Glioblastoma Pts Research Articles

Heterogeneity of response to bevacizumab in multifocal and multicentric glioblastomas.

2019 Background: Bevacizumab (BEV) is approved for treatment of recurrent glioblastoma (GBM). However, whether the subset of pts with multifocal (MF) or multicentric (MC) GBM respond favorably to BEV has not been described. Hypothesizing that MF/MC lesions may represent an initially more invasive and biologically heterogeneous phenotype in GBM which may exhibit varied responses to BEV, we examined their patterns of radiologic response and correlation with clinical outcome. Methods: Clinical and MRI data from pts with recurrent GBM who received BEV were reviewed to identify those with MC or MF disease who had adequate baseline and follow-up MRI. Enhancing and nonenhancing (FLAIR) disease were analyzed for response to BEV and categorized in each pt as homogeneous (similar changes in all lesions) or heterogeneous (varied across lesions) in response or progression. Progression-free survival (PFS) and overall survival (OS) were analyzed by the Kaplan Meier method and log rank test. Results: Of the 179 evaluable pts, 31 (17.3%) had multiple lesions (16 pts - MF, 15 pts - MC) prior to BEV treatment. Patients with MC/MF lesions had a significantly worse outcome than those with unifocal lesions in both PFS (Median 19 wks vs. 24.4 wks, p = 0.0124, HR 1.84, CI [1.14, 2.9]) and OS (Median 31 wks vs. 45.9 wks, p = 0.0011, HR 2.42, CI [1.42, 4.12]). Unlike previous reports of BEV response in recurrent GBM pts, MRI responders (homogenous or heterogeneous) among pts with MC/MF lesions did not have a better OS compared with nonresponders. In addition, no significant difference was seen in PFS or OS between patients with homogenous versus heterogeneous progression. There was no significant difference in median PFS (median 17.6 wks vs. 18.4 wks) or median OS (median 27.7 wks vs. 32.3 wks) between pts with MC versus MF GBM. Conclusions: The subset of GBM pts with MF/MC lesions had a worse outcome to BEV treatment compared with those with unifocal disease. Although a homogenous radiological response of all lesions to BEV was the commonest pattern on MRI, neither radiologic response nor pattern of radiologic progression correlated with improved PFS or OS. Our data also suggests that outcome of pts MC/MF GBM may need to be analyzed separately in studies with BEV containing regimens. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration IMEDEX Celgene, Lilly, MGI Pharma, Pfizer

Can OS-6 replace PFS-6 as a primary endpoint in phase II studies on glioblastoma patients given antiangiogenetic drugs?

2022 Background: In the last decade, progression-free survival (PFS) at 6 months (PFS-6) has been considered the best endpoint in phase II trials on recurrent glioblastoma (GBM). However, since a new standard of care was established by the EORTC/NCIC phase III trial, no data have reported on the PFS-6 or overall survival (OS) obtained with second line treatment following combined RT/TMZ. Moreover, antiangiogenic agents might alter data on response by repairing the blood-brain barrier and diminishing contrast enhancement, thus precluding the accurate assessment of any reduction in the tumor burden. The issue of a robust primary endpoint for phase II studies is therefore still open. The aim of the study was to evaluate outcome endpoints for second line treatment at recurrence. Methods: A retrospective analysis was made using a database on 653 GBM patients (pts) followed prospectively between 5/2001 and 8/2008. Eligibility criteria: age ≥ 18 years; PS 0-2; histological diagnosis of GBM; cytotoxic treatment at disease progression after RT/TMZ. The log-rank test was used to evaluate the significance of the prognostic variables, and the Cox model to ascertain any association between PFS and OS. Results: 150 pts (median age: 52 years, [24–76 y]) were enrolled. MGMT methylation status, evaluable in 110 pts, was present in 38% of cases. Median OS was 18.5 months. At disease progression, 40 pts (27%) received temozolomide, 92 pts (61%), nitrosourea-based chemotherapy, and 18 pts (12%) other treatments. At the time of recurrence, mPFS was 2.5 months (95%CI: 2.0–3.1), PFS-6 15% (95%CI: 9.5%–21.3%, mOS 7.6 months (95%CI: 6.9–8.3) and OS-6 64% (95%CI: 56.6%–72.2%). In the Cox proportional hazard model, PFS with the second line treatment was correlated with OS measured as from the start of the second line treatment (p < 0.0001). Conclusions: The findings made in the present study, the first in literature to evaluate outcome endpoints for a second line treatment after RT/TMZ in GBM pts, confirm that a PFS-6 of 15% should be considered the cut-off for active cytotoxic drugs in phase II studies on recurrent GBM. For antiangiogenic compounds, OS-6 can be considered as a sound endpoint. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Bristol-Myers Squibb, OncoMethylome Sciences, Roche, Schering-Plough GlaxoSmithKline, Merck Sharp & Dohme, Roche, sanofi-aventis, Schering-Plough

