Targeting immunoliposomes to EGFR-positive glioblastoma

Background

We assessed the capacity of epidermal growth factor receptor (EGFR)-targeted immunoliposomes to deliver cargo to brain tumor tissue in patients with relapsed glioblastoma harboring an EGFR amplification. We aimed to assess the tolerability and effectiveness of anti-EGFR immunoliposomes loaded with doxorubicin (anti-EGFR ILs-dox) in glioblastoma multiforme patients.

Patients and methods

Patients with EGFR-amplified, relapsed glioblastoma were included in this phase I pharmacokinetic trial. Patients received up to four cycles of anti-EGFR ILs-dox. Twenty-four hours later, plasma and cerebrospinal fluid (CSF) samples were obtained. In addition, we also treated three patients with anti-EGFR ILs-dox before resection of their relapsed glioblastoma. Doxorubicin concentrations were measured in plasma, CSF, and tumor tissue. Safety and efficacy parameters were also obtained.

Results

There were no or negligible levels of doxorubicin found in the CSF demonstrating that anti-EGFR ILs-dox are not able to cross the blood–brain barrier (BBB). However, significant levels were detected in glioblastoma tissue 24 h after the application, indicating that the disruption of BBB integrity present in high-grade gliomas might enable liposome delivery into tumor tissue. No new safety issues were observed. The median progression-free survival was 1.5 months and the median overall survival was 8 months. One patient undergoing surgery had a very long remission suggesting that neoadjuvant administration may have a positive effect on outcome.

Conclusions

We clearly demonstrate that anti-EGFR-immunoliposomes can be targeted to EGFR-amplified glioblastoma and cargo—in this case doxorubicin—can be delivered, although these immunoliposomes do not cross the intact BBB. (The GBM-LIPO trial was registered as NCT03603379).

•

Human pharmacokinetic and pharmacodynamic data for EGFR-targeted immunoliposomes.

•

Demonstration of delivery of immunoliposomes to glioblastoma tissue.

•

EGFR as a target to deliver drug-containing nanoparticles to glioma tissue.

The addition of deep hyperthermia to gemcitabine-based chemoradiation may achieve enhanced survival in unresectable locally advanced adenocarcinoma of the pancreas

•

Intensification of chemoradiation with hyperthermia was feasible in nine patients with LAPC.

•

Only one grade three toxicity was reported and two tumours became resectable.

•

The 24 months median OS and 100% 1 year OS are superior to historical series.

Introduction

Driven by the current unsatisfactory outcomes for patients with locally advanced pancreatic cancer (LAPC), a biologically intensified clinical protocol was developed to explore the feasibility and efficacy of FOLFORINOX chemotherapy followed by deep hyperthermia concomitant with chemoradiation and subsequent FOLFORINOX chemotherapy in patients with LAPC.

Methods

Nine patients with LAPC were treated according to the HEATPAC Phase II trial protocol which consists of 4 cycles of FOLFORINOX chemotherapy followed by gemcitabine-based chemoradiation to 56 Gy combined with weekly deep hyperthermia and then a further 8 cycles of FOLFORINOX chemotherapy.

Results

One grade three related toxicity was reported and two tumours became resectable. The median overall survival was 24 months and 1 year overall survival was 100%.

Conclusions

Intensification of chemoradiation with deep hyperthermia was feasible in nine consecutive patients with LAPC.

Abstract P2-16-19: Fulvestrant with or without selumetinib (Sel, AZD6244), a mitogen-activated protein kinase kinase (MEK) 1/2 inhibitor, in advanced stage breast cancer progressing after aromatase inhibitor (AI): A multicenter randomized placebo-controlled double-blind phase II trial, SAKK21/08

Background:

Mitogen-activated protein kinase (MAPK) pathway is one of the major signaling cascades responsible for resistance to endocrine therapy by promoting estrogen-independent tumor growth. Sel is a potent selective MEK1 and MEK2 inhibitor active in melanoma, ovarian, colorectal and non-small cell lung cancers. Fulvestrant (Fulv) is a pure antagonist and downregulator of nuclear and membranous/cytoplasmic estrogen receptor (ER). The combination of a MEK inhibitor with Fulv may improve the efficacy of endocrine therapy by reversing and/or delaying the development of resistance.

