INTRODUCTION — High-grade gliomas are malignant and often rapidly progressive brain tumors that are divided into anaplastic gliomas (anaplastic astrocytoma, anaplastic oligodendroglioma) and glioblastoma based upon their histopathologic and molecular features [1]. (See "Classification and pathologic diagnosis of gliomas, glioneuronal tumors, and neuronal tumors", section on 'Histopathologic and molecular classification'.)
Despite the survival benefit associated with adjuvant radiation and chemotherapy, the majority of patients relapse following initial therapy. Progressive disease can be difficult to distinguish from radiation necrosis or other radiation-induced imaging changes, and this distinction has important implications for further treatment.
Treatment decisions for patients with recurrent or progressive high-grade glioma must be individualized, since therapy is not curative and there are no randomized trials that directly compare active intervention versus supportive care. The benefit of reintervention must be balanced by the risk of iatrogenic neurotoxicity and its impact on quality of life.
The management of patients with recurrent or progressive high-grade glioma, including surgery, radiation therapy, and systemic therapy, is discussed here.
Other aspects of the management of high-grade gliomas that are covered separately include:
●The diagnostic approach to patients with suspected brain tumors (see "Overview of the clinical features and diagnosis of brain tumors in adults")
●Initial surgical management of high-grade gliomas (see "Clinical presentation, diagnosis, and initial surgical management of high-grade gliomas")
●Adjuvant radiation therapy and chemotherapy following surgery (see "Radiation therapy for high-grade gliomas" and "Initial treatment and prognosis of IDH-wildtype glioblastoma in adults")
●Management of anaplastic oligodendroglial tumors (see "Treatment and prognosis of IDH-mutant, 1p/19q-codeleted oligodendrogliomas in adults")
EARLY PROGRESSION VERSUS PSEUDOPROGRESSION — Distinguishing treatment-induced imaging changes from progressive disease has important implications to avoid premature and inappropriate discontinuation of a treatment regimen. This can be particularly difficult in the period immediately following completion of radiation in patients treated with chemoradiation. This phenomenon has been termed "pseudoprogression" [2,3].
Pseudoprogression is a subacute treatment-related effect with magnetic resonance imaging (MRI) features mimicking tumor progression, usually occurring within three months of completion of chemoradiation (image 1) [4]. The diagnosis is typically made retrospectively, based upon spontaneous improvement or stabilization of imaging findings in the setting of continuation of the original chemotherapy (temozolomide) for at least six months.
In retrospective series, the reported frequency of pseudoprogression after chemoradiation ranges from approximately 15 to 30 percent [3,5,6]. Of those patients with worsened imaging findings at the completion of chemoradiation, approximately 30 to 50 percent have pseudoprogression as determined clinically or with biopsy, and the remaining 50 to 70 percent have true disease progression [3,5-7]. While no single clinical or imaging feature can distinguish between pseudoprogression and true progression, the following features can be helpful:
●Presence of symptoms – Pseudoprogression is often asymptomatic, whereas true progression is more likely to be associated with clinical decline [7]. In one large retrospective series, symptoms were present in 67 percent of patients with true progression and 33 percent of those with pseudoprogression [3].
●MGMT status – An unmethylated O6-methylguanine-DNA methyltransferase (MGMT) promotor increases the likelihood of true progression. In a series of 103 patients with newly diagnosed glioblastoma, pseudoprogression was associated with the presence of MGMT methylation and with a better overall survival [8]. By contrast, early progression was significantly more common among those without MGMT methylation and was associated with a shorter survival.
●Specialized imaging – While specialized imaging (eg, magnetic resonance spectroscopy, tumor perfusion, positron emission tomography) does not reliably distinguish between pseudoprogression and true progression, features such as a lactate peak and decreased cerebral blood volume may increase the likelihood of pseudoprogression.
Patients with imaging evidence suggesting progression four to six weeks after completion of radiation and concurrent temozolomide are typically demonstrating early treatment effect rather than disease progression, and thus should be continued on their planned adjuvant chemotherapy unless there is evidence of clinical deterioration or until there is evidence of further changes suggestive of disease progression on imaging.
Surgery may be required to distinguish between treatment-induced tumor necrosis and progressive disease. Interpretation of biopsies in this setting can be challenging, however [9]. If the predominant finding at surgery is treatment-induced necrosis, continuation of adjuvant temozolomide should be considered [10].
ASSESSMENT OF RESPONSE AND PROGRESSION — Patient management decisions require an assessment of both response to treatment as well as subsequent evidence of progressive disease. Traditionally, this approach has used the Macdonald criteria, which rely upon measurement of areas of contrast enhancement [11]. Revised criteria by the Response Assessment in Neuro-Oncology (RANO) working group address problems in assessing patients with pseudoprogression or in assessing progressive disease in patients with nonenhancing lesions [12]. (See "Assessment of disease status and surveillance after treatment in patients with primary brain tumors".)
