Being sensitive: to specify when amino acid tracers accumulate in a brain lesion.

TO THE EDITOR: I read with considerable interest the recent report by Rapp et al. (1) demonstrating the diagnostic performance of O-(2-18F-fluoroethyl)-L-tyrosine (18F-FET) PET in newly diagnosed cerebral lesions suggestive of glioma. This retrospective but nevertheless convincing study provides us with substantial data on the accumulation of an amino acid PET tracer in glioma for a relatively large patient cohort. As the authors emphasize, their study meets the criteria of strict standardization of PET acquisition protocols. Considering that the clinical value of amino acid brain PET imaging for differential diagnosis still can be considered a work in progress, these precisely documented findings must not be underestimated. On the other hand, the quintessence of these evaluations can be found in previous publications, and of course this fact supports the merit of the data of Rapp et al. in showing that high-grade glioma nearly always exhibits intense accumulation of 18F-FET and that low uptake therefore excludes a high-grade tumor with high probability. Also, their finding of higher levels of 18F-FET uptake in high-grade glioma than in low-grade glioma is not new, and because of their observed marked overlap in uptake quantification—again supporting previous data—the authors’ conclusion that 18F-FET uptake ratios provide valuable additional information for grading of gliomas may be questionable at least in clinical practice. We also face the substantiality that on visual rating about two thirds of low-grade glioma are 18F-FET–positive and one third is 18FFET–negative. A study by our group published in 2010 (2) is cited by Rapp et al. as “the currently largest series” of patients. Additionally, they comment that “these results, however, were based only on a visual rating, and histology was available in only two thirds of patients.” This is right, as the images were of course rated visually by the reporting physicians. But we also published a lesion-to-brain ratio—correlated to histology when available—with results comparable to the findings of our German colleagues. Here comes my main message: evidentially, it is important to have valuable data on amino acid uptake in lesions that are very suggestive of glioma and subsequently proven to be glioma on pathologic examination. But we have to look forward. What is the nature of 18F-FET uptake in lesions that are possibly, but not very probably, glioma? Published data may focus on observational studies as well, as shown by the recent work of Hutterer et al. (3), in which only three quarters of patients had histology available. For the individual patient, the valuable information obtained from 18F-FET PET—additional to that from MR imaging in general—would allow for better decisions about medical management. We must know more about possible pitfalls, such as whether 18F-FET accumulates in abscesses, multiple sclerosis plaque, vasculitis (3), or radiation-induced astrogliosis (4). Then, we can make comparisons with a typical profile of 18F-FET uptake in glioma as shown by the retrospective data of Rapp et al. and others. The scientific community should also embrace the concept that negative or low-accumulating lesions—suggestive of low-grade glioma for example, by MR imaging—are related to a good prognosis even without specific therapy. Even more data are missing related to brain metastases from solid tumors, such as the already helpful data provide by Langen’s group (5).