Impact Factor

Therapeutic options in recurrent glioblastoma – an update

Standards of care are not yet defined for patients with recurrent glioblastoma. In this critical review, Katharina Seystahl and colleagues summarise the available literature for patients with recurrent (progressive) glioblastoma treated with repeat surgery, re-irradiation, chemotherapy or immunotherapy approaches.

This is an abridged version of K Seystahl et al (2016) Therapeutic options in recurrent glioblastoma –an update. Critical Reviews in Oncology/Hematology 99: 389-408. It was edited by Janet Fricker and is published with permission ©2016 Elsevier Ireland Ltd. doi:10.1016/j.critrevonc.2016.01.018

lioblastoma is a devastating temozolomide (TMZ) during and after bias, small sample sizes and disease disease with a median overall radiotherapy compared to radiotherapy heterogeneity. Gsurvival (OS) of 8.1 months for alone (12.1 months) (N Engl J the period 2000–2003 and 9.7 months Med 2005, 352:987–96). Promoter for 2005–2008 in population-based methylation of the O6-methylguanine- Diagnosis of progression and studies in the US (J Neurooncol 2012, DNA methyltransferase (MGMT) response 107:359–64). gene is a predictive biomarker for The current standard of care in benefit of TMZ in newly diagnosed The RANO (Response Assessment in newly diagnosed glioblastoma was glioblastoma (N Engl J Med 2005, Neuro-Oncology) criteria are considered established based on the trial of the 352:997–1003). Currently, no to be the most accepted approach for European Organisation for Research standard of care is established for diagnosis of progression and response and Treatment of Cancer (EORTC)/ recurrent or progressive glioblastoma in recurrent glioblastoma (J Clin Oncol National Cancer Institute of Canada (Lancet Oncol 2014, 15:e395–403). 2010, 28:1963–72). In suspected Clinical Trials Group (NCIC- Identification of effective therapies pseudoprogression, repeat MRI imaging in CTG), showing prolonged median has been complicated by lack of shortened time intervals is recommended, OS of 14.6 months by addition of appropriate control arms, selection while usually maintaining treatment. 62 September / October 2016 Impact Factor

Surgery at recurrence Approach for individualised treatment decisions in The role of repeat surgery in patients with glioblastoma progressive or recurrent glioblastoma remains controversial, underlining the need for prospective randomised trials. While some, mainly retrospective, studies suggest survival benefits for repeat surgery (J Neurooncol 2014, 117:147–52; World Neurosurg 2015, 84:301–7), others do not (Neuro Oncol 2014, 16:719–27; Eur J Cancer 2012, 48:1176–84). A post-hoc analysis of the prospective DIRECTOR trial in a subgroup of 59 evaluable patients stratifying for extent of resection showed superior survival only in those patients having received complete resection of gadolinium-enhancing tumours (Neuro Oncol 2016, 18:549–56). Beyond an expected therapeutic efficacy, acquiring tumour tissue at repeat surgery could distinguish between recurrent disease and radiation necrosis, and help biomarker-based decision making. Continuous arrows indicate evidence-based current clinical practice. Dashed arrows represent possibilities of individual decision-making which has still to be confirmed. CCNU – lomustine, KPS – Karnofsky performance scale, RT– radiotherapy, TMZ – Repeat radiotherapy temozolomide, TMZ/RT → TMZ – radiotherapy with concomitant and maintenance TMZ

Evidence for re-irradiation is limited, Five single-arm phase II trials and relevant single-agent activity of the ‘control’ highlighting the need for more randomised six randomised phase II or III trials agent or lack of efficacy of the experimental controlled trials. Concerns around repeat comprising one arm with nitrosourea agents. radiotherapy include radiation necrosis monotherapy were reviewed. Comparison The combination of lomustine plus and neurocognitive impairment as well as of the data is complicated by inclusion of bevacizumab showed prolonged median limited efficacy. TMZ-naïve or TMZ-pretreated patients PFS and OS and higher PFS-6 than the in some trials. Progression-free survival at single agents in the BELOB phase II 6 months (PFS-6) ranged between 17.5% trial (Lancet Oncol 2014, 15:943–53). Chemotherapy for recurrent and 61.5%, and median OS between The promising efficacy signal of this glioblastoma 6.0 and 11.1 months for monotherapy combination was not confirmed in the of nitrosourea agents. Notably, in the EORTC 26101 phase III trial comparing Nitrosoureas randomised studies, lomustine as lomustine plus bevacizumab with Nitrosoureas, such as carmustine monotherapy, commonly intended to be a lomustine alone in patients with recurrent (BCNU), lomustine (CCNU), nimustine ‘control’ agent, showed comparable results glioblastoma, which did not report a (ACNU), and fotemustine, are DNA with the investigational agents enzastaurin difference in OS (8.6 vs 9.1 months), alkylating agents and have been (J Clin Oncol 2010, 28:1168–74), although prolonged PFS (1.5 vs extensively used in glioma treatment. The cediranib (J Clin Oncol 2013, 31:3212–8), 4.2 months) was confirmed Neuro( Oncol use of nitrosoureas increased for recurrent galunisertib (J Clin Oncol 2015, 33:suppl, 2015, 17:suppl 5, abstr LB05). disease when TMZ became standard of abstr 2014) or bevacizumab (Lancet Oncol In summary, nitrosoureas remain care in newly diagnosed glioblastoma. 2014, 15:943–53), pointing towards one standard of care at least for current September / October 2016 63 Impact Factor

