Key Points
Question
Does the polyadenosine diphosphate-ribose polymerase (PARP) inhibitor veliparib enhance the efficacy of temozolomide in patients with newly diagnosed, MGMT promoter hypermethylated glioblastoma?
Findings
In this randomized, phase 2/3 clinical trial of 447 patients with newly diagnosed glioblastoma, there was no difference in overall survival for patients randomized to therapy with temozolomide and veliparib compared with temozolomide and placebo.
Meaning
The trial results suggest that trends for enhanced survival in subsets of patients may provide insights for designing subsequent trials with the next generation of more selective, brain-penetrant PARP inhibitors.
Abstract
Importance
The prognosis for patients with glioblastoma is poor following standard therapy with surgical resection, radiation, temozolomide, and tumor-treating fields.
Objectives
To evaluate the combination of veliparib and temozolomide in glioblastoma based on preclinical data demonstrating significant chemosensitizing effects of the polyadenosine diphosphate-ribose polymerase 1/2 inhibitor veliparib when combined with temozolomide.
Design, Setting, and Participants
Patients with newly diagnosed glioblastoma with MGMT promoter hypermethylation who had completed concomitant radiation and temozolomide were enrolled between December 15, 2014, and December 15, 2018, in this Alliance for Clinical Trials in Oncology trial. The data for this analysis were locked on April 21, 2023.
Interventions
Patients were randomized and treated with standard adjuvant temozolomide (150-200 mg/m2 orally, days 1-5) combined with either placebo or veliparib (40 mg orally, twice daily, days 1-7) for 6 cycles.
Main Outcomes and Measures
The primary end point for the phase 3 portion of the trial was overall survival (OS).
Results
There were 322 patients randomized during the phase 2 accrual period and an additional 125 patients randomized to complete the phase 3 accrual, for a total of 447 patients in the final phase 3 analysis. The median (range) age for patients was 60 (20-85) years and 190 patients (42.5%) were female. The median OS was 24.8 months (90% CI, 22.6-27.7) for the placebo arm and 28.1 months (90% CI, 24.3-33.3) for the veliparib arm (P = .17). The difference in survival did not meet the prespecified efficacy end point. However, there was a separation of the survival curves that favored the veliparib arm over 24 to 48 months of follow-up. The experimental combination was well tolerated with an acceptable elevation in grade 3 or 4 hematologic toxic effects.
Conclusions and Relevance
This trial found that adding veliparib to adjuvant temozolomide did not significantly extend OS in patients with newly diagnosed, MGMT-hypermethylated glioblastoma.
Trial Registration
ClinicalTrials.gov Identifier: NCT02152982
This randomized clinical trial evaluates the combination of veliparib and temozolomide in treating glioblastoma.
Introduction
Glioblastoma (GBM) is the most common primary malignant brain tumor and has a poor prognosis. Standard treatment includes surgery and radiotherapy with concurrent temozolomide followed by adjuvant temozolomide and tumor-treating fields (TTF). Sensitivity to temozolomide results from unrepaired DNA alkylation that triggers replication collapse. Base-excision repair (BER) and O6-methylguanine-DNA methyltransferase (MGMT) are the primary mechanisms for removing temozolomide-induced alkylation. While BER is highly efficient in GBM, MGMT activity is variable due to promoter hypermethylation–mediated suppression of MGMT expression.1 In MGMT hypermethylated tumors, mispairing of persistent O6-methylguanine with thymidine results in futile cycles of mismatch repair and replication fork collapse. The resulting DNA breaks are repaired by homologous recombination, and failure to repair these lesions results in cytotoxic effects.2 Unfortunately, even for patients with MGMT-hypermethylated GBM, the emergence of therapy resistance leads to inevitable tumor progression and death.
Polyadenosine diphosphate-ribose polymerase (PARP)–1 (PARP1) modulates multiple DNA repair pathways, and there is a strong rationale to evaluate the potential for PARP1 inhibitors as chemosensitizing agents. PARP inhibitor monotherapy significantly improves survival for patients with ovarian, breast, or prostate cancer with defects in homologous recombination.3 This synthetic lethal effect results from genetic defects in homologous recombination, coupled with drug-induced PARP1 trapping on DNA and suppression of BER and homologous recombination that contribute to replication collapse.4 Given the importance of BER and homologous recombination for recovery from temozolomide-induced damage, we performed a preclinical trial of temozolomide combined with the PARP inhibitor veliparib across 27 GBM patient-derived xenografts (PDXs).5 This demonstrated that deficiency in MGMT activity was required for effective chemosensitizing effects of veliparib. These data underpinned using MGMT hypermethylation as an inclusion criterion in this phase 2 to 3 double-blinded clinical trial that tested the hypothesis that veliparib would enhance the efficacy of adjuvant temozolomide for patients with newly diagnosed GBM.
