RECIST Progression Research Articles

Predictive and prognostic value of early response assessment using 18FDG-PET in advanced non-small cell lung cancer patients treated with erlotinib

[18F]fluorodeoxyglucose (FDG)-PET is being evaluated as a tool for the early detection of response to various targeted agents in solid tumors. The aim of this study was to evaluate the predictive value of PET response after 2 days of erlotinib in unselected pretreated patients with stage IV NSCLC. FDG-PET/CT scans were conducted at baseline and after 2 days of erlotinib, with a CT evaluation performed at baseline and after 45-60 days of therapy. PET responses were evaluated by quantitative changes on SUVmax tumor/non-tumor ratio and classified according to EORTC criteria. PET responses were compared with RECIST responses and related to progression-free (PFS) and overall (OS) survival. Erlotinib effects on glucose uptake were also studied in a panel of NSCLC cell lines. Fifty-three patients were enrolled. At 2 days of erlotinib, 20 (38 %) patients showed partial metabolic response (PMR), 25 (47 %) had stable metabolic disease (SMD) and 8 (15 %) had progressive metabolic disease (PMD). All patients with PMD had confirmed RECIST progression at 45-60 days. Patients with early PMR and SMD had significantly longer PFS (p < 0.001 and p = 0.001, respectively) and OS (p = 0.001 for both) than PMD patients. FDG-PET assessment after 2 days of erlotinib could be useful to identify early resistant patients and to predict survival in unselected NSCLC pretreated population.

EGFR tyrosine kinase inhibitors beyond focal progression obtain a prolonged disease control in patients with advanced adenocarcinoma of the lung

IntroductionRecent data show that EGFR pathway and its inhibition maintain their role after progression of disease during EGFR TKI therapy in NSCLCs. We conducted a retrospective study with the aim of evaluating efficacy and feasibility of prosecution of EGFR TKI therapy beyond focal progression associated to locoregional radiotherapy. MethodsWe retrospectively analyzed the data of all NSCLC patients treated with EGFR TKIs in our institution from 2004 to 2012. We included in the analysis patients that after a focal disease progression, meant as a single lesion RECIST progression, have been treated with definitive locoregional radiotherapy, associated to continuation of EGFR TKI therapy until further progression. Results15 out of 147 patients (10%) satisfied inclusion criteria. The median progression free survival, measured from the date of focal progression until further progression of disease or death by any cause, was 10,9 months (range 3–32 months). The corresponding 6 and 12 months PFS rates were 73% and 33%, respectively. ConclusionThe longer disease control observed in our patients suggests that continuation of EGFR TKI beyond focal progression associated to a locoregional treatment is an efficacious therapeutic strategy.

Volumetric tumor growth in advanced NSCLC patients (pts) with EGFR mutations during EGFR-TKI therapy: Developing criteria to define slow progression.

e19125 Background: EGFR-mutated advanced NSCLC pts treated with EGFR TKI typically progress after an initial tumor response. Most pts continue TKI beyond RECIST progression, and an objective guide for treatment decisions is needed. We analyzed the volumetric tumor growth in these pts as an initial step to develop criteria for slow progression to aid therapeutic decision making. Methods: The study population included 58 advanced NSCLC pts with sensitizing EGFR mutations, treated with first-line single-agent gefitinib or erlotinib between 2002-2011, who had baseline CT showing measurable lung lesion and at least two follow-up CTs while on TKI and experienced volumetric tumor growth. Tumor volume (mm3) of the dominant lung lesion was measured on baseline and all follow-up chest CT scans during therapy, using volume analysis software [Nishino et al. Acad Radiol. 2011]. A total of 405 volume measurements from nadir to the end of TKI therapy or last follow-up, with data closure on 6/1/12, were analyzed in a linear mixed effects model, fitting time as a random effect [Laird and Ware, Biometrics, 1982]. Results: Among 58 pts, 46 (79%) were female, median age was 62 (range: 35-84), 29 (50%) were never-smokers, 53 (91%) were stage IV at diagnosis, and 53 (91%) received erlotinib. The median time on TKI was 15.8 months. The median time to tumor nadir was 6.3 months. A linear mixed effects model was fitted to predict growth of the logarithm of tumor volume (logeV), adjusting for time in months from baseline. The growth rate of logeV, obtained as the regression coefficient for time, was 0.12/month (SE: 0.015; p&lt;0.001; logeV=0.12*months+7.68). The model provided a reference value for the volumetric tumor growth rate in EGFR-mutant NSCLC pts after they have achieved their nadir. Conclusions: Tumor volume analysis defined volumetric tumor growth after the nadir in EGFR-mutant advanced NSCLC pts receiving TKI. This provides a reference value for the tumor growth rate in pts progressing after the nadir on TKI. Based on these data which can be studied in additional cohorts, we are currently developing practical radiographic criteria to help define patients as slow progressors who can safely remain on EGFR TKIs.

Continuation of sunitinib following RECIST progression on first-line sunitinib.