GENERAL APPROACH
Prognostic assessment — Recurrent high-grade gliomas are associated with a median overall survival of less than one year, and many patients have significant tumor-related symptoms and morbidity. While interventions such as systemic therapy, reoperation, and reirradiation can benefit selected patients, all treatments are palliative and associated with risks and side effects. Further therapy, even when successful, rarely restores neurologic function that has already been lost. Regardless of whether subsequent treatment is pursued, patients should be offered maximal supportive care, including palliative care and hospice when appropriate [13].
One of the most important prognostic factors for benefit from reintervention is the pretreatment performance status [14-17]. Other factors that are useful in predicting the likelihood of benefit from second-line therapy include the extent of disease, the histologic grade (both at initial therapy and at recurrence), the relapse-free interval, and recurrence pattern (ie, local versus diffuse) [18-22]. Patients with a localized recurrence, particularly after a prolonged period of stability, are better candidates for interventions such as reoperation or reirradiation than those with primary refractory disease or diffuse, multifocal tumors. Similarly, patients with a prior low-grade tumor that has progressed to a high-grade glioma are more likely to respond to therapy than those with primary glioblastoma.
Often there is no better treatment than participation in a clinical trial. For patients who are unwilling or unable to participate in a clinical trial, treatment decisions at this stage must be individualized, taking into account patient preferences, prior therapies received, functional status, quality of life, and overall goals of care. The approach below is summarized in the algorithm (algorithm 1) and is generally consistent with consensus-based guidelines published by the National Comprehensive Cancer Network (NCCN) as well as the European Association for Neuro-Oncology [23-26].
Patients with good performance status — Patients who maintain a good performance status at the time of recurrence or progression are candidates for further therapy that must be tailored to the pattern of recurrence, the underlying histology, and prior treatments received. Referral to a multidisciplinary brain tumor center and/or review by a multidisciplinary tumor board are encouraged given the range of options spanning multiple specialties.
●The best candidates for reoperation are patients with large but well-circumscribed, symptomatic tumors that are amenable to complete or near-complete resection, particularly if the tumor has recurred or progressed after an extended interval. Surgical resection does not achieve durable tumor control and is typically followed by further systemic therapy, unless carmustine wafers are placed at the time of surgery. (See 'Reoperation' below.)
●The value of reirradiation of high-grade gliomas is uncertain but may benefit selected patients, especially those with a history of a low-grade tumor that has recurred as a high-grade tumor after an extended interval. Reirradiation is occasionally used in patients with a localized or out-of-field glioblastoma recurrence, particularly if there is a contraindication to further systemic therapy such as myelosuppression. (See 'Reirradiation' below.)
●Systemic options for patients with recurrent high-grade gliomas include bevacizumab (in some countries), second-line chemotherapy (eg, nitrosoureas, temozolomide rechallenge), and experimental therapies in clinical trials. For most patients with recurrent glioblastoma who do not choose to participate in a clinical trial, single-agent therapy with one of these agents is suggested as second-line therapy. Chemotherapy may be a better second-line option for patients with recurrent anaplastic gliomas with codeletion of chromosomes 1p and 19q, since these tumors are expected to be more chemosensitive. Alternating electric fields are available as a salvage therapy for interested patients with recurrent glioblastoma. (See 'Systemic therapy' below and 'Alternating electric fields' below and "Treatment and prognosis of IDH-mutant, 1p/19q-codeleted oligodendrogliomas in adults", section on 'Recurrent disease'.)
Patients with poor performance status — Patients with a poor performance status have a low likelihood of response to salvage therapy and an increased risk of toxicity. In many cases, the risks of pursuing subsequent treatment outweigh the benefits. One caveat is that patients whose decline has been driven primarily by peritumoral edema or steroid toxicities may benefit from a trial of bevacizumab, which can improve symptoms by virtue of its antiedema and steroid-sparing effects. (See 'Bevacizumab' below.)
LOCALIZED THERAPY
Reoperation — Surgery in the setting of recurrent disease may involve either biopsy (for diagnostic purposes) or repeat debulking of tumor. Only approximately 20 to 30 percent of patients with recurrent glioblastoma are candidates for a second surgery [24].
●Biopsy is frequently necessary to distinguish recurrent viable tumor from the effects of treatment, such as radiation necrosis. Both of these entities will enhance on postgadolinium magnetic resonance imaging (MRI) scanning, and biopsy can provide tissue to differentiate between these two entities. Small biopsy specimens may not be adequate to make the diagnosis due to sampling error, however, and larger samples are sometimes required. Even with adequate tissue it is not always possible to distinguish the two entities, as there may be a mix of both treatment effect and viable tumor.
●The indications for a debulking reoperation in a patient with recurrent disease are not firmly established. The median survival for patients undergoing surgery for recurrent glioblastoma ranges from 8 to 12 months in most series [16,27,28] and ranges from 12 to 18 months for patients with anaplastic astrocytoma [14,15,29]. The same imaging advances and intraoperative techniques used in the initial treatment of patients with high-grade gliomas are valuable in patients being treated for recurrent or progressive disease. (See "Clinical presentation, diagnosis, and initial surgical management of high-grade gliomas", section on 'Intraoperative techniques'.)