Methods
Trial Design
Between December 15, 2014, and October 21, 2016 (phase 2), and November 8, 2017, and May 18, 2018 (phase 3), this randomized, double-blind, phase 2/3 clinical trial compared the efficacy of veliparib or placebo combined with adjuvant temozolomide following completion of chemoradiotherapy (Supplement 1, Supplement 2, and Supplement 3). The protocol was approved by the Cancer Therapy Evaluation Program and the central institutional review board at the National Cancer Institute and sponsored under an agreement between AbbVie and the Cancer Therapy Evaluation Program. Patients were preregistered for central pathology review and MGMT testing. Patients with GBM and MGMT hypermethylation who met all other criteria provided written informed consent and were randomized to treatment. Randomization was stratified by age (≤70 vs >70 years), Eastern Cooperative Oncology Group (ECOG) performance status (0 or 1 vs 2), extent of resection (total vs subtotal or biopsy), and planned use of TTF (yes vs no). Patients were randomized 1:1 to veliparib + temozolomide vs placebo + temozolomide using dynamic allocation.6 The primary end point for the phase 2 portion was progression-free survival (PFS), and the primary end point for the phase 3 was overall survival (OS).
Eligibility
Patients 18 years or older were eligible if they met additional inclusion criteria, including (1) MGMT-hypermethylated GBM without 1p/19q co-deletion, (2) completion of concurrent temozolomide treatment (75 mg/m2 daily) and radiotherapy (59.4-60.0 Gy) with recovery from treatment, (3) platelet nadir levels of 75 000/mm3 or greater, (4) ECOG performance status of 2 or less, and (5) adequate organ function and seizure control. Central pathology review of histology slides was performed at the Mayo Clinic (C.G.) based on the World Health (WHO) Classification 2007 classification. Tumors were evaluated for MGMT promotor methylation using quantitative polymerase chain reaction of bisulfite-modified DNA (eMethods in Supplement 4) at the University of Texas MD Anderson Cancer Center.
Treatment and Follow-Up
Temozolomide (150-200 mg/m2 orally, days 1-5 every 28 days) was combined with placebo or veliparib (40 mg orally, twice daily, days 1-7) for 6 cycles. Extended temozolomide monotherapy was allowed for an additional 6 cycles. The protocol was amended on August 15, 2016, to allow for TTF use. History and physical examination, vital signs, fatigue/uniscale assessment,7 complete blood cell count, and chemistry results were obtained at baseline and with each cycle. Contrast-enhanced magnetic resonance imaging (MRI) and follow-up continued for 5 years after treatment using standardized MRI sequences.8
Trial Oversight
The trial was developed by the first author (J.N.S.) and the Alliance for Clinical Trials in Oncology, which also managed the execution of the trial. Clinical data were reviewed by the first author (J.N.S.) and other team members, and the data collection and analysis were performed by the Alliance Statistics and Data Management Center. The phase 2 and phase 3 portions of this study were monitored by the Alliance Data and Safety Monitoring Board, a National Cancer Institute–approved functioning body.
