The optimal duration of post-radiation temozolomide in newly diagnosed glioblastoma remains unclear, with no published phase III randomised trials. Standard-of-care stipulates 6 months. However, in routine care, it is often extended to 12 months, despite lacking robust supporting data. GEINO14-01 (Spain) and EX-TEM (Australia) studies enrolled glioblastoma patients without progression at the end of 6 months post-radiation temozolomide. Participants were randomised 1:1 to six additional months of temozolomide or observation. Primary endpoint was 6-month progression free survival from date of randomisation (6mPFS). Secondary endpoints included overall survival (OS) and toxicity. 204 patients were required to detect an improvement in 6mPFS from 50 to 60

2017 Background: MDM2 inhibits tumor suppressor p53. Brigimadlin, a potent MDM2–p53 antagonist, restores wild-type (wt) p53 function and has shown early efficacy in pts with solid tumors (LoRusso et al Cancer Disc 2023). GBM is an area of unmet need with 5-year survival <10%. In p53 wt GBM pt-derived xenograft models, brigimadlin promotes tumor cell apoptosis and extends survival in combination with RT. Methods: NCT05376800 is a Ph 0/Ia open-label, single-arm trial that aims to measure brigimadlin concentration in brain tumor tissue in pts with histologically or radiologically newly diagnosed GBM eligible for resection (Ph 0) and determine the maximum tolerated dose of brigimadlin plus RT in pts with TP53wt, IDH wt, MGMT promoter unmethylated GBM (Ph Ia). In Ph 0, pts received one brigimadlin dose (30 mg or 45 mg) ~12–24 h before resection. Ph 0 primary endpoints are the measured total concentration and the calculated unbound concentration of brigimadlin in brain tissue homogenate from non-contrast enhancing (NCE) and contrast enhancing (CE) regions. The predefined threshold for trial continuation is 0.5 nmol/L (corresponding to IC50 in GBM cell lines) unbound brigimadlin in CE samples in ≥50% of pts. Brigimadlin concentration was measured using LC/MS and corrected for amount of brigimadlin in residual blood. Unbound concentration was calculated using an in vitro estimate of unbound fraction (fu): 0.654% (rat brain slice). fu in human plasma was 0.22%. Kp,uu (ratio of unbound concentration in brain vs plasma) was calculated. Biomarker testing was performed. Results: Data are available for 11 pts (brigimadlin 30 mg: n=6; 45 mg: n=5). In the 30 mg group, median total brigimadlin concentration in NCE samples was 267 nmol/L (4 samples, range 86–316 nmol/L) and 332 nmol/L (6 samples, range 272–952 nmol/L) in CE samples. In the 45 mg group, median total brigimadlin concentration in NCE samples was 197 nmol/L (5 samples, range 140–347 nmol/L) and 603 nmol/L (5 samples, range 441–905 nmol/L) in CE samples. Unbound concentration exceeded the 0.5 nmol/L threshold in most cases (Table). Post-brigimadlin, an increase in selected p53 target gene expression was observed in CE vs NCE tissue. Conclusions: Unbound brigimadlin concentrations in CE regions in all pts receiving the low dose of 30 mg brigimadlin exceeded the 0.5 nmol/L threshold. Kp,uu in most patients was close to 1 in CE regions. Biomarker data support target engagement in brain tissue. Our findings support continued investigation of brigimadlin in GBM. Recruitment is ongoing. Updated data will be presented. Clinical trial information: NCT05376800 . [Table: see text]