Material and methods:

Postmenopausal patients with advanced stage endocrine-sensitive (ER and/or progesterone receptors ≥10%) breast cancer progressing after aromatase inhibitor were randomized to intramuscular Fulv 500 mg (day 1, 15, 29 then every 28±2 days) combined with either oral Sel 75 mg bid or placebo bid. The primary endpoint was disease control (DC = CR + PR + stable disease ≥ 24 weeks). Sample size was calculated according to Simon's optimal two-stage design. A DC rate of ≤30% in the experimental arm was considered as not interesting and ≥50% as promising. Using a 5% significance level and 80% power, 46 patients were needed, 15 in the first stage and 31 in the second stage. In the control arm a total of 43 patients was needed to estimate the DC rate with a confidence interval width of ≤30%. An interim efficacy analysis was planned according to Herndon's modification of Simon's optimal two-stage design.

Results:

Forty-six patients were included (23 per arm). Seventy percent of the patients in the experimental arm had measurable disease and 57% had visceral metastases. Patients had received AI in the metastatic (61%) or adjuvant (39%) setting (65% had also prior tamoxifen). Five of 23 patients (22%) reached DC in the experimental arm. DC rate being inferior to the statistical hypothesis the study was terminated following the planned interim efficacy analysis. Most frequently reported adverse events (AEs) were rash and other skin disorders, fatigue, diarrhea, hypertension, edema, nausea/vomiting, anorexia, musculoskeletal symptoms, dry mouth and paresthesia. AEs were mainly grade 1 and 2 (NCI-CTCAE v4.0), but they often precluded treatment continuation. Most of the first patients exposed to Sel 75 mg bid stopped treatment before disease progression. Seven of the 23 patients (30%) were exposed to Sel for only 1 cycle or less. After an amendment facilitating early dose-reduction, patients decreasing to 75 mg qd had much better tolerance and improved treatment duration with Sel. Four out of the 5 patients reaching DC had Sel dose reduction.

Conclusion:

The combination of Sel with fulvestrant did not reach the prespecified disease control rate. The toxicity of Sel 75 mg bid led to frequent treatment interruptions/dose reductions and poor exposure to the drug.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-16-19.

First-line temozolomide combined with bevacizumab in metastatic melanoma: a multicentre phase II trial (SAKK 50/07)

BACKGROUND

Oral temozolomide has shown similar efficacy to dacarbazine in phase III trials with median progression-free survival (PFS) of 2.1 months. Bevacizumab has an inhibitory effect on the proliferation of melanoma and sprouting endothelial cells. We evaluated the addition of bevacizumab to temozolomide to improve efficacy in stage IV melanoma.

PATIENTS AND METHODS

Previously untreated metastatic melanoma patients with Eastern Cooperative Oncology Group performance status of two or more were treated with temozolomide 150 mg/m(2) days 1-7 orally and bevacizumab 10 mg/kg body weight i.v. day 1 every 2 weeks until disease progression or unacceptable toxicity. The primary end point was disease stabilisation rate [complete response (CR), partial response (PR) or stable disease (SD)] at week 12 (DSR12); secondary end points were best overall response, PFS, overall survival (OS) and adverse events.

RESULTS

Sixty-two patients (median age 59 years) enrolled at nine Swiss centres. DSR12 was 52% (PR: 10 patients and SD: 22 patients). Confirmed overall response rate was 16.1% (CR: 1 patient and PR: 9 patients). Median PFS and OS were 4.2 and 9.6 months. OS (12.0 versus 9.2 months; P = 0.014) was higher in BRAF V600E wild-type patients.

CONCLUSIONS

The primary end point was surpassed showing promising activity of this bevacizumab/temozolomide combination with a favourable toxicity profile. Response and OS were significantly higher in BRAF wild-type patients.

Abstract P2-16-07: hMMP9 as Predictive Factor for Response and Progression Free Survival in Breast Cancer Patients Treated with Bevacizumab and Pegylated Liposomal Doxorubicin (PLD)

Background: The anti-angiogenic drug, bevacizumab (Bv), is currently used in the treatment of different malignancies including breast cancer. Many angiogenesis-associated molecules are found in the circulation of cancer patients. Until now, there are no prognostic or predictive factors identified in breast cancer patients treated with Bv. We present here the first results of the prospective monitoring of 6 angiogenesis-related molecules in the peripheral blood of breast cancer patients treated with a combination of Bv and PLD in the phase II trial, SAKK 24/06. Methods: Patients were treated with PLD (20 mg/m2) and Bv (10 mg/kg) on days 1 and 15 of each 4-week cycle for a maximum of 6 cycles, followed by Bv monotherapy maintenance (10 mg/m2 q2 weeks) until progression or severe toxicity. Plasma and serum samples were collected at baseline, after 2 months of therapy, then every 3 months and at treatment discontinuation. Enzyme-linked immunosorbent assays (Quantikine, R&D Systems and Reliatech) were used to measure the expression levels of human vascular endothelial growth factor (hVEGF), placental growth factor (hPlGF), matrix metalloproteinase 9 (hMMP9) and soluble VEGF receptors hsVEGFR-1, hsVEGFR-2 and hsVEGFR-3. The log-transformed data (to reduce the skewness) for each marker was analyzed using an analysis of variance (ANOVA) model to determine if there was a difference between the mean of the subgroups of interest (where α = 0.05). The untransformed data was also analyzed in the same manner as a “sensitivity” check.