4585 Background: In the last seven years the FDA and the EMA have approved seven agents for treatment of RCC. Five of these target the VEGF pathway. Methods: We conducted a detailed analysis of data from the sunitinib registration trial examining the growth and regression rate constants and the stability of the growth rate as measures of effectiveness and to understand development of resistance. Results: Sufficient data was available for the analysis of 350/374 patients enrolled. Statistically valid data was obtained in 321(91.7%). The median regression rate constant was 0.0048 days-1, and in 59 patients no evidence of growth was recorded while on study, only regression. The median growth rate was 0.00082 days-1 and this rate was stable a median of 267 days, remaining stable beyond 300 days in 172 patients, beyond 600 days in 95 patients, and beyond 900 days in 49 pts. A suggestion of a possible increase of the growth rate while sunitinib was administered could be discerned in only 15/321 pts. With a median growth rate 0.00082 days-1 the estimated time to progression were sunitinib discontinued and then re-started would have been a minimum of 7.3 months. Thus a meaningful outcome could be achieved provided continued sunitinib is tolerable. Finally with an estimated 47%, 27% and 13% of tumor still sensitive to sunitinib 100, 200 and 300 days after starting therapy, shrinkage with a new TKI in patients who discontinue sunitinib before day 300 for toxicity may not be a sign of non-cross resistance, but of residual sensitive tumor. Conclusions: Prolonged stability of the growth rate of RCC on sunitinib is consistent with intrinsic and not acquired resistance. Baring toxicity, continued sunitinib beyond RECIST criteria for progression may provide a beneficial outcome and can be considered a treatment alternative in selected patients. Randomized trials to assess the value of VEGF TKI’s in patients whose disease has “progressed” on sunitinib should consider including an arm that continues sunitinib to test this hypothesis.

Advanced Hepatocellular Carcinoma: Early evaluation of response to targeted therapy and prognostic value of Perfusion CT and Dynamic Contrast Enhanced-Ultrasound. Preliminary results

PurposeTo investigate whether there is any correlation between standard endpoints and tumor perfusion measurements with Perfusion CT and Dynamic Contrast-Enhanced Ultrasonography (DCE-US) in patients with advanced Hepatocellular Carcinoma (HCC) treated with targeted therapy. Materials and methodsNineteen patients were evaluated during targeted therapy (sorafenib n=16, sunitinib n=3). Changes in tumor perfusion measurements between baseline and month 1 were assessed and compared using RECIST progression criteria at month 2. ResultsMedian time to progression according to RECIST was 117 days and median time to death was 208 days. Perfusion CT values before treatment were significantly increased in HCC compared to the surrounding liver (n=17, P<.02). Eleven patients received complete examinations with both techniques at baseline and month 1. A non-significant decrease was found in all Perfusion CT values between RECIST nonprogressors (n=7) and progressors (n=4): mean Blood Volume: −27.9 vs. −11.1% and mean Blood Flow: −25.0 vs. −11.7% respectively. With DCE-US, opposite changes were found (mean Area Under the Curve AUC: −38.3 vs. 436.3%). RECIST progression at month 2 was significantly correlated with a threshold 40% decrease in AUC (P=.015). None of the patients with a decrease in AUC≥40% was a progressor at month 2. ConclusionDespite perfusion changes with both Perfusion CT and DCE-US in patients receiving treatment, only DCE-US at month 1 (with a decrease in the AUC of more than 40%) predicted non-progression at month 2 and may be a potential surrogate marker of tumor response during targeted therapy.

Radiographic assessment and therapeutic decisions at RECIST progression in EGFR-mutant NSCLC treated with EGFR tyrosine kinase inhibitors

PurposeAdvanced NSCLC harboring epidermal growth factor receptor (EGFR) mutations treated with EGFR tyrosine kinase inhibitors (TKIs) typically progresses after initial response due to acquired resistance. TKIs are often continued beyond progressive disease by RECIST. We investigated the practice of continuing EGFR-TKIs after RECIST-PD via CT findings. MethodsAmong 101 advanced NSCLC patients with sensitizing EGFR mutations treated with first-line EGFR-TKIs, 70 patients had baseline and at least one follow-up CT for retrospective radiographic assessments using RECIST1.1; 56 patients had experienced PD by the data closure date of June 2011. ResultsAmong 56 patients experiencing PD, 82% were female, median age was 63 years, 50% were never-smokers, 57% had distant metastasis, 57% had exon 19 deletion, and 89% were treated with erlotinib. 49 patients (88%) continued TKI therapy beyond retrospectively assessed PD. 31/32 (97%) patients who progressed by an increase in their target lesions continued TKI. 13/16 (81%) patients who progressed by appearance of a new lesion remained on TKI. 5/6 (83%) patients with both increase of target lesions and new lesion at PD continued TKI. Two patients with PD in non-target lesions discontinued therapy at PD. In 49 continuing patients, the median time from retrospectively assessed RECIST-PD to termination of TKI was 10.1 months. Conclusions88% of EFGR-mutant NSCLC patients who progressed on first-line TKI continued therapy beyond RECIST-PD, which is not the single determining factor for terminating TKI in EGFR-mutant NSCLC patients. Additional radiographically defined progression criteria are needed for this population.

1478O - Clinical Benefit with Regorafenib Across Subgroups and Post-Progression in Patients with Advanced Gastrointestinal Stromal Tumor (GIST) After Progression on Imatinib (IM) and Sunitinib (SU): Phase 3 Grid Trial Update