Abstract BACKGROUND Mouse double minute 2 (MDM2) inhibits the tumor suppressor protein 53 (p53). Brigimadlin, a potent oral MDM2-p53 antagonist, restores wild-type (wt) p53 function and has shown early efficacy in patients with solid tumors. Glioblastoma (GBM) is an area of unmet need with 5-year survival at <10%. In p53 wt GBM patient-derived xenograft models, brigimadlin promotes tumor cell apoptosis and extends survival in combination with radiotherapy (RT). METHODS NCT05376800 is a Phase 0/Ia non-randomized, open-label, single-arm trial that aims to measure brigimadlin concentration in brain tumor tissue in patients with histologically or radiologically newly diagnosed GBM eligible for resection (Phase 0), and determine the maximum tolerated dose of brigimadlin plus RT in patients with TP53 wt, IDH wt, MGMT promoter unmethylated GBM (Phase Ia). In Phase 0, patients received one brigimadlin dose (30 mg or 45 mg) ~12-24 hours before resection. Phase 0 primary endpoints are the measured total concentration and the calculated unbound concentration of brigimadlin in brain tissue homogenate from non-contrast enhancing (NCE) and contrast enhancing (CE) regions. The predefined threshold for trial continuation is 0.5 nmol/L (corresponding to IC50 in GBM cell lines) unbound brigimadlin in CE samples in ≥50% of patients. Brigimadlin concentration was measured using liquid chromatography-mass spectrometry and corrected for amount of brigimadlin in residual blood. Unbound concentration was calculated using an in vitro estimate of unbound fraction (fu): 0.654% (rat brain slice). fu in human plasma was 0.22%. Ratio of unbound concentration in brain versus plasma (Kp,uu) was calculated. Biomarker testing was performed. RESULTS Data are available for 11 patients (brigimadlin 30 mg: n=6; 45 mg: n=5). In the 30 mg group, median total brigimadlin concentration in NCE samples was 267 nmol/L (4 samples, range 86-316 nmol/L) and 332 nmol/L (6 samples, range 272-952 nmol/L) in CE samples. In the 45 mg group, median total brigimadlin concentration in NCE samples was 197 nmol/L (5 samples, range 140-347 nmol/L) and 603 nmol/L (5 samples, range 441-905 nmol/L) in CE samples. Unbound concentration exceeded the 0.5 nmol/L threshold in most cases. Post-brigimadlin, an increase in selected p53 target gene expression was observed in CE versus NCE tissue. CONCLUSION Unbound brigimadlin concentrations in CE regions in all patients receiving the low dose of 30 mg brigimadlin exceeded the 0.5 nmol/L threshold. Kp,uu in most patients was close to 1 in CE regions. Biomarker data support target engagement in brain tissue. Our findings support continued investigation of brigimadlin in GBM. Recruitment is ongoing. Updated data will be presented.The trial is funded by Boehringer Ingelheim.

The impact of age on optimal management of glioblastoma remains unclear. A recent combined analysis of two randomised trials, GEINO14-01 and EX-TEM, found no benefit from extending post-radiation temozolomide in newly diagnosed glioblastoma. Here, we explore the impact of age. Relevant intergroup statistics were used to identify differences in tumour, treatment and outcome characteristics based on age with elderly patients (EP) defined as age 65 years and over. Survival was estimated using the Kaplan Meier method. Of the combined 205 patients, 57 (28

Primary brain tumors (pBT) are a heterogeneous group of neoplasms. Disruption of TP53 tumor suppressor is a common event and MDM2 amplification is a major cause of functional loss of TP53 in patients (pts) with glioblastoma (GBM). We describe the characteristics of a pBT cohort with next-generation sequencing (NGS) and MDM2 alterations (MDM2alt). Patients with pBT who had NGS [Foundation Medicine®, local NGS, Caris®, Oncomine®, copy number (CN) and fusion panels] from the multicentric retrospective database ESCAT-pBTs were included. MDM2alt and the relation with other molecular and clinical findings was explored. In GBM, overall survival (OS) was estimated using Kaplan-Meier, with a comparative of survival distribution in MDM2alt and non-alt groups using log rank test. A total of 493 pts were included between Feb 2018 and Jan 2023 at 8 hospitals in Spain. Median age was 51.6 years (y) (3.3-83.8), 12.2% had ECOG ≥2, 39% were women, 27% were ≤40y and 69% had a diagnosis of GBM (WHO 2021). TP53 mutated(mt) was found in 32% of pBT (mutually exclusive with MDM2alt) and 22% had IDH1mt. Overall prevalence of MDM2alt was 8.5%, 11.8% in GBM and 1.3% in other pBT (p<0.001). Main coexisting alteration (alt) in MDM2alt pts was CDK4 gain in 71.4% compared to 5,7% in non-alt (p<0.001). A similar frequency of PI3KCA/PTENmt was observed in GBM MDM2alt and non-alt groups (38% vs 50%, p=0.15). No significant difference was found in GBM MDM2alt vs non-alt in EGFR amplifications (43% vs 32%; p=0.10) and EGFRmt (19% vs 19%; p=0.84). Median TMB was 5.2 mut/Mb in GBM pts. Median OS in GBM was 22.8 months (mo) (95%CI, 20.3-25.2). In MDM2alt group, median OS was 26.5 mo (95%CI 19.2-33.6), while median OS in non-alt was 22.6 mo (95%CI 20.2-25). No statistically significant differences were found in OS among MDM2alt and non-alt groups (p=0.07). MDM2alt is a common genomic event in molecularly defined GBM and has low prevalence in other pBT. MDM2alt did not appear to be prognostic. MDM2alt associated with high prevalence of CDK4 gain, suggesting a molecularly defined subgroup with chromosomal instability and the potential for targeted therapy with CDK4/6 plus MDM2 inhibitors in clinical trials. These findings enlighten further development of treatment strategies for MDM2alt GBM pts.