Statistical Analysis
This was a phase 2/3 trial, and randomization was performed centrally by the Alliance for Clinical Trials in Oncology statistical group. The primary end point for phase 2 was PFS, which was defined as the time from randomization until first disease progression/recurrence (site investigator determined) or death. Patients who were alive without documented disease progression were censored at the time of last disease evaluation. The primary end point for the phase 3 trial was OS as measured from the time of randomization until death. Patients who were not known to have died were censored at the time of their last follow-up. The phase 2 portion required a sample size of 293 patients for 90% power to detect a hazard ratio (HR) of 0.67 using a 1-sided log-rank test using a .20 significance level. Accrual was suspended pending the final phase 2 analysis (121 required PFS events). Sufficient PFS activity was demonstrated, and accrual was reopened for completion of the phase 3 portion. The phase 3 analysis included patients accrued during the phase 2 portion. A projected sample size of 400 was needed to obtain the 302 individuals who died required for the final phase 3 analysis, which yielded a 90% power to detect an HR of 0.71 using a 1-sided log-rank test with a .05 significance level. The primary phase 3 analysis was a rerandomization test in which 50 000 simulations were performed by randomizing patients (selected in random order) according to the dynamic allocation procedures used in the trial. The P value was computed as the number of instances the simulated test statistic (from a stratified log-rank test) was the same or more extreme than the observed test statistic from the trial data divided by 50 000. Secondary analyses were performed that compared PFS and OS between arms with a stratified log-rank test. Stratified Cox proportional hazard models were used to estimate the HR for treatment effects. The proportional hazards assumption was violated for OS (but not for PFS), so the OS HRs should be interpreted as the average across the follow-up time. 90% CIs were used to match the stated level of significance; 95% CIs are presented for secondary end points/analyses. Adverse events (AEs) were summarized using the maximum grade of a specific AE that a patient experienced. Mean fatigue scores were compared at each point between the treatment arms using a 2-group t test. The longitudinal fatigue scores were compared between the 2 arms, with the mean normalized area under the curve (AUC). Patient AUCs were normalized by the length of time between first and last fatigue score. Statistical analyses were performed using SAS, version 9.4M7 (SAS Institute).
Results
Study Patients
Between December 15, 2014, and October 21, 2016 (phase 2), and November 8, 2017, and May 18, 2018 (phase 3), 1364 patient cases underwent central pathology review, and 1292 samples underwent MGMT methylation testing. Of the centrally confirmed GBMs, 61% were MGMT hypermethylated, and 57% of those eligible were randomized to treatment: 322 patients (72.0%) were included in the phase 2, and an additional 125 patients (28.0%) in phase 3. A total of 447 patients comprised the intention-to-treat (ITT) population (Figure 1). The median (range) age for patients was 60 (20-85) years; 1 (0.2%) was American Indian or Alaskan Native, 8 (1.8%) were Asian, 17 (3.8%) were Black or African American, 16 (3.6%) were Hispanic, and 396 (88.6%) were White; and 190 (42.5%) were female and 257 (57.5%) male. The treatment groups were well balanced regarding baseline variables (Table 1).
Figure 1. CONSORT Diagram for Patient Enrollment.
MGMT indicates O6-methylguanine-DNA methyltransferase.
Table 1. Baseline Patient Demographic and Clinical Characteristics.
| Characteristic | Arm, No. (%) | Total (N = 447) | |
|---|---|---|---|
| Temozolomide + placebo (n = 224) | Temozolomide + veliparib (n = 223) | ||
| Age, y | |||
| Mean (SD) | 58.3 (12.40) | 58.8 (11.55) | 58.5 (11.97) |
| Median (range) | 60 (20-85) | 61 (22-82) | 60 (20-85) |
| <70 y | 187 (83.5) | 185 (83.0) | 372 (83.2) |
| ≥70 y | 37 (16.5) | 38 (17.0) | 75 (16.8) |
| Race | |||
| American Indian or Alaskan Native | 1 (0.4) | 0 | 1 (0.2) |
