Validity of the modified RECIST criteria and EORTC PET criteria evaluated based on the pathologic findings for patients with resectable malignant pleural mesothelioma

Abstract

e11639 Background: The unique growth pattern of malignant pleural mesothelioma (MPM) presents challenges for clinical investigators evaluating the responses to chemotherapy, which is an important surrogate endpoint for patient benefit, particularly in clinical trials. In patients with resectable MPM, multicenter clinical trials of neoadjuvant chemotherapy followed by extrapleral pneumonectomy (EPP) and subsequent radiotherapy have been attempted around the world. The applicability of modified RECIST based on the findings on CT images and EORTC (European Organization for Research and Treatment of Cancer) criteria based on the findings on FDG-PET images to resectable MPM would be challenging and significant, but their validity has never been examined. Methods: Between May 2006 and November 2008, 13 consecutive patients with resectable pathologically proven MPM were included in this study. All were initially treated with combination chemotherapy including cisplatin. EPP was successfully performed in all the patients. In addition to modified RECIST (CR vs PR vs SD vs PD), FDG uptake by the tumor on PET was also evaluated according to the EORTC PET criteria (CMR vs PMR vs SMD vs PMD). Also, pathologic findings (NT; no viable tumor vs MR; minimal residual vs GR; gross residual) were reviewed. Results: According to modified RECIST, in which the definition of measurable lesions is ≥10mm in diameter, 7 of the 13 patients investigated had no measurable lesion. Even when the definition of measurable lesions was changed to ≥5mm, 2 patients had no measurable lesion and 4 had only one lesion. In regard to the response, 4 of 11 patients with any measurable lesions were classified as PR, and 7 were classified as SD, while 8 patients were classified as PMR and 3 were classified as SMD according to the PET findings. Eight patients were classified as GR and 5 as MR. Kappa statistics suggested potential variation between the CT response and the pathologic findings (κ=0.214, 95% CI=-0.377 ∼ 0.806) and between the PET response and the pathologic findings (κ=0.286, 95% CI=-0.049 ∼ 0.620). Conclusions: Our data raise doubts about the validity of applying the modified RECIST criteria as well as EORTC PET criteria to resectable MPM. No significant financial relationships to disclose.