Post-progression Therapy Research Articles

Phase II trial of single agent PARP inhibitor ABT-888 (veliparib [vel]) followed by postprogression therapy of vel with carboplatin (carb) in patients (pts) with stage BRCA-associated metastatic breast cancer (MBC): California Cancer Consortium trial PHII-96.

1021 Background: We reported on a phase I trial showing 54% confirmed partial response (PR) with carb + vel. Here we describe single agent vel activity and, upon progression, the feasibility and ef...

Progression types after antiangiogenic therapy are related to outcome in recurrent glioblastoma.

This retrospective study analyzed whether the type of radiologic progression, classified according to contrast enhancement on MRI T1-weighted sequences and changes in T2-hyperintense signal, is relevant for outcome in patients with progressive glioblastoma (pGB) treated with bevacizumab. MRI scans of 83 patients with pGB treated with bevacizumab were evaluated prior to and at disease progression. Based on initial decrease in and subsequent flare-up of contrast enhancement in T1 and 2 patterns of T2-hyperintense tumor progression, progression types (PTs) were categorized as cT1 flare-up, T2-diffuse, T2-circumscribed, or primary nonresponder. Overall survival (OS), survival from start of bevacizumab therapy (OS_Bev), survival after bevacizumab failure (OS_PostBev), time from initial diagnosis until initiation of bevacizumab therapy (StartBevT), and time to bevacizumab progression were evaluated using Kaplan-Meier curves, log-rank test, and Cox regression analyses. The time observed for development of a T2-diffuse (n = 15) or a cT1 flare-up (n = 35) progression was longer than for progression in primary nonresponders (n = 16) or T2-circumscribed progression (n = 17). The T2-diffuse PT showed longer OS, OS_Bev, OS_PostBev, and StartBevT compared to the other PTs. Postprogression therapy tended to be relevant only for patients with a T2-circumscribed PT. Radiologic PTs following bevacizumab treatment failure show differences in time to development and are related to outcome. We therefore hypothesize that these PTs reflect a different glioma biology, including differential resistance mechanisms to bevacizumab, and may be associated with different responses to postprogression therapy.

Targeting Epidermal Growth Factor Receptor in the Management of Lung Cancer

The epidermal growth factor receptor (EGFR) mutation is a potent oncogenic driver that accounts for carcinogenesis and tumor growth of pulmonary adenocarcinoma. Targeting EGFR with tyrosine kinase inhibitors (TKIs) is highly effective in terms of tumor response rate, progression-free survival (PFS), and quality of life. Multiple randomized studies have confirmed the superiority of EGFR TKIs over platinum-based chemotherapy and established EGFR TKIs as standard first-line therapy for patients with EGFR mutation-positive non-small cell lung cancer (NSCLC). However, almost all patients will develop resistance to EGFR TKIs and post progression therapy may include a combination of local therapy, systemic chemotherapy, and second-generation EGFR TKIs.

Thalidomide Consolidation Post Autologous Stem Cell Transplant (ASCT) For Multiple Myeloma (MM) Is Cost-Effective With Durable Survival Benefit At 5 Years Post Randomisation: Final Analysis Of The ALLG MM6 Study