| Asian | 4 (1.8) | 4 (1.8) | 8 (1.8) |
| Black or African American | 11 (4.9) | 6 (2.7) | 17 (3.8) |
| White | 197 (87.9) | 199 (89.2) | 396 (88.6) |
| Unknown/not reported | 11 (4.9) | 14 (6.3) | 25 (5.6) |
| Ethnicity | |||
| Hispanic or Latino | 9 (4.0) | 7 (3.1) | 16 (3.6) |
| Not Hispanic or Latino | 208 (92.9) | 212 (95.1) | 420 (94.0) |
| Unknown/not reported | 7 (3.1) | 4 (1.8) | 11 (2.5) |
| Sex | |||
| Female | 94 (42.0) | 96 (43.0) | 190 (42.5) |
| Male | 130 (58.0) | 127 (57.0) | 257 (57.5) |
| Extent of resection | |||
| Gross total resection | 140 (62.5) | 140 (62.8) | 280 (62.6) |
| Subtotal resection or biopsy | 84 (37.5) | 83 (37.2) | 167 (37.4) |
| ECOG performance status | |||
| 0 or 1 | 205 (91.5) | 205 (91.9) | 410 (91.7) |
| 2 | 19 (8.5) | 18 (8.1) | 37 (8.3) |
| Planned concomitant use of Optune devicea | |||
| No | 191 (85.3) | 194 (87.0) | 385 (86.1) |
| Yes | 33 (14.7) | 29 (13.0) | 62 (13.9) |
| Side of lesion | |||
| Bilateral | 3 (1.3) | 8 (3.6) | 11 (2.5) |
| Left | 107 (47.8) | 106 (47.5) | 213 (47.7) |
| Midline | 1 (0.4) | 2 (0.9) | 3 (0.7) |
| Right | 113 (50.4) | 107 (48.0) | 220 (49.2) |
| Tumor location | |||
| Frontal | 67 (29.9) | 65 (29.1) | 132 (29.5) |
| Multiple locations | 82 (36.6) | 74 (33.2) | 156 (34.9) |
| Occipital | 3 (1.3) | 9 (4.0) | 12 (2.7) |
| Other | 1 (0.4) | 3 (1.3) | 4 (0.9) |
| Parietal | 26 (11.6) | 24 (10.8) | 50 (11.2) |
| Temporal | 42 (18.8) | 45 (20.2) | 87 (19.5) |
| Thalamus | 3 (1.3) | 3 (1.3) | 6 (1.3) |
Abbreviation: Eastern Cooperative Oncology Group.
Allowed per-protocol update 3 after 214 patients had been accrued: 106 were receiving temozolomide + placebo and 108 temozolomide + veliparib.
Study Treatment
A total of 430 randomized patients (96.2%) received at least 1 dose of the study drug, and 278 patients (62.2%) completed treatment per protocol (eTable 1 in Supplement 4). Disease progression was the most common reason for discontinuing study intervention: 41 (19.7%) and 51 patients (22.9%) in the placebo and veliparib arms, respectively. AEs were the next most common reason for discontinuing therapy for 7 (3.4%) and 12 (5.4%) patients, respectively. Sixty-two of 233 patients (26.6%) declared an intention to use TTF at the time of enrollment, and ultimately 35 (29.7%) and 36 (31.3%) patients used TTF in the placebo and veliparib arms, respectively. Treatment with temozolomide was extended beyond 6 cycles for 25 of 144 (17.4%) and 18 of 134 (13.4%) patients in the placebo and veliparib arms, respectively. All patients had completed study treatment by March 29, 2019.
Treatment Outcomes
Overall Survival
Survival analysis was based on the ITT population. At the time of final analysis (April 21, 2023), the median follow-up time was 73.6 months (95% CI, 71.6-78.9). There were 367 deaths, with 182 and 185 in the placebo and veliparib arms, respectively. The median OS was 24.8 months (90% CI, 22.6-27.7) for the placebo arm and 28.1 months (90% CI, 24.3-33.3) for the veliparib arm (Figure 2A). The difference in OS between arms was not statistically significant. A multivariable analysis that included the randomization stratification variables yielded an adjusted HR of 0.93 (90% CI, 0.78-1.12). There was a trend for better survival in the veliparib treatment arm from 24 to 48 months of follow-up; the Kaplan-Meier estimates of the OS rate for years 1 to 5 showed the largest difference at 3 years: 29.0% for placebo and 36.8% for veliparib (eTable 2 in Supplement 4). A sensitivity analysis using a modified ITT group that only included patients who received at least 1 dose of protocol treatment (n = 213 for placebo, n = 217 for veliparib) yielded similar results, with no significant difference in OS between arms.
Figure 2. Survival Estimates.
Kaplan-Meier estimates of survival for the intent-to-treat population. Overall survival (OS; A) was defined as the time from randomization to death for any reason (rerandomization P = .17). Progression-free survival (B) was defined as time from randomization to investigator-determined progression according to response assessment in neuro-oncology criteria (P = .26).HR indicates hazard ratio; TMZ, temozolomide.
