Abstract Background: Blood trough levels of 1100 ng/ml are thought to be required for effective treatment of CML with IM. In a solid tumor like GIST, there might be different access of tumors cells to the drug even depending on the harbouring organ. There are no data available comparing tissue and blood levels of IM in humans. Materials: 24 patients scheduled for tumor resection were treated with IM 400mg/d or bid for metastatic GIST and medication continued including the preop. day. Blood samples were drawn intraop. at the time of resection. Samples of tumor tissues (n=52) and healthy organ tissues the tumor grew in were taken from 28 locations: liver (n=12), omentum (n=7), stomach (n=5), small bowel (n=2), soft tissue (n=1) and lung (n=1). Methods: To quantify IM in serum and tissue specimens a LC-MS/MS assay linear for concentrations 10 – 5000 ng/ml was used. IM and internal standard D8-IM were extracted from plasma specimens and corresponding tissue of GIST-pats. with ion exchange cartridges (OASIS MCX). Extracts were injected into the HPLC system equipped with XBridge phenyl column (Waters Corp.) and the non-isocratic elution with a mobile phase consisting of methanol/ammonium formate (4mM, pH4) performed within 15 minutes at a flow rate of 0,3ml/min. Ion transitions [M + H]+ of IM (m/z = 494 –> 394), DM-IM (m/z 480–> 394) and of D8-IM (m/z 502–> 394) were monitored by MRM mode. The mass spectrometer HCT Ultra (Bruker Daltonics) was equipped with an electro spray ionization probe, operating in positive mode. Peak areas and respective concentrations of DM-IM were calculated with QuantAnalysis (Bruker Daltonics). A six-point calibration for IM was performed once a day before analytical runs and quality control specimens with 3 different concentration-ranges (20, 100 and 1000 ng/ml). The interday precisions for the low level quality control (20 ng/ml) was below 20% and below 10% for the medium (100 ng/ml) and high (1000ng/ml) – levels respectively. Tissue IM/DM-IM data are given as ng/mg tissue, and mean+/SD. Correlations were calculated using Spearman and Pearson tests. Results: IM blood levels were 997+/-265 ng/ml (range 225-2760), and DM-IM were 215 +/- 124 ng/ml (range, 29-399), correlated with r=0.74. There was a significant difference between organs: liver IM 16,8 +/6.2 (DM-IM 13.2 +/4.5), stomach IM 5.25 +/- 1.2 (DM-IM 4.12 +/- 1.1), omentum IM 1.25 (DM-IM 2.1), p<0.01 and only a weak correlation to blood IM levels (r=.45). Tumor concentrations of IM were 40+/-12 (DM-IM 18.6 +/- 14) There was only a moderate correlation (r=.486, p=.021,) between blood levels of IM and tumor tissue (DM-IM r=0.34, p 0.06). Conclusions: In humans, tissue concentrations of imatinib cannot be predicted from blood levels, neither in healthy organs, nor in GIST tumor tissue. Different tissue IM levels could explain tumor progression during therapy despite of the missing of secondary mutations. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1309. doi:10.1158/1538-7445.AM2011-1309
Read full abstract