REGOMA-OSS: a large, Italian, multicenter, prospective, observational study evaluating the efficacy and safety of regorafenib in patients with recurrent glioblastoma

BACKGROUND

In the randomized phase II REGOMA trial, regorafenib showed promising activity in patients with recurrent glioblastoma. We conducted a large, multicenter, prospective, observational study to confirm the REGOMA data in a real-world setting.

PATIENTS AND METHODS

The major inclusion criteria were histologically confirmed diagnosis of glioblastoma according to the World Health Organization (WHO) 2016 classification and relapse after radiotherapy with concurrent/adjuvant temozolomide treatment, good performance status [Eastern Cooperative Oncology Group performance status (ECOG PS 0-1)] and good liver function. Regorafenib was administered at the standard dose of 160 mg/day for 3 weeks on/1 week off. Brain magnetic resonance imaging was carried out within 14 days before starting regorafenib and every 8-12 weeks. The primary endpoint was overall survival (OS). The secondary endpoints were progression-free survival (PFS), objective response rate, disease control rate (DCR), safety and health-related quality of life. The Response Assessment in Neuro-Oncology (RANO) criteria were used for response evaluation and Common Terminology Criteria for Adverse Events (CTCAE) version 5 for assessment of adverse events (AEs).

RESULTS

From September 2020 to October 2022, 190 patients with recurrent glioblastoma were enrolled from 30 cancer centers in Italy: their median age was 58.5 years [interquartile range (IQR) 53-67 years], 68% were male and 85 (44.7%) were in optimal clinical condition (ECOG PS 0). The number of patients taking steroids at baseline was 113 (60%); the second surgery was carried out in 39 (20.5%). O6-methylguanine-DNA methyltransferase (MGMT) was methylated in 80 patients (50.3%) and 147 (92.4%) of the patients analyzed had isocitrate dehydrogenase (IDH) wild type. The median follow-up period was 20 months (IQR 15.6-25.5 months). The median OS was 7.9 months ([95% confidence interval (CI) 6.5-9.2 months] and the median PFS was 2.6 months (95% CI 2.3-2.9 months). Radiological response was partial response and stable disease in 13 (7.3%) and 26 (14.6%) patients, respectively, with a DCR of 21.9%. The median number of regorafenib cycles per patient was 3 (IQR 2.0-4.0). Grade 3-4 drug-related adverse events were reported in 22.6% of patients. A dose reduction due to AEs was required in 36% of patients. No deaths were considered as treatment-related AEs.

CONCLUSIONS

This large, real-world observational study showed similar OS with better tolerability of regorafenib in patients with relapsed glioblastoma compared with the REGOMA study.

Abstract P5-17-06: The deficient eNOS c.894G>T genotype is not associated with increased severity of hypertension and proteinuria in breast cancer patients receiving bevacizumab

Background. Tumor angiogenesis is a complex process involving a wide array of effector molecules and stromal cells. In tumor tissue, vasculature is structurally and functionally abnormal, causing elevated interstitial pressure and irregular perfusion. The expression of vascular endothelial growth factor (VEGF), the most important angiogenic factor, is enhanced in many tumors. VEGF may induce nitric oxide (NO) production via up-regulation of the endothelial NO synthase (eNOS, NOS3) and the resultant overproduction of NO is associated with vasodilation and edema within tumors (Goel S et al. Physiol Rev 2011;91:1071). eNOS plays an important physiologic role in maintaining blood pressure homeostasis and vascular integrity by providing constitutive release of NO in endothelial cells. Functional variants of the eNOS gene, including the single-nucleotide polymorphism rs1799983 (c.894G>T, p. Asp298Glu) at codon 298, have been associated with reduced function of eNOS and higher incidence of hypertension (HT) (Niu W, Qi Y. PLoS One 2011;6:e24266).