Trotabresib (TROTA) is a novel bromodomain and extraterminal protein inhibitor that has shown brain tumor tissue penetration, encouraging tolerability and preliminary efficacy in combination with standard of care (SOC) concomitant TMZ + RT and adjuvant TMZ in pts with ndGBM. We present long-term follow-up for part A (dose escalation) and an update on part B (expansion) of the ph 1b/2 CC-90010-GBM-002 study (NCT04324840). Study design has been previously described (Vieito et al. SNO 2022 . Abstr CTNI-21). Part B enrolled pts with IDH wild-type ndGBM; pts were randomized 2:1 to TROTA + SOC then maintenance TROTA (arm A) vs SOC alone (arm B). As of 20 Jan 2023, part A has closed with 32 pts enrolled (concomitant, n = 14; adjuvant, n = 18); part B has enrolled 136 pts (arm A, n = 91; arm B, n = 45). In part A, no new safety events were observed with longer follow-up; median follow-up was 21.6 mo (range 3.4–27.6). The most frequent grade (G) 3/4 treatment-related adverse event (TRAE) was thrombocytopenia (8/14 pts [57%] in the concomitant group and 9/18 pts [50%] in the adjuvant group). Efficacy in part A is shown in the table. At last follow-up, 8 pts (4 per group) remained on treatment, including 1 pt with ongoing complete response at cycle 24. In part B, the most common all-cause G 3/4 AE was thrombocytopenia occurring in 21/88 (24%) and 5/43 (12%) patients in arms A and B, respectively. TRAEs related to TROTA led to treatment discontinuation in 3 pts (arm A), and TRAEs related to TMZ led to discontinuation in 4 pts (2 in each arm); no treatment-related deaths reported. Enrollment of part B was recently completed; efficacy data are not yet mature. Addition of TROTA to SOC followed by maintenance TROTA in pts with ndGBM was well tolerated with no new safety signals in parts A and B, and promising efficacy in part A. Follow-up is ongoing.

