Phase II Study of Iniparib with Concurrent Chemoradiation in Patients with Newly Diagnosed Glioblastoma

The addition of temozolomide to radiotherapy changed the therapeutic landscape for patients with newly diagnosed glioblastoma (GBM) by demonstrating the activity of temozolomide and the predictive importance of the O6-methylguanine–DNA methyltransferase (MGMT) repair enzyme (1–3). Unfortunately, no drugs have reliably improved median overall survival (mOS) for people with GBM since then. Given that more than 75% of patients with GBM who are well enough to enroll on clinical trials die within 2 years of diagnosis, there are multiple efforts underway to improve therapies for patients with GBM (3–5).

An attractive approach is to identify therapies that enhance the activity of temozolomide without increasing its toxicity. Iniparib (4-iodo-3-nitrobenzamide) is a prodrug with demonstrated clinical activity in cancers mediated by mismatch repair defects such as triple-negative breast cancer and BRCA2-mutated pancreatic cancer (6, 7). Development of iniparib was initially focused on these tumors as it was thought to be a specific inhibitor of PARP. PARP inhibitors are hypothesized to enhance the efficacy of alkylating therapies such as radiotherapy and temozolomide by impairing DNA repair (6, 7). Ultimately, iniparib was shown to have anticancer activity through a prodrug mechanism in which an active nitro radical ion is released through one- and two-electron cytosolic activation, rather than direct PARP inhibition (8–10). The activated nitro radical ion binds to cysteine residues on enzymes critical for reduction–oxidation reactions, including selenoproteins such as thioredoxin reductase (TrxR). This mechanism has the potential to be selectively cytotoxic to cells high in enzymes required for prodrug metabolism or high in targets of the generated metabolites. The proposed mechanism of action paired with extensive clinical data showing iniparib has nonoverlapping toxicities with alkylating therapies as well as good access to brain tissue (likely due to the fact that it is lipophilic and MW = 292), make it a compelling choice for addition to GBM standard therapy (7, 11, 12).

Iniparib was assessed in a dose-escalation study in people with newly diagnosed malignant gliomas after completion of chemoradiation. In the setting of adjuvant temozolomide, iniparib was well tolerated at doses up to 17.2 mg/kg/week (13). The most common toxicities were fatigue and low blood counts. In addition to encouraging tolerability, the phase I study showed signs of clinical activity with an estimated mOS of 18.9 months [95% confidence interval (CI), 16.2–23.4 months] across all enrolled patients (13). The current study evaluated the tolerability and efficacy (via a primary endpoint of mOS) of iniparib when given concurrently with radiotherapy and temozolomide followed by adjuvant temozolomide in people with newly diagnosed GBM (Fig. 1).

The study was sponsored by the Cancer Therapy Evaluation Program (CTEP) of the NCI and conducted by the Adult Brain Tumor Consortium (ABTC, http://www.abtconsortium.org). The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of each participating institution. All patients provided written informed consent as a condition for participating in the study. Patients eligible for enrollment met the following criteria: ≥ 18 years old, histologically proven newly diagnosed supratentorial GBM; absolute neutrophil count ≥ 1,500/μL; platelet count ≥ 100,000/μL; serum creatinine ≤ 1.7 mg/dL; total bilirubin ≤ 1.5 mg/dL; aspartate and alanine aminotransferase ≤ 4 times the upper limit of normal; stable dexamethasone dose for ≥ 5 days prior to the start of treatment; Karnofsky Performance Status (KPS) ≥ 60%. Exclusion criteria included enzyme-inducing antiepileptic medications; other malignancy within 5 years; pregnant or nursing women; serious concurrent medical condition or other condition that would compromise safety or compliance. Agreement to practice adequate birth control methods was required.

This was an open-label, single-arm, multicenter study to estimate the mOS when iniparib is given concurrent with chemoradiation and adjuvant temozolomide. The MTD in the phase I study of iniparib given with standard dosing of temozolomide after completion of chemoradiation (5/28 days 150–200 mg/m²) was 8.6 mg/kg i.v. twice weekly (days 1 and 2). The MTD of iniparib when given with metronomic temozolomide (75 mg/m² continuous daily dosing for 6 weeks) was 8 mg/kg i.v. twice weekly (13). Because these doses were determined in people who had already completed chemoradiation, a safety run-in with iniparib at 1 dose below the MTD (6.8 mg/kg i.v. twice weekly) was first completed to ensure tolerability with concurrent radiotherapy and temozolomide. Safety evaluation was completed at 10 weeks to assess for tolerability before escalating to 8 mg/kg i.v. twice weekly during radiotherapy (Fig. 1).