Secondary End Point Analyses
PFS and objective response rate were secondary end points for the phase 3 trial. There was no significant difference in PFS (Figure 2B; HR, 1.07; 95% CI, 0.88-1.30). An objective response was a complete response or partial response according to response assessment in neuro-oncology criteria and was only evaluable for patients with measurable disease. There was no meaningful difference in the fraction of patients, with an objective response observed for 34 of 118 (28.8%) and 37 of 119 (31.1%) patients in the placebo or veliparib arms, respectively. Based on site investigator–reported incidence of pseudoprogression, 33 patients (14.8%) in the veliparib arm and 47 patients (20.1%) in the placebo arm experienced pseudoprogression (Fisher exact P = .11). A secondary analysis did not identify statistically significant differences regarding OS across stratification variables of age, ECOG performance status, extent of resection, and TTF use (Figure 3), although there was a trend for increased survival in patients 70 years and older treated with veliparib + temozolomide.
Figure 3. Stratification Variables of Age, Eastern Cooperative Oncology Group Performance Status (PS), Extent of Resection, and Planned Use of Tumor-Treating Fields (TTF).
HR indicates hazard ratio.
Adverse Events
Analysis of AEs included patients who received at least 1 dose of the protocol treatment (n = 430). Veliparib therapy was associated with higher rates of hematologic AE: grade 3 or 4 hematologic AEs occurred in 50 of 213 (23.5%) and 108 of 217 (49.8%) patients in the placebo and veliparib arms, respectively (Table 2). Nonhematologic grade 3 or higher AEs were slightly lower in the veliparib arm (49 of 271 [22.6%]) compared with placebo (59 of 213 [27.7%]). There were 2 grade 5 AEs observed in the placebo arm and 4 grade 5 AEs in the veliparib arm (Table 2; eData in Supplement 4). Only 1 grade 5 AE (thromboembolic event) that occurred in the veliparib arm was deemed at least possibly related to treatment. Fatigue was evaluated using a uniscale assessment over time (eTable 3 in Supplement 4). There was no difference in fatigue levels at any point and no difference in the means of the normalized AUC for the fatigue values of each patient over time (placebo mean AUC, 3.61; 95% CI, 3.35-3.87; veliparib mean AUC, 3.76; 95% CI, 3.51-4.01). Overall, the combination of temozolomide and veliparib was well tolerated.
Table 2. Grade 3 or Greater Adverse Events that Occurred in 5% of More of Patients and All Grade 5 Adverse Events Regardless of Attribution.
| Adverse events | Arm | No. (%) | ||
|---|---|---|---|---|
| Grade 3 | Grade 4 | Grade 5 | ||
| Hematologic adverse events | ||||
| Lymphopenia | 1a | 30 (14) | 0 | 0 |
| 2b | 30 (14) | 2 (1) | 0 | |
| Neutropenia | 1 | 6 (3) | 2 (1) | 0 |
| 2 | 45 (21) | 24 (11) | 0 | |
| Thrombocytopenia | 1 | 16 (8) | 6 (3) | 0 |
| 2 | 42 (20) | 24 (11) | 0 | |
| White blood cell count decreased | 1 | 6 (3) | 0 | 0 |
| 2 | 23 (11) | 9 (4) | 0 | |
| Nonhematologic adverse events constitutional symptoms | ||||
| Death not associated with CTCAE term | 1 | 0 | 0 | 1 (0.5) |
| 2 | 0 | 0 (0) | 3 (1) | |
| Other | ||||
| Thromboembolic event | 1 | 5 (2) | 1 (0.5) | 0 |
| 2 | 3 (1) | 2 (1) | 1 (0.5) | |
| Aspiration | 1 | 0 | 0 | 1 (0.5) |
| 2 | 0 | 0 | 0 | |
Abbreviation: CTCAE, Common Terminology Criteria for Adverse Events.
Arm 1 was the placebo group (213 evaluable patients).
Arm 2 was the veliparib group (218 evaluable patients).
Discussion
This randomized clinical trial did not demonstrate a significant overall survival benefit for patients with MGMT-hypermethylated GBM treated with adjuvant temozolomide combined with veliparib. While PFS was superimposable, there was a trend for higher survival with veliparib + temozolomide treatment at intermediate points. Similar discordance in PFS and OS results was observed in several trials in advanced or metastatic colorectal cancer.9,10,11,12 The results for each of these studies could be explained if a more effective experimental treatment produced a deeper nadir in tumor cell burden. If death results from tumor burden exceeding a lethal threshold, then the more cytotoxic therapy could provide longer survival time, while the time from the nadir to detectable tumor progression could be relatively similar. Especially in the context of GBM, in which treatment effects on tumors and the brain can obfuscate interpretation of anatomic imaging, changes in OS are considered a more robust end point. Consistent with this explanation, analyzing only those patients with documented progression, the time to death after progression was 12.5 months and 10.0 months in the veliparib and placebo arms, respectively (HR, 0.81; 95% CI, 0.67-0.98; eFigure 1 in Supplement 4). Thus, despite the inexorable deterioration expected of GBM that resulted in equivalent survival with extended follow-up, the trend toward increased survival at intermediate points in the veliparib + temozolomide arm warrants careful consideration.