Glioblastoma (GB) resistance against anticancer therapies is enhanced by Smoothened (SMO) signaling. Glasdegib (GLG), a SMO inhibitor, may lead to improved efficacy of the Stupp scheme. Newly diagnosed GB patients (pts) received GLG with standard radiotherapy (RT)/ temozolomide (TMZ) followed by maintenance with GLG monotherapy. The study was carried out in two phases. In phase Ib the primary objective was the recommended phase 2 dose (RP2D) in a 3+3 dose escalation strategy. The dose of 75 mg/QD of GLG was declared the RP2D. The primary objective in phase II was 15-m overall survival (OS) rate. The study established a futility threshold of 60% for 15-m OS to consider the trial positive; accrual required: 70 evaluable pts in phase II. Secondary objectives included progression-free survival (PFS) according to RANO criteria, safety, changes in performance status, pharmacokinetic and exploratory biomarker analysis. Between 2018 and 2020, 79 GB pts were enrolled, and 74 (98.7%) pts received GLG at 75mg/QD. The median age was 55 years (range: 28-78), 54% were male, 58.1% were ECOG 1, 39.2% were MGMT methylated, and 1 pts had an IDH1/2 mutation. Complete surgical resection was achieved in 35 (47.3%) pts. GLG treatment lasted a median of 6.9 m (range 0.7-24). 68 (91.9%) finalized concomitant period with RT/TMZ, 64 (86.5%) started adjuvant therapy with TMZ, 33 (51.6%) finalized adjuvant period, and 28 (37.8%) continued GLG monotherapy. 53 pts (71.6%) presented stable disease. After a median follow up of 14.8 m (range 0.7-36.4), mPFS was 7 m (95% CI: 6.1-8.6), and the 12-m PFS rate was 22.2% (95% CI: 14.4-34.2). The mPFS based on MGMT status was 8.4 m (95% CI: 6.5-16.9) and 7.3 m (95% CI: 4.9-9.1) for MGMT methylated and unmethylated respectively (p= 0.044). The mOS was 15.3 m (95% CI: 14-21.2). At data cut-off, 22 (29.7%) pts alive. The mOS was 25.6 m (95% CI: 16-NR) and 14.3 m (95% CI: 11.7-19) for MGMT methylated and unmethylated respectively (p= 0.005). The addition of GLG to standard RT and TMZ showed preliminary efficacy for newly diagnosed GBM, with almost 30% pts still alive at data cutoff. Long-term survival will be updated. No new safety alerts were reported.

2060 Background: Hedgehog signaling through Smoothened (SMO) protein in gliomas promotes cell cycle progression and leads to glioma stem cells (GSCs) maintenance, which constitutes one of the key hallmarks for glioblastoma (GB) resistance against anticancer therapies. Glasdegib, a SMO inhibitor, may disrupt GSCs and lead to enhanced efficacy of the Stupp scheme. Methods: Newly diagnosed GB pts received glasdegib with standard radiotherapy (RT)/ temozolomide (TMZ) followed by maintenance with glasdegib monotherapy. The primary objective was to determine the recommended phase 2 dose (RP2D) in a 3+3 dose escalation (DE) strategy in phase Ib and overall survival (OS) in phase II. Secondary objectives included progression-free (PFS) according to RANO criteria, safety, changes in performance status, and exploratory biomarker analysis. Results: Between 2018 and 2020, 79 GB pts were enrolled and 78 received at least one dose of glasdegib. In DE, 4 pts received Glasdegib at 100 mg/QD and 6 pts received 75 mg/QD. DLTs were reported in 3/4 pts in 100 mg dose level, and 1/6 pts in 75 mg dose level, declaring 75 mg/QD of glasdegib as RP2D. For phase II, 68 additional pts were treated at 75 mg/QD dose. The median age was 55 years (range: 28-78), 54% were male, 45% were MGMT unmethylated, and 1 pts had an IDH1/2 mutation. Glasdegib treatment lasted a median of 6 m (range: 0.5-21.9). Overall, 72 (97.3%) pts completed concomitant therapy, 65 (87.8%) started adjuvant therapy, 28 (37.8%) completed adjuvant therapy, and 23 (31.1%) continued glasdegib monotherapy. Treatment combination was discontinued due to treatment-related adverse events (TRAEs) in 9 (12.2%) pts. For those pts that received RT/TMZ combined with glasdegib at 75 mg/QD dose, 7 (9.5%) presented grade (G) ≥3 hematological TRAEs during concomitant treatment and 2 (3.1%) during the adjuvant treatment. There were no G≥3 TRAEs of any type during the maintenance phase. Neutrophil count decrease G≥3 was reported in 6 (8.1%) pts and platelet count decrease G≥3 in 7 (9.5%) pts. Cutaneous events G≥3 were reported in 3 (4.1%) pts. ECOG, Minimental and Barthel indexes, were maintained when comparing baseline with end of treatment (p = 0.181, 0.25 and 0.346 respectively). Stabilization was the most common response, reported in 60 (81.1%) pts. After a median follow up of 7.8 m (range 0.7-25.9), median PFS was 6.9 m (95% CI: 6.1-8.5). The 6-m PFS rate was 62.1% (95% CI: 50.9-75.8) and the 18 m OS rate was 63.3% (95% CI: 47.5-84.4). Conclusions: The addition of glasdegib to standard RT and TMZ was safe. Glasdegib monotherapy showed no G ≥3 TRAEs. Most patients had disease stabilization, with a promising preliminary PFS and OS for newly diagnosed GBM. Final survival results are awaited. Clinical trial information: NCT03466450.