ABSTRACT Background Regorafenib (REG) therapy significantly improved progression-free survival (PFS) in patients (pts) with GIST after failure of both IM and SU (J Clin Oncol 2012; 30: suppl; abstr LBA10008). Results of subgroup and post-progression analyses are presented here. Methods Pts were randomized (2:1) to receive best supportive care plus either REG 160 mg or placebo (PL) po once daily (3 wks on/1 wk off). The primary endpoint was PFS. Upon progression in either arm, unblinding could occur. Pts on PL were eligible for crossover to open-label (OL) REG and progressive pts on REG were allowed to continue REG upon local physician's choice. Results GRID screened 234 pts and randomized 199 (REG: 133, PL: 66), well balanced for baseline characteristics. The PFS primary endpoint (according to RECIST) was met both per central review (median PFS 4.8 mo for REG vs. 0.9 mo for PL), and by investigator assessment (median PFS 7.4 mo for REG vs. 1.7 mo for PL). Exploratory sensitivity analyses demonstrate similar positive impact on PFS across pre-specified subgroups by number of prior systemic therapy, geographical region, age, baseline ECOG score, duration of prior treatment with imatinib, or KIT/PDGFRA mutation. Median overall survival has not been reached in either arm. Taking the double-blind and OL periods together, 188 pts received REG. Post-progression PFS per investigator assessment was 5.0 mo for pts in the PL arm who crossed over to REG (n = 56), and 4.5 mo for pts in the REG arm who continued to receive REG after unblinding (n = 41). Conclusion REG induced significant PFS benefit in GIST pts following resistance to both IM and SU across all pre-specified subgroups. Interestingly, 41 pts on REG (out of 133) whose physicians decided to continue REG after RECIST progression had an additional PFS benefit which was in the same range. This suggests that continuous kinase inhibition after progression may benefit pts by slowing tumor progression and/or that RECIST criteria for progression applied by blind central review have clinical shortcomings. Disclosure P.G. Casali: Consultant/advisory: Bayer, MSD, GSK, Infinity, Novartis, Pfizer, PharmaMar, Sanofi Honoraria: Novartis, Pfizer, PharmaMar Research funds: Amgen Dompe, Bayer, MSD, GSK, Imclone, Infinity, Lilly, Molmed, Novartis, Pfizer, PharmaMar, Sanofi. P. Reichardt: Consultant/advisory: Bayer. Y. Kang: Consultant/advisory: Bayer Research funding: Bayer. J. Blay: Honoraria: Pharmama, Novartis, Pfizer, Roche, GSK Research funding: Pharmama, Novartis, Pfizer, Roche, GSK. P. Rutkowski: Consultant/advisory: Novartis, MSD, BMS Honoraria: Novartis, Pfizer, BMS, MSD, Roche. M. Leahy: Research funding as PI for GRID study at Christie Hospital, Manchester, UK. M. von Mehren Consultant/advisory: Novartis, Pfizer, Astex; Honoraria: Novartis, Pfizer. H. Joensuu Consultant/advisory: Novartis, Boehringer-Ingelheim. A. Le Cesne Honoraria: Novartis, Pfizer, Pharmamar. P. Schoffski: Research funding: Bayer. R. Maki: Consultant/advisory: Bayer, Pfizer, Novartis; Honoraria: Bayer, Pfizer, Novartis; Research funding: Pfizer; Expert testimony: Pfizer. I. Kuss: Employment: Bayer. D. Laurent: Employment: Bayer Stock ownership: Bayer. G.D. Demetri: Consultant/advisory: Novartis, Pfizer, GSK, Roche, Deciphra (uncompensated), Infinity Research funding: Novartis, Pfizer, GSK, Infinity Expert testimony (uncompensated): Infinity, Pfizer, Novartis, Bayer. All other authors have declared no conflicts of interest.

584P - Fdg-Pet Tumor Load Parameters Measured by Total Lesion Glycolysis as Predictors of Treatment Response and Outcome in Metastatic Colorectal Cancer

ABSTRACT Objectives Early metabolic response measured by the change of FDG maximum Standardized Uptake Value (SUVmax) on PET-CT after 1 cycle of chemotherapy in patients with metastatic colorectal cancer (mCRC) can predict negative objective response and is correlated to Overall Survival (OS) (Hendlisz et al. 2011). The aim of this work is to assess the correlation between the change in FDG-PET whole body metabolic tumoral load parameters and the RECIST response, progression free survival (PFS), OS. Methods FDG-PET/CT scans were performed at baseline and on Day 14, in 41 patients with unresectable mCRC undergoing a biweekly chemotherapy regimen. Patients were followed-up for up to 28 months. The change in Functional Tumor Volume (FTV) and Total Lesion Glycolysis (TLG) (TLG = FTV x SUVmean) of all the lesions was calculated using a fixed threshold segmentation algorithm (threshold for segmentation = SUV mean of 30 mm sphere in the right lobe of the liver plus 3 standard deviations) corrected to lean body mass (Wahl et al 2009). Predictive and prognostic value was tested through correlation with RECIST response (using independent dedicated CT), PFS and OS. A cutoff value for TLG responding patients was set at 40% or more decrease in the baseline value. Results 39 patients were evaluable for analysis. The median OS was 20 months (95% CI, 10 to 28), the change in whole body FTV and TLG were significantly related to RECIST response (p-value: 0.015 and 0.0038), to PFS (P-value: 0.009, 0.018) and to OS (p-value: 0.002, 0.0027). Patients with > 40% decrease in TLG have better PFS: median 12 months compared to median 3 months in the rest of the patients [log rank test p-value 0.004; HR 0.31 (95% CI, 0.14 to 0.72)], and a better median OS 27 months compared to 11 months [log rank test p-value 0.05; HR 0.38 (95% CI, 0.14 to 1.04)]. Conclusions Changes in the metabolic tumoral load parameters after one cycle of standard chemotherapy for mCRC are predictors of RECIST response, progression free and overall survival. Disclosure All authors have declared no conflicts of interest.