Prescribed radiotherapy was 6,000 cGy in 30 daily fractions of 200 cGy each with temozolomide 75 mg/m²/daily with iniparib 8 mg/kg i.v. twice weekly for 6 weeks followed by 1 month with no therapy. This period of treatment was termed initiation. Subsequently, temozolomide was given at 150 mg/m² for 5/28 days (cycle 1). If well tolerated, temozolomide was increased to 200 mg/m² 5/28 days for cycles 2–6. Iniparib 8.6 mg/kg i.v. twice weekly was given with temozolomide for all cycles independent of temozolomide dose. This period of treatment was termed maintenance. All centers participating in the protocol completed accreditation for conformal techniques (3DCRT/IMRT) according to the procedures outlined in the ABTC Radiation Oncology plan (www.abtconsortium.org).

Iniparib was provided by Sanofi Pharmaceuticals, Inc. and administered intravenously on 2 consecutive days weekly beginning with day 1 of temozolomide. If a patient had to stop temozolomide for any reason, they stopped iniparib as well. P. carinii prophylaxis was mandatory during initiation and continued until resolution of lymphocytopenia to ≤ grade 1. The use of antiemetics was at the discretion of the treating physician during chemoradiation but required during maintenance temozolomide. Corticosteroids were prescribed at the discretion of the treating physician, but held constant prior to surveillance imaging and doses were recorded with concurrent medications.

Baseline evaluations included brain MRI, medical history and examination, KPS; complete blood count (CBC); serum chemistry profile; and pregnancy test when appropriate. After initiating treatment, CBC and adverse event (AE) reports were obtained weekly; vital signs and serum chemistries were obtained before each cycle. Brain MRI, clinical examination, and KPS were repeated every other cycle.

Patients who completed the initiation phase of concurrent chemoradiation without tumor progression or dose-limiting toxicity were prescribed 6 cycles of iniparib 8.6 mg/kg i.v. twice weekly with adjuvant temozolomide on a 5/28-day schedule (in both the safety run-in and phase II portions of the study). Progression was assessed every 2 months and defined as: (i) progressive neurologic abnormalities not explained by causes unrelated to tumor progression; (ii) greater than 25% increase in the measurement of the contrast enhancing tumor mass by MRI scan on a stable or increasing dose of corticosteroids or the presence of new lesions on MRI outside of the treatment field. Of note, patients for whom therapy was stopped due to radiographic progression and were subsequently histologically determined to have radionecrosis ≤ grade 3 severity were permitted to restart treatment. Patients stopped treatment in the setting of progression, toxicity (a dose-limiting toxicity restricting treatment > 7 days in initiation or > 21 days in maintenance), noncompliance, or if the patient chose to discontinue treatment for any reason. All patients were followed for survival calculated from the date of diagnosis until death from any cause. All enrolled patients were asked to provide tissue for analysis of MGMT promoter methylation status. MGMT status was centrally determined at LabCorp via quantitative methylation-specific PCR.

This was a single-arm, multi-center, open-label phase II study to estimate the safety and efficacy of concurrent iniparib with chemoradiation followed by adjuvant temozolomide and iniparib in newly diagnosed adult GBM patients with results compared with the EORTC/NCIC phase III trial (1). The primary endpoint was mOS and the secondary endpoint was frequency of toxicity associated with the regimen. The trial was designed to have a total of 55 death events to detect an HR of 0.75, a 25% reduction in hazard of death compared with the EORTC/NCIC trial, with 80% power at an alpha level of 0.1 to be statistically significant (1). The HR was approximated as 0.45 versus a null of 0.6 and defined as the number of death events per person-year of follow-up based on the HR from the EORTC/NCIC trial (1).

Eighty-three patients were accrued into the trial from June, 2011 to December, 2012. The trial database was closed on April 2017 for the final analysis with a total of 75 death events. Eighty-one patients were included in the efficacy analysis (intention-to-treat principle); 2 were ineligible. All 83 treated patients were included in the safety analysis. Survival time was calculated from the date of initial histologic diagnosis to the date of death or censored at the time of the analysis if patients were alive. Overall survival was estimated using the Kaplan–Meier method (14) and the CI of median survival was constructed by the method of Brookmeyer–Crowley (15). The hazard rate was expressed as the hazard of death per person-year of follow-up. Cox regression model was used for survival analysis. Patient baseline characteristics and frequency of observed toxicities attributable to the treatment were summarized using descriptive statistics. Tumor MGMT status was obtained retrospectively. All subgroup analyses were exploratory. The P values reported are 2 sided. All analyses were performed using the SAS software (version 9.3; SAS Institute).