Phase II study of XL184 (BMS 907351), an inhibitor of MET, VEGFR2, and RET, in patients (pts) with progressive glioblastoma (GB).

2006 Background: XL184 is an oral, potent inhibitor of MET, VEGFR2 and RET that has demonstrated clinical antitumor activity. In solid tumors, the MTD was 175 mg qd. Elevated levels of VEGFR2 and MET are found in GB, the latter correlated with poor prognosis. Methods: XL184 was studied in two cohorts of previously treated GB pts at 175 and 125 mg qd. The 125 mg cohort was added to further optimize tolerability while maintaining activity. Pts with and without prior antiangiogenic therapy (AAT) were allowed. Primary objectives include independent centrally confirmed (IRF) overall response rate (ORR), 6-month progression-free survival (PFS6) and safety. Secondary objectives include duration of response, survival, and changes in steroid usage. Plasma biomarkers were assessed for evidence of target inhibition. Results: Interim results are presented for 124 pts enrolled with a median age 55 yrs (range: 21-73); 42 (34%) had prior AAT, 42 (34%) were in second relapse, and 48 (39%) required corticosteroids at baseline. In the 175 mg qd cohort (n = 46) PFS6 (all pts) was 21%; ORR in AAT-naïve (n = 34) and AAT-pretreated pts (n = 12) was 21% and 8%, respectively; median duration of response was 5.9 months (range: 1.9-12.8). In the 125 mg qd cohort, ORR in response- evaluable AAT-naïve (n = 25) pts was 32%, with a median duration of response not yet reached. The rate of ≥ 50% decline in size of enhancing lesions at ≥ 1 timepoint, in response-evaluable AAT-naïve pts for the 175 mg and 125 mg qd cohorts was 41% and 60%, respectively. Decrease in size of enhancing lesions has been observed in some AAT-pretreated pts. Most frequent Gr3/4 AEs were fatigue (23%), hypophosphatemia (10%), serum lipase elevation (10%) and ALT, headache, lymphopenia and convulsion (9% each). Dose interruption/reduction, AEs related to TKI/VEGFR-inhibition (hypertension, proteinuria, bleeding, hand-foot syndrome), and modulation of plasma biomarkers for VEGFR and MET inhibition (VEGF-A, PlGF, sVEGFR2, sMET) were observed at both doses. Conclusions: XL184 demonstrates clinical activity and target inhibition of MET and VEGFR2 at 175 mg and 125 mg qd in pts with progressive GB, including pts with prior AAT. Longer follow-up will allow better risk/benefit evaluation. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Exelixis Exelixis Exelixis

Change in MGMT methylation status between first and second surgery for recurrence: Clinical implications

