The primary finding of this study is that CSF NSE levels are elevated in ALS. Based on a previous study, they are influenced by age and sex22. Moreover, this research showed elevated CSF NSE levels in ALS via the subgroup analyses of male and female patients and those aged < 70 and ≥ 70 years, which were performed to control the confounding effects of age and sex. To the best of our knowledge, this study, at least among those written in English, first showed elevated CSF NSE levels in ALS. Another main finding is that CSF NSE levels are higher in ALS than in CSM; therefore, they are useful in distinguishing ALS from CSM. In relation to the finding, the CSF NSE levels were not elevated in CSM. In the literature, the CSF NSE levels in CSM are controversial. That is, a previous report showed high CSF NSE levels in CSM23. Meanwhile, another revealed normal levels24. Taken together, CSF NSE levels in CSM may not be as elevated as those in ALS. Hence, they can be used to distinguish ALS from CSM.
There are several explanations why the ALS group had higher CSF NSE levels than not only the control but also CSM groups, even though NSE is generally a non-specific marker of neural damage13. First, the difference in CSF NSE levels may reflect different degrees of neural damage. A widespread and aggressive neural damage in ALS can result in significantly elevated CSF NSE levels. However, a limited and non-aggressive neural damage in CSM may not. Second, differences in CSF NSE levels can reflect varying affected areas. For example, the grey matter, which has high NSE levels, could be involved in ALS but not in CSM25. Third, differences in CSF NSE levels may reflect varying pathologic processes. That is, NSE may not be a non-specific marker of neural damage. However, it may play a role in several pathologic processes, and the mechanism might occur in ALS. In fact, NSE has been involved in pathologic processes such as neuroinflammation, particularly in the expression of pro-inflammatory cytokines and the proliferation of inflammatory glial cells25,26. In addition, the importance of neuroinflammation in ALS has been reported27. However, specific pathologic processes leading to elevated CSF NSE levels in ALS is not addressed.
Distinguishing ALS from CSM is challenging, particularly when patients with ALS present with cervical cord compression on MRI and they do not experience bulbar symptoms and do not fulfill the criteria on ALS. In this study, patients with ALS and such features had significantly higher CSF NSE levels than those with ALS without such features, or the CSF NSE levels of the former group was as high as those of the latter group. In detail, patients with ALS who do not fulfil the criteria on definite, probable, or PLS ALS had significantly higher CSF NSE levels than those who fulfilled the criteria. The CSF NSE levels of patients with ALS with cervical cord compression was as high as those of patients with ALS without compression. Moreover, the CSF NSE levels of patients with ALS without bulbar symptoms was as high as those of patients with ALS with the symptoms. Consequently, the diagnostic values of CSF NSE levels in distinguishing ALS with such features from CSM were higher or as high as those of CSF NSE levels in distinguishing whole ALS from CSM. In cases in which patients with ALS are challenging to distinguish from those with CSM, CSF NSE can be used. Hence, it may be an effective biomarker.
The reason why patients with ALS who do not fulfil the criteria had significantly higher CSF NSE levels than those who fulfilled the criteria is uncertain. However, it could be explained by a hypothesis that CSF NSE levels might decrease with disease progression at a certain stage because it could be accompanied by a decreased number of motor neurons, which might be the source of CSF NSE. Notably, the hypothesis could also explain our findings that patients with mild ALS had higher CSF NSE levels than other patients.
Our study had several limitations. First, it had a small sample size and was retrospective in nature. Hence, further large prospective studies should be conducted. Second, the control group only comprised unhealthy patients who underwent lumbar puncture, which is an invasive test, because of suspected neurological disorders that were ruled out after extensive investigations.
CSF NSE levels are elevated in ALS. Further, they can effectively distinguish ALS from CSM and prevent the misdiagnosis of CSM in patients with ALS. Thus, unnecessary surgery and subsequent rapid deterioration may be prevented. Notably, numerous physicians including those in general medical institutions can benefit from the use of this biomarker in daily clinical practice because NSE is a common tumor marker for diseases including small lung cancer and can be measured in general medical institutions. In addition, because elevated CSF NSE levels in ALS may reflect a specific pathologic process, this finding could provide new perspectives regarding the understanding of ALS pathogenesis and could facilitate the development of appropriate treatments.