clinical trials. It is likely to expect that Bevacizumab In conclusion, bevacizumab has clinical clinical efficacy will be more prominent Bevacizumab, an antibody to the activity with prolonged PFS in recurrent in patients with tumours with MGMT vascular endothelial growth factor glioblastoma, but an effect on OS remains promoter methylation (Lancet Oncol (VEGF), was approved by the FDA uncertain. 2014, 15:943–53; N Engl J Med 2000, in 2009 for treatment of recurrent 343:1350–4). glioblastoma based on two phase II trials showing an overall response rate of around Targeted therapy Temozolomide (TMZ) one third and PFS-6 rates of 42.6% and TMZ was approved for recurrent 29% (J Clin Oncol 2009, 27:4733–40; There is plethora of clinical trials, glioblastoma in 1999 based on two J Clin Oncol 2009, 27:740–45). Approval mainly single-arm studies, evaluating phase II trials, which both used a in Europe was refused due to the lack agents aiming to target receptors or soluble schedule of TMZ 150–200mg/m2 for five of a bevacizumab-free control arm. In factors involved in angiogenesis, oncogenic out of 28 days. In one of these trials, TMZ nine phase II trials with a bevacizumab pathways or factors involved in tumour cell was superior to procarbazine in patients, monotherapy arm, PFS-6 rates ranged stemness or tumour invasiveness. Agents 60% of whom were pretreated with from 18% to 42.6%, with a median OS tested in a randomised design include nitrosoureas, with a PFS-6 rate of 21% from 6.5 to 9.2 months. The BELOB cilengitide (J Clin Oncol 2008, 26:5610– versus 8% and median OS prolonged by phase II trial, comprising a bevacizumab- 7; J Neurooncol 2012, 106:147–53), 1.5 months (Br J Cancer 2000, 83:588– free control arm, showed comparable erlotinib (J Clin Oncol 2009, 27:1268–74), 93). The second trial, conducted as a activity of bevacizumab versus lomustine cediranib (J Clin Oncol 2013, 31:3212– single-arm study, showed a PFS-6 rate as single agents, and increased OS of 8; J Clin Oncol 2009, 27: 1268–74), of 18% (Ann Oncol 2001, 12:259–66). the combination of bevacizumab and enzastaurin (J Clin Oncol 2010, 28:1168– PFS-6 rates of other prospective studies, lomustine (Lancet Oncol 2014, 15:943– 74), galunisertib (J Clin Oncol 2015, mainly without previous TMZ treatment, 53). In contrast, the EORTC 26101 33:suppl, abstr 2014), vorinostat (J Clin using this schedule ranged from 21% to phase III trial showed no difference in Oncol 2015, 33 suppl; abstr 2012) and 24% (Jpn J Clin Oncol 2007, 37:897–906; OS of the combination bevacizumab plus dasatinib (J Clin Oncol 2015, 33:suppl, Ann Oncol 2001, 12:255–7; Oncology lomustine versus lomustine alone (Neuro abstr 2004), with disappointing results. 2002, 63:38–41; Hong Kong Med J 2005, Oncol 2015, 17:suppl 5, LB05). EGFR-targeting agents such as 11:452–6). Several mainly single-arm More than a dozen prospective trials gefitinib or erlotinib showed poor results trials evaluated alternative TMZ dosing combining bevacizumab with other in glioblastoma (J Clin Oncol 2009, schedules aiming at overcoming TMZ agents failed to show an efficacy signal 27:1268–74; Neuro Oncol 2015, 17:430– resistance. Yet, it seems very unlikely that beyond single-agent activity. Agents 9; J Neuro Oncol 2009, 92:99–105; there are relevant differences between the tested include irinotecan (J Clin Oncol Neuro Oncol 2013, 15:490–6). However, various dose-intensified TMZ regimens, 2009, 27:4733–40), carboplatin (Neuro efficacy of EGFR-targeted agents might and their superiority over standard- Oncol 2015, 17:1504–13), the histone- be improved in target-selected patient dose TMZ, for patients experiencing deacetylase inhibitor vorinostat (J Clin populations, since a subgroup analysis of recurrence after a TMZ-free interval, Oncol 2015, 33:suppl, abstr 2012), the afatinib in a phase II study showed longer has not been demonstrated either. The multikinase inhibitor dasatinib (J Clin median PFS for patients with EFGRvIII- DIRECTOR trial demonstrating no Oncol 2015, 33:suppl, abstr 2004), positive than negative tumours (Neuro outcome differences for two alternative etoposide (Br J Cancer 2009, 101:1986– Oncol 2015, 17: 430–9). TMZ dosing schedules established the 94), the mTOR inhibitor temsirolimus role of MGMT promotor methylation as a (Anticancer Res 2013, 33:1657–60), the prognostic marker for benefit of TMZ in EGFR-targeted tyrosine kinase inhibitor Immunotherapeutic recurrent glioblastoma (Clin Cancer Res erlotinib (Neuro Oncol 2010, 12:1300– approaches 2015, 21:2057–64). 10), the multikinase inhibitor sorafenib Prospective trials evaluating TMZ- (Clin Cancer Res 2013, 19:816–23) Therapeutic principles of immuno- based combination regimens, mainly or the histone deacetylase inhibitors therapy include immunomodulatory drugs conducted as single-arm studies, have panobinostat (Neuro Oncol 2015, aiming at activating the immune system failed to provide convincing efficacy 17:862–7) or vorinostat (J Clin Oncol against the tumour, treatment with on- signals beyond single-agent activity. 2015, 33:suppl, abstr 2034). colytic viruses, and different vaccination 64 September / October 2016 Impact Factor