●Surgical adjuncts that have been used for primary resection have also demonstrated utility in resection or ablation of recurrent glioma. 5-aminolevulinic acid (5-ALA) guided resection showed benefit in the recurrent glioma setting, with similar diagnostic accuracy in recurrent high-grade glioma compared with newly diagnosed high-grade glioma [30]. Results from the ongoing LAANTERN registry showed that laser interstitial thermal therapy (LITT) stabilizes or improves quality of life from baseline levels in patients with recurrent glioma and radiation necrosis [31].
There is no evidence to suggest that these results are better than can be expected with radiation and/or chemotherapy alone. However, selected patients may benefit from reoperation (eg, those with a bulky tumor exerting symptomatic mass effect). The most significant predictor of longer survival after reoperation is a good performance status [14,15,27,29,32]. Other favorable prognostic variables include young age, a longer interval since the original surgery, and the extent of the second surgical resection [14,15,22,29,32-34]. One study found that ependymal involvement was the most important negative prognostic factor in patients undergoing reoperation for glioblastoma, independent of performance status, tumor size, and extent of resection [35].
Particular caution is required in patients who had been treated with bevacizumab preoperatively because of the risk of wound-healing complications. (See 'Side effects' below.)
Carmustine polymer wafers may prolong survival following reoperation for locally recurrent disease [36,37]. In a phase III trial, 222 patients were randomly assigned to have either carmustine or placebo wafers implanted into the tumor site after surgery for locally recurrent disease [36]. The carmustine wafer group had a significantly longer median survival (31 versus 23 weeks with placebo). Central nervous system (CNS) toxicity was not worse than in the placebo group.
Reirradiation — The role of reirradiation in patients with recurrent high-grade glioma is uncertain, and there is a paucity of prospective data. Based on mostly retrospective series, selected patients with small recurrent tumors and a good performance status may benefit from repeat radiation using modern high-precision techniques to deliver total doses of 30 to 35 Gy in 5 to 15 fractions [23,38-40]. Participation in clinical trials is encouraged.
By contrast, reirradiation with conventional involved field radiation at therapeutic doses used for new-diagnosis high-grade gliomas (54 to 60 Gy) is not recommended in patients with relapsed disease. The administration of a tumoricidal dose is limited by potentially fatal or severe treatment-related toxicity, since most patients will have undergone adjuvant radiation therapy with maximally tolerated doses at the time of primary therapy. Exceptions can sometimes be made for the rare patient who relapses four or more years after initial radiation therapy, but even then, reduced dose and conservative volumes are suggested to minimize toxicity.
External beam delivery techniques — A variety of radiation-delivery modalities can be used to deliver fractionated reirradiation to the brain, most commonly in the form of fractionated radiosurgery or hypofractionated radiotherapy (eg, 30 to 35 Gy in 5 to 15 fractions). Selection is based on preference of the treating radiation oncologist and local availability, since there are no clear differences in efficacy.
Reirradiation can be given both with and without concurrent administration of bevacizumab [38-48]. The available data in patients with recurrent glioblastoma generally suggest that reirradiation modestly improves progression-free survival compared with systemic therapy alone, but overall survival is similar with various approaches [25].
●In an ongoing randomized phase II study sponsored by the Radiation Therapy Oncology Group (RTOG), patients with recurrent glioblastoma are treated with repeat external beam irradiation (35 Gy in 10 fractions) with or without bevacizumab (RTOG 1205, NCT01730950). Preliminary results found that hypofractionated reirradiation improved progression-free survival (7.1 versus 3.8 months) but not overall survival compared with bevacizumab alone [49].
●A smaller single-center trial compared bevacizumab-based chemotherapy with or without fractionated radiosurgery (32 Gy in 4 fractions) in 35 patients with recurrent high-grade glioma [50]. Patients randomly assigned to reirradiation had improved progression-free survival (5.1 versus 1.8 months) and a nonsignificant trend towards benefit in overall survival (7.2 versus 4.8 months, p = 0.11). Adverse effects were similar between groups, and there were no cases of radiation necrosis.
Single-fraction stereotactic radiosurgery (SRS) has been studied in the past for recurrent glioblastoma but has generally been replaced by delivery in multiple fractions. In a series of 114 consecutive patients with recurrent high-grade gliomas treated with single-fraction SRS with a median marginal dose of 16 Gy (range 12 to 50 Gy), radiation necrosis was observed in 24 percent of cases at a median follow-up of 11.2 months [44]. Late effects of normal tissue injury, specifically to the brain in these cases, continue to increase in months to years following reirradiation, and therefore these novel approaches should be carefully considered and cautiously employed.
Brachytherapy — Interstitial brachytherapy has been used in patients with recurrent high-grade gliomas, with several observational studies suggesting a survival benefit [20,51-53]. In the largest series, the median survival in 66 patients with recurrent glioblastoma was 49 weeks from the date of implantation and 52 weeks in the 67 patients with anaplastic astrocytoma [51]. The three-year survival rates in patients with glioblastoma and anaplastic astrocytoma were 15 and 24 percent, respectively.