Based on our proposal for the first ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) classification for brain tumors [Mirallas et al. ESMO 2022 ] and the descriptive analysis of real-world data in gliomas [Mirallas et al. ASCO 2023 ], we aim to define the clinical actionability of molecular alterations (MA) detected by next-generation sequencing (NGS) in a multicentric glioma cohort. Patients (pts) with glioma who had NGS at seven Spanish institutions were included between Jan 2018 and Dec 2022. For predictive associations, pts were classified into Tier1/2, Tier3/4 or wild type (WT) group according to their ESCAT score. Groups were compared using Kruskal-Wallis or Chi-squared tests. To explore prognostic markers, pts with grade 4 glioblastoma (GBM) IDH WT tested within 1 year from diagnosis who had TERT or TP53 genomic results were analyzed. Overall survival (OS) was estimated using the Kaplan-Meier method and Cox hazard ratio were fitted. Of 580 pts included, 567 pts (98%) underwent NGS testing. A total of 395/567 (71%) had a diagnosis of GBM and 165 (29%) harbored a Tier 1/2 MA. Median age was 51y (IQR: 39-61), 381 (67%) had ECOG ≥1, 223 (40%) were women, and 148 (26%) were 40 or younger. Most common MAs in the entire cohort were PI3KCA (29%), CDKN2A/B loss (28%), EGFR gain (26%), IDH1 (22.4%), EGFRvIII and ATRX (both 17%). Tier 1/2 MAs were more prevalent in younger pts (56% vs 10% in <40y vs >40y), non-GBM (65% vs 9% in non-GBM vs GBM), and TP53 mut (52% vs 24% in mutant vs WT) [all p<0.001]. Meanwhile, Tier 3/4 MAs were more typical for grade 4 gliomas with EGFR MAs (57% vs 2% in WT), TERT mut (55% vs 33% in WT) and CDKN2A/B loss (43% vs 13% in WT) [all p<0.001]. Among pts with grade 4 GBM-IDH WT, 53% had TERT mut and 26% had TP53 mut. In this subset of pts, median OS in case of incomplete resection (R1) was 29.5 months [95%CI 20-40.3] for those with TERT WT and 17 months [95% CI 14-22] in TERT mut pts (p=0.01), but no differences in OS were found in TP53 mut vs WT pts. Most common actionable MA were IDH, PIK3CA, EGFR, ATRX, suggesting NGS value for identifying potential candidates for targeted treatment. TERT mut was prognostic of worse OS in pts with grade 4 GBM-IDH WT after incomplete tumor resection.

Recent advances in molecular profiling, have reclassified medulloblastoma, an undifferentiated tumor of the posterior fossa, in at least four diseases, each one with differences in prognosis, epidemiology and sensibility to different treatments. The recommended management of a lesion with radiological characteristics suggestive of MB includes maximum safe resection followed by a post-surgical MR < 48 h, LCR cytology and MR of the neuroaxis. Prognostic factors, such as presence of a residual tumor volume > 1.5 cm2, presence of micro- or macroscopic dissemination, and age > 3 years as well as pathological (presence of anaplastic or large cell features) and molecular findings (group, 4, 3 or p53 SHH mutated subgroup) determine the risk of relapse and should guide adjuvant management. Although there is evidence that both high-risk patients and to a lesser degree, standard-risk patients benefit from adjuvant craneoespinal radiation followed by consolidation chemotherapy, tolerability is a concern in adult patients, leading invariably to dose reductions. Treatment after relapse is to be considered palliative and inclusion on clinical trials, focusing on the molecular alterations that define each subgroup, should be encouraged. Selected patients can benefit from surgical rescue or targeted radiation or high-dose chemotherapy followed by autologous self-transplant. Even in patients that are cured by chemorradiation presence of significant sequelae is common and patients must undergo lifelong follow-up.