445PD - Clinical Activity and Pharmacokinetics (PK) of Cabozantinib (XL184) in Patients with Progressive Medullary Thyroid Carcinoma (MTC)

ABSTRACT Background Cabozantinib (cabo) is a potent oral therapy that inhibits MET, VEGFR2, and RET. In a phase 1 study, cabo demonstrated anti-tumor activity in patients with MTC, and a long terminal half-life of 120 hours. We report clinical activity and PK analyses from a Phase 3 study of cabo versus placebo (P) in patients with progressive, unresectable, locally advanced or metastatic MTC. Methods Patients with MTC and documented RECIST progression within 14 months (mo) of screening were randomized 2:1 to receive cabo (140 mg freebase qd, n = 219) or placebo (n = 111). Blood samples for PK were collected on days 1 and 29. Baseline tumor and blood samples were evaluated for RET mutation status. Tumor assessments occurred every 12 weeks, and the primary efficacy measure was progression-free survival (PFS), assessed by an independent radiology committee using RECIST. Results Statistically significant PFS prolongation was observed, with median PFS for cabo of 11.2 mo versus 4.0 mo for P (HR 0.28, 95% CI 0.19-0.40, p Conclusions Cabo significantly prolonged PFS compared to placebo in a patient population with progressive MTC. PK analysis supports administration of cabo as a fixed dose without adjustment for patient covariates. Disclosure M.J. Schlumberger: MJ Schlumberger has been a compensated consultant for Exelixis. M. Brose: Dr. Brose has received research funding and honoraria from Exelixis. M. Shah: Dr. Shah has received research funding from Exelixis. D.R. Miles: D. Miles is an employee and stockholder of Exelixis. L.T. Nguyen: Linh Nguyen is an employee and stockholder in Exelixis. S. Sherman: Dr. Sherman is a compensated consultant to Exelixis. All other authors have declared no conflicts of interest.

Radiographic assessment and therapeutic decisions at RECIST progression in EGFR-mutant NSCLC treated with EGFR tyrosine kinase inhibitors.

7553 Background: EGFR mutated advanced NSCLC treated with EGFR TKIs typically progresses after initial response due to acquired resistance. TKI therapy is often continued beyond RECIST progression (PD). We investigated the frequency of this practice and patterns of RECIST PD via imaging findings, as well as the association between patient characteristics and discontinuation of TKI among patients (pts) who progressed while on TKI. Methods: Among a cohort of 101 advanced NSCLC pts with sensitizing EGFR mutations treated with first-line erlotinib or gefitinib at DFCI, 70 pts treated between 2002 and 2010 had at least two CT scans for retrospective radiographic assessments using RECIST1.1; 56 pts had experienced PD by the data closure date of June 2011. Results: Among 56 pts experiencing PD, 46 (82%) were female, median age was 63 (range 35-79), 28 (50%) were never-smokers, 32 (57%) had distant mets, 32(57%) had exon 19 deletion, and 50 (89%) received erlotinib. 49 pts (88%) continued TKI therapy for at least 2 mos beyond retrospectively assessed PD. 31/32 (97%) pts who progressed by increase of target lesions continued TKI. 13/16 (81%) pts who progressed by new lesion remained on TKI. Two pts with PD in non-target lesions discontinued therapy at PD. 5/6 (83%) pts with both increase of target lesions and new lesion at PD continued TKI. In 49 continuing pts, the median time from RECIST PD to termination of TKI was 10.1 mos (range: 2.2-64.2 mos). 15/49 (31%) pts who continued TKI received additional chemo compared to 0/7 pts who discontinued (Fisher’s p=0.17). Pts who discontinued therapy (n=7) were significantly younger (median 48 yrs) than those who continued TKI at PD (median 64 yrs, Wilcoxon p=0.003). Median OS beyond RECIST PD among those who continued TKI was 31.8 mos (95% CI 15.9- not reached) and though underpowered, this did not appear to be impacted by TTP when adjusted in a Cox model (p=0.84). Conclusions: 88% of EFGR-mutant NSCLC pts who progressed on first-line TKI continued therapy beyond RECIST PD, which is not the single determining factor for terminating TKI in EGFR-mutant NSCLC pts. Additional progression criteria specific to this population are needed to better guide therapeutic decision making.

Biomarker analysis and final results of INT70/09 phase II proof-of-concept study of pazopanib (PZP) in refractory urothelial cancer (UC).

4507 Background: Discouraging results have been achieved with standard and novel compounds in refractory UC. The final results with biomarker analysis of INT70/09 trial with pazopanib are presented. Methods: Pts failing ≥ 1 chemotherapy regimen for metastatic disease underwent PZP 800 mg once daily until PD or unacceptable toxicity. CT scan and PET scan were planned at baseline and q4weeks thereafter. Independent review of all CT scans was made and ≥ 5 RECIST CR+PR were required to conclude for activity according to Simon’s 2-stage design. 50 mL of EDTA blood samples were collected at baseline and q4wks in all pts to analyze plasma VEGF, sVEGFR-1,-2 and -3, c-Kit, IL-6, 8 and 12 by multiplex ELISA plates. Results: 41 pts were enrolled from 02/2010 to 07/2011. 15 pts (37%) had UC of the upper urinary tract. 20/41 pts were treated in 3rd line or beyond. 19 pts (46%) were CDDP-refractory and 22 (54%) had hepatic metastases. 27 (66%) had ECOG PS 1-2. 7 pts (17%) had a confirmed PR, 24 had a SD (76% clinical benefit). 20 pts (49%) had a confirmed necrotic evolution of metastases and/or a decreased SUV at PET consistent with PR. Median PFS and OS (95% CL) were 2.6 (1.7-3.7) and 4.7 mos (4.2-7.3), respectively but 7 pts (17%) had long-term cure for &gt;10mos (4/7 beyond 2nd line). G3 hypertension occurred in 2 pts, diarrhoea in 5, anemia and hand-foot syndrome in 3 pts each. Significant increase from T0 (baseline) to T1 (+4wks) level was observed for VEGF (p&lt;0.0001), IL6 (p&lt;0.0129) and IL8 (p&lt;0.0013) and decrease for VEGFR2 and c-Kit (p&lt;0.0001 each). Rising IL8T1 levels significantly associated with RECIST progression at covariance analysis (p=0.0104). Elevated IL8T1 levels (IQ range, HR=2.11), liver mets (HR=2.33), PS (HR=3.73) and upper tract UC (HR=0.33) were significant variables for OS at multivariate Cox analysis. Conclusions: This is the 1st trial ever presented to meet the primary endpoint with a targeted drug in UC. A role for antiangiogenesis in UC has been set. Increasing levels of IL8 were associated with PD and worse outcome and may be included in a new prognostic model. Future investigation should aim at targeting IL8 to improve efficacy results by prolonging responses.