However, brachytherapy is associated with a high incidence of radiation necrosis [51,54]. Although 92 percent of patients in the above series had no acute toxicity, 40 percent required later reoperation for radiation necrosis [51]. Some series suggest a lower rate of brachytherapy-related radiation necrosis in patients who undergo preimplantation surgical debulking [53,55]. (See "Delayed complications of cranial irradiation", section on 'Brain tissue necrosis'.)
An alternate form of brachytherapy uses an inflatable balloon catheter containing a liquid I-125 radioisotope (GliaSite) inserted at the time of surgical resection [56-58]. This approach allows delivery of a quantifiable dose of radiation to the tissue at highest risk for tumor recurrence. No randomized clinical trials have been reported comparing this form of brachytherapy with other approaches.
Although these data suggest a long-term survival benefit with brachytherapy for patients with local recurrence, the role of brachytherapy is diminishing as experience with SRS and fractionated localized limited field radiation evolves. Forms of external beam radiation are not limited to dimensions, location, and thickness of target, which are all limitations for brachytherapy. (See 'External beam delivery techniques' above.)
Other approaches — Laser interstitial thermal therapy (LITT) is increasingly being studied and delivered in the context of recurrent high-grade gliomas [59] as well as for brain tissue necrosis from radiation therapy [60].
Randomized studies of LITT for recurrent high-grade glioma are lacking. However, available data suggest that it may be a reasonable option for patients with brain lesions that are deep seated or harder to access through craniotomy or for those who may not be good open-surgery candidates [61].
SYSTEMIC THERAPY
Choice of therapy — The most commonly used systemic agents in recurrent or progressive high-grade gliomas are bevacizumab, nitrosoureas, and temozolomide rechallenge (since the majority of patients will have received temozolomide at part of initial therapy). For patients with an adequate performance status, participation in a research clinical study is the preferred approach whenever possible.
Outside of a clinical trial, we suggest single-agent therapy with either bevacizumab, lomustine, or temozolomide in most patients who are selected for systemic therapy. All therapies are considered palliative for patients with recurrent glioblastoma, and none have been directly compared with best supportive care to determine whether survival is improved by active therapy in this setting. Similarly, no single agent has been shown to be clearly superior to another. In the largest randomized trial to test whether combination therapy is superior to single-agent therapy, the addition of bevacizumab to lomustine improved progression-free but not overall survival compared with lomustine alone, and toxicity was higher with combination therapy [62]. Across trials of various systemic therapies, the median survival of patients with recurrent glioblastoma is approximately eight to nine months. (See 'Efficacy' below.)
Absent a major difference in efficacy, the choice among therapies is therefore individualized based on treatment history, performance status, side effect profiles, tumor size and extent of surrounding edema, steroid requirements, and patient preferences. Bevacizumab is often favored in patients with large amounts of symptomatic edema or steroid toxicities, whereas temozolomide rechallenge may be most reasonable in patients who have recurred after a period of observation, especially if the tumor has a methylated O6-methylguanine-DNA methyltransferase (MGMT) promotor. Treatment availability also impacts choice in some regions; bevacizumab has not been granted regulatory approval in Europe for high-grade glioma, for example, and it is not available outside of the context of a clinical trial in some countries.
Bevacizumab — Bevacizumab is a monoclonal antibody that binds to circulating vascular endothelial growth factor (VEGF). It is approved for use in recurrent high-grade glioma in the United States but not in Europe.
Efficacy — Bevacizumab has demonstrated radiographic response rates of approximately 30 to 40 percent as a single agent or in combination with chemotherapy agents such as lomustine for patients with recurrent high-grade gliomas [63-67]. In most but not all studies, it has also shown important steroid-sparing effects in many patients that can positively impact quality of life. Despite these effects, however, bevacizumab has not been demonstrated to improve overall survival in patients with newly diagnosed or recurrent glioblastoma [62,68,69]. (See "Initial treatment and prognosis of IDH-wildtype glioblastoma in adults", section on 'Limited role of bevacizumab'.)
In the largest randomized trial to date in the recurrent setting, 437 patients at first progression of glioblastoma after chemoradiation were randomly assigned to receive lomustine plus bevacizumab or lomustine alone [62]. The initial dose of lomustine was 90 mg/m2 in the combination arm, then increased to 110 mg/m2 on cycle 2 as tolerated. The monotherapy dose of lomustine was 110 mg/m2, and there was no placebo. Compared with lomustine alone, the combination of lomustine plus bevacizumab improved both locally determined objective response rate (41.5 versus 13.9 percent) and progression-free survival (4.2 versus 1.5 months), but not overall survival (9.1 versus 8.6 months; hazard ratio [HR] 0.95, 95% CI 0.74-1.21). The rate of adverse effects was higher in the combination arm, although the median number of cycles of lomustine was also higher in this group (three versus one). Bevacizumab did not delay or reduce the rate of initiation of glucocorticoids, and time to deterioration in health-related quality of life was similar between groups. Approximately one-third of patients in the lomustine-alone arm received bevacizumab at progression.