BackgroundStudies investigating thalidomide consolidation/maintenance strategies post ASCT in patients with MM have consistently demonstrated improvement in duration of myeloma control, however not consistently shown improvements in overall survival. Cost effectiveness of thalidomide in the post-ASCT consolidation/maintenance setting in Australian clinical practice has not been previously established AimTo determine whether progression free survival (PFS) and overall survival (OS) advantages for thalidomide consolidation post ASCT at 3 years post randomisation in the ALLG MM6 study are durable at later follow-up. To compare overall response rate (ORR) to salvage therapy and incidence of second primary malignancy (SPM). To investigate the cost-effectiveness of thalidomide in the post-ASCT consolidation setting. MethodsPhase III, randomised, multi-centre, open label study. 243 newly diagnosed MM patients 6 weeks following a single MEL200 ASCT as part of their first-line therapy were randomly assigned to receive indefinite prednisolone maintenance (50mg alternate days) alone (CA = 129 patients, 1 patient withdrew consent) or in combination with 12 months of thalidomide consolidation (100mg/d increasing to 200mg/d after 2/52) (TA = 114 patients). This was a post hoc analysis, PFS and OS were measured from date of randomization; these endpoints were compared using intention-to-treat analyses. Data for ORR to salvage therapy was collected retrospectively for 96 of 187 relapsed/progressed patients only (TA = 42/81 relapsed, CA = 54/109 relapsed), as was data for SPM (207/238 patients, TA = 104/112, CA = 103/126). All statistical analyses were performed using SPSS version 19.Economic analysis incorporated treatment exposure (primary drug and co-therapies), the use of ongoing medical and diagnostic services, occurrence of SAEs and use of therapies to treat those events, post progression therapy and duration of survival, and was represented as incremental cost per discounted mean life year gained (LYG) – incremental cost effectiveness ration (ICER). All costs and outcomes beyond 12 months were discounted at 5% per annum, and were calculated in Australian dollars. ResultsAfter a median follow-up of 5.4 years post randomization, 2 patients per arm were lost to follow-up (TA = 112, CA = 126). Post randomization estimated 5 year PFS rates were 27% versus 15% (p=0.005; hazard ratio [HR] 0.16: 95% CI 0.044 to 0.582) and OS rates were 66% versus 47% (p=0.007; HR 0.12: 95% CI 0.028 to 0.558) in TA and CA respectively. Thalidomide remained beneficial irrespective of pre-ASCT B2m level <4mg/L (p=0.002) and ≥4mg/L (p=0.049), however TA patients who achieved VGPR/CR post ASCT no longer had a PFS advantage over CA patients who achieved VGPR/CR. Patients required at least 8 months of thalidomide exposure to gain a PFS and OS advantage (p<0.001). Landmark analysis confirmed that PFS/OS benefit was gained within the first 8-12m of therapy. There was no difference in ORR to salvage therapy (62% versus 69%, p=0.5), survival post-progression or incidence of SPM for TA versus CA.Discounted mean LYG for TA patients was 0.92 years (95%CI 0.32 to 1.52), and estimated treatment for TA patients cost $67,911 compared to $42,999 for CA patients, with a resultant incremental cost of $24,912 for TA compared with CA. ICER was $26,996 per mean LYG for TA versus CA. Cost of thalidomide accounted for 56% of the overall incremental cost difference between TA and CA, post-progression therapy 21%, and SAEs contributed the least to estimated difference in costs (3%). ConclusionPFS and OS advantages ascribed to thalidomide consolidation post ASCT remain highly significant at 5 years. At least 8 months of thalidomide exposure was required to attain the PFS/OS benefit. Further recapitulating previous findings, thalidomide did not impact on ORR to salvage therapy or survival following relapse in the context of salvage with alternate novel therapies. Thalidomide consolidation is cost effective in terms of ICER, with cost of thalidomide the key contributor to the cost difference. Disclosures:Spencer:Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees.

Analysis of survivorship in high-risk patients on treated on GOG-218

Objective: Bevacizumab (B) added to paclitaxel/carboplatin (PC) followed by maintenance B improved progression-free survival (PFS) in 2 contemporaneous phase III trials (GOG-0218, ICON7). While preliminary analysis of overall survival (OS) was not affected significantly in either study, OS was significantly improved within a prospectively stratified high-risk subgroup (stage III >1cm postoperative residuum and stage IV) in ICON7. We hypothesized that a similar effect would not be observed in GOG-0218 due to eligibility differences and differential use of post-progression therapy.

Olaparib plus paclitaxel in patients with recurrent or metastatic gastric cancer: A randomized, double-blind phase II study.

4013 Background: Our multicenter study compared the efficacy of the oral PARP inhibitor olaparib plus paclitaxel (O/P) vs paclitaxel alone (P) as second-line therapy in pts with recurrent/metastatic gastric cancer (GC) (NCT01063517). As initial preclinical data suggested that responsiveness of GC cell lines to olaparib was associated with low ATM protein levels, our study was enriched for pts with low ATM tumors (ATM–) by IHC (50% randomized vs 14% screening prevalence). Methods: Eligible pts were randomized 1:1 (stratified by ATM status) to receiveolaparib 100 mg bid (tablet form) plus paclitaxel (80 mg/m2 iv on days 1, 8, 15 per 28-day cycle) or placebo plus paclitaxel until progression or investigator decision. After combination therapy, pts could take olaparib 200 mg bid monotherapy or placebo until progression. Co-primary endpoints: progression-free survival (PFS; RECIST v1.1) in all pts and ATM– pts. Secondary endpoints: overall survival (OS), objective response rate (ORR), safety. Results: 123/124 randomized pts were treated (O/P=61; P=62). Baseline characteristics were generally well balanced. Use of post-progression therapy was similar in both arms (O/P=48.4%; P=43.5%) as was median paclitaxel duration (O/P=17 wks; P=16 wks); 18 pts received monotherapy (O/P=11; P=7). More pts in the O/P than P arm had delays (79 vs 63%) and reductions (41 vs 27%) in paclitaxel dosing. The most common grade ≥3 AEs in the O/P and P arms were neutropenia (56 vs 39%) and anemia (11 vs 11%). Conclusions: Olaparib plus paclitaxel was well tolerated and led to a statistically significant improvement in OS, but not PFS, vs paclitaxel alone in both all pts and ATM– pts, with a larger benefit in ATM– pts. Clinical trial information: NCT01063517. [Table: see text]

Outcomes of Patients with Refractory/Relapsed Diffuse Large B-Cell Lymphoma Who Progress After Autologous Stem Cell Transplantation in the Rituximab Era.