Does RECIST-defined progression correlate with lack of further sunitinib (SU) benefit in advanced renal cell carcinoma (aRCC)?

e15093 Background: RECIST defined radiographic progression on SU may be confounded by the intermittent dosing schedule and scan timing. Response rates to second line agents are typically lower and of shorter duration We analysed the potential for continued clinical benefit from continuation on Sunitinib following RECIST-defined progression in selected pts . Methods: Our institutional database was reviewed to identify aRCC pts who: 1) received SU, 2) achieved response or stable disease, and 3) had continuation of SU after documentation of RECIST-defined progression. Treatment history was obtained from medical chart review. CT scans were retrospectively assessed by RECIST criteria, pattern of progression was documented. SU cessation reasons were identified. Progression-free survival (PFS), post-progression treatment duration (PPD) and overall survival (OS) estimated by Kaplan Meier method. Results: Of 39 patients (pts) treated with SU between 2006 and 2009, 13 met entry criteria. Median age was 62 years (range: 33 – 66) with 11(85%) males. Eleven (85%) had prior nephrectomy and 4 (31%) had prior immunotherapy. Number of pts with MSKCC good/intermediate/poor risk groups were 5/7/1. Best radiographic response achieved was complete response/partial response/stable disease in two, five and six pts. Median PFS was 16.4 mos (11.8 – 23.8). Seven (54%) had progression of existing lesions while the six developed new lesions. Pts continued on SU as progression was relatively minor and clinical status was maintained. Median PPD was 4.2 months (0.6 – 40.5). Most common reasons for SU cessation were on-going radiographic progression (71%) and toxicity (18%). Only 2 pts stopped SU secondary to clinical deterioration. 62% of pts received subsequent lines of treatment, with half receiving 3rd line therapy. Median OS for the cohort was 31.7 mos. Conclusions: Continuation of treatment with SU following RECIST progression appears feasible and safe with prolonged disease control in selected pts . This approach should be further validated prospectively in a larger patient cohort. In this group, current standards for disease response assessment may underestimate the duration of clinical benefit conferred.

A phase I dose-escalation study of the HSP90 inhibitor AUY922 and erlotinib for patients with EGFR-mutant lung cancer with acquired resistance (AR) to EGFR tyrosine kinase inhibitors (EGFR TKIs).

3083 Background: AUY922 is a highly potent, non-geldanamycin analog, HSP90 inhibitor that degrades mutated EGFR and MET. Preclinical studies demonstrate that HSP90 inhibitors have anti-tumor activity in EGFR mutant, EGFR TKI-sensitive and TKI-resistant lung cancer xenograft models. To prevent disease flare after stopping EGFR TKIs in patients (pts) with EGFR mutated lung cancer, erlotinib (E) is often continued with subsequent lines of treatment for AR. This phase I study will determine the maximally tolerated dose (MTD) of AUY922 and E for pts with mutated EGFR and RECIST progression on EGFR TKIs. Methods: All pts have EGFR mutant lung cancer, with development of AR (per Jackman, JCO 2010) after treatment with an EGFR TKI. Pts underwent repeat tumor biopsies after development of AR and prior to study entry. Pts receive AUY922 IV weekly and E oral daily in 28-day cycles, with dose escalation using standard 3+3 design. Pharmacokinetics, serial ophthalmology evaluations and weekly assessment for toxicity are required. Tumor tissue from re-biopsy at study entry is analyzed for EGFR T790M and MET. Results: Since April 2011, 11 pts have been enrolled, in 4 of 5 planned cohorts (AUY922 25 mg/m2 + E 75 mg; AUY922 25 mg/m2+ E 150 mg; AUY922 37.5 mg/m2+ E 150 mg; AUY922 55 mg/m2+E 150 mg). Prior to developing AR, pts were treated with EGFR TKIs for a median of 9 months (range 8-32). Of 9 pts with tissue analyzed thus far, 6 had EGFR T790M. Pts have received a median of two cycles (range 1-7), and 5 remain on study. There have been no dose-limiting toxicities. Adverse events reported in ≥ 20% of pts were diarrhea, nausea, vomiting, fatigue, and rash, all graded 1 or 2. No cardiovascular, renal or hepatic toxicities have been observed. Three pts have reported “flashing lights” and 2 pts transient night blindness, all grade 1. Dose escalation continues. No partial responses have been seen in 6 evaluable pts, 4 patients have confirmed stable disease at 8 wks. Conclusions: AUY922 and E is a well-tolerated combination at dose levels up to AUY922 55 mg/m2+ E 150 mg. Further dose-escalation, MTD, and correlation with markers of AR will be reported. Supported by Novartis, Inc.