The cytotoxic effects of PARP inhibitors are linked to synthetic lethal effects on DNA repair. Originally a concept from developmental genetics, synthetic lethality defines an effect that requires simultaneous genetic or pharmacologic disruption of parallel pathways for lethal effects. The efficacy of PARP inhibitor monotherapy in breast, ovarian and prostate cancer depends on genetic defects in BRCA-associated homologous recombination pathways coupled with suppression of PARP-mediated DNA repair pathways.13,14,15 However, at least partially due to complex synthetic lethal effects when combined with cytotoxic therapies, multiple trials have failed to demonstrate a definitive clinical efficacy for PARP inhibitor/chemotherapy combinations.15,16 Reflecting on the preclinical data, veliparib + temozolomide resulted in a near doubling of survival in 5 MGMT-hypermethylated GBM PDXs, but had nonsignificant effects in another 15 MGMT-hypermethylated PDXs.5 Combined with a trend for improved survival with veliparib + temozolomide in the clinical trial, the collective data suggest that a subset of GBM are uniquely sensitive to the combination. We hypothesized that molecular defects in a key repair pathway, coupled with inhibition of parallel, PARP1-dependent repair pathways, would prevent efficient repair of temozolomide-induced DNA damage in this subset of sensitive GBM. Expansion of our PDX preclinical trial is ongoing to identify additional molecular determinants of sensitivity to PARP inhibition in MGMT-hypermethylated tumors to specifically evaluate this hypothesis.
The trial design used in this study enabled efficient evaluation of the hypothesis that veliparib could enhance the efficacy of temozolomide in GBM patients. Veliparib was selected because combinations with temozolomide were well tolerated in mice and veliparib is brain penetrant.5,17 In the preclinical PDX trial, robust efficacy of the combination was lost if MGMT activity was restored in hypermethylated tumors, which provided a biology-based, a priori identification of MGMT hypermethylation as an inclusion criterion for the trial. Consistent with the import of MGMT methylation status for efficacy, 2 clinical trials testing temozolomide and veliparib in MGMT-unmethylated GBM or recurrent GBM failed to demonstrate any survival signal.18,19 The trial included a phase 2 to phase 3 transition based on PFS and inclusion of the phase 2 patients in the phase 3 analysis of OS. To our knowledge, this was the first seamless phase 2/3 clinical trial used in neuro-oncology, and integration of this efficient trial design, coupled with the biomarker enrichment strategy, enabled complete accrual on this phase 3 trial in less than 3.5 years vs an expected 7 to 10 years if a more conventional sequential randomized phase 2 followed by phase 3 trials were used.
The toxic effect profile for veliparib + temozolomide was consistent with prior studies that evaluated similar combinations. Veliparib accentuated hematologic toxic effects of temozolomide, with a higher incidence of neutropenia, thrombocytopenia, and anemia. These results were similar to other trials evaluating this combination in central nervous system and peripheral solid cancers.18,19,20,21,22,23 In contrast, daily dosing of veliparib with temozolomide and radiotherapy for 6 weeks was poorly tolerated (NCT00770471), which highlights a distinct toxic effect profile associated with protracted temozolomide + veliparib dosing. In the present study, a key entry criterion was a platelet nadir level of 75 000/mm3 or more during radiotherapy + temozolomide, which was intended to exclude patients predisposed to hematologic toxic effects. While combinations with some other nonselective PARP1 inhibitors are overtly toxic, temozolomide combinations with next-generation, PARP1-selective inhibitors are anticipated to have a favorable therapeutic window.
Limitations
Enrollment on this trial was based on the WHO 2007 pathologic classification, so isocitrate dehydrogenase mutation status was not determined for the patients enrolled. While isocitrate dehydrogenase–mutant tumors will be present for a small subset of the enrolled population, their inclusion may skew the survival results reported compared with updated WHO diagnostic criteria. Radiotherapy was performed before trial enrollment, and a retrospective quality assurance review for the radiotherapy administered was not performed.
