**6. Future perspectives**

More than a decade PET imaging has seen a revival in the ALS field using either new hardware and software technologies or novel tracers. In the near future, PET will hopefully find applications in both clinical practices, namely clinical trials and neurobiological research.

First, PET imaging could become of value in the ALS diagnosis in the future. Although conventional imaging (e.g. MRI) is only intended to rule out other diseases, FDG PET has the potential to be used as a positive argument to make a diagnosis of ALS at the single patient level. Both hypometabolism in the motor cortex as in the frontotemporal cortex will be of diagnostic value and could be considered to incorporate in clinical criteria for ALS, similar to the inclusion of frontotemporal hypometabolism in the diagnostic criteria of FTD [90]. However, additional research comparing ALS patients with ALS mimicking diseases is needed to reliably assess the sensitivity and especially the specificity in the real-life clinical setting of early diagnosis.

Second, PET imaging will be further investigated as a biomarker of disease. More studies assessing the correlation with severity of disease need to be undertaken. Similarly, more longitudinal studies are needed to relate early PET abnormalities with clinical course, hope‐ fully fulfilling the high need for a prognostic marker in the clinic.

Third, PET imaging could become valuable in clinical trials of ALS patients. By increasing the diagnostic yield, especially early in the disease, it will be possible to include patients early after disease onset, hence increasing the power to obtain positive results in pharmaceutical trials. Additionally, several tracers (neuroinflammation, neuronal loss …) could be used as a read out to demonstrate target engagement or even to assess the effect of treatments. Of particular interest, a tracer for TDP-43 would revolutionize the field, as PET detection and quantification of misfolded proteins have done in Alzheimer's disease with beta-amyloid and tau imaging.

Fourth, tracers assessing neuroinflammation will be further investigated, and aside from TSPO other targets such as type 2 cannabinoid receptors, the purinergic receptor P2X7, and matrix metalloproteinases are investigated currently. The power of protein or receptor tracers mainly lies in the selectivity of their target, that is they selectively reflect one aspect of ALS patho‐ genesis which increases with disease progression. Hence, in contrast to FDG PET which reflects general glucose metabolism, these tracers have the potential to be used as 'positive tracers' for disease severity and progression. They are also of high interest to gain insight in the patho‐ physiology of ALS. Probably, several tracers assessing other aspects of ALS pathology will be developed and investigated as well.