This trial was unable to confirm the survival advantage of lomustine plus bevacizumab that was observed in the noncomparative phase II trial in which 153 patients with a first recurrence of glioblastoma were randomly assigned to receive bevacizumab, lomustine (110 mg/m2), or combined bevacizumab plus lomustine [66].
Other previous studies have shown similar response rates and survival outcomes between bevacizumab monotherapy and bevacizumab combined with chemotherapy. As an example, initial approval of bevacizumab for recurrent high-grade glioma in the United States was based on a noncomparative phase II trial, in which 167 patients with recurrent glioblastoma were randomly assigned to bevacizumab (10 mg/kg), either as a single agent or at the same dose in conjunction with irinotecan (125 mg/m2 for those not taking enzyme-inducing antiseizure agents and 340 mg/m2 for those taking these agents) [64]. Treatment cycles were repeated every two weeks. All patients had received prior chemotherapy with temozolomide. The objective response rates with bevacizumab alone or in combination with irinotecan were 28 and 38 percent, respectively, and the six-month progression-free survival and overall survival rates were 43 and 50 percent, and 9.2 and 8.7 months, respectively. Treatment with bevacizumab or bevacizumab plus irinotecan was generally well tolerated, and toxicity was limited to that expected with these agents. Steroid doses were stable or reduced in almost all patients. With longer follow-up, the 12- and 24-month survival rates were 38 and 16 to 17 percent on both treatment arms, which appeared to be better than historical control series [65].
Although studies are limited, tumor progression while on bevacizumab is associated with a poor prognosis and reduced likelihood of response to other agents [63,70-75]. There are no controlled trials in this setting. In patients who progress while receiving bevacizumab plus chemotherapy, continuation of bevacizumab and a change of chemotherapy agent to a drug with a different mechanism of action may be attempted in patients with good performance status. In patients who progress on bevacizumab monotherapy, some experts recommend continuation of bevacizumab and addition of a cytotoxic agent.
Dose — The two clinical trials that led to the approval of bevacizumab for recurrent high-grade glioma used a dose of 10 mg/kg every two weeks. This is the suggested dose for glioblastoma listed in the US Food and Drug Administration (FDA)-approved label and the one we use most commonly. However, dose-response studies have not been performed in high-grade glioma, and a variety of dosing schedules are used in other cancer types (ranging from 5 mg/kg every two weeks to 15 mg/kg every three weeks).
Limited observational data suggest that there is no difference in progression-free or overall survival when patients with high-grade glioma are treated with doses ranging from 5 to 15 mg/kg every two to three weeks [76-78]. In fact, at least two retrospective studies have suggested that a lower dose (eg, 7.5 mg/kg every three weeks) may be associated with improved survival compared with the standard dose of 10 mg/kg every two weeks in patients with progressive glioblastoma [79,80].
Less-frequent dosing can be more convenient for patients, many of whom have neurologic disability that makes travel difficult. Based on these considerations and the lack of compelling data to support a dose-response effect, an every-three-week dosing schedule (using 5, 10, or 15 mg/kg per dose) is an alternative to every-two-week dosing.
As is the case for other weight-based oncologic therapeutics, there are no data suggesting increased toxicity for underweight or obese individuals receiving bevacizumab dosed by actual weight. Thus, in the absence of data for harm, doses should be based on actual body weight. (See "Dosing of anticancer agents in adults", section on 'Weight-based dosing'.)
Side effects — Bevacizumab is associated with a wide spectrum of toxicities, including cardiovascular effects, such as hypertension, thromboembolism, and left ventricular dysfunction, and noncardiovascular effects, such as proteinuria, delayed wound healing, and bleeding [81]. The full spectrum of toxicities associated with bevacizumab and their management are discussed separately. (See "Toxicity of molecularly targeted antiangiogenic agents: Non-cardiovascular effects" and "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects".)
Bevacizumab monotherapy is associated with an approximately 2 to 3 percent risk of intracranial hemorrhage (ICH) in patients with recurrent high-grade glioma who are not receiving anticoagulation [64,82-84]. In patients receiving anticoagulants, the risk appears to be higher [82,85]. This was illustrated by a retrospective review of 282 glioma patients treated with bevacizumab, of whom 64 received concurrent anticoagulation [82]. Both overall ICH rate (11 versus 3 percent) and serious ICH rate (3 versus 1 percent) were increased in patients who received anticoagulants compared with those who did not. Therefore, the decision regarding whether to treat patients with concurrent bevacizumab and anticoagulants must be based on a careful assessment of the risk-to-benefit ratio in individual patients. In many cases, however, the benefits of concurrent therapy are felt to justify the risks when progressive tumor and symptomatic venous thromboembolism warrant the use of both.
Hypertension is a frequent side effect of bevacizumab. Guidelines for pretreatment assessment, monitoring, and management of elevated blood pressure in patients receiving bevacizumab are available (table 1A-B). This subject is discussed elsewhere. (See "Toxicity of molecularly targeted antiangiogenic agents: Cardiovascular effects", section on 'Hypertension'.)