21-gene Recurrence Score Assay (OncotypeDx®, RSA) testing is strongly recommended to guide adjuvant chemotherapy in HR+/HER2- early breast cancer (eBC). However, it is not available for all patients. At our site, RSA has been offered since 2012, initially according to a set of regionally defined criteria and with a limited number of tests. Following TAILORx reports, criteria broadened and the number of tests expanded. Here, we investigated whether a high agreement between oncologists might overcome the added value of RSA testing in a real-world scenario. Six staff oncologists of the Breast Cancer Section independently completed a survey on a retrospective series of consecutive patients with HR+/HER2- eBC. Patients had been surgically treated at Hospital del Mar between 2016 and 2019. We addressed three questions: 1) The degree of agreement between oncologists (based on patient age, pTNM, histology, grade, ER/PR levels, and Ki67) on whether chemotherapy (CHT) should have been discussed with the patient besides hormone therapy (HT). High agreement (HiA) was considered if ≥ 5 oncologists agreed either for CHT or for HT, and the rest was considered low agreement (LoA); 2) The proportion of RSA testing in HiA compared to LoA cases; and 3) In patients with HiA that had undergone RSA testing, the rate of change in adjuvant therapy recommendation. We included 291 patients, of which 44% had undergone RSA testing. Mean age was 61 (range 35-91), 75% had stage I and 18% N1 disease. HiA was reached in 80% of cases (234/291): 72% for HT and 28% for CHT. RSA was ordered for 35% (82/234) of HiA cases (49% for HT and 51% for CHT) compared with 82% (47/57) of LoA cases (p<0.001). In the HiA subgroup that had undergone RSA testing, 79% (33/42) of cases with HiA for CHT received HT alone, while 13% (5/40) with HiA for HT received CHT, leading to an overall rate of change of 46% in therapy recommendation after RSA. Oncologists highly agreed on adjuvant therapy recommendations in 80% of cases. RSA testing was more frequent in LoA cases. High agreement between oncologists was not sufficient to overcome the need for RSA testing in driving adjuvant treatment in eBC.

Abstract BACKGROUND Standard treatment of glioblastoma (GBM) is focal radiation with concomitant and adjuvant temozolomide (TMZ) for 6 cycles. The GEINO-14-01 trial (NCT02209948) investigated the role of extending adjuvant TMZ to 12 cycles in a randomized multicenter study. MATERIAL AND METHODS Between Aug/2014 and Nov/2018, 166 patients (p) were screened and 159 randomized to extend (80p) or not (79p) TMZ treatment to 12 cycles after proving stable disease in the MRI performed before inclusion. The trial was stratified by MGMT status and presence or absence of residual disease (defined as a residual enhancement larger than 1cm on the MRI). The primary endpoint was differences in 6monthsPFS, secondary endpoints were differences in PFS, OS, toxicity, between arms and per stratification factors. RESULTS Median age was 60.4 (range 29–83), 97p (61%) were methylated and 83 p (52.2%) were reported with residual disease. Median (m) PFS was 7.9 months (95%CI: 6.1–9.8) and mOS: 20.9 (95%CI: 17.6–24.1). A methylated status was a factor of better PFS (HR=0.29, 95% CI 0.46–0.95; p=0.029) and better OS (HR= 0.43: 95% CI 0.28–0.66; p=0.000) as well as the absence of residual disease (PFS: HR = 0.84: 95% CI =0.71–1.01; p=0.068; OS: HR=0.77, 95%CI 0.63–0.96; p=0.019). We didn’t find any difference in PFS (HR=1.02, 95%CI 0.85–1.21; p=0.82), or OS (HR=0.90; 0.73–1.11; p=0.34) on extending treatment with temozolomide longer than 6 cycles. CONCLUSION There is no benefit of continuing TMZ treatment for more than 6 cycles in the adjuvant treatment of glioblastoma. Final data will be presented at the congress. Supported by a Grant of the ISCIII: PI13/01751