2027 Background: MGMT promoter methylation status is known to be a potent prognostic factor in newly diagnosed glioblastoma (GBM) patients (pts). However, it is not yet clear whether and, if so, how MGMT methylation status may change; nor is it known whether the prognostic role of this epigenetic feature is retained during the disease course. Methods: A retrospective analysis was made using a database of 614 GBM pts treated prospectively from January 2000 to August 2008. We evaluated only patients who met the following inclusion criteria: age ≥18; PS 0–2; two distinct surgical procedures; histological diagnosis of GBM both at first and at second surgery for recurrence; postoperative treatment consisting of: a) radiotherapy (RT) followed by temozolomide (TMZ) until 2005, and b) TMZ concurrent with and adjuvant to RT after 2005; a time interval ≥3 month between first and second surgery. The study aim was to evaluate changes of MGMT status during the course of GBM. The log-rank test was employed to evaluate the significance of the prognostic variables. The percentages of MGMT methylated cases at first and second surgery were compared using the McNemar test. Results: MGMT status, evaluated at first and second surgery in all 44 pts (M:F 32:12, median age: 49 years, range: 27–67), was assessable in 38 (86.4%) cases: MGMT promoter was methylated in 13 (34.2%) pts at first surgery. MGMT methylation status, unchanged in 63.2% of second surgery samples, changed more frequently in methylated than in unmethylated pts (61.5% vs 24%, p = 0.03). The median survival was 24.3 months (95% CI: 20.8–27.7), being 35.2 months (95% CI: 10.1–60.2) and 21.9 months (95% CI: 17.3–26.5) for pts with methylated and unmethylated MGMT assessed at first surgery, respectively (p = 0.04). However, MGMT status at second surgery was no longer prognostic for survival (p = 0.1). Conclusions: Significant changes in MGMT methylation status during the course of GBM occur more frequently in MGMT methylated than unmethylated cases. Moreover, while MGMT methylation status is prognostic at first surgery, it appears to be of no prognostic utility at the time of second surgery. No significant financial relationships to disclose.

Recurrence pattern after concomitant radio-chemotherapy in newly diagnosed glioblastoma patients: Correlation with MGMT promoter methylation status

2027 Background: Temozolomide (TMZ), concomitant with and adjuvant to radiotherapy (RT), has become the standard treatment for newly diagnosed glioblastoma (GBM). The aim of the present analysis was to evaluate the recurrence pattern in GBM patients (pts) treated with this treatment and to assess its correlation with MGMT promoter methylation status. Methods: A review was made of brain MRI images available in 95 GBM pts enrolled from January 2005 to January 2007, all of whom had been prospectively treated with TMZ (75 mg/m2/day) concomitant with standard RT (60 Gy/30F), followed by adjuvant TMZ (150–200 mg/m2 days 1–5, q28, maximum 12 cycles if MRI findings were negative or until progression if MRI appeared positive. An analysis was made of the correlation between MGMT promoter methylation status (assessed with methylation specific PCR) and the recurrence pattern. Results: After a median follow-up of 18 months (range: 6.6 - 44.8), 79 pts (83%) presented GBM recurrence: 57 pts (72%) had recurrence within, and 17 pts (22%) exclusively outside, the RT field; 5 pts (6%) had recurrence both within and outside the field. The median progression free survival (PFS) of pts with recurrence within, within plus outside, and only outside the RT field was 9.2, 8.7, and 14.9 months, respectively (in+in&out vs out p=0.025). The median survival time (MST) was 17.3 months (95%CI: 15.1–20), 14.8 months (95%CI: 10.8-NA), and 26.1 months (95%CI: 19-NA) in pts with recurrence within, within and outside, and only outside the RT field respectively (in+in&out vs out: p=0.013). Recurrence outside the RT field was reported in 15% of unmethylated pts and 42% of methylated pts (p=0.01). About 20% of GBM pts treated with TMZ concomitant and adjuvant to RT may develop recurrence outside the RT field. In pts with methylated MGMT promoter status with a longer PFS and OS, the incidence of recurrence outside the RT field is significantly higher. Conclusions: These data, of value in the development of surveillance strategies for GBM patients, may be an useful aid in selecting treatment for GBM pts in clinical trials. No significant financial relationships to disclose.

