Tatsuya Katayama

Reciprocal and Complementary Role of MET Amplification and EGFR T790M Mutation in Acquired Resistance to Kinase Inhibitors in Lung Cancer

Purpose: In epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) therapy for lung cancer patients, acquired resistance develops almost inevitably and this limits the improvement in patient outcomes. EGFR T790M mutation and MET amplification are the two main mechanisms underlying this resistance, but the relationship between these two mechanisms is unclear. In this study, we explored their relationship using in vitro models and autopsy specimens. Experimental Design: Erlotinib-resistant HCC827 (HCC827ER) cells were developed by chronic exposure to erlotinib at increasing concentrations. HCC827EPR cells were also developed by chronic exposure to erlotinib in the presence of PHA-665,752 (a MET TKI). The erlotinib-resistant mechanisms of these cells were analyzed. In addition, 33 autopsy tumor samples from 6 lung adenocarcinoma patients harboring multiple gefitinib-refractory tumors were analyzed. Results: HCC827ER developed MET amplification, and clinically relevant resistance occurred at ≥4-fold MET gene copy number gain (CNG). By contrast, HCC827EPR developed T790M without MET CNG. Of six patients harboring gefitinib-refractory tumors, three exhibited T790M only, one exhibited MET amplification only, and the other two exhibited T790M and/or MET amplification depending on the lesion sites. In these gefitinib-refractory tumors, T790M developed in 93% (14 of 15) of tumors without MET gene CNGs, in 80% (4 of 5) of tumors with moderate MET gene CNGs (<4-fold), and in only 8% (1 of 13) of tumors with MET amplification (≥4-fold). Conclusions: These results indicate a reciprocal and complementary relationship between T790M and MET amplification and the necessity of concurrent inhibition of both for further improving patient outcomes. Clin Cancer Res; 16(22); 5489–98. ©2010 AACR.

Efficacy of Erlotinib for Brain and Leptomeningeal Metastases in Patients with Lung Adenocarcinoma Who Showed Initial Good Response to Gefitinib

Introduction: The efficacy of high-dose (1250 mg/d) gefitinib for the treatment of leptomeningeal metastasis in a patient with lung cancer harboring a mutation in the epidermal growth factor receptor (EGFR) gene was previously reported. We speculate that erlotinib, instead of high dose of gefitinib, may be also effective for the treatment of central nervous system (CNS) lesions, as trough serum concentration of erlotinib is nine times higher than that of gefitinib. Patients and Methods: Patients with lung cancer in whom CNS lesions developed after an initial good response to gefitinib for extra CNS lesions were enrolled in the study. Tumor response, performance status, neurologic symptoms, and survival were retrospectively evaluated. Results: All seven patients had EGFR mutations in their primary tumors except one patient. The median interval between gefitinib withdrawal and erlotinib administration was 5 days. Three patients showed partial response, three had stable disease, and one had progressive disease. Performance status and symptoms improved in five patients. The overall survival from the initiation of erlotinib treatment ranged from 15 to 530 days (median, 88 days). Conclusions: Erlotinib was a reasonable option for the treatment of CNS diseases that appeared after a good initial response of extra CNS disease to gefitinib.

Clinicopathological features of small-sized non-small cell lung cancer with mediastinal lymph node metastasis

INTRODUCTION In clinical practice, peripheral small-sized lung cancers with positive mediastinal lymph nodes are sometimes detected. To understand the characteristics of these aggressive tumors, we reviewed the clinicopathological features of small-sized non-small cell lung cancer patients with mediastinal lymph node metastasis resected in our institution. METHODS We studied 360 patients with small-sized lung lesions with a maximum diameter of 2 cm or less. The clinicopathological characteristics of each patient were reviewed and compared among the subgroups, which were stratified according to pathological nodal status. RESULTS 21 patients (5.8%) had a positive mediastinal lymph node. Among them, 17 patients had lung lesions larger than 1.5 cm. No mediastinal nodal involvement was found in patients with squamous cell carcinomas. In contrast, mediastinal nodal involvement was significantly common in patients with poorly differentiated carcinoma (P=0.004) and high serum carcinoembryonic antigen levels detected during preoperative evaluation (P=0.006). None of the 14 patients with upper lobe tumor had a positive subcarinal lymph node. Lower lobe tumors frequently developed extensive multiple-level involvement, which included the upper mediastinum. Radiographic evaluation of pN2 patients using computed tomography revealed a total absence of ground-glass opacity, or the presence of a small area of ground-glass opacity. CONCLUSIONS Most small-sized non-small cell lung cancer cases with mediastinal lymph node metastasis were invasive adenocarcinoma with poor differentiation, which usually showed a solid shadow without ground-glass opacity on computed tomography.

Comparison of methods for placing and managing a silastic drain after pulmonary resection

We have been using a silastic drain [Blake drain (BD)] after pulmonary resection by different placement methods and reviewed the daily amount of drainage in each patient. A 19-Fr BD was placed for each of 110 patients. First, a drain was inserted from the anterior chest wall and the tip reached the dorsal part of the diaphragm [anterior-to-posterior (AP)]. For the others [posterior-to-anterior (PA); n=37], we inserted a drain from the lower intercostal space, turned it around the apex and placed its tip in the lower front. Patients in the AP group included those placed under a water seal (AP-WS; n=43) or suction (AP-SC; n=30). The reference group consisted of 68 patients with a 32-Fr plastic drain during the same period [conventional drains (CD)]. The amount of drainage on the day of surgery in the PA group was significantly higher than that in the AP-WS group (P<0.0001) and similar to that in the CD group (P=0.54). The mean amount of drainage on postoperative day 1 and total amounts accumulating during drain placement showed no significant differences between the four groups. A BD placed using a PA approach with suction might be efficient for drainage.

In reply: Paravertebral block: the first procedure with ''no contraindications''…really?

We appreciate the comments and concerns expressed by Bronshteyn and Modest regarding our article titled ‘‘Safety of the paravertebral block in patients ineligible for epidural block undergoing pulmonary resection’’ [1]. A catheter is certainly inserted in the operative field by surgeons; catheter placement usually induces bleeding, which is directly visualized and monitored several times for approximately 10 min before closing the chest cavity. If bleeding occurs within this time, it can be directly stopped by applying pressure or coagulation with a cautery knife. As rightly pointed out by them [2], the paravertebral and epidural spaces are contiguous. Hence, we believe that our procedure via an intrathoracic approach is one of the safest methods to place a paravertebral catheter with respect to bleeding risk. In our procedure, the pleura are separated to the paravertebral space starting from the main incision at the side of the chest cavity. Therefore, the worst possible complication that can arise due to postoperative hemorrhage from the paravertebral space will be hemothorax and not spinal hematoma (and also epidural hematoma) as with the existing procedures. Thus, we consider that this procedure is certainly useful rather than being an absolute contraindication for postoperative pain control among patients on anticoagulation therapy. As per their claim [2], although our study size remained small despite the subsequent accumulation of over 100 cases, spinal hematoma (and also epidural hematoma) was still a relatively rare complication. While spinal hematoma can potentially occur, its pathophysiology remains mostly unknown [3]. Furthermore, the absolute safety of patients on anticoagulation therapy with respect to bleeding risk cannot be stated. Therefore, we provide detailed information to such patients and their relatives/guardians regarding bleeding risk and attempt to obtain informed consent. Currently, this paravertebral catheter placement trial is still carefully performed as a clinical study under approval from our hospital’s research ethics committee.