RESUMEN
FDG (18F-deoxy-glucose) is the current gold standard for PET imaging. FLT (3'-deoxy-3'-(18F-fluorothymidine), a PET imaging marker of proliferation, has been proposed as an alternative to FDG for the assessment of therapeutic response. We examined the therapeutic predictive value of FLT-PET and FDG-PET using CALU-6, a human, p53-null, non-small cell lung cancer cell line with comparison of combined targeted therapy, TRAIL and sorafenib, versus combined conventional chemotherapy, docetaxel and cisplatin. CALU-6 tumor-bearing nu/nu mice (n=46) were evaluated in 3 therapeutic trials measuring FLT and FDG prediction of tumor response at 72 h following initiation of daily combination therapy with targeted agents, TRAIL (200 µg i.v.) and sorafenib (30 mg/kg i.p.) and compared to conventional chemotherapeutics cisplatin (3 mg/kg i.p.) and docetaxel (7.5 mg/kg i.p.). PET imaging response was compared to morphological and histological indicators of therapeutic response, including decreased vascularity (in vivo AngioSense imaging and anti-CD31 staining), slowed tumor growth (caliper measurements), decreased cellular proliferation (Ki-67 staining) and increased apoptosis (TUNEL staining). Decreases in tumor accumulation of FLT (FLTMAX -30%, p=0.03) at 72 h post treatment were observed in response to TRAIL and sorafenib combination therapy resulting in smaller, less vascular, more apoptotic tumors. No similar reduction in tumor accumulation of FLT (FLTMAX -2%, p=0.67) was observed 72 h following initiation of cisplatin and docetaxel combination therapy, despite histological and morphological evidence of drug response. In contrast, tumor imaging with FDG did demonstrate a decrease in accumulation in both treatment groups, -21% (p=0.30) in response to cisplatin/docetaxel and -8% (p=0.59) in response to TRAIL/sorafenib, but did not reach statistical significance. FLT, but not FDG, is predictive of therapeutic response to the targeted regimen TRAIL/sorafenib. However, FLT-PET may not predict therapeutic response to DNA damaging agents in p53-null tumors, likely due to loss of cell cycle control of thymidine kinase 1 (TK1). Thus, tumor imaging response by FLT may be limited in human tumors without functional p53.