Conclusions
There are at least 2 highly brain-penetrant PARP1-selective inhibitors in clinical development. With preclinical confirmation of robust in vivo temozolomide-sensitizing effects, there are ongoing efforts to evaluate these in combination with temozolomide in GBM. The results reported in this randomized clinical trial highlight the importance of developing more precise predictive biomarkers to identify the subset of patients with the greatest temozolomide-sensitizing potential. Using such an a priori–defined inclusion biomarker(s) and a phase 2/3 trial design, the present study potentially provides a roadmap for a highly efficient clinical evaluation strategy for PARP1-selective or other novel chemosensitizing strategies in newly diagnosed GBM.
Clinical trial amendments
Statistics analysis
Clinical trial update
eFigure. Plot of time to death from time of progression
eTable 1. Treatment profile
eTable 2. Estimates of overall survival rates at yearly points
eTable 3. Longitudinal fatigue assessment
eData. Details surrounding coded Grade 5 toxicities
eMethods.
Data sharing statement
References
- 1.Hegi ME, Sciuscio D, Murat A, Levivier M, Stupp R. Epigenetic deregulation of DNA repair and its potential for therapy. Clin Cancer Res. 2009;15(16):5026-5031. doi: 10.1158/1078-0432.CCR-08-1169 [DOI] [PubMed] [Google Scholar]
- 2.Strobel H, Baisch T, Fitzel R, et al. Temozolomide and other alkylating agents in glioblastoma therapy. Biomedicines. 2019;7(3):69. doi: 10.3390/biomedicines7030069 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kim D, Nam HJ. PARP inhibitors: clinical limitations and recent attempts to overcome them. Int J Mol Sci. 2022;23(15):8412. doi: 10.3390/ijms23158412 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ray Chaudhuri A, Nussenzweig A. The multifaceted roles of PARP1 in DNA repair and chromatin remodelling. Nat Rev Mol Cell Biol. 2017;18(10):610-621. doi: 10.1038/nrm.2017.53 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gupta SK, Kizilbash SH, Carlson BL, et al. Delineation of MGMT hypermethylation as a biomarker for veliparib-mediated temozolomide-sensitizing therapy of glioblastoma. J Natl Cancer Inst. 2015;108(5):djv369. doi: 10.1093/jnci/djv369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics. 1975;31(1):103-115. doi: 10.2307/2529712 [DOI] [PubMed] [Google Scholar]
- 7.Sloan JA, Liu H, Satele DV, et al. Prognostic significance of baseline fatigue for overall survival: a patient-level meta-analysis of 43 oncology clinical trials with 3915 patients. Trends Cancer Res. 2017;12:97-110. [PMC free article] [PubMed] [Google Scholar]
- 8.Ellingson BM, Bendszus M, Boxerman J, et al. ; Jumpstarting Brain Tumor Drug Development Coalition Imaging Standardization Steering Committee . Consensus recommendations for a standardized Brain Tumor Imaging Protocol in clinical trials. Neuro Oncol. 2015;17(9):1188-1198. doi: 10.1093/neuonc/nov095 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bujko K, Wyrwicz L, Rutkowski A, et al. ; Polish Colorectal Study Group . Long-course oxaliplatin-based preoperative chemoradiation versus 5 × 5 Gy and consolidation chemotherapy for cT4 or fixed cT3 rectal cancer: results of a randomized phase III study. Ann Oncol. 2016;27(5):834-842. doi: 10.1093/annonc/mdw062 [DOI] [PubMed] [Google Scholar]
- 10.Ciseł B, Pietrzak L, Michalski W, et al. ; Polish Colorectal Study Group . Long-course preoperative chemoradiation versus 5 × 5 Gy and consolidation chemotherapy for clinical T4 and fixed clinical T3 rectal cancer: long-term results of the randomized Polish II study. Ann Oncol. 2019;30(8):1298-1303. doi: 10.1093/annonc/mdz186 [DOI] [PubMed] [Google Scholar]
- 11.Robinson HR, Lieu CH. Anti-EGFR therapy for left-sided RAS wild-type colorectal cancer—PARADIGM shift. JAMA Oncol. 2023;9(6):767-769. doi: 10.1001/jamaoncol.2023.1088 [DOI] [PubMed] [Google Scholar]
