Optimizing Serum GFAP and NfL Timing Enhances Accuracy in Assessing NMOSD Activity, Study Finds

Su-Hyun Kim, MD, PhD

A recently published retrospective study has shed light on the distinct dynamics of serum glial fibrillary acidic protein (sGFAP) and serum neurofilament light (NfL) after attacks in patients with neuromyelitis optica spectrum disorder (NMOSD). In the study, results showed that sGFAP levels peaked within the first week after symptom onset, while sNfL rose more gradually, with some patients showing normal levels within the first week and peaking at around 5 weeks.¹

"sGFAP may help differentiate genuine attacks from pseudoattacks, particularly in challenging clinical scenarios with preexisting T2 signal abnormalities or limited gadolinium enhancement on MRI," lead author Su-Hyun Kim, MD, PhD, neurologist at the National Cancer Center, Goyang, South Korea, et al wrote.¹ "Furthermore, sNfL’s correlation with neuronal damage severity in NMOSD attacks offers insights into predicting recovery after an attack. In conclusion, this study significantly advances our comprehension of sNfL and sGFAP as biomarkers for NMOSD."

The cohort study, published in JAMA Neurology, included 181 patients with aquaporin-4-positive (AQP4) NMOSD, 78 of which were from a discovery cohort in Korea and 103 from a validation cohort comprising German and Brazilian patients. Discovery cohort samples were collected from February 2008 to October 2023 and validation cohort samples were collected from January 2013 to October 2023. The analysis began by examining the temporal patterns of biomarker levels in 202 serial blood samples collected from 74 patients within the discovery cohort, with samples available within 6 months postattack.

Kim et al found that the median z score of sGFAP peaked at 4.5 (95% CI, 4.1-5.0) in the first week after symptom onset (week 0) and then rapidly declined over the next 12 weeks. In contrast, sNfL levels rose gradually, reaching a peak z score of 2.8 (95% CI, 2.4-3.2) at 5 weeks after symptom onset, with a slower decline in z scores over the following 20 weeks compared to sGFAP. Based on area under the receiver curve (AUC) results, investigators established that the optimal time frames for assessing sGFAP and sNfL levels during an attack were within the first week and from 1 to 8 weeks following the onset of attack symptoms, respectively.

READ MORE: Dried Blood Spot Testing Rivals Conventional Methods for NMOSD Diagnosis

Overall, the optimal cutoff for the sGFAP z score for distinguishing attack from remission was 3.0 (99.9^th percentile) with an AUC of 0.95 (95% CI, 0.88-1.02). This cutoff achieved a sensitivity of 85%, specificity of 96%, positive predictive value (PPV) of 87%, negative predictive value (NPV) of 96%, and an overall accuracy rate of 94%. For sNfL, the optimal cutoff was determined at a z score of 2.1 (98.3^rd percentile), with an AUC of 0.87 (95% CI, 0.82-0.91). This cutoff demonstrated a sensitivity of 71%, specificity of 89%, PPV of 77%, NPV of 85%, and an accuracy of 83%.

In the validation cohort, 7 sNfL samples were collected within the optimal attack window (1 to 8 weeks post-attack) and 245 during remission (at least 6 months after the last attack), while 1 sGFAP sample was collected within the 7-day post-attack window and 184 during remission, ultimately yielding 189 sGFAP and 252 sNfL samples for validation analysis. All told, a single sample from the attack phase showed an sGFAP z score of 5.0, exceeding the cutoff of 3.0, while all 196 remission-phase samples had values below this threshold, resulting in 100% sensitivity and specificity.

Using a z score cutoff of 2.1 to differentiate attacks from remission, sNfL achieved 71% sensitivity, 94% specificity, 26% PPV, 99% NPV, and 94% accuracy. During remission, 5 sGFAP samples (3%) and 11 sNfL samples (4%) showed z scores above 2.

At the conclusion of the study, the authors stressed that the data was strengthened by the international multicenter collaboration and diverse racial and ethnic groups involved. In addition, they wrote that "By refining their predictive accuracy and enhancing their reliability, our research contributes to improving the use of these biomarkers in clinical settings, facilitating more informed and precise management of patients with NMOSD."

The study also looked at clinical factors associated with sGFAP and sNfL levels during attack and remission. While there were more cases of myelitis (n = 67) than optic neuritis (ON; n = 10) in the discovery cohort, results showed that myelitis had a higher median sGFAP z score (4.9; 95% CI, 4.0-5.2) compared to ON (3.1; 95% CI, 1.9-4.3; P = .047) whereas the sNfL z score did not differ. Major attacks were associated with higher median sNfL z scores (2.95 [1.47-3.54]) compared to minor attacks (1.82 [0.85-2.73]; P = .003), while sGFAP z scores did not differ significantly between major and minor attacks based on the opticospinal impairment scale; during remission (at least 6 months after attacks), 24 of 164 sGFAP samples (16%) and 20 of 164 sNfL samples (12%) showed elevated z scores greater than 2.

No significant differences in disease duration, IST type, Expanded Disabilty Status Scale (EDSS) score, or time since the last relapse were found between patients with and without elevated sGFAP or sNfL levels during remission. In a Cox regression of 164 remission samples, 54 relapsed, with a median time to relapse of 12 months (IQR, 3-91). Higher EDSS scores at remission were linked to early relapse (HR, 1.26), while rituximab reduced relapse risk compared to oral drugs (HR, 0.14). However, sGFAP and sNfL levels in remission were not associated with early relapse.

REFERENCE
1. Kim SH, Gomes ABAGR, Schindler P, et al. Blood-Based Biomarkers for Identifying Disease Activity in AQP4-IgG–Positive Neuromyelitis Optica Spectrum Disorder. JAMA Neurol. Published online December 23, 2024. doi:10.1001/jamaneurol.2024.4400