Abstract 2689Salvage chemotherapy and autologous stem cell transplant (ASCT) remain the current standard of care for patients with diffuse large B-cell lymphoma (DLBCL) who have primary refractory disease or relapse after initial therapy. The addition of rituximab results in improved overall survival (OS) after first line treatment, but cure rates of salvage therapy and ASCT are inferior when compared to historical controls (Gisselbrecht et al, JCO 2010). In the pre-rituximab era, patients with DLBCL who progressed after ASCT had an extremely poor prognosis, with a median OS of 3 months from the time of progression (Vose et al, Blood 1992). There is a paucity of data regarding outcomes and clinical patterns following progression after ASCT in the rituximab era.We conducted a retrospective analysis using our institutional database of DLBCL patients who underwent ASCT for primary refractory or relapsed disease. For those who progressed after ASCT, we evaluated OS defined as time from date of progression after ASCT to date of death from any cause or last follow-up. We also analyzed OS for various subgroups based on their clinical characteristics and post-progression therapy. The impact of post-progression therapy on outcome was assessed by calculating OS from the time of exposure to a particular treatment to date of death from any cause or last follow-up. Median OS was estimated by the Kaplan-Meier method and subgroup comparisons were performed using the log rank test.We identified 215 patients with primary refractory or relapsed DLBCL who underwent ASCT between January 1, 2005 and December 31, 2011. All patients received rituximab as part of their first line treatment. Fifty-six patients (26% of total) were found to have progression after ASCT. Eight patients were excluded from the analysis because they were not re-biopsied at relapse (n=5) or they had indolent disease at relapse (n=3). Data on the remaining 48 patients (median age 57 years; range 20–74) were further analyzed. The median OS from progression after ASCT for the entire cohort was 10.7 months (95% CI: 6.8–18.0 months). Patients who progressed < 1 year from ASCT (n=39) had a significantly shorter OS than those who progressed ≥ 1 year from ASCT (n=9): 9.5 vs. 26.7 months (p=0.02). Patients with at least stable disease as first assessment after ASCT (n=29) had a significantly longer OS than those who progressed immediately after ASCT (n=19): 18 vs. 6 months (p=0.01). Patients who were exposed to at least one novel agent (lenalidomide, radioimmunotherapy, non-rituximab monoclonal antibodies or tyrosine kinase inhibitors) as part of their post-progression therapy after ASCT (n=20) had a median OS of 11.2 months from treatment initiation. In contrast, patients who were treated with conventional cytotoxic chemotherapy but not novel agents (n=20) had a median OS of 6.7 months. In particular, patients who underwent radioimmunotherapy at some point after progression post-ASCT (n=9) had a median OS of 17.5 months (95% CI: 2.6-n/a months) with one patient in complete response > 48 months after the treatment. Patients who underwent allogeneic stem cell transplant following progression after ASCT (n=6) had a median OS of only 7.1 months (95% CI: 0.8-n/a months).In the rituximab era, the median OS of DLBCL patients who progress after ASCT appears improved when compared to historical controls, especially in those who progress > 1 year from ASCT. Patients with DLBCL progressing after ASCT should be considered for therapy with novel agents, which may result in long term survival. [Display omitted] Disclosures:No relevant conflicts of interest to declare.

LBA33_PR - Monotherapy Administration of Sorafenib in Patients with Non-Small Cell Lung Cancer: Phase III, Randomized, Double-Blind, Placebo-Controlled Mission Trial