Purpose. Since suppression of VEGF-eNOS axis by anti-angiogenic therapies is considered a causative factor of HT in patients, the purpose of this study was to examine whether the major eNOS non-synonymous variant c.894G>T may be associated with increased risk of developing hypertension (HT) and proteinuria (PU) in breast cancer patients treated with bevacizumab.

Patients and methods. Forty-one metastatic breast cancer patients given bevacizumab as per standard of care were enrolled. Main characteristics were: median age 49.5 years (range 29–73) at first diagnosis, 53 years (range 34–74) at metastatic disease; PS 0–1 in all patients; 4 subjects with hypertension and 1 patient with compensated cardiovascular disease at diagnosis. Twenty-six subjects had received neoadjuvant or adjuvant chemotherapy based on anthracycline and taxane; first-line chemotherapy for metastatic disease was paclitaxel plus bevacizumab for all patients; 14 subjects received hormone-therapy for metastatic disease (range 1–5 lines). Germline DNA was extracted from peripheral blood and used to screen patients for eNOS c.894G>T variant by automatic sequencing. The study was approved by the local Ethics Committee.

Results. Three patients (7.3%) were homozygous variant c.894TT, 12 (29.3%) homozygous wild-type c.894GG and the remaining 26 (63.4%) were heterozygous c.894GT. The c.894TT patients had no HT or PU at baseline and developed grade (G) 1, 2, 2 HT, respectively, and in one case PU during treatment. G1, 2 and 3 HT developed in 4, 5 and 2 c.894GG subjects, respectively, while PU was observed in 7/12 (58%) patients. The full range of HT grades and PU were observed in heterozygous subjects. Thirty-seven patients achieved one of the following: partial remission, minimal response or stable disease upon treatment with bevacizumab in combination with chemotherapy; 3 subjects had progressive disease and 1 was not evaluable.

Conclusions. The presence of the mutant T allele of c.894G>T is not associated with increased severity of HT and PU; therefore, bevacizumab can be administered at no increased risk in TT patients with respect to the wild-type GG population.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-17-06.

KRAS codon 61, 146 and BRAF mutations predict resistance to cetuximab plus irinotecan in KRAS codon 12 and 13 wild-type metastatic colorectal cancer

Background:

KRAS codons 12 and 13 mutations predict resistance to anti-EGFR monoclonal antibodies (moAbs) in metastatic colorectal cancer. Also, BRAF V600E mutation has been associated with resistance. Additional KRAS mutations are described in CRC.

Methods:

We investigated the role of KRAS codons 61 and 146 and BRAF V600E mutations in predicting resistance to cetuximab plus irinotecan in a cohort of KRAS codons 12 and 13 wild-type patients.

Results:

Among 87 KRAS codons 12 and 13 wild-type patients, KRAS codons 61 and 146 were mutated in 7 and 1 case, respectively. None of mutated patients responded vs 22 of 68 wild type (P=0.096). Eleven patients were not evaluable. KRAS mutations were associated with shorter progression-free survival (PFS, HR: 0.46, P=0.028). None of 13 BRAF-mutated patients responded vs 24 of 74 BRAF wild type (P=0.016). BRAF mutation was associated with a trend towards shorter PFS (HR: 0.59, P=0.073). In the subgroup of BRAF wild-type patients, KRAS codons 61/146 mutations determined a lower response rate (0 vs 37%, P=0.047) and worse PFS (HR: 0.45, P=0.023). Patients bearing KRAS or BRAF mutations had poorer response rate (0 vs 37%, P=0.0005) and PFS (HR: 0.51, P=0.006) compared with KRAS and BRAF wild-type patients.

Conclusion:

Assessing KRAS codons 61/146 and BRAF V600E mutations might help optimising the selection of the candidate patients to receive anti-EGFR moAbs.

