Effective chemotherapy is expected to induce various metabolic changes in tumours that are followed by alterations in morphology and growth inhibition, unregulated cell death or apoptosis. It can be assumed that the physiological changes occur before changes in the tumour size can be observed. This is the principle behind the concept that an efficient chemotherapy can be detected with PET at earlier stages than with other types of imaging modalities. However, the question remains which metabolic modifications will occur after a specific chemotherapy, and which PET tracer is most sensitive in detecting those modifications. Of equal importance is ascertaining when a treatment induces metabolic changes that are recordable with PET that do not translate into an anti-tumoural action such as inhibition of growth.
Although there is a relatively good understanding of the mechanisms by which certain anticancer drugs exert their effect, and it is relatively well known by which mechanisms certain PET tracers accumulate in tumour cells, it is usually less obvious how these mechanisms are interrelated. The correlation is especially difficult to predict with PET tracers that reflect rather general physiological phenomena, e.g. there is no obvious connection between microtubule inhibition and expression of glucose transporters. Hence, without an understanding of these links, we must revert to empirical evaluations of correlations. This increases the complexity of the situation, as correlations observed in a certain assay with its own experimental conditions might not be relevant for other conditions. For example, observations made with a certain time of drug exposure might not be reproduced with another drug schedule. In this sense, we believe that the proposed assay is an important methodological contribution because it allows a work-efficient way of exploring different experimental schemes, and the spheroid model allows variable scheduling to be simulated.
The aim of the present study was to explore our assumptions and to complement our previous studies on use of MTS and monitoring of PET tracer uptake by illustrating its ease and potential using five different drugs and five different PET tracers available at multiple PET centres, all of which have shown usefulness in the diagnosis of different types of cancer. Of special importance has been the use of the semi-automated size determination method (SASDM), as it allows an accurate and precise assessment of the volume of viable cells in the spheroids. This in turn gives us an opportunity to define tracer uptake in relation to viable volume and not only the total volume, which is especially important in long-term studies and drug effect studies where the proportion between viable volume and necrosis can be expected to change. Furthermore, it allows a better estimate of growth rate and effect on growth rate by treatment.
The use of ACE in PET has been suggested in urologic malignancies, hepatocellular carcinoma (HCC) and cardiac disease. The mechanism of tumour retention of this tracer differs from that of myocardium, in which ACE is incorporated into the citric acid cycle and rapidly metabolised. The dominant process of ACE incorporation in tumours is thought to be participation in lipid synthesis [17
]. In this study, an increase of ACE uptake, from non-significant with docetaxel to 40% with tamoxifen, to a decrease by 30% with doxorubicin, was observed. Hence, compared with other tracers, ACE seemed to have less utility for the monitoring of breast cancer treatment.
In the human body, choline is needed for the synthesis of phospholipids in cell membranes, methyl metabolism, transmembrane signalling, and lipid-cholesterol transport and metabolism. The uptake is transport mediated and intracellular choline is rapidly metabolised to phosphorylcholine (PC); it can also undergo acetylation to form acetylcholine or oxidation to form betaine (mainly in liver and kidney). The phosphorylation is catalysed by the enzyme choline kinase. After phosphorylation, the polar PC molecule is trapped within the cell. Various studies have revealed an increased choline uptake as well as an up-regulated activity of choline kinase and elevated levels of PC in cancer cells [18
]. Although a primary use is in prostate cancer, it has also been suggested that CHO could be a potential PET tracer for imaging of breast cancer [19
]. In this study, a 10 to 20% decrease of CHO uptake in MTS treated with doxorubicin and imatinib was observed. However, for doxorubicin treatment the decrease in uptake was much less than growth inhibition and for imatinib, the decrease was a false indication of treatment efficiency.
MET has been suggested as a tumour imaging agent [20
]. The uptake of MET reflects increased amino acid transport and, in part, protein synthesis; it also is related to cellular proliferation activity. In MTS treated with paclitaxel, docetaxel and doxorubicin, MET uptake decreased slightly but to a lesser extent than the growth inhibition.
FLT has become one of the promising PET agents that reflect nucleoside transport into the cell and has been proposed as a marker for cellular proliferation [23
]. FLT is phosphorylated by thymidine kinase-1 and then trapped intracellularly by entering the salvage pathway of DNA synthesis without incorporation into the DNA molecule. The FLT uptake was considerably decreased during 3 days of treatment with paclitaxel, docetaxel, doxorubicin and high dose of tamoxifen where also a growth inhibition was observed. The modest FLT uptake increase observed in imatinib treatment was correlated to the moderate growth enhancement. Ki67 staining illustrated and confirmed the anti-proliferative effect of the drugs monitored with FLT.
FDG is the most commonly used PET tracer in cancer diagnosis and has shown utility for treatment monitoring for selected agents. FDG uptake reflects glucose transporters expression and hexokinase in the cells, dual mechanisms that lead to fluorodeoxy-glucose-phosphate accumulating without further metabolism. It is the most commonly used PET tracer for diagnosis and treatment follow-up of breast cancer. FDG uptake reduced as a result of treatment with taxanes, doxorubicin [16
] and the higher dose of tamoxifen. FDG uptake reduction correlated with growth inhibition, but not as clearly as FLT.
Our results showed high sensitivity of FLT, compared with FDG and other tracers, for monitoring the effects of the drugs in MTS. This is an example where PET tracers other than FDG seem necessary to improve the ability to measure and predict response and hopefully help to tailor therapy to individual patients. This study only used one cell line, MCF-7, and aimed primarily to show the methodology. To allow more general conclusions with respect to treatment monitoring of breast cancer, it would be necessary to explore further breast cancer cell lines, evaluate more concentrations of the drugs, investigate the time courses of drug effects by using pulse treatment, and probe more carefully the uptake pattern of the PET tracers.
There are also many additional factors that should be considered, e.g. concentrations of glucose, methionine and choline in the cell culture medium that are 5–10-fold higher compared to blood.