ASPIRATION: Phase II study of continued erlotinib beyond RECIST progression in Asian patients (pts) with epidermal growth factor receptor (EGFR) mutation-positive non-small cell lung cancer (NSCLC).

TPS7614 Background: First-line erlotinib (an EGFR tyrosine-kinase inhibitor) significantly increased progression-free survival (PFS) vs chemotherapy in phase III trials of pts with EGFR mutation-positive NSCLC. Discontinuation of erlotinib on RECIST disease progression (PD) may lead to rapid disease flare-up; continued erlotinib beyond RECIST PD may extend clinical benefit by slowing progression of this life-threatening disease. We describe ASPIRATION, a large, Asian, multicenter, single-arm, open-label, phase II trial (NCT01310036), which will increase understanding of first-line erlotinib and erlotinib continuation beyond RECIST PD in pts with EGFR-mutated NSCLC. Methods: Pts (n=204) ≥18 yrs with stage IV/recurrent NSCLC, ≥1 measurable lesion (≥10mm), ECOG performance status (PS) 0-2 and positive EGFR mutation status established by local pathology laboratory (that underwent voluntarily QA/QC) are eligible. All pts receive erlotinib 150mg/day. Tumor response is evaluated using RECIST (v1.1). The primary endpoint is PFS. At investigator's discretion, pts may continue on erlotinib beyond RECIST PD, e.g. if they have slow PD (&gt;6 months of partial response/stable disease), asymptomatic minimal PD, or new brain metastasis controlled locally. Pts should not continue erlotinib if they have extracranial PD with symptoms; rapid PD and/or worsening of PS; or life-threatening complications. Pts continuing erlotinib who present with second RECIST PD will discontinue. Secondary endpoints include objective response rate, disease control rate, overall survival, and safety. For the exploratory biomarker study, pre-treatment tumor tissue blocks are collected; remaining tissue (after EGFR mutation testing for eligibility) will be analyzed centrally to study the association of biomarkers and clinical outcomes. Pre-treatment and post-treatment plasma and serum samples will be obtained at various time points for biomarker assays, including EGFR mutations and other candidate NSCLC biomarkers. Recruitment began in Apr 2011; the estimated final data collection for the primary endpoint is Dec 2014.

Temsirolimus (TEM) maintenance therapy after response of castration-resistant prostate cancer (CRPC) to docetaxel (TAX).

e15177^ Background: There are no standard treatments for men who have demonstrated a good response to TAX without signs of progression. The mTOR pathway is involved in many aspects of CRPC, and mTOR inhibitors have shown significant anti-CRPC activity in preclinical testing. Therefore, we designed a phase II trial to explore whether maintenance therapy with TEM has the potential to prolong response to TAX without compromising quality of life. Methods: For this single-arm, multicenter phase II trial we recruited CRPC pts with documented treatment response by PSA (&gt; 50% decline from baseline) or RECIST (1.0) criteria following ≥ 6 to 10 cycles of TAX (75 mg/m2 q3wks). Subjects received weekly TEM (25 mg iv x 4/cycle). The primary endpoint was time to treatment failure (TTF; RECIST or symptomatic progression). Secondary endpoints included safety (NCI-CTCAE v3.0), quality of life (FACT-P, PPI), changes in PSA, objective tumor response rate, and overall survival. We were also studying surrogate markers of TEM activity, including the enumeration of circulating endothelial cells (CEC) as a potential indicator of the antiangiogenic effects of TEM. Results: 19 pts have been enrolled to data after a median of 7 cycles of TAX. The mean age was 68 years (range 51-82), and the median baseline PSA 17.3 (0.02-380.7). 15 pts had received prior local radiation therapy, 6 surgery. 17 pts had bone metastases, 9 visceral and 8 nodal disease. Pts received a median of 4 cycles of TEM (maximum 23 and continuing), resulting in a median TTF of 5.8 (95%CI 3.8-9.2) months. No PSA responses were seen (time to PSA progression 1.8 months), but 1 PR. 8 pts achieved SD. 12 pts have been discontinued due to treatment failure (RECIST [4], symptomatic [7], combined [1]), and 3 for other reasons, including TEM-associated toxicity (lymphedema [1]). TEM has been generally well tolerated without unexpected adverse events or negative impact on quality of life. The median overall survival has not yet been reached. There is a trend for TEM to reduce viable CEC amid large inter/intrapatient CEC variability. Conclusions: TEM maintenance therapy in CRPC pts that have responded to TAX appears to be well tolerated and results in meaningful prolonged TTF.

An international, double-blind, randomized, placebo-controlled phase III trial (EXAM) of cabozantinib (XL184) in medullary thyroid carcinoma (MTC) patients (pts) with documented RECIST progression at baseline.