Nitrosoureas — For patients who initially were treated with temozolomide and are not candidates for a clinical study or bevacizumab, nitrosourea-based chemotherapy is a reasonable alternative. Nitrosoureas (eg, carmustine, fotemustine) either as single agents or in combination regimens such as procarbazine, lomustine (CCNU), and vincristine (PCV) (table 2) have shown evidence of activity in phase II studies of previously treated patients and are reasonable options [67,86-89].
In prospective studies in patients with recurrent glioblastoma, single-agent lomustine has been associated with a response rate of 9 to 14 percent and a median progression-free survival of 1.5 to 2.7 months [62,90,91]. The most common starting dose when used as a single agent is 110 mg/m2 (maximum 200 mg). The rate of grade 3 and higher hematologic toxicity is approximately 50 percent.
The combination regimen of PCV was compared with temozolomide in a phase III trial in 447 patients with high-grade glioma at first recurrence following initial treatment with radiation alone [92]. There was no statistically significant difference in progression-free survival or in overall survival when patients treated with PCV were compared with those treated with temozolomide (3.6 versus 4.7 months and 6.7 versus 7.2 months, respectively).
PCV has been shown to improve survival in patients with newly diagnosed anaplastic gliomas when given along with radiation therapy, and PCV is a good salvage option in patients who have not previously received it, particularly for tumors with 1p/19q-codeletion. (See "Treatment and prognosis of IDH-mutant, 1p/19q-codeleted oligodendrogliomas in adults".)
Temozolomide rechallenge — Temozolomide using a variety of dosing schedules has been studied in several phase II studies of patients with recurrent high-grade glioma with mixed results [71,93-100]. In general, patients who have relapsed some months after completion of adjuvant temozolomide and whose tumors have a methylated MGMT promotor are the best candidates for rechallenge with temozolomide. There does not appear to be an advantage of dose-intensive regimens over standard temozolomide dosing in patients who are temozolomide naïve [92].
In the largest study (RESCUE), continuous daily temozolomide (50 mg/m2/day for up to one year) was evaluated in 120 patients [93]. For patients with glioblastoma, the six-month progression-free survival ranged from 15 to 29 percent, depending upon whether progression occurred during or after the original adjuvant temozolomide treatment. Patients who were rechallenged after completing an adjuvant regimen were the most likely to respond. For patients with an anaplastic glioma, the six-month progression-free survival was 36 percent with dose-intense temozolomide.
In a randomized phase II trial of dose-intense temozolomide (150 mg/m2/day one week on, one week off or 100 mg/m2/day three weeks on, one week off) in patients with recurrent glioblastoma that was stopped early due to loss of funding, the two different dosing regimens performed similarly [101]. The most important predictor of efficacy was MGMT status. Regardless of dosing regimen, six-month progression-free survival was significantly better in patients with MGMT methylated versus unmethylated tumors (40 versus 7 percent).
As with other agents, the use of dose-intense temozolomide in patients who have progressed on bevacizumab-containing regimens is associated with a poor response rate and overall survival [71,102,103].
Genotype-directed therapies — A growing number of somatic gene alterations have emerged as driver mutations across a range of cancers, including gliomas. Methods for screening patients for actionable mutations and oncogenic gene fusions are evolving, and there is no single standard platform for testing. When feasible, we encourage next-generation sequencing or other methods of testing of tumor samples in advance of recurrence or progression on standard therapies and ideally at the time of diagnosis in patients with glioblastoma.
Testing can identify clinical trial candidates as well as potentially actionable targets for currently approved drugs in a small subset of patients with malignant gliomas, and regulatory agencies have begun to approve drugs in a tissue-agnostic, genotype-specific manner.
TRK fusion-positive tumors — Fusions involving one of the neurotrophic receptor tyrosine kinase (NTRK) genes result in a tropomyosin receptor kinase (TRK) fusion oncoprotein that drives downstream signaling and tumor growth. The reported prevalence of NTRK fusion events is approximately 1 to 2 percent in adult glioblastoma, 4 percent in pediatric gliomas, and up to 40 percent in infantile high-grade glioma (age <3 years) [104-106].
Two TRK inhibitors, larotrectinib and entrectinib, are approved in the United States and elsewhere for use in adults and children with TRK fusion-positive solid tumors without a known acquired resistance mutation, that are either metastatic or where surgical resection is likely to result in severe morbidity, and who have no satisfactory alternative treatments or whose cancer has progressed following treatment. The high response rates and durable remissions seen with TRK inhibitors in TRK fusion-positive tumors have led some to recommend a TRK inhibitor for first-line therapy in patients with TRK fusion-positive advanced cancers. Data on efficacy and a suggested testing algorithm to identify TRK fusions are presented elsewhere. (See "TRK fusion-positive cancers and TRK inhibitor therapy".)
BRAF V600E-mutant tumors — BRAF V600E mutations are reported in approximately 3 percent of glioblastomas [107]. In such tumors, targeted therapy with dual BRAF and mitogen-activated protein kinase kinase (MEK) inhibition is a reasonable strategy in the recurrent setting.
