KRASmtin ~30% of patients (pt). We previously presented clinical results of the Phase I trial of OXIRI (GI ASCO 18 #411). We now present the first prospective evaluation of ctDNA in APC by an amplicon-based NGS approach in this Phase I trial. Methods: Paired ctDNA and CA19-9 samples were taken at baseline, C2D1, C3D1 and end of trial. A targeted panel with error-correction (Lucence Diagnostics) was used to detect for mtin KRAS, TP53, SMAD4, CDKN2A, CTNNB1, GNAS, APC and MYC. CT scans were performed every 2 cycles. Survival curves by Kaplan Meier were compared by mutational status, ctDNA and CA19-9 response using the log-rank test. Spearman correlation of ctDNA and CA19-9 changes was performed. Results: ctDNA mtwas detected at baseline in 19/23 (83%) samples, comprising KRAS 17/23 (73%), TP53 (61%), SMAD4 (48%) and CDKN2A (30%). KRAS mtand SMAD4 mt conferred a negative prognosis for overall survival with a hazard ratio of 4.2 (CI: 1.6-10.4; p = 0.01) and 2.8 (CI: 0.9-8.65; p = 0.01) respectively. Drop in ctDNA and CA19-9 was associated with a trend for longer progression free survival at C2D1 (both) and C3D1 (ctDNA only). Radiological partial response (PR) was associated with lower ctDNA in 5/5 pt and CA 19-9 in 4/5 pt. Decrease in ctDNA and CA19-9 was associated with disease control (PR/SD) at C2D1 in 11/14 pt and 10/10 pt; at C3D1 in 11/12 pt and 6/7 respectively. No significant correlation between the amplitude of CA19-9 and ctDNA changes was found. Conclusions: ctDNA could be detected in 83% of pts of whom KRAS mtrates were similar to reports using tissue NGS. Determination of RAS mtand SMAD4 mtin ctDNA may aid in the prognostication of pts and decrease in ctDNA levels may predict for treatment benefit, similar in extent to CA 19-9. This may be particularly useful in non-CA19-9 secreting APC as an adjunct to imaging. Clinical trial information: NCT02368860.

65 Background: Trastuzumab (T) + chemo is standard first-line therapy (tx) for HER2+ gastroesophageal adenocarcinoma (GEA) pts, though progression ensues in 6-8 months. The approved second-line tx is ramucirumab +/- paclitaxel (R+PAC). Pts with GC are less responsive to R+PAC than gastroesophageal junction (GEJ) pts, in particular HER2+ GC, and no anti-HER2 agents are approved in post-T setting. We report results of combination M+P in HER2+ GC pts and describe a biological rationale for this population. M is an anti-HER2 mAb Fc optimized for enhanced binding to activating FcgRIIIa (CD16A) and decreased binding to inhibitory FcgRIIb (CD32B). M demonstrated an enhanced Fc-dependent MoA, including enhanced ADCC. Methods: HER2+, PD-L1-unselected, second-line GEA pts post T progression received M (15 mg/kg) + P (200 mg) Q3wk. Safety, objective response rate (ORR), median overall & progression-free survival (mOS, mPFS), disease control rate (DCR), circulating tumor DNA, & tumor PD-L1 expression were assessed. Results: To date, 66 GEA pts were dosed; 35 (53%) GC and 31 (47%) GEJ. Overall, 12/66 (18.2%) had tx-related adverse events ≥ grade 3; 5 had drug-related SAEs: dehydration, diabetic ketoacidosis, hypotension and pneumonitis, each a single event, and 2 events of autoimmune hepatitis. Eligibility was based on archival HER2 expression; an exploratory endpoint measured retention of HER2 expression post-T by ERBB2 ctDNA. HER2 expression was lost in 23/56 (41.1%) of pts tested post T. HER2 retention was higher in pts with GC versus GEJ (65.8% vs. 44.8%) and in GEA pts with IHC 3+ vs 2+ archival tumors (61.7% vs 47.4%, respectively). Furthermore, GC had higher PD-L1 expression than GEJ, 53.3 vs. 33.3%, respectively. This coincided with more responses in IHC3+ GC pts, ORR 12/29 (41.4%; 95% CI 23.5-61.1), DCR 21/29 (72.4%; 95% CI 52.8-87.3), mPFS 5.5 months (95% CI 2.3-7.6), mOS not reached, with lower bound of 9.1 months for 95% CI. Enrollment of an additional 25 pts enriched for IHC3+ GC is ongoing. Conclusions: Results suggest that M+P, a chemo-free regimen, demonstrates acceptable tolerability and has encouraging preliminary activity in second-line HER2+ GEA, specifically in GC pts who retain ERBB2 amp prior to second-line tx. Clinical trial information: NCT02689284.