Treatment failure due to intracerebral stem cell-like behavior of glioblastoma (GBM) cells: A reason for targeted maintenance therapy with imatinib (I) and hydroxyurea (H)? An analysis of study DE21 and DE40 (Ambrosia)

2048 Background: As the role of normal glia cells is not totally understood the behavior of glioblastoma cells can be described as stem cell like, able to migrate along intra cerebral cords and eluding established treatment including surgery, irradiation and chemotherapy. As PDGF-R signaling seems to play a role in GBM proliferation, PDGF-R signal transduction inhibition with I was analyzed. While monotherapy of I was not effective in GBM, the combination of I plus H (I/H) showed efficacy in a small number of progressive patients (pts), with stable disease (SD) being the main result. The role of H in this concept and the optimal treatment setting for I plus H is not clear so far. Methods: We conducted two trials: study DE21, a single-center, Phase II study to analyze the effect of I/H as maintenance treatment (MT) in patients with SD; and the multi-center, randomized, phase III, open-label, cross-over design study DE40 to analyze the efficacy of I/H versus H as induction treatment in pts with progressive disease (PD). In DE21, 30 SD GBM pts post-radiotherpy (RT) and at least one chemotherapy regimen received I 600 mg/day with H 1,000 mg/day. In DE40, 240 recurrent pts, post-temozolomide failure, received either I 600 mg/day plus H 1,000 mg/day, or H 1,500 mg/day. Results: Characteristcs of both patient populations were typical. Performance status was 0, 1 or 2. Toxicity was moderate without grade IV toxicity in both studies. In DE21 22/30 pts were in stage of SD after at least one progression. In DE21 mean progression-free survival (PFS) was 7.5 months, 36 months PFS was 17%, 4 of these 5 pts with at least one prior relapse. In DE40 mean PFS was 7 weeks, 12 months PFS less than 5%. Conclusions: The results show that MT with I/H in SD pts is superior to the treatment of PD pts. As at least in part GBM cells show a stem cell-like behavior in the brain, treatment strategies can be divided into a remission inducing strategy followed by a MT for detectable or undetectable residual disease, thus preventing these GBM cells from proliferating. For some GBM pts I/H seems to be a durable MT while the role of I/H in remission induction is inferior. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Novartis Pharma Novartis Pharma Novartis Pharma Novartis Pharma

Patterns of care for 952 patients with newly diagnosed glioblastoma

2052 Background: This work aims to describe oncological patterns of care, prognostic factors and survival for patients (pts) with histologically confirmed glioblastoma (GB) newly diagnosed in France in 2004. Methods: The French Brain Tumor Data Bank identified newly diagnosed and histologically confirmed GB pts in 2004, and collected prospectively initial data. A specific data card was compiled to collect retrospectively the management and the follow up care of these pts from 2004/01/01 to 2006/12/01. Results: 952 pts with newly diagnosed and histologically confirmed GB [males/females 1.6, median age 63.9 yrs, median preoperative Karnofsky performance status (KPS) 80] have been recorded from academic and private institutions in France. Surgery consisted of resection (RS): 541 pts and biopsy (B): 411 pts, with 183 pts with no subsequent oncological treatment. After surgery, first line treatment (n=769) consisted of radiotherapy (RT) and temozolomide (TMZ) concomitant ± adjuvant (RT-TMZ) in 331 pts, RT alone in 236 pts, chemotherapy (CT) alone in 157 pts, and other modality of treatment in 45 pts. The median overall survival of the cohort was 286 days [95% confidence intervals (95%CI) 266- 314] significantly impacted by age, KPS and tumor location (unilateral versus bilateral). Median survival (days, [95%CI]) associated to these main strategies, when analyzed by surgical group, was as followed: RS+RT-TMZ (n=237): 474 [427–506], B+RT-TMZ (n=94): 329 [301–371], RS+RT (n=147): 363 [331–431], B+RT (n=89): 178 [153–237], RS+CT (n=61): 245 [190–361], B+CT (n=96): 244 [198–280], and B only (n=118): 55 [46–71]. Conclusions: This population study confirms in everyday practice results of the pivotal study that define RT and TMZ combination as the new standard of care for GB, and emphasis the role of surgery in this tumor type. No significant financial relationships to disclose.