Benefit from the combination of dabrafenib and trametinib was shown in the phase 2 open-label ROAR trial, which included 45 patients with recurrent/progressive high-grade glioma (69 percent glioblastoma) previously treated with radiation therapy and first-line chemotherapy [108]. With a median follow-up of 12.7 months, the objective response rate was 31 percent (3 complete and 11 partial) and median duration of response was 13.6 months by independent radiology review; median progression-free survival was 4.5 months and median overall survival was 17.6 months. Grade 3 or 4 adverse effects occurred in 53 percent of patients in the trial, most commonly fatigue, neutropenia, and headache. Based on these and other data, the combination of dabrafenib plus trametinib was granted tissue-agnostic approval by the FDA in June 2022 in patients ≥6 years of age with unresectable or metastatic BRAF V600E-mutant solid tumors that have progressed on previous treatment [109].
ALTERNATING ELECTRIC FIELDS — A portable medical device that generates low-intensity alternating electric fields (tumor treating fields [TTFields]) is also available for treatment of recurrent glioblastoma. Clinicians must be trained and certified to prescribe the device. The device is applied to a shaved scalp, with four transducer arrays connected to a portable battery or power supply operated device; continuous treatment is recommended. The device is an alternative to other salvage therapies for interested patients.
Approval in the United States, Europe, and elsewhere was based on results of a clinical trial that randomly assigned 237 patients with recurrent glioblastoma to TTFields or clinician's choice chemotherapy [110]. The majority of patients were enrolled at the time of second or greater recurrence and approximately 20 percent had received prior bevacizumab. Median progression-free and overall survival were similar in those treated with TTFields versus chemotherapy (2.2 versus 2.1 months and 6.6 versus 6 months, respectively). The objective response rate was nonsignificantly higher in patients treated with TTFields compared with chemotherapy (14 versus 10 percent). Quality-of-life data available in only 27 percent of patients were similar between groups. Mild to moderate scalp dermatitis related to transducer arrays was the most common device-related side effect (16 percent). Hematologic and gastrointestinal adverse events, primarily mild or moderate, occurred in 20 percent of those treated with chemotherapy and less than 3 percent of device-treated patients.
A subsequent open-label randomized trial in patients with newly diagnosed glioblastoma showed prolonged progression-free and overall survival in the group assigned to TTFields when used in combination with postradiation temozolomide [111]. (See "Initial treatment and prognosis of IDH-wildtype glioblastoma in adults", section on 'Alternating electric fields'.)
Consensus-based guidelines published by the National Comprehensive Cancer Network (NCCN) include alternating electric field therapy as a treatment option for patients with recurrent glioblastoma based on the trial data reviewed above as well as postmarketing analysis of >450 patients treated commercially in the United States [23,112].
EXPERIMENTAL THERAPIES — Patients with recurrent glioblastoma are encouraged to consider participation in clinical trials of new therapies. A large number of surgical, radiation, and systemic approaches are under active investigation. Tumor genotyping can be useful to identify mutations for which clinical trial options may be available. Additional information and instructions for referring a patient to an appropriate research center can be obtained from the United States National Institutes of Health (www.clinicaltrials.gov).
Immunotherapy — Early experience with checkpoint inhibitors such as pembrolizumab and nivolumab in patients with recurrent high-grade glioma has shown only modest activity [113-118], and use of checkpoint inhibitors in unselected patients with recurrent high-grade glioma is not recommended outside of a clinical trial. The level of expression of programmed death ligand 1 (PD-L1) does not help to identify patients who are more likely to respond to treatment [115,118].
The largest study of single-agent nivolumab was the open-label CheckMate 143 trial, in which 369 patients with glioblastoma at first recurrence were randomly assigned to receive nivolumab (3 mg/kg) or bevacizumab (10 mg/kg) every two weeks [119]. With a median follow-up of 9.5 months, overall survival was similar in the nivolumab and bevacizumab groups (9.8 versus 10.0 months, hazard ratio [HR] 1.04, 95% CI 0.83-1.3), and objective response rate was higher for bevacizumab (25 versus 8 percent). A randomized pilot trial in patients undergoing resection of recurrent glioblastoma suggested that initiating pembrolizumab presurgically might improve outcomes compared with initiating therapy postoperatively, but further study is needed [120]. Two large trials of nivolumab in the upfront setting have completed enrollment [121,122].
Combination therapy has also been explored. In a phase I trial of nivolumab with and without ipilimumab in 40 patients with recurrent glioblastoma, nivolumab was better tolerated than combination therapy, and a partial response was seen in three patients (one treated with monotherapy, two with combination therapy) [115]. An additional eight patients had stable disease for ≥12 weeks (two treated with nivolumab monotherapy). PD-L1 expression ≥1 percent was present in 68 percent of tumor specimens but did not correlate with clinical response. Another phase I trial demonstrated the feasibility of combining pembrolizumab and bevacizumab and hypofractionated stereotactic reirradiation [123].