4030Background: T + chemo is standard 1st line therapy (tx) for HER2+ GEA patients (pts). However, pts typically progress within 6-8 months, with up to 30% demonstrating loss of HER2 positivity. No...

411 Background: The FOLFIRINOX regimen is highly active in APC, but its use is limited by toxicity. Irinotecan (IR) toxicity is correlated with UGT1A1 genotype status and chronomodulated delivery of fluoropyrimidines may reduce toxicity and increase efficacy. We have presented the dose escalation (DES) of OXIRI. In dose expansion (DXP), several pts required dose reduction after cycle 1 and we evaluated a lower starting dose. We now present data of the recommended Phase II dose (RP2D) and DES combined. Methods: Pts with APC and ECOG performance status ≤2, adequate organ function were enrolled. Previous OX or IR exposure was disallowed. During DES, APC pts homozygous for UGT1A1*6/*6 or UGT1A1*28/*28 were excluded. During DXP, only pts progressing on ≥ 1 line prior chemo was allowed and IR was dosed according to UGT1A1 genotype. At RP2D, pts received iv OX 37.5mg/m2 and IR 50mg/m2 on D1, 8 and po capecitabine 2650mg once at midnight on D1-14 every 21D until progression. Results: A total of 25 pts were enrolled, 17 in DES and 8 in DXP, 21 had metastatic and 4 had locally advanced APC. Male:Female 11:14, median (range) age and ECOG PS was 65yrs (50-77) and 0 (0-1) respectively. Pts with 0, 1 or 2 lines of prior chemo were 4 (16%) , 14 (56%) and 7(28%) respectively. Grade 3/4 AEs (≥5%) related to OXIRI were neutropenia (32%), anaemia (12%), diarrhoea (16%). No events of febrile neutropenia were reported. At data cut-off on 31 Aug 2017, 4 pts are on study. Median progression-free and overall survival was 22 and 35 wks respectively. Response rate (RR) was 5/25 (20%, overall); 5/20 (25%, evaluable pts) including 1 complete response and 1 unconfirmed partial response. Disease control rate for ≥ 12wks was 16/25 (64%) with 3 pts > 52wks. CA19-9 decrease of ≥ 20% was seen in 9/25 pts. Conclusions: OXIRI is an active regimen with RR of 20% despite 84% of pts having prior chemo, with disease control for ≥ 12wks in 64% of pts. Predominant toxicity was neutropenia. This study was supported by the National Medical Research Council Singapore. Clinical trial information: NCT02368860.

140 Background: Trastuzumab (T) + chemotherapy (ctx) is standard for 1st line advanced HER2+ GEA, yet subsequent targeted options are lacking. M is an anti-Her 2 monoclonal antibody with an optimized Fc domain to increase affinity for activating CD16A Fc-receptors (FcR) on NK cells. Outcomes for T-treated patients (pts) carrying the low-affinity CD16A-F allele are generally worse than pts homozygous for the high-affinity V allele. M is designed to be FcR genotype independent. Evidence of clinical activity of M alone has been seen in HER2+ GEA pts post T, while P has demonstrated durable activity. Loss of HER2 amplification may occur after T failure in a subset of initially HER2+ GEA pts. Preclinical studies suggest that engagement of innate and adaptive immunity with the combination of anti-HER-2 antibodies and T-cell checkpoint inhibition could achieve greater antitumor activity than either agent alone. Methods: Advanced HER2+, PD-L1-unselected GEA pts post T failure were eligible. Dose escalation evaluated 10 and 15 mg/kg M and 200 mg P for 2nd line or higher pts. Cohort expansion evaluates safety and objective response rate (ORR) by RECIST v1.1 in 2nd line pts. M + P is given every 21 days; response assessed every 6 weeks. HER2 amplification status was assessed in a subset of pts by plasma circulating tumor (ct) DNA analysis prior to Cycle 1 of M+P. Results: Dose escalation enrolled 9 pts; cohort expansion 48 pts at 15 mg/kg M: 30 in North America (NA) and 18 in Asia (A). Treatment was well tolerated, with 1 drug-related serious adverse event. Of 38 evaluable pts to date in expansion (24 NA and 14 A), the best overall responses include 7 pts (18.4%) with PR (4 confirmed and 3 unconfirmed) and 11 (28.9%) with SD. Higher ORR trends were observed in A vs NA (35.7% vs 8.3%) and G vs GEJ (31.6% vs 5.3%). Of 25 pts with ctDNA results, HER2 amplification detection was higher in GC than GEJ (80% vs 53%). Responses were independent of FcR genotype; CD16A genotype for evaluable pts with PR: 1 V/V, 2 V/F, 2 F/F with similar allelic distribution among non-responders. Conclusions: M+P is a well-tolerated ctx-free regimen that has shown preliminary antitumor activity in 2nd line pts with advanced/metastatic GEA. Clinical trial information: NCT02689284.