- 12.Watanabe J, Muro K, Shitara K, et al. Panitumumab vs bevacizumab added to standard first-line chemotherapy and overall survival among patients with RAS wild-type, left-sided metastatic colorectal cancer: a randomized clinical trial. JAMA. 2023;329(15):1271-1282. doi: 10.1001/jama.2023.4428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.D’Andrea AD. Mechanisms of PARP inhibitor sensitivity and resistance. DNA Repair (Amst). 2018;71:172-176. doi: 10.1016/j.dnarep.2018.08.021 [DOI] [PubMed] [Google Scholar]
- 14.Mekonnen N, Yang H, Shin YK. Homologous recombination deficiency in ovarian, breast, colorectal, pancreatic, non–small cell lung and prostate cancers, and the mechanisms of resistance to PARP inhibitors. Front Oncol. 2022;12:880643. doi: 10.3389/fonc.2022.880643 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Groelly FJ, Fawkes M, Dagg RA, Blackford AN, Tarsounas M. Targeting DNA damage response pathways in cancer. Nat Rev Cancer. 2023;23(2):78-94. doi: 10.1038/s41568-022-00535-5 [DOI] [PubMed] [Google Scholar]
- 16.Li C, Hao M, Fang Z, et al. PARP inhibitor plus chemotherapy versus chemotherapy alone in patients with triple-negative breast cancer: a systematic review and meta-analysis based on randomized controlled trials. Cancer Chemother Pharmacol. 2023;91(3):203-217. doi: 10.1007/s00280-023-04506-x [DOI] [PubMed] [Google Scholar]
- 17.Lin F, de Gooijer MC, Roig EM, et al. ABCB1, ABCG2, and PTEN determine the response of glioblastoma to temozolomide and ABT-888 therapy. Clin Cancer Res. 2014;20(10):2703-2713. doi: 10.1158/1078-0432.CCR-14-0084 [DOI] [PubMed] [Google Scholar]
- 18.Sim HW, McDonald KL, Lwin Z, et al. A randomized phase II trial of veliparib, radiotherapy, and temozolomide in patients with unmethylated MGMT glioblastoma: the VERTU study. Neuro Oncol. 2021;23(10):1736-1749. doi: 10.1093/neuonc/noab111 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Robins HI, Zhang P, Gilbert MR, et al. A randomized phase I/II study of ABT-888 in combination with temozolomide in recurrent temozolomide resistant glioblastoma: an NRG oncology RTOG group study. J Neurooncol. 2016;126(2):309-316. doi: 10.1007/s11060-015-1966-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Xu J, Keenan TE, Overmoyer B, et al. Phase II trial of veliparib and temozolomide in metastatic breast cancer patients with and without BRCA1/2 mutations. Breast Cancer Res Treat. 2021;189(3):641-651. doi: 10.1007/s10549-021-06292-7 [DOI] [PubMed] [Google Scholar]
- 21.Han HS, Diéras V, Robson M, et al. Veliparib with temozolomide or carboplatin/paclitaxel versus placebo with carboplatin/paclitaxel in patients with BRCA1/2 locally recurrent/metastatic breast cancer: randomized phase II study. Ann Oncol. 2018;29(1):154-161. doi: 10.1093/annonc/mdx505 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pishvaian MJ, Slack RS, Jiang W, et al. A phase 2 study of the PARP inhibitor veliparib plus temozolomide in patients with heavily pretreated metastatic colorectal cancer. Cancer. 2018;124(11):2337-2346. doi: 10.1002/cncr.31309 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pietanza MC, Waqar SN, Krug LM, et al. Randomized, double-blind, phase II study of temozolomide in combination with either veliparib or placebo in patients with relapsed-sensitive or refractory small-cell lung cancer. J Clin Oncol. 2018;36(23):2386-2394. doi: 10.1200/JCO.2018.77.7672 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Clinical trial amendments
Statistics analysis
Clinical trial update
eFigure. Plot of time to death from time of progression
eTable 1. Treatment profile
eTable 2. Estimates of overall survival rates at yearly points
eTable 3. Longitudinal fatigue assessment
eData. Details surrounding coded Grade 5 toxicities
eMethods.
Data sharing statement