ABSTRACT Background Sorafenib monotherapy showed activity in a randomized discontinuation phase II trial of patients with advanced NSCLC who failed 2–3 chemotherapy regimens (ECOG 2501). The Phase III MISSION trial assessed whether 3rd- or 4th-line treatment with sorafenib plus best supportive care (BSC) would improve overall survival (OS), relative to placebo plus BSC, in patients with advanced relapsed/refractory NSCLC of predominantly non-squamous histology. Methods Eligible patients were randomized 1:1 to oral sorafenib (S) 400 mg bid or placebo (P) and stratified by geographic region, number of prior lines of treatment, brain metastases, and prior anti-EGFR therapy. The primary endpoint was OS; secondary endpoints included progression free survival (PFS), time to progression (TTP), overall response rate (ORR), disease control rate (DCR), and safety; endpoints were tested with one-sided significance level of 0.025. Results A total of 703 patients were randomized (S = 350; P = 353). Baseline demographic factors and prior treatments were generally balanced, with minor imbalances (S vs P) in female gender (47% vs 41%) and never smokers (46% vs 38%). Fewer patients received post-progression therapy in the S arm (44%) than in the P arm (56%). Assessment of OS was based on a total of 579 events (S = 285, P = 294). Median OS was similar in the two groups (248 vs 253 d; HR 0.99, p = 0.4687), whereas median PFS (84 vs 43 d; HR 0.61; p Conclusion The phase III MISSION trial did not meet its primary endpoint of improving OS as 3rd- or 4th-line treatment in patients with advanced NSCLC. Sorafenib treatment significantly enhanced PFS, TTP, ORR, and DCR compared to BSC alone. Safety and tolerability data were as expected. Supported by Bayer HealthCare Pharmaceuticals and Onyx Pharmaceuticals. Disclosure L. Paz-Ares: Luis Paz-Ares has received honoraria from Bayer HealthCare Pharmaceuticals, Lilly, Roche and Pfizer V. Hirsh: Member of the steering committee for the MISSION trial. L. Zhang: Dr. Zhang has received research support from Bayer, Aventis, AstraZeneca, Henrui Pharm and Eli Lilly and has received consulting and lecture fees from Roche, Aventis, AstraZeneca and Eli Lilly. J.C. Yang: Dr. Yang has acted in an advisory capacity for and received honoraria from Bayer Healthcare H. Wakelee: Dr. Wakelee has received research funding from Bayer Healthcare T. Seto: Dr. Seto has received research funding from Bayer Healthcare T. Schmelter: Dr. Schmelter is an employee of Bayer HealthCare and owns stock in the corporation. T.J. Ong: Dr. Ong is an employee of Bayer HealthCare and owns stock in the corporation. T.S.K. Mok: Dr. Mok has received honoraria from AstraZeneca, Roche, Eli Lilly, Merck Serono, Eisai, BMS, BeiGene, AVEO, Pfizer, Taiho, Boehringer Ingelheim, GSK Biologicals. He speaks for Astrazeneca, Roche, Eli Lilly, Boehringer Ingelheim, Merck Serono, and received research funding from Astrazeneca. All other authors have declared no conflicts of interest.

Cost-effectiveness of treating patients with advanced progressive pancreatic neuroendocrine tumors with everolimus versus sunitinib in the United States.

e14525 Background: Everolimus and sunitinib were recently approved to treat patients with advanced, progressive pancreatic neuroendocrine tumors (pNETs). This analysis examined the projected cost-effectiveness of everolimus versus sunitinib in this disease setting from a US payer perspective. Methods: A lifetime (20-year) Markov model was developed to simulate a cohort of advanced, progressive pNET patients and to estimate the cost per life-year (LY) gained and quality-adjusted life-year (QALY) gained when treating with everolimus as compared to sunitinib. Absent head-to-head trials, efficacy data were based on a weight-adjusted indirect comparison of the two agents using the respective phase 3 trial data. Disease or health states considered in the model included: stable disease with no adverse events, stable disease with adverse events, disease progression, and death. Costs of anti-tumor therapies, symptomatic care drugs, and post-progression therapy were based on wholesale acquisition cost. Other costs including physician visits, tests, infusions, hospitalizations, adverse event costs, and end-of-life care costs were obtained from literature and/or standard sources such as the Healthcare Cost &amp; Utilization Project and Medicare reimbursement rates. Utility inputs were based on a time trade-off study conducted in the United Kingdom. Sensitivity analyses were conducted to test the model’s robustness. Results: In the base case analysis, the estimated gain of everolimus over sunitinib was 0.448 LYs (0.304 QALYs) at an incremental cost of $12,673, resulting in an incremental cost-effectiveness ratio (ICER) of $28,281/LY ($41,702/QALY gained), which fell within the cost per QALY range for many other widely used oncology drugs. The ICER was most sensitive to the uncertainty of the sunitinib trial outcomes: however, a probabilistic sensitivity analysis found consistent results across simulations. Conclusions: While the analysis is limited by its reliance on an indirect comparison of two phase 3 studies rather than a single head-to-head trial, everolimus is projected to be cost-effective relative to sunitinib in advanced pNET.

Post-progression survival (PPS) according to treatment line, type, and time in phase III trials in advanced colorectal cancer (ACC).