Background MET signaling has a role in gliomagenesis and glioma stem cell maintenance and midkine (ALK ligand) promotes resistance of glioma cells to anticancer therapies. Crizotinib is an ALK and c-MET inhibitor with a preclinical rationale to be tested in newly diagnosed GB. Methods Elegible patients received crizotinib in addition to standard RT and TMZ and then adjuvant TMZ. Maintenance treatment with crizotinib beyond 6 TMZ cycles was allowed. The primary objective was safety evaluation. Secondary objectives included efficacy (progression free survival (PFS) and overall survival (OS)) and an exploratory biomarker analysis. PFS and OS were estimated with Kaplan–Meier method. The results of the dose-escalation cohort (DE) have been reported previously. 250 mg/d was the crizotinib dose selected for the expansion cohort (EC). We report here safety and efficacy for the whole cohort. Results 38 patients (pts) were enrolled, 37 evaluable for safety and 36 for efficacy (12 included in DE). Median age 52 years (33-76). 44%were male. Median KPS 90%, Barthel 100%. 44% were MGMT methylated and 3 pts had IDH1/2 mutation. Most common related adverse events (AE) (all grades) included: nausea (67.6%), asthenia (62.2%), transaminase elevation (40.5%), neutropenia (32.4%) thrombocytopenia (29.7%), diarrhea (29.7%), anorexia (29.7%), vomiting (27%) and constipation (24.3%). In the EC 8/25 pts (32%) presented grade ≥3 AEs (transaminase elevation, thrombocytopenia). 97.2% finished concomitant therapy. 94.4% initiated adjuvant treatment, 67.7% completed 6 TMZ cycles. 18 pts (50%) started maintenance therapy. 8 pts are still on treatment. At the time of this analysis 24 pts have progressed and 1 died without progression. Median follow up was 13.7 months (m), median PFS was 10.78m (95% CI, 7.61-13.94), with 6 month PFS and 12 month PFS of 71.6% and 40.2% respectively. Median OS was 31.4 m(95% CI, 12.64-50.10) with 12month OS of 78.9% and 24month OS of 56.6%. Conclusions In this phase Ib study addition of crizotinib to standard RT and TMZ was safe and resulted in highly promising efficacy for newly diagnosed GB, deserving further investigation. Clinical trial identification NCT02270034. Legal entity responsible for the study GEINO. Funding PFIZER. Disclosure M. Martinez Garcia: Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: ROCHE; Travel / Accommodation / Expenses: PFIZER. E. Pineda: Advisory / Consultancy: Celgene; Travel / Accommodation / Expenses: Sanofi; Travel / Accommodation / Expenses: Amgen. All other authors have declared no conflicts of interest.

Abstract This abstract was withdrawn by the authors. Citation Format: Cortes J, Martin M, Pernas S, Gomez Pardo P, Lopez-Tarruella S, Gil-Martin M, Manso L, Ciruelos E, Perez-Fidalgo JA, Hernando C, Ademuyiwa FO, Weilbaecher K, Mayer I, Pluard TJ, Martinez Garcia M, Vahdat L, Wach A, Barker D, Romagnoli B, Kaufman PA. Withdrawn [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr PD1-02.

PurposeWe retrospectively examined the potential effect on overall survival (OS) of delaying radiotherapy to administer neoadjuvant therapy in unresected glioblastoma patients.Patients and methodsWe compared OS in 119 patients receiving neoadjuvant therapy followed by standard treatment (NA group) and 96 patients receiving standard treatment without neoadjuvant therapy (NoNA group). The MaxStat package of R identified the optimal cut-off point for waiting time to radiotherapy.ResultsOS was similar in the NA and NoNA groups. Median waiting time to radiotherapy after surgery was 13weeks for the NA group and 4.2weeks for the NoNA group. The longest OS was attained by patients who started radiotherapy after 12weeks and the shortest by patients who started radiotherapy within 4weeks (12.3 vs 6.6months) (P=0.05). OS was 6.6months for patients who started radiotherapy before the optimal cutoff of 6.43weeks and 19.1months for those who started after this time (P=0.005). Patients who completed radiotherapy had longer OS than those who did not, in all 215 patients and in the NA and NoNA groups (P=0.000). In several multivariate analyses, completing radiotherapy was a universally favorable prognostic factor, while neoadjuvant therapy was never identified as a negative prognostic factor.ConclusionIn our series of unresected patients receiving neoadjuvant treatment, in spite of the delay in starting radiotherapy, OS was not inferior to that of a similar group of patients with no delay in starting radiotherapy.