Abstract
178
Objectives: Accurately identifying nociceptive sources of low back pain has remained to be a great diagnostic challenge despite its pressing need due to the high prevalence and social impact of the condition1. Unfortunately, findings from structural imaging methods, such as MRI, have been often fraught with false positives and non-specific to pain2, which can result in ineffective and unnecessary treatments3. Sigma-1 receptor (S1R) is a very promising alternative biomarker of painful pathologies because it plays a key role in the ion channel modulation for nociceptive processes4. Our group recently demonstrated the feasibility of in vivo PET/MR imaging of S1R using a novel S1R radioligand (S1RR), 18F-FTC-1465,6. The objective of this study is to introduce early results from our PET/MRI of S1R patient study focusing on radiating low back pain. Methods: We recruited seven patients presenting pain radiating from the low back down to buttocks or legs. Our prospective imaging study was approved by the institutional review board. An informed consent form signed by the patient was obtained prior to imaging. We conducted a whole-body PET/MRI 40 minutes after the intravenous injection of 10-mCi dose of S1RR (18F-FTC-146). A GE SIGNA PET/MRI scanner (GE Healthcare, Waukesha, U.S.) was used for collecting the PET/MRI images simultaneously. PET/MRI images of the lumbar spine to the mid-thigh regions were reviewed by two musculoskeletal radiologists. The MRI image review was to find lesions with abnormal signal intensity on T2-weighted contrast or structural changes. The PET image review was mainly to identify focal hotspots of S1RR uptake. The SUVmax of unilateral hotspots in nerves and muscles was compared with that of the contralateral side using the Wilcoxon signed rank test. The identified abnormalities were grouped depending on the source imaging modality: 1) identified in MRI only, 2) identified in PET only, 3) identified in both MRI and PET.
Results: All patients presented abnormalities on either PET (all 7 patients) or MRI (6 patients) while more diverse abnormalities were found on PET images (Table 1). MRI-only abnormalities were mostly increased signal on the peripheral nerves (3 patients) while structural abnormality was found on only two patients as shown in Figure 1 (stenosis and disc abnormalities). On the other hand, S1R PET images showed increased S1RR uptake on a wider variety of tissues such as spinal joints (3 patients), paraspinal/leg muscles (6 patients), and peripheral nerves (4 patients). The SUVmax of unilateral nerve and muscle hotspots was higher than that of the contralateral side with a 5% significance level (Table 2). Except one patient, all other patients presented abnormally increased uptake of S1RR on the muscles in the symptomatic area with any apparent sign of degeneration on MRI (Figure 2). Lesions presenting abnormalities on both S1R PET and MRI were located at either peripheral nerves as shown in Figure 3 or structures close to nerves (patient 6). Conclusion: In summary, we have presented early results of S1R PET/MRI on patients with radiating low back pain. While the MRI abnormalities were mostly found involving the peripheral nerves, the increased S1RR uptake was detected on various tissues including joints, nerves, and muscles around the lumbar spine. The S1RR uptake at the site of the structural abnormalities on MRI can form a stronger evidence in determining the cause of pain, considering the high false-positive rate of the structural abnormalities on MRI in general. Also, the increased S1RR uptake on muscles without the gross structural change may indicate the sensitivity of our S1R PET method to early inflammatory processes. Therefore, our approach demonstrated the promise for the improved confidence and sensitivity in detecting local sources of radiating low back pain. Acknowledgement: GE Healthcare, NIH P41 EB015891.