A comparative study of circulating endothelial cells (CECs) and circulating progenitor cells (CPCs) kinetics in four multidisciplinary phase II studies of antiangiogenic agents

3544 Background: CECs and CPCs are being evaluated as generic biomarkers of antiangiogenic therapy in cancer patients (pts). We hypothesized that: 1. CEC and CPC kinetics depend on the antiangiogenic agent used; and 2. CECs and CPCs have differential biomarker value. Methods: Using a standard flow cytometry protocol, we measured the fraction of cells with CEC and CPC phenotype among blood mononuclear cells at multiple time-points during treatment in 4 phase 2 studies of anti-VEGF agents: 1. sunitinib in hepatocellular carcinoma (HCC); 2. cediranib in glioblastoma (GBM); 3. bevacizumab (BV) with chemoradiation in rectal cancer (RC); and 4. BV with chemotherapy in ovarian cancer (OC). The changes posttreatment were evaluated using the Wilcoxon test. Correlations between CECs and CPCs with measures of tumor response (imaging, histology, RECIST) were analyzed with a Spearman’s correlation test in a mixed effects model or using the Wald test in a proportional hazards model. Results: Sunitinib induced a sustained decrease in CPCs (p<0.05), but not CECs in HCC pts. Cediranib did not significantly change the CECs or CPCs over time in GBM pts. BV decreased the CECs and CPCs on day 3 (p<0.05) but not on day 12 when given alone, and not when combined with chemoradiation in RC pts. BV with chemotherapy induced a sustained increase in CPCs (p<0.01), but not CECs in OC pts. The number of CECs correlated with RC response to BV/chemoradiation when evaluated at pre- surgery (p<0.05) and tended to correlate with GBM radiographic relapse on cediranib (p=0.06). In contrast to results from preclinical studies, best tumor responses to BV/chemotherapy in OC correlated with higher pretreatment CPCs (p<0.05). In HCC pts, the time-dependent CPC changes positively correlated with mortality risk on sunitinib (p<0.05). Conclusions: CEC and CPC kinetics depend on the type of antiangiogenic agent. Similarly, their biomarker value is differential. Further characterization of the CECs and CPCs in an agent-specific manner might provide ways to use them to optimize antiangiogenic therapies. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration AstraZeneca, Bayer, Dyax, Eli Lilly and Company, Genentech, GlaxoSmithKline, Pfizer, Takeda Pharmaceuticals Genentech AstraZeneca, Bayer HealthCare Pharmaceuticals, CuraGen Corporation, DARA Biosciences, Genentech, Inc., MGI Pharma, Inc.

Correlation of response with survival endpoints in patients with newly diagnosed and recurrent glioblastoma (GBM) treated on prospective North Central Cancer Treatment Group (NCCTG) clinical trials

2024 Background: The high response rates observed in recent trials of GBM patients (pts) treated with VEGF inhibitors, and the ongoing search for early endpoints which facilitate the efficient evaluation of new agents, have renewed the focus on tumor response as a potential surrogate for survival. Methods: In newly diagnosed (N=1,359) and recurrent (N=357) GBM pts treated prospectively on NCCTG phase II or III trials, we correlated best response (BR) category (responder or non-responder) with survival endpoints (PFS, TTP and OS). A responder was defined by standard NCCTG criteria as CR, PR or REGR (regression), and non-responder defined as STAB (stable) or PROG (progression) following treatment. Log-rank tests and Cox proportional hazard models were used to compare differences between survival and progression for each category. Landmark analyses at different cut-points were used for OS (newly diagnosed: 26 wks; recurrent: 9 and 13 wks) and TTP/PFS (newly diagnosed: 13 wks; recurrent: 6 and 9 wks). Results: For newly diagnosed GBM, best response was CR (n=9), PR (n=17) and REGR (n=150) (176 pts, 13%). For recurrent GBM, BR was CR (n=1) and REGR (n=26) (27 pts, 7.6%). For responders, median time to response for newly diagnosed and recurrent pts were 14 and 7 wks. For newly diagnosed pts, 988 pts were alive (176 responders, 812 non-responders) at wk 26. Median OS for responders and non-responders were 36 and 28 wks, respectively (p=0.01). At week 13, 1173 pts were alive without PROG (189 responders, 984 non-responders). Median TTP for responders and non-responders were 47 and 34 wks, respectively (p=0.0002). For recurrent pts, 251 were alive at wk 9 (227 non-responders, 24 responders), with median OS of 18 and 31 wks respectively (p=0.13); 195 pts were alive at wk 6 without PROG (171 non-responders, 24 responders), and median TTP were 7 and 21 wks (p=0.02). Results were not significantly altered after adjustment for age, gender and ECOG status. Conclusions: In newly diagnosed GBM, tumor response correlated significantly with TTP, but the predictive effect for OS was less robust, albeit reaching significance. For recurrent pts, response was an acceptable surrogate for TTP/PFS, but not OS. No significant financial relationships to disclose.

