Abstract
Introduction/Aims
There is limited information on the potential effects of repeated intrathecal antisense oligonucleotide drug delivery on cerebrospinal fluid (CSF) biochemical and blood cell profiles. This study aimed to examine longitudinal changes in the biochemical components (glucose, protein) and blood cell counts in the CSF of spinal muscular atrophy (SMA) patients treated with intrathecal nusinersen.
Methods
We collected and analyzed clinical and CSF parameters (cell count, protein, glucose, culture) of 50 individuals with SMA during nusinersen treatment (22 type 1, 17 type 2, and 11 type 3).
Results
The median protein concentration at baseline and during treatment was within the normal range but rose during treatment and was significantly above baseline at the time of the ninth intrathecal injection (p = 0.02, two‐tailed Wilcoxon matched‐pairs test, and p = 0.0015, Friedman test for repeated measures). Further analysis showed that the increase in CSF protein concentration was evident for SMA types 2 and 3 patients, but not for type 1. This observation was also demonstrated by a significant correlation between the SMN2 gene copy number and the increase in CSF protein concentration (Spearman rank correlation test).
Discussion
Our results demonstrate that a delayed increase in CSF protein concentration is expected during nusinersen treatment for SMA types 2 and 3. This might reflect the medication's effect and a possible therapeutic biochemical marker.
Keywords: biomarker, cerebrospinal fluid, nusinersen, pediatric, spinal muscular atrophy
1. INTRODUCTION
The development of nusinersen, an antisense oligonucleotide that modifies pre‐mRNA SMN2 splicing, has changed the course and prognosis of spinal muscular atrophy (SMA).1, 2, 3, 4, 5 Pre‐mRNA SMN2 modification induces better expression of the functional SMN protein, which is necessary for the survival of anterior horn cells, playing an essential role in RNA processing and metabolism.6, 7
The intrathecal administration of nusinersen, requiring the removal of cerebrospinal fluid (CSF) before injecting the medication, has led to the emerging field of CSF biomarker studies in SMA.8, 9 Ever since the US Food and Drug Administration (FDA) approved nusinersen in 2016, numerous studies attempting to identify biomarkers of clinical significance have been published.10, 11, 12, 13, 14, 15, 16 The CSF protein composition is unique: while most plasma proteins are excluded by the blood‐CSF barrier, the choroid plexus and the brain parenchyma comprise an exclusive source for certain proteins, such as transthyretin, tau, NSE, S‐100B, cystatin, and more. 17
The influence and effects of repeated intrathecal injections of a synthetic oligonucleotide, such as nusinersen, on blood–brain‐barrier function and CSF content, are not well established. Studies by Wurster et al. showed an increase in the total CSF protein in the spinal fluid of adults diagnosed with SMA types 2 and 3.15, 18
This study aimed to evaluate routine biochemical components (glucose, protein) and cell counts in the CSF of SMA patients treated with nusinersen. Our hypothesis is that repeated intrathecal administration of nusinersen might cause detectable changes within basic CSF parameters in relation to protein synthesis pathways induced by the medication or, alternatively, by inflammation/blood–brain‐barrier breakdown caused by the repeated punctures.
2. METHODS
2.1. Patients and samples
The medical records of genetically confirmed 5q‐associated SMA patients treated with nusinersen in the Dana‐Dwek Children's Hospital and routinely followed‐up at the SMA multidisciplinary clinic were reviewed. Clinical and routine laboratory spinal fluid data, as well as specific functional motor test scores, were collected at baseline and after approximately 1 y of treatment. The functional motor tests were performed by the multidisciplinary SMA clinic physical therapists (done routinely every 4 mo) and included the Children's Hospital of Philadelphia Infant Test of Neuromuscular disorders (CHOP INTEND) and The Hammersmith Infant Neurological Examination (HINE) for the SMA type 1 patients, and the Hammersmith Functional Rating Scale Expanded (HFMSE) and the Revised Upper Limb Module (RULM) for the SMA types 2 and 3 patients.
The intervals between CSF sampling followed a globally recognized treatment protocol for loading and maintenance intrathecal administration of nusinersen.1, 19
CSF samples were collected by routine inter laminar lumbar puncture, or, if necessary, by a computerized tomographic (CT)‐guided transforaminal approach.
2.2. Sample exclusion
To assure the quality of the results, we excluded samples with more than 500 red blood cells ([RBCs]/μL) or samples with missing RBC information.
2.3. Statistical analyses
Statistical analyses were performed by GraphPad prism version 8.1.2 (San Diego, CA) to recognize exceptions compared to known norms and individual changes in CSF protein, glucose, and cell counts concentrations during treatment. Considering the non‐normal distribution of the measured variables (Kolmogorov–Smirnov test), non‐parametric tests (paired Wilcoxon matched‐paired signed Rank test, Friedman test, and Spearman rank test) were applied for analytic statistics with a two‐sided p value ≤0.05 as the threshold for statistical significance.