Take home message from the authors

Katharina Seystahl (left ) and Michael Weller (right ) are from the Department of Neurology at the University Hospital Zurich, in Switzerland; Wolfgang Wick (centre) is from the Department of Neurology and Neurooncology Program of the National Center of Tumor Diseases, at University Hospital Heidelberg, in Germany.

here is little evidence for effective treatment options in Clinical implications recurrent glioblastoma due to the paucity of randomised Treatment for patients with recurrent glioblastoma should “Tcontrolled trials and, more importantly, active agents. Most be somewhat individualised. Age, general condition of the clinical trials are single-arm studies lacking a control arm. patient, previous therapy and response to the respective Based on the available data, alkylating chemotherapy with treatment should be taken into account as well as molecular temozolomide or nitrosoureas represents the currently markers, especially MGMT promoter methylation and most widely accepted option for systemic therapy at tumour quality of life with regard to expected toxicities. recurrence. O6-methylguanine DNA methyltransferase Furthermore, we should aim to treat patients within clinical (MGMT ) promoter methylation may serve as a biomarker trials in order to improve the knowledge on effective predicting benefit from chemotherapy with temozolomide or therapies for the future. nitrosoureas, not only for newly diagnosed glioblastoma but also at recurrence. Bevacizumab, the antibody to the vascular Future studies endothelial growth factor, has clinical activity with prolonged Instead of small uncontrolled trials, novel therapeutic progression-free survival in recurrent glioblastoma, but an concepts should be tested in a randomised fashion effect on overall survival is uncertain. The BELOB phase II already at an early stage of the development of the drug. trial pointed towards efficacy regarding overall survival of the Identification of predictive biomarkers will help to further combination of bevacizumab with nitrosoureas; however, develop evidence-based concepts for patients with this was not confirmed in the EORTC26101 phase III trial. recurrent glioblastoma. ” approaches, either cell-based or antigen- a control vaccine plus bevacizumab comparable results for PFS-6 and median based or both. All approaches theoreti- in bevacizumab-naïve patients. The OS in both arms (Eur J Cancer 2012, cally should work best if applied early in rindopepimut arm had a higher overall 48:2192-202). Another ‘chemotherapy- the course of the disease to patients with response rate (24% vs 17%), prolonged free’ approach, evaluating a ketogenic diet minimal residual disease. This is why the PFS-6 (27% vs 11%) and median OS in 20 patients with recurrent glioblastoma, majority of immunotherapeutic studies in (12 vs 8.8 months) (J Clin Oncol 2015, reported disappointing results (Int J Oncol glioblastoma today are conducted in the 33:suppl, abstr 2009). 2014, 44:1843–52). first-line setting and no longer in recurrent glioblastoma. Immune checkpoint inhibition, Other approaches Conclusion interfering with inhibitory T cell signalling via programmed death 1 (PD-1), the An alternative treatment concept Treatment should be individualised, PD-1 ligand or cytotoxic T-lymphocyte- comprises a portable device, called and take into account factors such as associated antigen 4 (CTLA-4), offers a tumour-treating alternating electric fields patient age, performance status, MGMT promising approach. (TTFields/NovoTTF), delivering low- promoter methylation, response to Regarding vaccination, encouraging intensity, intermediate-frequency electric previous regimens, and quality of life results were reported for rindopepimut, a fields, aiming to physically interfere with with regard to toxicities (see figure, p 51). vaccine consisting of a peptide sequence cell division. A phase III trial randomising Further investigations are needed to of EGFRvIII, which was evaluated in patients to NovoTTF versus best improve the prognosis for patients with combination with bevacizumab versus physician’s choice of chemotherapy showed recurrent glioblastoma. September / October 2016 65