2.2.1. Positron Emission Tomography (PET) Imaging Biomarkers

Fluorodeoxyglucose-F18 (18F-FDG) is the most commonly used PET imaging radiotracer, being an indicator of tumour activity via glucose metabolism, and has an established role in tumour staging and treatment response monitoring. Its role in the assessment of tumour proliferation was also researched, with a considerable number of studies examining the potential of 18F-FDG in predicting EGFR mutation status in NSCLC patients. In a retrospective clinical study involving 109 NSCLC patients, Chen et al. showed that EGFR mutation decreases cellular accumulation of FDG via the NOX4/ROS/GLUT1 axis ^[16]. The SUVmax values in the cohort with EGFR mutations were significantly lower (6.52 mean value) than in the wild-type EGFR cohort (9.37 mean value, p < 0.001). Similarly, in a study of 102 NSCLC patients with EGFR mutation (22%), KRAS mutation (27%) and wild-type profiles (51%), it was observed that 18F-FDG uptake was significantly higher in those harbouring KRAS mutations as compared to EGFR+ or wild-type (SUVmean 9.5 vs. 5.7 vs. 6.6, p < 0.001) ^[17]. These findings are corroborated by a much larger study, encompassing 849 NSCLC patients with 45.9% identified with EGFR mutation, that also showed low SUVmax association with EGFR mutation status ^[18]. This result could be combined with other clinical factors to improve patient stratification, particularly when EGFR testing is not available ^[18][19].

A recent study reported on the development of a new PET tracer with high specificity to activating EGFR mutant kinase showing significant correlation between tracer uptake and the EGFR mutation status in both preclinical animal models and in patients with NSCLC ^[20]. The study aimed to identify, via a new imaging tracer—18F-MPG (N-(3-chloro-4-fluorophenyl)-7-(2-(2-(2-(2-¹⁸F-fluoroethoxy) ethoxy) ethoxy) ethoxy)-6-methoxyquinazolin-4-amine)—, those patients that are sensitive to EGFR-TKIs and to monitor the efficiency of EGFR-TKI therapy. The cut-off value for SUVmax was set at 2.23, showing a greater response to EGFR-TKI in those presenting with SUVmax ≥2.23 as compared to patients with values <2.23 (81.58% vs. 6.06%). Furthermore, 18F-MPG uptake positively correlated with median progression-free survival ^[20].

While 18F-FDG has its own merits in the functional imaging of lung cancer, it is not the optimal indicator of proliferation, showing poorer correlation with cellular proliferation markers than other PET tracers. Fluoro-3′-deoxythymidine-F18 (18F-FLT), a successfully used imaging marker of cellular proliferation, is a radiolabeled structural analog of a DNA nucleoside—thymidine—and its uptake relates to the activity of thymidine kinase 1 (TK1) that is expressed during DNA synthesis in the S-phase of the cell cycle ^[21]. The uptake of 18F-FLT in tumour cells is lower as compared to 18F-FDG, as it only accumulates in cells during the S-phase ^[15]. Yet, several studies demonstrated the superior correlation of 18F-FLT with cellular proliferation markers when compared to the traditional 18F-FDG ^[22][23]. In one of the first comparative studies that involved a cohort of 26 lung cancer patients, Buck et al. showed high correlation between 18F-FLT uptake and Ki-67 index (p < 0.0001; r = 0.92), and concluded that 18F-FLT may be a better imaging marker than FDG for response assessment and outcome prediction ^[22]. These observations are supported by a recent meta-analysis that assessed 1213 patients from 22 imaging studies that correlated the Ki-67 labeling index with FDG and FLT uptake, respectively, showing that the latter is a more robust marker of tumour proliferation in lung cancer ^[23].

In a recent pilot study, Kairemo et al. demonstrated the feasibility of 18F-FLT PET in monitoring treatment response by early signal activity in NSCLC patients receiving targeted therapies (c-MET inhibitors) ^[24]. Several others have confirmed the potential of 18F-FLT PET imaging to monitor and guide molecular targeted therapies in NSCLC ^[25][26][27].

