Identifying erlotinib-sensitive non-small cell lung carcinoma tumors in mice using [(11)C]erlotinib PET

Abstract

Background: Non-small cell lung carcinoma (NSCLC) represents approximately 80% of lung cancer cases, and over 60% of these tumors express the epidermal growth factor receptor (EGFR). Activating mutations in the tyrosine kinase (TK) domain of the EGFR are detected in 10% to 30% of NSCLC patients, and evidence of their presence is a prerequisite for initiation of first-line therapy with selective TK inhibitors (TKIs), such as gefitinib and erlotinib. To date, the selection of candidate patients for first-line treatment with EGFR TKIs requires an invasive tumor biopsy to affirm the mutational status of the receptor. This study was designed to evaluate whether positron emission tomography (PET) of NSCLC tumor-bearing mice using [(11)C]erlotinib could distinguish erlotinib-sensitive from erlotinib-insensitive or erlotinib-resistant tumors.

Methods: Four human NSCLC cell lines were employed, expressing either of the following forms of the EGFR: (i) the wild-type receptor (QG56 cells), (ii) a mutant with an exon 19 in-frame deletion (HCC827 cells), (iii) a mutant with the exon 21 L858R point mutation (NCI-H3255 cells), and (iv) a double mutant harboring the L858R and T790M mutations (NCI-H1975 cells). Sensitivity of each cell line to the anti-proliferative effect of erlotinib was determined in vitro. In vivo PET imaging studies following i.v. injection of [(11)C]erlotinib were carried out in nude mice bearing subcutaneous (s.c.) xenografts of the four cell lines.

Results: Cells harboring activating mutations in the EGFR TK domain (HCC827 and NCI-H3255) were approximately 1,000- and 100-fold more sensitive to erlotinib treatment in vitro, respectively, compared to the other two cell lines. [(11)C]Erlotinib PET scans could differentiate erlotinib-sensitive tumors from insensitive (QG56) or resistant (NCI-H1975) tumors already at 12 min after injection. Nonetheless, the uptake in HCC827 tumors was significantly higher than that in NCI-H3255, possibly reflecting differences in ATP and erlotinib affinities between the EGFR mutants.

Conclusions: [(11)C]Erlotinib imaging in mice differentiates erlotinib-sensitive NSCLC tumors from erlotinib-insensitive or erlotinib-resistant ones.

Keywords: EGFR; Imaging; NSCLC; PET; TKI; [11C]Erlotinib.

Figure 1

Representative Western blots of four human NSCLC cell lysates comparing the extent of EGFR and phospho-EGFR expression. β-actin served as a reference for equal loading.

Figure 2

Representative PET/CT slice images of NSCLC tumor-bearing mice. The images were taken following sequential injections of [¹¹C]erlotinib (a-d) and [¹⁸F]FDG (e-h) into each mouse, demonstrating (arrowheads) [¹¹C]erlotinib uptake in erlotinib-sensitive tumors (b, c) and in erlotinib-insensitive ones (a, d). [¹¹C]Erlotinib and [¹⁸F]FDG images depict the summation of radioactivity uptake from 30 to 60 min and 40 to 60 min after injection, respectively. Each set of [¹¹C]erlotinib and [¹⁸F]FDG images was normalized to the same scale.

Figure 3

TACs representing NSCLC tumor kinetics following injection of [ ¹¹ C]erlotinib. Tumor TACs were obtained after i.v. injection of [¹¹C]erlotinib into NSCLC tumor-bearing mice, demonstrating increased radioactivity uptake in erlotinib-sensitive tumors (HCC827 and NCI-H3255) compared to insensitive (QG56) or resistant (NCI-H1975) ones. The number of scanned tumors is indicated in brackets. **p < 0.01 and ***p < 0.001.

Figure 4

Representative Western blots of NSCLC tumor extracts, demonstrating total EGFR and phospho-EGFR levels in tumors. n = 3 per group. β-actin served as a reference for equal loading.

Figure 5

Inverse correlation between [ ¹¹ C]erlotinib uptake in HCC827 tumors at 60 min after injection and injected carrier mass. n = 12. The last point on the right represents the administered blocking dose (6.7 mg/kg, n = 5).