Overall survival of primary glioblastoma (GBM) patients (pts) receiving carmustine (BCNU) wafers followed by radiation (RT) and concurrent temozolomide (TMZ) plus rotational multi-agent chemotherapy

2070 Background: Treatment of pts with primary GBM using BCNU wafer implantation at initial surgery (Westphal, 2003) as well as the administration of TMZ concurrently and after RT (Stupp, 2005) have each been shown to increase survival. Thus, we hypothesize that the use of the combination of BCNU wafer followed by RT and concurrent (TMZ) plus rotational multi-agent chemotherapy (CCNU, TMZ, CPT-11) will increase survival compared to pts that do not receive BCNU wafers. Methods: Retrospective IRB-approved analysis was conducted on 85 eligible GBM pts who received surgery with (36 pts) and without (49 pts) BCNU wafer insertion followed by RT and concurrent (TMZ) plus adjuvant rotational multi-agent chemotherapy. Survival was estimated within treatment groups using the Kaplan-Meier method. Cox regression was used to evaluate the effect of BCNU wafers on survival after controlling for covariates such as RPA (Recursive Partition Analysis) class, race, and gender. Results: Within the non-BCNU wafer cohort, 39% of patients were younger than 50 and 89% had KPS&gt;70%. Within the BCNU cohort, 31% of patients were younger than 50 and 94% had KPS&gt;70. Overall 1- and 2-year survival for the non-BCNU wafer cohort was 67% (95% CI: 55%, 82%) and 27% (95% CI: 17%, 42%), respectively, with a median survival of 72.7 wks (median follow-up: 123.9; range 14.3–198). Overall 1-year survival for the BCNU wafer cohort was 78% (95% CI: 65%, 93%) with a median survival of 89.6 wks (median follow-up: 118.6; range 26.7–192.1). Given the trends towards significance (p=0.092) between the two groups using a log-rank test, a Cox regression model to control for known prognostic factors was examined: Conclusions: After controlling for the effects due to RPA class, a significant difference in survival is found between treatment groups (p = 0.0325) and patients receiving BCNU wafers have a lower risk of death (HR = 0.548) than patients not receiving BCNU wafers. [Table: see text] No significant financial relationships to disclose.

Single center phase II trial analysing the role of imatinib/hydroxyurea in patients (pts) with pretreated non-progressive glioblastoma (GBM) as maintenance treatment