Next to the most common Fluor-based radiotracers employed in PET for tumour proliferation imaging, copper is another successful candidate. Functional imaging with PET employing 64Cu-ATSM (Cu-labeled diacetyl-bis(N(4)-methylthiosemicarbazone) and 18F-FDG was undertaken for the intratumoral distribution assessment of the two radionuclides in Lewis lung carcinoma tumour cells implanted in mice ^[28]. Both proliferation markers (Ki-67 and BrdU-bromodeoxyuridine) and the hypoxic marker, pimonidazole, were used to compare radionuclide uptake with immunohistochemical staining patterns. The association of staining with radionuclide accumulation revealed an increase in Ki-67 positive areas with 18F-FDG uptake increase and, at the same time, a decrease with 64Cu-ATSM accumulation. Conversely, the other proliferation marker, BrdU, showed an opposite behaviour, with the number of BrdU-positive cells being positively correlated with 64Cu-ATSM uptake and negatively related to 18F-FDG accumulation. Given that BrdU is a marker for proliferation by way of DNA synthesis, the fact that cells with high 64Cu-ATSM uptake positively correlated with the number of BrdU cells indicates that they are able to undergo DNA synthesis, though not during the proliferation process (denoted by the low Ki-67 levels which are specific to G1 and early S phase). This result suggests that cells in regions with high 64Cu-ATSM uptake were quiescent, yet sustained DNA synthesis and were sensitive to progression factors, just like quiescent cancer stem cells. Clonogenic assays within the same study have proven the stem-like properties of cells originating from high 64Cu-ATSM uptake tumour areas ^[28]. Furthermore, pimonidazole-positive areas were specific to regions with low 64Cu-ATSM accumulation, suggestive of mild hypoxic conditions, still optimal for the thriving of clonogenic tumour cells.

This study is a clear illustration of the complexity of the tumour microenvironment and of the many factors that influence tumour development and response to therapy (hypoxia, proliferation, cancer stem cells). Based on the above results, 64Cu-ATSM could potentially serve as a complex imaging biomarker to supply prognostic information for treatment adaptation and optimisation.

2.2.2. Single Photon Emission Computed Tomography (SPECT) Imaging Biomarkers

Beside PET tracers, a number of researchers attempted to develop SPECT radioisotopes for novel insights into EGFR targeting. The capacity of 99mTc-HYNIC-MPG ((2-(2-(2-(2-(4-(3-chloro-4- fluorophenylamino)-6-methoxyquinazolin-7-yloxy)ethoxy)ethoxy)ethoxy)ethyl-6-hydrazinylnicotinate hydrochloride) was evaluated in detecting EGFR-activating mutations both in vitro and in vivo, using human NSCLC cell lines ^[29]. The study showed that of the four cell lines (EGFR+, EGFR− and wild-type), 99mTc-HYNIC-MPG uptake was the highest in the cell line with exon 19 deletion (PC9), probably due to the activating mutations in EGFR tyrosine kinase domain. The results could serve to further stratify NSCLC patients by identifying the subgroup that would benefit the most from targeted therapies with EGFR-TKIs ^[29].

Table 1 is a compilation of different functional imaging agents tested as markers for tumour proliferation in NSCLC.

2.2.3. Magnetic Resonance Imaging (MRI) Biomarkers

The latest advances in biomaterials, specifically in nanomedicine, have greatly increased the sensitivity of imaging techniques using magnetic resonance to perform accurate and non-invasive functional imaging. In this regard, one of the recent developments is in the field of superparamagnetic iron oxide (SPIO) nanoparticles (40–50 nm), whereby polyethylene glycol-coated SPIO nanoparticles (PEG-SPIO) were synthesized and further labeled with high affinity anti-EGFR monoclonal antibody (cetuximab) for targeted delivery to lung cancer that overexpresses EGFR ^[30]. The targeting efficiency, MRI contrast enhancement and cytotoxicity of this nanocomposite was evaluated in both H460 lung cancer cells (in vitro) and tumour-bearing rats (H460 lung xenografts) in vivo. The uptake of the nanocomposite in the cell lines was evaluated by Prussian blue staining which showed an increased cellular uptake of anti-EGFR targeted NPs compared to non-targeting NPs at the same iron concentration, suggesting that the high cellular accumulation of anti-EFGR NPs is due to the EGFR receptor-mediated endocytosis pathway. This was also illustrated by TEM (transmission electron microscopy) imaging, where cells incubated with anti-EGFR targeting NPs showed the presence of electron-dense particles in the cell endosome, in contrast with those incubated with non-targeting NPs, which showed no such uptake. To further confirm these results, MRI-based investigation was undertaken by measuring the T2 weighted signal intensity of lung cells after incubation with NPs having various iron concentrations. It was observed that the T2 signal decreased with the increasing iron concentrations in the EGFR targeting NPs group. Furthermore, the signal intensity of lung cancer cells that overexpressed EGFR and were targeted with anti-EGFR-PEG-SPIO decreased more significantly than in the PEG-SPIO (non-targeting NPs) group. The study concluded that efficient identification and targeting of lung cells overexpressing EGFR can be achieved by means of anti-EGFR-PEG-SPIO nanocomposite, under MRI monitoring ^[30].