from the SUV, these are all relative parameters and sensitive for “inter-corporal” properties (blood volume, perfusion, clearance rate), relying on the definition of representative reference volumes (blood or muscle). Also, the parameters may vary among different protocols (e.g., amount of tracer injected, reconstruction method, correction for breathing motion). Furthermore, the time after administration influences the values, since at later time points the blood will be further cleared, while the uptake in hypoxic tissue increases as long as there is supply of the tracer.
Since hypoxia is in general defined by a signal increase of only 20–40%, the signal-to-noise ratio needs to be maximal, especially for the detection of small hypoxic subvolumes. Besides taking advantage of the optimal imaging timeframe according to the clearance and decay, image noise can be reduced by increasing the acquisition time or the administered activity. The first is at the cost of time and patient discomfort, while the second is at the cost of a higher imaging-induced radiation dose. Compared to often-used 18F-FDG-PET imaging protocols, the time per bed position is much longer. While 18F-FDG-PET is frequently used as a whole body imaging method (e.g., 6 bed positions), tumor hypoxia imaging can mostly be limited to 1 or 2 bed positions only, which reduces the overall time required for imaging.
Mechanism of Cu-ATSM Uptake in Hypoxic Tissue
The other group of PET-based hypoxia tracers consists of Cu-ATSM compounds. Although the uptake mechanism of these tracers is not fully understood, it is known that the lipophilic molecule diffuses through the cell membrane, and within the cell the copper compound undergoes reduction by thiols: Cu(II)-ATSM is converted to Cu(I)-ATSM. In the case of hypoxia, this unstable complex undergoes further reduction and the resulting free Cu(I) becomes rapidly entrapped in intracellular proteins [32].
Experiences with Cu-ATSM
Experiences with Cu-ATSM vary between promising and disappointing. The advantage of Cu-ATSM compared to nitroimidazole compounds is the higher uptake in target tissue (T/M 3.0) within only 10–15 min [32]. However, there are doubts on the hypoxic selectivity since the Cu-ATSM uptake does not always correspond to the distribution of immunohistochemical markers, especially at early imaging timepoints (<16 h after injection) [39, 40]. In future it will be elucidated whether further development of Cu-ATSM tracers can overcome these doubts [10].
Advantages and Disadvantages of PET-Based Hypoxia Imaging
Besides being a time-consuming procedure with hours between administration and imaging and a poor contrast between hypoxic and normoxic tissue, PET-based hypoxia imaging has the major disadvantage of a low resolution, resulting in problematic detection of small but potentially relevant volumes of hypoxic tissue [23]. A more general challenge of hypoxia imaging is its dynamic nature; it remains questionable whether a single PET can provide enough reliable information to make the treatment dependent on this biomarker [41–43]. The advantage of imaging biomarkers for hypoxia is that they provide 3D information about the tumor tissue, in a noninvasive manner. This enables longitudinal monitoring and facilitates further research in this field.
MRI-Based Hypoxia Imaging
Besides PET, hypoxia and perfusion may also be visualized with MRI. Different MRI protocols have been suggested for this, for instance dynamic contrast enhanced (DCE) MRI in which gadolinium is typically used as a paramagnetic contrast agent. With DCE MRI, different parameters can be retrieved (e.g., ktrans/kep, ve) for the visualization of tissue perfusion.
Reduced perfusion might indicate tumor hypoxia, and therefore DCE MRI can be used as a surrogate for hypoxia imaging. However, increased vascular permeability leads to higher intratumoral contrast agent levels, while it may be associated with increased hypoxia [37, 44].
Other examples of interesting sequences/protocols, although not (yet) with a proven predictive value, are blood-oxygen level-dependent (BOLD) MRI [45], which visualizes the ratio between paramagnetic deoxyhemoglobin and diamagnetic oxyhemoglobin, and the less mature tissue oxygen level dependent (TOLD) MRI, and mapping of oxygen by imaging lipid relaxation enhancement (MOBILE). These could be used for the visualization of tissue oxygenation in water and in lipids, respectively [44].
In MRI, the acquisition itself can be performed with many different settings, and many different parameters can be retrieved from the acquired data. Without denying the value of this field of research, in depth discussion about the advantages and disadvantages of these different techniques and applications is not included here.
Imaging of Proliferation
Proliferation and Radioresistance
In cancer tissue, a typical misbalance between the rate of cell death and the rate of cell proliferation occurs, resulting in tumor growth. Generally, within tumor categories the tumors with the highest proliferation rate have (without treatment) the worst prognosis [12], although conflicting study results have been published [46]. Generally, a higher proliferation rate before treatment and early during treatment may be related to a worse response to radiotherapy [8, 16], and especially accelerated proliferation during the course of radiotherapy is related to a poorer outcome.
Causes of Accelerated Proliferation
Irradiation results in the killing of well-oxygenated tumor cells, and therefore the tumor volume reduces. For the remaining malignant cells this leads to improved supply of oxygen and nutrition, a process referred to as “reoxygenation.” Before therapy, these cells had to compete with many more tumor cells, which apparently would have led to spontaneous cell loss. Instead, due to radiotherapy, these cells are enabled for further proliferation. Proliferation itself is also stimulated in response to ionizing radiation [47]. This mechanism of proliferation of tumor cells, stimulated or enabled by radiotherapy, is called accelerated repopulation [48].
Counteracting Radioresistance due to Proliferation
Especially in fast-responding tumors, accelerated repopulation should be considered in the treatment design. Practically, specifically for head-and-neck squamous cell, and non-small-cell and small-cell lung cancer, this risk of repopulation during the treatment course is known [48]. In case of a longer treatment time, in these cancers accelerated repopulation plays a more prominent role and reduces the effectiveness of the radiotherapy. In this situation, tumor cell kill due