3557 Background: Since post-progression therapy influences overall survival (OS), PPS is of interest as a determinant of OS. We quantified the relationship between several median times to event in ACC: progression-free survival (PFS), PPS and OS. We looked for factors with an impact on these times. Methods: We searched PubMed for phase III trials published between January 1998 and December 2010 in 12 leading journals. PPS is the difference between median OS and median PFS or time to tumor progression. The statistics of interest (median times [PFS, OS and PPS] and the ratio of PPS/OS) were compared using t-tests. Results: We retrieved 70 trials that yielded 71 comparisons and 156 arms (one trial was factorial) involving 38,504 patients. Trials from the period 2005-10 enrolled more patients than trials from the period 1998-2004 (medians of 246 vs 174 per arm; P=.026). PPS could be calculated for 60 trials and 135 arms (Table). Treatment type did not have an impact on any statistic. Treatment line had a significant impact on all times and on the ratio of PPS/OS, although the difference on the latter statistic was small. Treatment period had a significant impact on all times but not on the ratio of PPS/OS. Conclusions: None of the factors examined have a major impact on the ratio of PPS/OS. This suggests that in ACC, gains in PFS are translated into proportional gains in OS, one of the conditions required for PFS to be considered a surrogate for OS. [Table: see text]

Progression-free survival (PFS) as a surrogate end point for overall survival (OS) in first-line treatment of advanced epithelial ovarian cancer (EOC).

5081 Background: The duration and cost of clinical trials of first-line chemotherapy in advanced EOC could be reduced if a surrogate endpoint were used in place of OS. The consensus of the Gynecological Cancer Intergroup was that PFS is the preferred primary end point for these trials due to potential confounding of post-progression therapy on OS. We performed a systematic review to assess the extent to which treatment effect on PFS is predictive of OS in this patient population. Methods: Randomised trials of first-line chemotherapy comparing platinum containing regimens in advanced EOC were identified; trials with non-platinum backbone, maintenance strategies or biological containing therapies were excluded. Summary data (hazard ratios (HR), median PFS and OS and the ratios of medians between treatment arms) were extracted. Weighted least-square (WLS) R2 by trial sample size derived using linear regression was used to report correlation. Results: 15 eligible trials with 19 treatment comparisons were identified comprising a total of 16,598 patients. There was a good correlation between treatment effect on PFS and OS (correlation of HR: r, 0.88, WLS R2, 0.71; correlation of ratios of medians: r, 0.84, WLS R2, 0.72). Good correlation between treatment effect on PFS and OS was also observed in various prognostic subgroups (Table). Conclusions: In clinical trials of first-line platinum based chemotherapy without biological agents in advanced EOC, treatment effect on PFS and OS is highly correlated. PFS could be considered as a primary end point when evaluating future first line strategies in advanced EOC. [Table: see text]

Vandetanib Versus Placebo in Patients With Advanced Non–Small-Cell Lung Cancer After Prior Therapy With an Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitor: A Randomized, Double-Blind Phase III Trial (ZEPHYR)

Vandetanib is a once-daily oral inhibitor of vascular endothelial growth factor receptor, epidermal growth factor receptor (EGFR), and RET signaling. This placebo-controlled trial assessed whether vandetanib conferred an overall survival benefit in patients with advanced non-small-cell lung cancer (NSCLC) after prior treatment with an EGFR tyrosine kinase inhibitor and one or two chemotherapy regimens. Eligible patients were randomly assigned 2:1 to receive vandetanib 300 mg/d or placebo until disease progression or unacceptable toxicity. The primary objective was to compare the outcomes between the two arms with respect to overall survival. Overall, 924 patients received vandetanib (n = 617) or placebo (n = 307). No significant increase in overall survival was detected in the vandetanib cohort compared with placebo (hazard ratio = 0.95; 95.2% CI, 0.81 to 1.11; P = .527); median overall survival was 8.5 months versus 7.8 months for vandetanib and placebo patients, respectively. Statistically significant advantages favoring vandetanib were observed for progression-free survival (hazard ratio = 0.63; P < .001) and objective response rate (2.6% v 0.7%; P = .028). Postprogression therapy was balanced across the cohorts in both number and type. Adverse events were generally consistent with previous NSCLC studies of vandetanib 300 mg; common events occurring with a greater frequency in the vandetanib arm versus placebo included diarrhea (46% v 11%), rash (42% v 11%), and hypertension (26% v 3%). The study did not demonstrate an overall survival benefit for vandetanib versus placebo. There was a higher incidence of some adverse events with vandetanib.