The use of trastuzumab, a monoclonal antibody targeting the HER2 protein, in combination with 5-fluorouracil/platinum-based chemotherapy improves survival in patients with HER2-positive advanced gastric cancer. In addition, TS-one (S-1)/platinum is also used as a standard of care in Asian countries. However, little is known about the combination of S-1/cisplatin chemotherapy and trastuzumab in patients with HER2-positive advanced gastric/gastroesophageal junction (GEJ) cancer. We conducted a single-arm, two-stage, open-label, multicenter phase II study. Trastuzumab was administered intravenously on day 1 of the first cycle at 8 mg/kg and 6 mg/kg on day 1 of subsequent cycles. Cisplatin was administered intravenously at 60 mg/m2 on day 1 of each cycle after trastuzumab. S-1 was administered orally [based on body surface area (BSA)] twice a day for 14 days in a 3-weekly cycle. Patients with BSA of

There is currently no standard salvage chemotherapy regimen in relapsed and refractory lymphoma. Gemcitabine is a novel nucleoside analogue, which acts synergistically with cisplatin both in vitro and in clinical studies. We evaluated the combination of gemcitabine, cisplatin and methylprednisolone (GEM-P) in 41 heavily pretreated patients with relapsed and refractory Hodgkin's and non-Hodgkin's lymphoma. The best-achieved response rate (RR) was 79% (95% CI 64–91), with a complete RR of 21%. In patients with chemo-resistant disease, the RR was 63%. Myelosuppression was the main toxicity, the incidence of Grade 3 or 4 anaemia, neutropenia and thrombocytopenia was 17.1, 61.0 and 53.7% respectively. Only one patient had neutropenic sepsis and none of the patients suffered from haemorrhage. Grade 3 or 4 nonhaematological toxicity was minimal and stem cell mobilisation was not inhibited. GEM-P is an effective salvage regimen and its use prior to autologous stem cell transplant warrants further investigation.

48 Background: Theoxaliplatin/S-1 (SOX) and cisplatin/S-1 (SP) regimens are interchangeably used in the management of advanced gastric cancer (AGC). We previously developed a classification tool using gene expression signatures which successfully stratified gastric cancer cell lines and primary tumour samples according to their sensitivity to SOX or SP ( Tan et al, Gastroenterology 2011). We now report the first prospective study to evaluate the feasibility and efficacy of using a genomic classifier to tailor treatment in this setting. Methods: Pts with histologically-confirmed locally-advanced, metastatic and recurrent GC were recruited from 3 centres in Singapore and South Korea. Tumours were analysed using the Affymetrix HG-U133 Plus 2.0 array and results were used to classify pts as G1 (oxaliplatin-sensitive), G2 (cisplatin-sensitive) and G3 (status unclear or gene expression not available). G1 and G2 pts were matched to SOX and SP regimens respectively, while G3 pts were assigned SOX. The primary endpoint was best overall response; secondary endpoints were turnaround time and biomarker analyses. Results: Between July 2, 2010, and Apr 2, 2015, we screened 85 AGC pts. 74 pts received at least 1 cycle of treatment and were evaluable for analysis. Median turnaround time was 7 working days (IQR, 5–9). The misclassification rate was 6%. After an initial 30 pts in the G1 subgroup were treated with SOX, subsequent pts ( N = 13) classified as G1 received the SP regimen. The ORR were 44.8%, 8.3%, 26.7% and 55.6% for G1 SOX, G1 SP, G2 SP ( N = 19), G3 SOX ( N = 12) respectively; and was higher in G1 pts treated with SOX compared with SP ( P = 0.033). Post hoc genomic reclassification based on Lei et al (Gastroenterology 2013) confirmed the utility of the metabolic subtype as a predictive marker of benefit from chemotherapy (log rank P value for PFS = 0.004). Conclusions: This bench-to-bedside effort establishes that molecular profiling to direct choice of conventional chemotherapy for AGC is possible within a reasonable timeframe. The clinical utility of our genomic classifiers in question are promising but warrant further investigation. Clinical trial information: NCT01100801.

