No abstract

In this early development study with a limited number of patients, postoperative MAGE-A3 immunization proved to be feasible with minimal toxicity. These results are being investigated further in a large phase III study.

The MAGE-A3 expression rate showed that a promising proportion of operable patients with early-stage non-small cell lung cancers are possible candidates for trials investigating adjuvant therapy with MAGE-A3 immunization. Currently, a phase two trial of adjuvant MAGE-A3 vaccination is in progress.

We have previously demonstrated expression of cytochrome P 450 3A (CYP3A) protein in pulmonary carcinomas and surrounding normal tissue, using immunohistochemistry. These results suggested that different CYP3A enzymes may be expressed in normal and tumour tissue. Therefore, the aim of the present study was to identify specific CYP3A enzymes expressed in normal human lung and lung tumours. Both normal lung tissue and tumour tissue from eight patients was analyzed for CYP3A4, CYP3A5 and CYP3A7 mRNA using a specific RT-PCR (reverse transcriptase-polymerase chain reaction) method. Identical samples were subjected to immunohistochemical analysis of CYP3A protein. CYP3A5 was the major enzyme of the CYP3A subfamily present at the mRNA level in both normal human lung and lung tumours. CYP3A5 mRNA was detected in normal lung tissue in all eight cases and in tumour tissue in four cases. CYP3A7 mRNA was detected in five cases in normal tissue and in one tumour. Notably, no CYP3A4 mRNA was found in any of the samples. Immunohistochemical staining for CYP3A protein was found in normal lung tissue in each case. Interestingly, all pulmonary carcinomas showed immunostaining for CYP3A, while mRNA for CYP3A enzymes was found in only four cases. In summary, our study indicates a specific expression pattern of the members of the CYP3A subfamily in normal human lung and lung tumours. These findings have potential clinical significance, since it has been recently shown that CYP3A5 catalyzes the activation of the anticancer pro-drugs cyclophosphamide and ifosfamide. Thus, local activation of these agents may take place in pulmonary carcinomas and surrounding normal tissues.

7554 Background: After complete resection, about 50% of patients with stages IB-II NSCLC disease die within 5 years. Adjuvant chemotherapy improves overall survival at the expense of substantial toxicity. Activity of MAGE-A3 immunotherapeutic (i.e. recombinant MAGE-A3 protein and a potent GlaxoSmithKline adjuvant) was previously demonstrated in metastatic melanoma. As about 35% of NSCLCs express MAGE-A3 antigen, post-operative MAGE-treatment may be a tumor-specific, well tolerated, and effective adjuvant therapy. Methods: Patients with completely resected, MAGE-A3 (+), stage pIB or pII were randomly assigned to postoperative MAGE-A3 or placebo (2:1), with 5 administration at 3-week intervals, followed by 8 administrations every 3 months. Randomization was stratified for stage (IB vs. II), histology (squamous vs. other), and lymph-node (LN) procedure (sampling vs. dissection). Primary endpoint was disease-free interval (DFI); other endpoints were safety, disease-free survival (DFS), and overall survival (OS). This exploratory Phase II study was designed to detect a clinically relevant HR with a 10% one-sided a. Results: 182 patients (122 stage IB, 60 stage II) from 59 centers in 14 countries were randomized over 2 years: Median age 63 (45–81); 87% male; 65% squamous cell carcinoma; 65% lymph-node dissection. After a median follow-up of 28 months, 67 recurrences and 45 deaths were recorded. Group comparisons of DFI, DFS and OS gave respectively a hazard ratio (HR) of 0.74 (95% CI 0.44–1.20, p=0.107), 0.73 (95% CI 0.45–1.16) and 0.66 (95% CI 0.36–1.20) in favor of the MAGE-A3 group. Overall, treatment was well tolerated, with excellent protocol compliance. Subset analysis also suggests that LN dissection may have an effect on survival. Conclusions: The final analysis of this randomized phase II study shows a positive trend for activity of MAGE-A3 treatement in NSCLC with a relative improvement of DFI and DFS of 27%. Further phase III evaluation is planned. This study also suggests that complete lymph-node dissection may have an effect on survival and should be confirmed prospectively. [Table: see text]

No abstract

The cytochrome P450 (CYP) enzymes metabolize drugs and other xenobiotics in liver and also in some extrahepatic tissues. We have studied the expression and localization of CYP3A in primary lung tumours and normal lung tissue from the same patients. Thirty-two patients undergoing partial or total lung resection for therapy of primary pulmonary carcinoma were included in this study. Immunohistochemical staining for CYP3A was performed with a modification of the ABC technique. Eight of the 32 cases of primary pulmonary carcinoma showed expression of CYP3A. In 12 of the 32 cases of normal tissue, the seromucous glands were positive for CYP3A. The bronchial epithelium was positive for CYP3A in 11 cases. We observed no correlation between CYP3A expression in tumour tissue and that in seromucous glands or bronchial epithelium. We conclude that CYP3A is present in both normal and cancerous lung tissue. Our findings suggest, however, no co-expression of CYP3A in lung cancer.

An ex-vivo isolated, perfused, and ventilated human lung (IPHL) model is well suited for many kinds of physiological, pharmacological, and surgical studies, when the physiological and biochemical conditions in the lung can be maintained near to those in vivo. The aim of this work was to develop such a model. The lung preparations used were available after resection because of bronchial carcinoma. Since the tumor remains intact in these anatomical preparations, this model is particularly suitable for investigation of the pharmacokinetics and effects of anticancer agents. Carrying out a series of 52 IPHL experiments (with 11 whole-lung preparations and 41 lobe preparations), we have established an IPHL model which allows extracorporeal perfusion and ventilation of the resected lungs in physiological conditions for 2-3 hours. The net weight gain during the experiment, wet-to-dry weight ratio for lung tissue, angiography of the pulmonary artery, pulmonary vascular resistance, color and fluorescence of the lung surface, and alveolar gas diffusion into the perfusate proved to be useful parameters to assess the stability of the preparations and the quality of the experiments. To confirm that an intraparenchymal tumor was perfused via the pulmonary artery, methods to detect avidin and dextran-biotin in tumor tissue after administration into the perfusion solution were employed. Histological examination of bronchial as well as tumor tissue, a computerized histoanalyzation, and a tumor grading program demonstrated that IPHL experiments did not interfere with the grading and staging of the tumors-an important ethical precondition for the use of human preparations in an extracorporeal perfusion model.