Cost-effectiveness of treating patients with advanced progressive pancreatic neuroendocrine tumor with everolimus versus sunitinib in the United States.

226 Background: Everolimus and sunitinib were recently approved to treat patients with advanced, progressive pancreatic neuroendocrine tumors (pNETs). This analysis examines the projected cost-effectiveness of everolimus versus sunitinib in this setting from a US payer perspective. Methods: A lifetime Markov model was developed to simulate a cohort of advanced, progressive pNET patients and to estimate the cost per incremental life-years gained (LYG) and quality-adjusted life years (QALYs) gained when treating with everolimus as compared to sunitinib. Absent head-to-head trials, efficacy data were based on a weight-adjusted indirect comparison of the two agents using the respective phase 3 trial data. Disease or health states considered in the model included: stable disease without adverse events, stable disease with adverse events, disease progression, and death. Costs of anti-tumor therapies, symptomatic care drugs, and post-progression therapy were based on wholesale acquisition cost. Other costs including physician visits, tests, infusions, hospitalizations, adverse event costs, and end-of-life care costs were obtained from literature and/or standard sources such as the Healthcare Cost and Utilization Project and Medicare reimbursement rates. Utility inputs were based on a UK time trade-off study. Sensitivity analyses were conducted to test the model’s robustness. Results: In the base case analysis, the estimated gain of everolimus over sunitinib was 0.448 LYs (0.304 QALYs) at an incremental cost of $12,673, resulting in an incremental cost-effectiveness ratio (ICER) of $28,281/LYG ($41,702/QALY gained), which fell within the cost per QALY range for many other widely used oncology drugs. The analysis was sensitive to the uncertainty of the sunitinib trial results; however, a probabilistic sensitivity analysis showed the results were consistent across simulations. Conclusions: While the analysis is limited by its reliance on an indirect comparison of two phase 3 studies rather than a single head-to-head trial, everolimus is expected to be cost-effective relative to sunitinib in advanced pNET.

Final results from study 181: Randomized phase III study of FOLFIRI with or without panitumumab (pmab) for the treatment of second-line metastatic colorectal cancer (mCRC).

387 Background: In the primary analysis of study 181, pmab+FOLFIRI significantly improved progression-free survival (PFS) vs FOLFIRI as second-line therapy in patients (pts) with wild-type (WT) KRAS mCRC. Here, we report the results of a prespecified final descriptive analysis planned for 30 months (mos) after the last pt was enrolled. Methods: Pts were randomised 1:1 to pmab 6.0 mg/kg Q2W+FOLFIRI (Arm 1) vs FOLFIRI alone (Arm 2). Pts had one prior fluoropyrimidine-based chemotherapy regimen for mCRC and ECOG 0-2. The co-primary endpoints were PFS (central assessment) and OS, and were independently tested. Secondary endpoints included objective response rate (ORR), and safety. KRAS status was determined by a blinded central lab. Results: 1,186 pts were randomised and received treatment (tx): 591 in Arm 1, 595 In Arm 2. 1,083/1,186 pts (91%) had KRAS results. Adverse event rates were consistent with the primary analysis. Results are shown in the table . Conclusions: In Arm 1, PFS (standard and on-treatment definition) and ORR were improved, and there was a trend toward improved OS in pts with WT KRAS mCRC. The large proportion of pts receiving post-progression anti-EGFR therapy may have affected the ability to observe a difference in OS between the tx arms. In pts with MT KRAS there was no difference in efficacy. KRAS testing is critical to select appropriate pts for tx with pmab. [Table: see text]

Economic Evaluation of Everolimus versus Sorafenib for the Treatment of Metastatic Renal Cell Carcinoma after Failure of First-Line Sunitinib

BackgroundA recent indirect comparison study showed that sunitinib-refractory metastatic renal cell carcinoma (mRCC) patients treated with everolimus are expected to have improved overall survival outcomes compared to patients treated with sorafenib. This analysis examines the likely cost-effectiveness of everolimus versus sorafenib in this setting from a US payer perspective. MethodsA Markov model was developed to simulate a cohort of sunitinib-refractory mRCC patients and to estimate the cost per incremental life-years gained (LYG) and quality-adjusted life-years (QALYs) gained. Markov states included are stable disease without adverse events, stable disease with adverse events, disease progression, and death. Transition probabilities were estimated using a subset of the RECORD-1 patient population receiving everolimus after sunitinib, and a comparable population receiving sorafenib in a single-arm phase II study. Costs of antitumor therapies were based on wholesale acquisition cost. Health state costs accounted for physician visits, tests, adverse events, postprogression therapy, and end-of-life care. The model extrapolated beyond the trial time horizon for up to 6 years based on published trial data. Deterministic and probabilistic sensitivity analyses were conducted. ResultsThe estimated gain over sorafenib treatment was 1.273 LYs (0.916 QALYs) at an incremental cost of $81,643. The deterministic analysis resulted in an incremental cost-effectiveness ratio (ICER) of $64,155/LYG ($89,160/QALY). The probabilistic sensitivity analysis demonstrated that results were highly consistent across simulations. ConclusionsAs the ICER fell within the cost per QALY range for many other widely used oncology medicines, everolimus is projected to be a cost-effective treatment relative to sorafenib for sunitinib-refractory mRCC.