5508 Background: MTC arises from parafollicular cells of the thyroid gland, accounts for 5-8% of thyroid cancers and represents an unmet medical need. Cabozantinib (cabo) is an oral inhibitor of MET, VEGFR2, and RET. We conducted a phase III study of cabo vs placebo (P) in pts with progressive, unresectable, locally advanced or metastatic MTC. Methods: Eligible pts were required to have documented RECIST progression within 14 months of screening. The primary efficacy measure was progression-free survival (PFS) as assessed by an independent review facility (IRF) using RECIST. Secondary efficacy measures included objective response rate (ORR) and overall survival (OS). The study has 90% power to detect a 75% increase in PFS and 80% power to detect a 50% increase in OS. Tumor assessments occurred every 12 weeks. Crossover between treatment arms was not allowed. Results: Between Sept 2008 and Feb 2011, 330 pts (median age 55 yrs; 67% male; 96% measureable disease; RET mutation status: pos 48%; neg 12%; unknown 39%; prior TKI exposure: yes 21%, no 78%, unknown 2%) were randomized 2:1 to cabo (140 mg free base [175 mg salt form] qd; n=219) or P (n=111). The planned primary PFS analysis included events through the date of the 138th event. As of 15June2011, 44.7% of pts on cabo and 13.5% on P were still receiving study treatment. Statistically significant PFS prolongation of 7.2 mo was observed; median PFS for cabo was 11.2 mo vs 4.0 mo for P (HR 0.28, 95% CI 0.19-0.40, p&lt;0.0001). PFS results favored the cabo group across subset analyses including RET status and prior TKI use. ORR was 28% for cabo vs 0% for P (p&lt;0.0001). An interim analysis of OS (44% of the 217 required events) did not show a difference between cabo and P. The most frequent grade ≥3 adverse events (cabo vs P) were diarrhea (15.9 vs 1.8%), palmar-plantar erythrodysesthesia (12.6 vs 0%), fatigue (9.3 vs 2.8%), hypocalcemia (9.3 vs 0%), and hypertension (7.9 vs 0%). Conclusions: This phase III study met its primary objective of demonstrating substantial PFS prolongation with cabo vs. P in a patient population with MTC and documented progressive disease in need of therapeutic intervention.

Abstract LB-433: Biomarker analysis and final results of INT70/09 Phase II proof-of-concept study of Pazopanib (PZP) in refractory urothelial cancer (UC)

Abstract Introduction: Discouraging results have been achieved with standard and novel compounds in refractory UC. Promising interim results with PZP, a multitargeted drug with distinct anti-angiogenic activity, were recently reported (ESMO 2010, ASCO 2011). The final results corroborated by biomarker analysis is presented. Material and Methods: Eligible pts relapsing or progressing after at least 1 chemotherapy regimen for metastatic disease underwent PZP 800 mg once daily until disease progression or unacceptable toxicity. Whole-body contrast-enhanced CT scan with densitometric analysis of target lesions and PET scan were planned at baseline and q4weeks thereafter. Independent review of all CT scans was made. ≥5 RECIST complete + partial responses (CR+PR) were required overall to declare drug activity according to Simon's Optimal 2-stage design. 50 mL of EDTA blood samples were collected at baseline and q4wks (concurrently with CT-PET scans) in all pts to analyze drug-induced changes in the amount of plasma VEGF, sVEGFR-1,-2 and -3, c-Kit, IL-6, 8 and 12 by multiplex ELISA plates. Results: 41 pts were enrolled from 02/2010 to 07/2011. Median age was 67 yrs (40-84). 15 pts (37%) had UC of the upper urinary tract. 20/41 pts were treated in 3rd line or beyond. 19 pts (46%) were CDDP-refractory and 22 (54%) had hepatic metastases. 27 (66%) had ECOG PS 1-2. 7 pts (17%) had a confirmed RECIST-defined PR, 24 had a stable disease (76% clinical benefit). 20 pts (49%) had a clear necrotic evolution of multiple metastases and/or a decreased SUV at PET consistent with PR. Median progression-free (PFS) and overall survival (OS) were 2 (1-14) and 4 mos (2-19), respectively but 7 pts (17%) had long-term cure for &amp;gt;10mos (4/7 beyond 2nd line). 4 cases of cavitation-fistulization of large tumor masses were observed. G3 hypertension occurred in 2 pts, G1-2 asthenia in 11, diarrhoea in 5, anemia and hand-foot syndrome in 3 pts each. No discontinuations/dose reductions were needed. Significant increase from T0 (baseline) to T1 (+4wks) level was observed for VEGF (p&amp;lt;0.0001), IL6 (p&amp;lt;0.0129) and IL8 (p&amp;lt;0.0013) and decrease for VEGFR2 and c-Kit (p&amp;lt;0.0001 each). Rising IL8(T1) levels significantly associated with RECIST progression at covariance analysis (p=0.0104) and both elevated IL8(T0) and IL8(T1) levels associated with shorter OS at Cox model analysis (p=0.0170 and 0.0107, respectively). Early (+4wks) PET and densitometric response associated with longer PFS at Gehan-Breslow-Wilcoxon test (p=0.008 and p=0.0472, respectively). Conclusions: This is the 1st trial ever presented to meet the primary endpoint with a targeted drug in UC. A role for antiangiogenesis in UC has been set in a population of highly pre-treated pts. Increasing levels of IL8 were associated with PD and worse outcome. Future investigation should aim at targeting IL8 to improve efficacy results by prolonging responses. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-433. doi:1538-7445.AM2012-LB-433