For the safety run-in, 5 patients were enrolled (3 female), median age 62 years (range 27–77), all with KPS 80 to 90, all with prior tumor resection. Two patients stopped treatment for grade 3 and grade 4 thrombocytopenia during the maintenance period and 1 patient had progressive disease at the end of cycle 5 of the maintenance period. Two patients completed all prescribed treatment without toxicity. For the efficacy portion of the study, 78 patients were enrolled across 11 centers. However, 2 of these patients later had pathology inconsistent with GBM on central pathology review and hence were not eligible for efficacy analysis. All 83 patients were evaluated for safety analyses. The median age was 58 years (range 27–81), 63% male, and the median KPS 90 (range 60–100) with 95% of patients undergoing resection (Table 1).

Overall HR was 0.44 (95% CI, 0.35–0.55) per person-year of follow-up with intention-to-treat analysis and the mOS across all patients was 21.6 months (95% CI, 17.4–23.7 months; Fig. 2). An estimated hazard reduction was 27% (HR 0.73, 0.44 vs. 0.6). Known prognostic factors such as age, performance status, and MGMT status had significant association with outcome on OS (Table 2). However MGMT promoter methylation status had the greatest impact on overall survival. MGMT was methylated in 29 (36%), unmethylated in 37 (46%), and unknown in 15 (18%) of patients. Overall, 66 patients (81%) had successful centralized MGMT assessment. Of the 15 patients whose tissue could not be assessed, in 5 instances, this was due to technical failures and in 10 instances tissue was not provided for testing. The mOS was 30 months (95% CI, 21.7–37.2), 15.8 months (95% CI, 14.1–19.4), and 25.9 months (95% CI, 11.7–36.7) for patients with MGMT methylated, unmethylated, and unknown, respectively (P = 0.0015). Percent survival at 2 and 3 years was estimated at 38.3% (95% CI, 27.7–49.7) and 24.7% (95% CI, 15.8–35.5), respectively, across all patients. In people whose tumors had MGMT methylation, 2- and 3-year survival rates were 58.6% (95% CI, 38.9–76.5) and 37.9% (95% CI, 20.7–57.7), respectively (Table 3).

There were 67 AEs of ≥ grade 3 with attribution to the combination treatment (radiotherapy + temozolomide +iniparib; Table 4; Supplementary Table S1). The most common AEs overall were thrombocytopenia (18%), neutropenia (10%), fatigue (5%), and rash (4%; Table 4). Nine patients stopped treatment because of toxicity, and 11 patients refused further treatment or were removed from treatment for noncompliance (7 after completing the initiation phase).

This study reached its primary endpoint with an HR across all patients of 0.44 (95% CI, 0.35–0.55) per person-year of follow-up compared with 0.6 from the landmark EORTC/NCIC data (1). This translates into an approximately 27% reduction in hazard of death with the addition of iniparib to standard radiotherapy and temozolomide. As in the EORTC/NCIC trial and subsequent studies, MGMT methylation status was an important prognostic factor. Patients with MGMT methylation had substantially better survival than those with unmethylated MGMT in this study (mOS 30 months methylated vs. 15.8 months unmethylated, P = 0.0015). In addition, iniparib showed a favorable tolerability and safety profile when added to radiotherapy and temozolomide. Specifically, the rates of fatigue and leukopenia in this study were similar to those seen in studies assessing radiotherapy and temozolomide alone (1, 4).

These encouraging results, however, must be considered in the context of recent similar efforts to define optimal therapies for people with newly diagnosed GBM. Over the last 10 years, multiple studies have shown improved survival with new agents and schedules added to the standard treatment established by EORTC/NCIC (4, 5, 16). Such improvement may be attributable to improved overall supportive care and experience rather than to any one active agent (5). This possibility may confound the interpretation of modern single-arm studies. For example, although it is notable that the mOS with iniparib added to radiotherapy and temozolomide exceeds the mOS for radiotherapy and temozolomide alone either in the original EORTC/NCIC or the more modern RTOG 0525 study (mOS 21.6 months with iniparib and temozolomide vs. 16.6 months with temozolomide alone), the 2-year survival rates are relatively overlapping across these studies (Table 3). This is particularly notable when considering the 2-year survival results of the CENTRIC EORTC study that showed no difference in mOS between patients with newly diagnosed MGMT methylated GBM randomized to radiotherapy and temozolomide alone versus cilengitide added to radiotherapy and temozolomide in a large, randomized, and blinded study (16). In addition, the report of mOS is influenced by the timepoint from which survival is calculated. This study and the single radiotherapy and temozolomide studies reported by NABTT (or ABTC) calculate survival from the date of diagnosis (5). In contrast, randomized studies calculate overall survival from the time of randomization (1, 4, 16). This may result in up to a month difference in calculated mOS and is an additional caveat that must be considered when interpreting the mOS from this study. Hence, although the results of this single-arm phase II study suggest important biological activity and safety for iniparib in adults with GBM based on reduction in hazard of death and percent survival at 2 years, more work is required before advancing to confirmatory efficacy studies.