How Long Is Long Enough?

How Long Is Long Enough?

Everolimus plus octreotide LAR (E+O) versus placebo plus octreotide LAR (P+O) in patients with advanced neuroendocrine tumors (NET): Updated results of a randomized, double-blind, placebo-controlled, multicenter phase III trial (RADIANT-2).

159 Background: Octreotide LAR has been the foundation of NET therapy; however, additional treatment options are needed. Everolimus, an oral inhibitor of mTOR, demonstrated promising antitumor activity in patients with NET as a single agent and in combination with octreotide LAR in two phase II studies. Methods: Patients (n = 429) with well or moderately differentiated advanced NET and history of carcinoid symptoms received oral everolimus 10 mg/d + octreotide LAR 30 mg IM q 28 days (n = 216) or placebo + octreotide LAR (n = 213). Common primary sites included the small intestine, lung, and colon. The primary endpoint was progression-free survival (PFS) per central review by RECIST. Crossover from P+O to open-label E+O was allowed at disease progression. Results: Median PFS (95% CI) with E+O was 16.4 months (13.67-21.19) vs. 11.3 months (8.44-14.59) for P+O. E+O was associated with a 23% reduction in risk of progression: HR = 0.77; 95% CI: 0.59-1.00; one-sided p-value = .026 (pre-specified significance boundary is p ≤ .0246). A high rate of informative censoring resulted in loss of power in central review results. Correcting for the informative censoring bias, the pre-specified marginal Cox model using inverse probability of censoring weights (IPCW) analysis showed a significant 5.5-month improvement in median PFS with E+O; HR = 0.60; 95% CI: 0.44-0.84, with one-sided p-value = .0014. The benefit of E+O was seen across all patient subgroups. Updated analyses of the impact of pre-study and post-progression octreotide LAR therapy will be reported. Most frequent drug-related adverse events (AEs) with E+O were stomatitis, rash, fatigue, and diarrhea; mostly grade 1-2. Grade 3-4 AEs reported in &gt;5% were stomatitis, fatigue, diarrhea, infections, and hyperglycemia. Conclusions: In this large phase III trial, E+O provided a 5.1-month clinically meaningful increase in median PFS compared with P+O in the central adjudicated review. Correcting for the informative censoring bias, a significant PFS improvement of 5.5 months was seen, with a p value = .0014. The safety profile of E+O was acceptable. [Table: see text]

Post-Progression Survival (PPS) and Overall Survival (OS) According to Treatment Type in Contemporary Phase III Trials in Advanced Breast Cancer (ABC).

Abstract Objective: Progression-free survival (PFS) and time to tumor progression (TTP) have been the most frequent primary endpoints in recent phase III trials in ABC. OS, the most frequent secondary endpoint (Katz ASCO 2009), is largely influenced by subsequent-line therapy, and the role of PFS/TTP as surrogates for OS is still uncertain, perhaps due to the influence of post-progression therapy. Thus, PPS is of interest as a determinant of OS, and may vary according to treatment type.Methods: Using the medical subject headings “breast neoplasms” and “drug therapy”, we searched PubMed for phase III trials on systemic antineoplastic therapies published between January 1, 1998, and December 31, 2007 in one of 11 leading medical journals (Ann Oncol, BCRT, Br J Cancer, Cancer, Clin Cancer Res, Eur J Cancer, JCO, JNCI, Lancet Oncol, Lancet and NEJM). We excluded trials with factorial design, studies on high-dose chemotherapy, papers reporting combined analyses of two or more trials, and companion studies on correlative biology or prognostic factors. PPS was calculated as median OS minus median PFS/TTP for each trial arm. The proportion of OS accounted for by PPS (PPS%OS) was calculated as 100 - [median PFS/TTP/median OS]% for each trial arm.Results: The search yielded 75 trials enrolling a total of 28,973 evaluable patients in 159 trial arms. The following table displays the results, considering trial arms with available medians for both PFS/TTP and OS. PPS%OS is larger in hormone-therapy trials than in chemotherapy trials (P&amp;lt;0.01). Overall, PPS has a stronger correlation with OS than do PFS/TTP (R-squares of 0.83 and 0.24, respectively; both P&amp;lt;0.01), at least in part because PPS is longer than PFS/TTP.Main results Mean ofmedians(months) Trial typeN armsPFS/TTPOSPPSMedian PPS%OSOVERALL1276.920.513.665.7%1. Chemotherapy96718.311.463.7%1.1 Chemo 1st line only597.820.712.964.4%1.2 Chemo 2nd line only146.715.28.559.8%1.3 Other (mixed lines)235.014.19.164.4%2. Hormone therapy296.927.420.574.4%2.1 Hormone 1st line only139.331.121.869.0%2.2 Hormone 2nd line only135.023.218.177.9% Conclusion: The contemporary average median OS in ABC is 20.7 months for first-line chemotherapy ± targeted agents and 31.1 months for first-line hormone therapy ± chemotherapy trials. Subsequent lines of therapy play a major role in determining OS in ABC, and further studies are needed to understand the role of PFS/TTP versus PPS in determining OS, since first-line PFS/TTP have not been convincingly shown to be good surrogates for OS in the literature on ABC. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 5116.