Results: 132 blood samples were collected in 41 out of 43 enrolled patients. Baseline levels of the molecules were compared to disease status according to RECIST. There was a statistically significant difference in the mean of the log-transformed levels of hMMP9 between responders [CR+PR] versus the mean in patients with PD (p-value=0.0004, log fold change=0.7536), and between patients with disease control [CR+PR+SD] and those with PD (p-value=<0.0001, log fold change=0.81559), with the log-transformed level of hMMP9 being higher for the responder group. The mean of the log-transformed levels of hsVEGFR-1 was statistically significantly different between patients with disease control [CR+PR+SD] and those with PD (p-value=0.0068, log fold change=-0.6089), where the log-transformed level of hsVEGFR-1 was lower for the responder group. The log-transformed level of hMMP9 at baseline was identified as a significant prognostic factor in terms of progression free survival (PFS): p-value=0.0417, hazard ratio (HR)=0.574 with a corresponding 95% confidence interval (0.336 — 0.979)). No strong correlation was shown either between the log-transformed levels of hsVEGF, hPlGF, hsVEGFR-2 or hsVEGFR-3 and clinical response or the occurrence of severe toxicity, or between the levels of the different molecules.

Conclusions: Our results suggest that baseline plasma level of the matrix metalloproteinase, hMMP9, could predict tumor response and PFS in patients treated with a combination of Bv and PLD. These data justify further investigation in breast cancer patients treated with anti-angiogenic therapy.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-16-07.

Infrequent mutation of the tumour-suppressor gene Smad4 in early-stage colorectal cancer

Smad4 is a candidate tumour-suppressor gene identified recently on chromosome 18q21.1. Both alleles are inactivated in nearly one-half of pancreatic carcinomas, but its role in the tumorigenesis of other tumours is still unknown. The aim of this study was to investigate the potential involvement of the Smad4 locus in early-stage colorectal cancers (stages I-III) in tumour samples from a randomised multicentre trial. Of a large collection of DNA samples, 73 with a loss of one allele of the Smad4 gene were analysed for the presence of point mutations in the remaining gene. Patients, from whom biopsies were isolated, were part of a previous randomised multicentre study of the Swiss Group for Clinical Cancer Research on the benefit of adjuvant chemotherapy (SAKK study 40/81). Mutation analysis was restricted to the highly conserved C-terminal domain (exons 8, 9, 10 and 11) of Smad4, using PCR and single-strand conformational variant analysis. Two of the 73 patients (3%) with loss of one allele of Smad4 had a point mutation in the remaining allele. These results indicate that whereas Smad4 point mutations are prevalent in pancreatic carcinoma, they are infrequent in early stages (I-III) of colorectal cancer.

Evidence for the involvement of 5-lipoxygenase products in the regulation of the expression of the proenkephalin gene in cultured astroglial cells

Cultured astroglial cells secrete eicosanoids which are produced by the cyclooxygenase and lipoxygenases. These cells also transcribe the proenkephalin gene. In the present study, it was investigated whether agents which inhibit the metabolism of arachidonic acid affect the basal and stimulated expression of the gene. Tetradecanoyl phorbol acetate (TPA; 1-1000 nmol/l) increases the concentration of proenkephalin mRNA in these cells by activating protein kinase C. The enhancement in proenkephalin mRNA caused by TPA (10 nmol/l) was not affected by the cyclooxygenase inhibitor indomethacin (5 mumol/l). However, nordihydroguaiaretic acid, which blocks cyclooxygenase and lipoxygenases, potentiated the effect of TPA on proenkephalin mRNA, when used at concentrations of 0.5-50 mumol/l. Two selective inhibitors of 5-lipoxygenase, i.e. MK886 (5 mumol/l) and BAY X1005 (1 mumol/l), also enhanced the effect of TPA (10 nmol/l) without affecting the basal expression of the gene. When added to the incubation medium, leukotriene E4 (10-1000 nmol/l) diminished in a dose-dependent manner the basal and TPA-induced expression of the proenkephalin gene. It is concluded that in astroglial cells derived from cortex of new-born rats products of 5-lipoxygenase can diminish the action of protein kinase C on the proenkephalin gene.