We examined the cardiotoxicity in 153 patients treated with capecitabine and oxaliplatin in two prospective trials for advanced colorectal cancer. Ten patients (6.5%) developed cardiac events. One patient (0.7%) had sudden death, one patient developed cardiac failure with raised troponin I while another developed ventricular tachycardia (VT). The remaining seven patients (4.6%) experienced angina and three of the seven patients had raised troponin I, one of which developed ventricular fibrillation. Eight events occurred within cycle 1 (median cycle 1 day 10). Four patients with angina and one patient with VT recovered on stopping capecitabine, four patients required additional medical management and the remaining patient died suddenly at home. Patients with ischaemic heart disease appeared to be at increased risk. Physicians and patients need to be aware of these complications, so that prompt discontinuation of treatment and appropriate interventions may be instituted.

127 Background: The addition of trastuzumab to cisplatin and 5-fluorouracil or capecitabine has been shown to have superior survival benefit. In this study, the addition of trastuzumab to chemotherapy combination, S-1 (tegafur, gimeracil, oteracil potassium; TS-ONER) and cisplatin, a standard first line therapy for advanced gastric cancer in Japan, was evaluated for efficacy and safety. Methods: This was a single-arm, phase II trial recruiting patients with advanced HER-2 positive (IHC 3+ or FISH +) gastric adenocarcinoma who had not had prior systemic therapy. Patients included had measurable disease, ECOG 0-2, creatinine clearance >60ml/min, LVEF at least 50%. Patients accrued received cisplatin at 60mg/m2 on day 1 of each cycle, trastuzumab at 8mg/kg as loading dose followed by 6mg/kg on day 1 of each cycle and oral S-1 for 14 days ( dosing based on BSA), every 3 weeks. Treatment continued till progressive disease or unbearable toxicity. Primary endpoint was response rate (ORR) and secondary endpoints were PFS, OS, CBR and safety. The study was designed to distinguish a favorable ORR of 55% from a null rate of 30% at 5% significance level and 80% power. Results: A total of 30 patients were recruited. Median age was 61.9 years, 73% were males and 90% had no prior resection of primary tumor. 56.7% were Chinese, 23.3% Japanese and 20% Korean. One patient achieved complete response, 16 had partial response and 2 had stable disease >24 weeks (ORR 63%; CBR 70.4%). Three patients did not have evaluable disease. With a median follow-up of 7.2 months, the median PFS was 7.4 months with 67.7% free of disease progression at the 6-month mark. Median OS was 14.6 months. The most common AEs were anorexia (50%), fatigue (47%), diarrhea ( 47%), nausea (37%), neutropenia (33%), anemia (27%) and mucositis ( 27%). LVEF decreased

2608 Background: AR-12 (OSU-03012) is an oral celecoxib analogue lacking COX-2 inhibitory activity that inhibits pyruvate dehydrogenase kinase isoenzyme 1 (PDK-1), AKT and impacts the endoplastic reticulum stress pathway. Preclinical studies indicate antitumor activity of AR-12 in various models and enhanced activity in combination. We completed a first in human clinical trial to determine its safety and tolerability, maximum tolerated dose (MTD) and recommended phase II dose (RD). Secondary objectives included assessment of tumor response, pharmacokinetics (PK) and pharmacodynamics (PD) including food effect. Methods: Patients (pts) with advanced solid tumors, ECOG PS 0-1, and adequate organ function were recruited in a modified 3+3 dose-escalation study. Pts received a run-in dose of AR-12 to analyze PK-PD and food effect, followed by continuous daily (QD) dosing in 28-day cycles. A twice daily (BID) cohort was initiated based on safety data. PD analysis was performed in platelet-rich plasma (PRP) and paired tumor biopsies when feasible. Results: 35 pts received at least one dose of study drug; 30 were evaluable for dose limiting toxicities (DLT) at dose ranges 100-3200mg QD and 800-1600mg BID. No DLT were observed in the QD cohort; DLT in the BID cohort are listed in table 1. Drug-related events (NCI-CTCAE v3) included rash (G2-2pts; G3-1pt), fatigue (G2-2pts; G3-4pts), nausea (G2-7pts; G3-1pt) and bloating (G2-1pt). Cmax after single dose was dose-proportional but high PK variability was observed, likely due to inadequate disintegration and dissolution of the formulation in the stomach. At RD, partial GSK3ß inhibition in PRP after 4 hours suggests AKT-pathway modulation. Best response (RECIST v1.0) was stable disease >6 cycles for 2 pts. Conclusions: The RD based on safety data is 800mg BID. Signs of pathway modulation were observed in concordance with the expected mechanism of action but were short-lasting. Considering limited drug absorption and PK variability, a new formulation of the drug will be developed to overcome these limitations. Clinical trial information: NCT00978523. [Table: see text]