Sequential bevacizumab and oral cyclophosphamide for recurrent ovarian cancer

ObjectiveTest the safety and efficacy of sequentially blocking angiogenesis by adding oral cyclophosphamide to bevacizumab following cancer progression on bevacizumab in patients with recurrent ovarian cancer. MethodsEligibility included ≤2 lines of treatment for recurrence and measurable cancer by RECIST 1.0. Patients received bevacizumab (15mg/kg every 3weeks IV) and upon RECIST progression, oral cyclophosphamide (50mg orally daily) was added. Objectives included safety, toxicities, 3- and 6-month PFS rates, response rate, PFS, and OS. Results20 patients were enrolled. Overall response rate was 10%, and 65% of patients had confirmed stable disease (SD). Thirteen of 20 patients received oral cyclophosphamide added to bevacizumab upon bevacizumab progression. Of these 13 patients, 1 patient subsequently achieved a PR (this patient had SD as best response during bevacizumab) and 3 patients had a confirmed SD. For all patients, median PFS was 8.41months, 6month PFS rate was 65%, duration of response (DOR) was 7.3months, and median OS was 22.72months. Median DOR for patients receiving both bevacizumab and cyclophosphamide was 8.4months. Most toxicities were grades 1 and 2 and manageable. Grades 3 and grade 4 toxicities included 1 myocardial infarction, 1 gastrointestinal perforation (GIP), and 12/20 patients (60%) developed grade 3 HTN. ConclusionsAddition of oral cyclophosphamide to bevacizumab at the time of cancer progression on bevacizumab appears to have continued anti-cancer effects in a subgroup of patients and appears to be safe. Randomized trials testing combination versus sequential anti-angiogenic therapy for recurrent ovarian cancer are warranted.

Intermittent chemotherapy (CT) plus continuous or intermittent cetuximab (C) in the first-line treatment of advanced colorectal cancer (aCRC): Results of the two-arm phase II randomized MRC COIN-B trial.

536^ Background: COIN-B is a trial of intermittent chemotherapy (ICT) plus intermittent cetuximab (C) vs ICT plus continuous C in the first-line treatment of aCRC. It complements the COIN trial by investigating how C might safely and effectively be added to an ICT strategy. Methods: Patients (pts) had measurable aCRC; no prior CT for metastases; WHO PS 0-2 and good organ function. Randomisation was: Arm D - continuous OxFU + weekly C for 12 wks then a planned break from all therapy; Arm E - OxFU + weekly C for 12 wks then weekly C. Upon RECIST progression on either arm, OxFU (or FU) plus C was restarted and continued until progression on maximal tolerated therapy. Prospective KRAS testing was introduced in May 2008. Primary outcome measure is Failure-Free Survival (FFS) at 10 months (mo) in KRAS-wt pts who had not progressed, died or failed the treatment strategy within 3 mo of randomisation. The trial was powered to differentiate between a desired 10-mo FFS rate of 50% and a minimum of 35%, in 136 pts (168 allowing for drop-outs). Secondary outcome measures included safety, overall survival (OS) and toxicity. Results: 169 KRAS-wt pts were randomised 07/07 to 06/10, 77 arm D / 92 arm E. Median age 64 years (IQR 55-70); 92% PS 0-1. In Arms D and E respectively, 65 (84%) and 67 (73%) pts were eligible for the primary analysis; 10-mo FFS rates were 48% vs 54% (one-sided 95% confidence limit 37% and 43% respectively). Median FFS was 12.0 vs 13.7 mo respectively (IQR 6.1-20.3 and 8.6-23.2). Median OS was 20.1 vs 18.4 mo. First CFI length was 3.7 mo vs 5.1 mo (IQR 2.5-6.2 and 2.5-8.9). In pre-planned exploratory analysis, median time to progression/death after chemo break was 3.1 mo (IQR 2.1-8.1) in Arm D and 6.0 mo (IQR 2.9-10.9) in Arm E. Toxicity was similar and only 1 arm D pt had G 3 hypersensitivity following C reintroduction. Analyses by BRAF &amp; NRAS (tested retrospectively) will be presented. Conclusions: C was safely incorporated in 2 ICT strategies. Continuous C as maintenance was associated with a longer CFI and longer time to progression/death. This encouraging strategy of incorporating biological maintenance therapy needs validation in phase III trials.

Disease Flare after Tyrosine Kinase Inhibitor Discontinuation in Patients with EGFR-Mutant Lung Cancer and Acquired Resistance to Erlotinib or Gefitinib: Implications for Clinical Trial Design

Treatment of patients with oncogene-addicted cancers with tyrosine kinase inhibitors (TKI) is biologically and clinically different than with cytotoxic chemotherapy. We have observed that some patients with EGFR-mutant lung cancer and acquired resistance to erlotinib or gefitinib (RECIST progression after initial benefit) have accelerated progression of disease after discontinuation of TKI. To examine this observation and define the course of patients following TKI discontinuation, we systematically evaluated patients enrolled on clinical trials of agents to treat acquired resistance to erlotinib or gefitinib. We evaluated patients with EGFR-mutant lung cancer who participated in trials for patients with acquired resistance that mandated TKI discontinuation before administration of study therapy. Disease flare was defined as hospitalization or death attributable to disease progression during the washout period. Fourteen of 61 patients (23%; 95% CI: 14-35) experienced a disease flare. The median time to disease flare after TKI discontinuation was 8 days (range 3-21). Factors associated with disease flare included shorter time to progression on initial TKI (P = 0.002) and the presence of pleural (P = 0.03) or CNS disease (P = 0.01). There was no association between disease flare and the presence of T790M at the time of acquired resistance. In patients with EGFR-mutant lung cancer and acquired resistance to epidermal growth factor receptor TKIs, discontinuation of erlotinib or gefitinib before initiation of study treatment is associated with a clinically significant risk of accelerated disease progression. Clinical trials in this patient population must minimize protocol-mandated washout periods.