In addition, there remains lack of clarity about the mechanism of action of iniparib in conjunction with radiotherapy and temozolomide. Iniparib was initially selected for combination with radiotherapy and temozolomide as it was thought to be a PARP inhibitor that would enhance the efficacy of these therapies by preventing DNA repair (6, 7, 13). However, subsequent investigations indicate that iniparib is not a dedicated PARP inhibitor, but rather a prodrug that when activated, modulates enzymatic activity critical for maintaining redox homeostasis in cells (8, 9, 17, 18). Cancer cells are particularly susceptible to shifts in reactive oxygen species (ROS), and it is possible that iniparib is making both dividing and quiescent malignant glioma cells more vulnerable to radiotherapy and temozolomide by increasing ROS intracellularly (8, 17). Furthermore, there is evidence suggesting that iniparib cytotoxicty is potentiated in states of low glutathione (8, 18). This is compelling, as there are new data indicating that mutations in isocitrate dehydrogenase 1 (IDH1) may mediate chemosensitivity via reduced ROS response and decreased glutathione (19). Further studies are needed to evaluate whether iniparib enhances radiotherapy and temozolomide effects in a similar manner, allowing IDH1 wild-type tumors to behave more like IDH1-mutant tumors or if the effect is limited to IDH1-mutant tumors.

The initial enthusiasm for iniparib as an active anticancer agent dwindled after disappointing late-stage studies and evidence that its activity is not as a specific PARP inhibitor (7, 9, 10). The emerging data that iniparib may act to enhance radiotherapy and temozolomide by increasing intracellular ROS needs to be evaluated in GBM model systems. Future studies should evaluate markers of oxidative stress such as 4-HNE-His adducts and IHC for protein nitration in pre- and posttreatment GBM tissue to assess the impact iniparib has on oxidative stress in these cells alone and in combination with radiotherapy and temozolomide (20). Finally, preclinical studies are needed to determine whether iniparib has increased activity in settings of low glutathione. If confirmed, this would be highly relevant to both low- and high-grade gliomas as it would present an opportunity to sequence iniparib with radiotherapy and temozolomide to modulate mutant IDH1 effects in malignant gliomas.

J.O. Blakeley is a consultant/advisory board member for AbbVie and AstraZeneca. M.S. Ahluwalia is a consultant/advisory board member for AbbVie; AstraZeneca; Bristol, Myers, Squibb; Kadmon Pharmaceuticals; CBT Pharmaceuticals; and VBI Pharmaceuticals.I. Garcia-Ribas is an employee of Sanofi. No potential conflicts of interest were disclosedby the other authors.

Conception and design: J.O. Blakeley, S.A. Grossman, L.B. Nabors, X. Ye

Development of methodology: J.O. Blakeley, S.A. Grossman, X. Ye

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.O. Blakeley, S.A. Grossman, A.S. Chi, T. Mikkelsen, M.R. Rosenfeld, M.S. Ahluwalia, L.B. Nabors, A. Eichler

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.O. Blakeley, S.A. Grossman, M.S. Ahluwalia, L.B. Nabors, A. Eichler, X. Ye

Writing, review, and/or revision of the manuscript: J.O. Blakeley, S.A. Grossman, A.S. Chi, T. Mikkelsen, M.R. Rosenfeld, M.S. Ahluwalia, L.B. Nabors, A. Eichler, I.Garcia-Ribas, X. Ye

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J.O. Blakeley, S.A. Grossman, L.B. Nabors, S. Desideri

Study supervision: J.O. Blakeley, L.B. Nabors

The authors would like to thank Joy Fisher for assistance with data collection oversight on behalf of the Adult Brain Tumor Consortium, the team at Cancer Therapy Evaluation Program, NCI for their oversight and collaboration through the Adult Brain Tumor Consortium and Sanofi for the provision of the drug. This work was supported by grants from the NIH (U01 CA-62475, to S.A. Grossman; U01 CA-137443, to S.A. Grossman) and Sanofi Pharmaceuticals, Inc.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.