At least one case report describes a favorable response to nivolumab in two children with recurrent glioblastoma related to constitutional mismatch repair-deficiency syndrome [124]. By contrast, a hypermutation phenotype in recurrent glioblastoma does not appear to be associated with response to checkpoint inhibitors, at least experimentally and in retrospective patient cohorts [125]. (See "Risk factors for brain tumors", section on 'Mismatch repair deficiency'.)
Others — A variety of other therapies are in development involving both systemic and local delivery methods.
The antibody-drug conjugate depatuxizumab mafodotin (composed of an epidermal growth factor receptor [EGFR] monoclonal antibody conjugated to a tubulin inhibitor) plus temozolomide was associated with a trend towards improved survival compared with lomustine or temozolomide in patients with recurrent EGFR-amplified glioblastoma (HR 0.71, 95% CI 0.50-1.02) [126]. However, these findings were not supported by a separate trial in newly diagnosed glioblastoma, in which the addition of depatuxizumab mafodotin to temozolomide and radiation did not improve outcomes at the time of an interim analysis, leading to early stopping for futility [127].
Some of the genetically engineered local viral therapies, such as recombinant poliovirus, vocimagene amiretrorepvec, and herpes simplex virus type 1 (HSV-1) G207, have also been associated with an encouraging number of durable responses in early-phase studies in adults or children [128-130]. The phase II/III trial of vocimagene amiretrorepvec in over 403 patients with recurrent high-grade glioma failed to confirm a benefit over standard-of-care single-agent therapy, however [131]. (See "Overview of gene therapy, gene editing, and gene silencing", section on 'Cancer therapy'.)
SUPPORTIVE CARE — Optimal supportive care is critical in the management of all patients with recurrent or progressive high-grade glioma, whether or not they elect to proceed with subsequent therapy [13].
Corticosteroids and antiseizure medications are commonly used in brain tumor patients for management of tumoral edema and seizures, respectively. Each of these classes of drugs can have unintended side effects and toxicity in brain tumor patients, particularly in older patients. These issues are reviewed separately. (See "Management of vasogenic edema in patients with primary and metastatic brain tumors" and "Seizures in patients with primary and metastatic brain tumors" and "Seizures and epilepsy in older adults: Treatment and prognosis".)
Patients with a very low functional status, including those who are nonambulatory and fully dependent for activities of daily living, have a very poor prognosis and are best managed with maximal supportive care alone. Many aspects of palliative care for patients with recurrent or progressive high-grade glioma are common to adults suffering from a variety of advanced-stage illnesses, including management of urinary incontinence, delirium, and increased risk for falls. An approach to the management of these issues is discussed separately. (See "Overview of comprehensive patient assessment in palliative care" and "Overview of managing common non-pain symptoms in palliative care".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Primary brain tumors".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topic (see "Patient education: High-grade glioma in adults (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Progression versus pseudoprogression – An accurate diagnosis of recurrent or progressive disease in patients with previously treated gliomas is essential. The initial treatment may induce imaging changes that are difficult to distinguish from progressive disease. Continued treatment with adjuvant temozolomide is warranted until there is evidence that such imaging changes actually represent disease progression. (See 'Early progression versus pseudoprogression' above.)
●Individualized approach to progression – Despite the use of a combined-modality approach, most patients eventually relapse. The management of patients with recurrent or progressive high-grade glioma is difficult, and active reintervention has not been proven to prolong survival. Treatment decisions at this stage must be individualized, taking into account patient preferences, prior therapies received, functional status, quality of life, and overall goals of care (algorithm 1). (See 'General approach' above.)
●Candidates for local therapy – Surgery may be useful in carefully selected patients to distinguish between tumor recurrence and treatment-induced necrosis, to debulk a localized tumor recurrence, or to provide symptom palliation. Focal reirradiation may also be useful in selected patients with a localized recurrence, particularly after a prolonged period of stability. (See 'Localized therapy' above.)
●Systemic therapies – For most patients with recurrent or progressive high-grade glioma who are selected for systemic therapy, we suggest single-agent therapy with bevacizumab, lomustine, or temozolomide (Grade 2C). Absent a major difference in efficacy, the choice among therapies is based on consideration of treatment history, side effect profiles, tumor size and extent of surrounding edema, steroid requirements, and patient preferences. (See 'Choice of therapy' above.)
•Bevacizumab has important antiedema and steroid-sparing effects in many patients and may be most likely to benefit patients with symptomatic edema or steroid toxicities. (See 'Bevacizumab' above.)
•Lomustine alone or in combination with procarbazine and vincristine (PCV) should be considered in patients with recurrent anaplastic gliomas, particularly those with 1p/19q-codeletion. (See 'Nitrosoureas' above.)
•Patients who have relapsed some months after completion of adjuvant temozolomide and whose tumors have a methylated O6-methylguanine-DNA methyltransferase (MGMT) promotor may be the best candidates for rechallenge with temozolomide. (See 'Temozolomide rechallenge' above.)
●Supportive care – Regardless of whether subsequent treatment is pursued, patients should be offered maximal supportive care, including palliative care and hospice when appropriate. (See 'Supportive care' above.)
