ESTRO 29: PET in radiotherapy planning: particularly exquisite test or pending and experimental tool?

14 Sep 2010

By Vincent GREGOIRE (1), MD, PhD, FRCR and Arturo CHITI (2), MD

In medicine, clinical practice cannot resist fashion: in the year 2000, 9 publications dealt with the "use of PET in radiotherapy planning"; by the end of 2005, 187 articles have been published on this subject, and by the end of June 2010, PubMed indicated 565 hits using the same keywords!

But are we sure that this is fashion? Isn't it a reflection of a medical progress that translated into practice change? And if so, what would have made the use of PET so exquisite in radiotherapy planning? Answering this question requires a profound understanding of the process of radiotherapy planning and the key elements that are likely to impact on it.

Radiotherapy planning is a multiple step process that starts with a decision of a multidisciplinary tumor board to irradiate a malignant tumor with a given T-N-M stage, and end with the signature of a board certified Radiation Oncologist on a treatment plan comprising a dose prescription, a set of dose distribution together with a set of technical parameters that will be transferred to the linear accelerator (or equivalent machine) for the plan execution.

In this process, the appropriate selection and delineation of target volumes (TV) and Organs At Risk (OAR) -typically performed on contrasted CT and/or MRI- by a trained Radiation Oncologist is of paramount importance. It requires the proper use of both extremely sensitive (very few false negative examination, or with a high negative predictive value) and specific (very few false positive examination, or with a high positive predictive value) imaging modalities. In other words, we want the images to give us the truth, all the truth and only the truth... In addition, for the delineation aspect, one needs imaging modalities that have a high spatial resolution, typically in the order of 1-2 mm. Both CT and MRI fulfill this criterion.

How is the picture with PET? PET has an intrinsic advantage, which is that it could image any physiopathological pathways within a tumor or a normal tissue, providing an ad hoc tracer can be injected to the patient. And tracers of metabolism, proliferation, hypoxia, apoptosis, gene expression (e.g. EGFR), etc... have been validated and are available if not all for routine clinical practice, at least for clinical studies. But PET has potential limitations, which are its relatively lower spatial resolution and the low signal-to-background ratio observed with some tracers; both limitations could compromise the overall sensitivity and specificity of PET compared to CT and/or MRI. The lower spatial resolution is a complex issue resulting from several factors, and among them the energy of the positron, the size of the detector and the low counting efficiency are the most important. These factors may require the use of appropriate acquisition protocols (e.g. correction for movement artifacts) and special reconstruction algorithms to sharpen the edges of the volume of interest. The low signal-to-background ratio typically results from the tracer metabolism, which ideally should distribute evenly into the body, get fixed to the site(s) of interest while the unfixed tracer should get cleared from the body. And all of these processes should be done as quickly as possible, and for sure within a time frame compatible with the half-life of the tracer. But as one knows, the reality is far from this ideal picture...

Several articles have been recently published on the use of PET for radiotherapy planning in brain, head and neck, oesophageal, lung, prostate, and anal canal tumors. No definitive conclusion could however be drawn from these publications, PET influencing target volume selection and delineation in some studies, not in others. But what seems consistently observed in all these publication is the importance of the use of well-defined protocols for image acquisition, reconstruction and segmentation.

It is expected that ESTRO 29 will contribute to increase the awareness of Radiation Oncologist, Nuclear Medicine physicians and other related specialists on the proper use of PET in radiotherapy planning. The question remains is PET a "Particularly Exquisite Test" or a "Pending and Experimental Tool"? Obviously, the answer will be in the middle...

Source: ESTRO 29

(1) Radiation Oncology Dept. and Laboratory of Molecular Imaging and Experimental Radiotherapy, Université catholique de Louvain and St-Luc University Hospital, Brussels, Belgium.
(2) Nuclear Medicine Dept., Istituto Clinico Humanitas, Milan, Italy