2055 Background: In GBM the highest malignant brain tumor with a median survival of about 15 months dysregulated signalling of platelet derived growth factor receptors (PDGF-Rs) is implicated in pathogenesis. I/HU showed impressive efficacy and tolerability in patients (pts) with recurrent progressing GBM. In 30 pilot pts with recurrent GBM the progression free survival (PFS) at 6 and 24 months (m) was 32% and 16% respectively. Disease stabilisation (SD) in 37% was an important result. SD for more than 2 years was possible. In GBM a short period of SD after primary or relapse treatment is typical. In this phase II study the efficacy of I/HU as maintenance treatment in pts with GBM in stage of SD was analysed. Methods: From December 2003 up to June 2005 30 GBM pts with documented SD for more than 6 weeks following prior treatment, including surgery, radiotherapy and at least one chemotherapeutic regimen were included,. No enzyme-inducing anticonvulsive drugs were allowed. I dose was 600 mg od, HU dose was 1000 mg (500 mg bid) as continuous oral treatment, pts were followed up by blood cell count weekly and magnetic resonance imaging every 6 weeks. Results: All 30 pts were eligible for safety and for 6, 12 and 18 m PFS and overall survival (OS); 25 pts were male, 5 pts female, median age was 44 years (32 to 71), 24 pts had primary (de novo) and 6 pts secondary GBM. All 30 pts had prior radiotherapy, 21 pts had prior temozolomide and 9 pts non- temozolomide containing regimens. 8 pts were free from relapse, 17 pts after first and 5 pts after second relapse. The median observation time is 28 m. 6, 12 and 18 m PFS was 60% (18/30), 40% (12/30) and 30% (9/30) respectively. 6, 12 and 18 m OS is 90% (27/30), 67% (20/30) and 53% (16/30) so far. PFS for more than 24 m occurred in 3/6 pts with secondary and in 2/24 pts with primary GBM. Toxicity was low (anemia grade 3: 2 pt; grade 2: 4 pts; leucocytopenia grade 3:2 pts; grade 2: 7 pts; thrombocytopenia grade 2: 4 pts), HU dose was reduced in 8 pts, I dose in 1 pt and G-CSF was given in 8 pts, no treatment related death occurred. Conclusion: I (600 mg/d)/ HU (1000 mg/d) was well tolerated in this study. Long-term disease stabilisation was possible - especially in pts with secondary GBM. Confirmation in further investigation is required. [Table: see text]

Imatinib plus hydroxyurea in pretreated non-progressive glioblastoma (GBM)—A single center phase II study

1583 Background: GBM is a highly malignant tumor with a median survival of less than 15 months. Dysregulated signalling of platelet derived growth factor receptors (PDGF-Rs) is implicated in pathogenesis. Imatinib (I) plus Hydroxyurea (HU) is effective in patients (pts) with recurrent progressing GBM. In a pilot series of 30 pts with recurrent GBM progression free survival (PFS) at 6 and 24 months was 32 % and 16 % respectively. 37 % of pts achieved disease stabilisation (SD). Despite the aggressive course of GBM, short periods without progression after primary treatment or effective treatment of relapse is common. The current Phase II study is designed to analyse the efficacy of I plus HU treatment in GBM pts with SD as sequential treatment. Methods: From Dec 2003 up to June 2005 30 pts were included. No enzyme-inducing anticonvulsive drugs were allowed. I at the dose of 600 mg and 1000 mg of HU were given as a continuous daily treatment, follow up included blood cell count weekly and magnetic resonance imaging (MRI) every 6 weeks. The primary endpoint is 12 months PFS. Results: All pts are eligible for safety and 28 pts of them for 6 months PFS and 6 months overall survival (OS) so far; 25 pts were male, 5 pts female, median age was 44 years (32 to 71). All 30 pts had prior irradiation, 21 pts had temozolomid containing and 9 pts non-temozolomid containing regimens. Eight pts were free from relapse, 17 pts after first and 5 pts after second relapse. 25 pts had measurable disease in MRI scan, 5 pts had no evidance of disease. The median follow up was 14 months. The best response was partial remission in 4 pts. SD for more than 3 months in 20 pts. 6 months PFS was 64% (18/28) and 6 months OS was 93 % (26/28). Hematotoxicity grade 2 and 3 occurred in 11 out of 30 pts (anemia grade 3: 2 pt; anemia grade 2: 4 pts; leucopenia grade 3:2 pts; grade 2: 7 pts; thrombopenia grade 2: 4 pts) and required dose reduction of HU in 8 pts, dose reduction of I in 1 patient and G-CSF subcutaniously in 8 pts. Conclusions: I (600 mg/d) and HU (1000 mg/d) showed significant efficacy as sequential treatment. Hematotoxicity was higher compared to I (400 mg/d) and HU (1000mg/d). Six months PFS and OS data are promising so far. Complete analysis according to the primary endpoint will be presented at the 2006 ASCO meeting. [Table: see text]