Amphiregulin (AR) expression in the prediction of benefit from cetuximab plus irinotecan in KRAS wild-type metastatic colorectal cancer (mCRC) patients

4021

Background: AR is an endogenous ligand of Epidermal Growth Factor Receptor (EGFR), whose binding is prevented in the presence of cetuximab. Methods: We retrospectively assessed KRAS mutational status and AR expression at immunohistochemistry (IHC) in 86 irinotecan-refractory EGFR-positive mCRC patients treated with cetuximab plus irinotecan. AR-IHC was performed on tissue sections from paraffin-embedded primary tumors. Specimens were defined AR-low or AR-high according to a score (range 0–300) obtained multiplying intensity (0 to 3+) by percentage of stained cells (0–100%). According to the results of a ROC analysis, we identified a cut-off value of 18. The association between AR-IHC and treatment outcome in terms of response rate (RR), PFS, and OS was investigated in the subgroup of KRAS wild-type patients. Results: Eighty-six consecutive patients were included. M/F = 44/42, median age = 67 (41–78), median number of previous lines of chemotherapy = 2 (1–5). Among 51 (59%) KRAS wild-type patients, 12 PRs and 1 CR were observed, for an overall RR of 25% (13/51). AR-IHC was high, low or unconclusive in 27, 22 and 2 cases respectively. AR-low patients reported a significantly worse RR (2/22, 9%) compared with AR-high (10/27, 37%) (p = 0.024) and a trend toward shorter PFS (3.5 vs 5.3 months, HR 0.88 [95%CI: 0.46–1.60], p = 0.628) and OS (8.8 vs 15.1 months, HR 0.60 [95%CI: 0.30–1.10], p = 0.106). Conclusions: These results underline the potential role of endogenous ligands in influencing the activity of anti-EGFR monoclonal antibodies. Absent or low AR expression at IHC may be related to resistance to cetuximab plus irinotecan. Further data regarding the prognostic impact of AR expression are needed.

No significant financial relationships to disclose.

Analysis of single nucleotide polymorphisms (SNPs) of cytidine deaminase (CDA) and xeroderma pigmentosum group D (XPD) genes for the prediction of clinical response to gemcitabine and cisplatin in advanced non-small cell lung cancer (NSCLC) patients

7219

Background: Detection of genomic differences predictive of drug response may allow to customize chemotherapy to the characteristics of individual patients. Gemcitabine alone or in combination with cisplatin are two of the most common regimens used in the treatment of advanced NSCLC. We investigated the correlation between selected SNPs of genes encoding enzymes involved in the activity of gemcitabine and cisplatin, including CDA, XPD and ERCC1, and the response to chemotherapy. Methods: Forty advanced NSCLC chemo-naïve patients were enrolled into the study; 16 patients, aged > 70 years, were treated with gemcitabine; 24 patients, aged < 70 years, were treated with cisplatin plus gemcitabine. SNPs CDA A79C, XPD A751C, XPD G312A and ERCC1 C118T were analyzed with the ABI PRISM 7900HT Sequence Detection System using TaqMan probe-based assay in DNA obtained from peripheral blood samples. Statistical analysis was performed using Pearson χ2 test, Kaplan-Meier curve and log rank test. Results: Up to now 31 patients, 11 treated with gemcitabine and 20 with cisplatin plus gemcitabine, are evaluable for response with RECIST criteria; correlation between response and genotype are showed in the table. Actuarial progression free survival (PFS) was longer in patients treated with gemcitabine ± cisplatin with CDA A79A genotype than in others (respectively 11.0 vs 2.5 months, p = 0.0092), while no correlation was found between the SNPs of XPD and ERCC1 and PFS in patients receiving cisplatin and gemcitabine. Conclusions: CDA A79A genotype resulted predictive of response and longer PFS in NSCLC patients treated with gemcitabine ± cisplatin. There was also a trend toward significant correlation between XPD A751A and better response in cisplatin-treated patients, while the SNPs in XPD 312 and ERCC1 118 were not associated with response to chemotherapy.

[Table: see text]

No significant financial relationships to disclose.