Treatment center (TC) variability in survival of patients (pts) with colorectal cancer (mCRC) treated with bevacizumab (BV): A subanalysis from the BRiTE Observational Study (OCS)

e17532 Background: Treatment options and overall survival (OS) rates for pts with mCRC have increased markedly over the past decade. However, the variability in OS across TCs, and possible contributing factors (measured and unmeasured), remain unclear. We present an analysis from the BRiTE OCS assessing variability in OS as a function of TC, and sought to identify site factors that may predict better outcomes. Methods: BRiTE enrolled 1,953 pts with mCRC receiving a first-line BV-based regimen, from 248 U.S. sites, mostly community practices (Grothey et al. J Clin Oncol. 2008 26:5326). A Cox proportional hazard model was used to identify predictors of OS (age, PS, albumin, alkaline phosphatase, site of primary tumor) and calculate the expected number of deaths (E) compared to the observed number of deaths (O) within each TC. The ratio O/E was calculated and its distribution is presented. Sites with high O/E (top 25%) vs low O/E (lowest 25%) were titled “worst” (W) and “best” (B), respectively. B and W sites were compared by type (academic vs community, hospital vs clinic, urban vs rural), number of pts treated, choice of first-line chemotherapy, postprogression (PD) therapy, and duration of BV. TCs with ≥3 pts were included in the analysis (198). Results: Significant variation in the O/E ratio was observed across TCs ( Table ). The differences between W and B TCs were that pts at W TCs were less likely to receive a first-line oxali containing regimen (65% vs 67%), had shorter overall duration of BV (8 m vs 12 m) and were less likely to use BV following PD (33% vs 39%). Measured site, pt, and treatment variables explained ∼7% of the variability in O/E, suggesting most TC variability was due to unmeasured factors. Conclusions: Significant variability in OS exists across BRiTE TCs, which appears partly related to the specific therapies received by pts. Further work is needed to determine whether W TCs can be optimized and improve outcomes in pts with mCRC. [Table: see text] [Table: see text]

Assessing the role of combination therapy in mCRC

Doublet chemotherapy regimens (i.e. FOLFOX and FOLFIRI) have been shown to improve objective response rate and progression-free survival compared with fluoropyrimidine monotherapy, and may also improve overall survival (OS) in patients with metastatic colorectal carcinoma (mCRC). Further improvements in efficacy parameters can be achieved by the addition of a third cytotoxic agent, such as in FOLFOXIRI, or a targeted agent (e.g. bevacizumab or cetuximab) to doublet chemotherapy. Data suggests that two groups of patients, defined by clinical characteristics, may be particularly appropriate for aggressive combination treatment: patients at risk of rapid disease progression, and those with liver metastases, which may become resectable following such therapy. Furthermore, theoretical considerations suggest that use of combination regimens first-line increases the number of patients who are able to benefit from exposure to multiple agents, enabling improvement of overall survival for all patient groups. First-line combination therapy can be envisaged as the first phase in a programme of treatment consisting of intensive induction therapy given for a limited period, followed by less intensive maintenance therapy given until disease progression. On disease progression, intensive therapy with either the induction regimen or another regimen may be considered to regain disease control. While accumulating data have established the efficacy of aggressive induction therapy, further research is required to determine the value and feasibility of maintenance therapy and post-progression reinduction therapy, together with the identification of the most appropriate agents to use in these settings.