The study was approved by the Ethics Committee of the Tel Aviv Sourasky Medical Center, which waived informed consent for this retrospective analysis.
3. RESULTS
The cohort included 50 SMA patients, of whom 22 (44%) were diagnosed with SMA type 1, 17 (34%) with SMA type 2, and 11 (22%) with SMA type 3. Their demographic and clinical characteristics are given in Table 1. The median number of intrathecal injections per patient was nine (Table 1). Considering all standing quality criteria samples, a total of 402 CSF samples comprised the basis for the descriptive/analytical statistics. Table 2 provides the number of patients included in each subgroup analysis.
TABLE 1.
Summary of the clinical characteristics of the study group
| Total cohort | SMA type 1 | SMA type 2 | SMA type 3 | |
|---|---|---|---|---|
| (n = 50) | (n = 22) | (n = 17) | (n = 11) | |
| No. of intrathecal injections, median (range) | 9 (2–13) | 9 (5–13) | 9 (2–11) | 9 (7–10) |
| Gender (male, %) | 21 (42) | 8 (36) | 9 (53) | 4 (36) |
| Age of symptoms onset, y median (range) | 0.5 (0–2.2) | 0.08 (0–0.5) | 0.7 (0.3–1.3) | 1.5 (0.8–2.2) |
| Age at treatment initiation, y median (range) | 7 (0.08–21) | 1.7 (0.08–20.8) | 11.5 (1–19.3) | 10.8 (3.8–17.3) |
| SMN 2 copies, n (range) | 3 (2–4), n = 33 | 2 (2–3), n = 13 | 3 (2–4), n = 13 | 4 (2–4), n = 7 |
| Gastrostomy tube feeding, n (%) | 10 (20) | 9 (41) | 1 (6) | 0 (0) |
| Tracheostomy, n (%) | 9 (18) | 9 (41) | 0 (0) | 0 (0) |
| CT guided intrathecal drug administration, n (%) | 7 (14) | 0 (0) | 7 (41) | 0 (0) |
| CHOP S1, median (range) | – | 17 (1–45), n = 21 | – | – |
| CHOP S5‐S7, median (range) | – | 31 (1–54), n = 20 | – | – |
| HINE S1, median (range) | – | 1 (0–4), n = 19 | – | – |
| HINE S5‐S7, median (range) | – | 2 (0–12), n = 20 | – | – |
| HFMSE S1, median (range) | – | – | 5 (0–21), n = 15 | 49.5 (12–55), n = 10 |
| HFMSE S5‐S7, median (range) | – | – | 7 (1–21), n = 17 | 55 (19–61), n = 11 |
| RULM S1, median (range) | – | – | 18 (3–22), n = 7 | 35 (25–37), n = 5 |
| RULM S5–S7, median (range) | – | – | 17 (2–24), n = 17 | 35 (29–37), n = 11 |
Abbreviations: S1–S7, from baseline to seventh injection.
TABLE 2.
CSF protein concentration (mg/dL) in SMA patients over time
| CSF sample | Total SMA (n = 50) | SMA type 1 (n = 22) | SMA type 2 (n = 17) | SMA type 3 (n = 11) | p value a |
|---|---|---|---|---|---|
| S1 | 15 (14–21.5) | 18 (14–28.5) | 16 (12–17.5) | 16.5 (14–20.25) | 0.17 (n = 45) |
| S2 | 18 (15–25) | 18 (15–31) | 18 (13.75–22) | 17 (15.25–24.75) | 0.83 (n = 43) |
| S3 | 17 (14–22.5) | 18 (13.5–24) | 15 (12.5–19) | 17 (15–22) | 0.52 (n = 45) |
| S4 | 17 (14–21) | 17 (12.5–23.5) | 16 (13.5–20) | 18.5 (15.75–23.25) | 0.33 (n = 44) |
| S5 | 17 (15–23.5) | 17 (13.5–22.75) | 17 (14.75–21.5) | 18 (15–27.75) | 0.71 (n = 46) |
| S6 | 20 (16.25–23.75) | 17 (14–21) | 20.5 (18.75–33.5) | 22 (19–28) | 0.06 (n = 44) |
| S7 | 19.5 (16–24.75) | 17 (13.5–20) | 21 (16.25–25) | 24 (21–28) | 0.009 (n = 40) |
| S8 | 20 (18–25) | 18 (16–21.5) | 23 (18.5–23) | 22.5 (19.25–28.75) | 0.04 (n = 35) |
| S9 | 23 (18–26) | 19 (17–22) | 25 (23.5–31.5) | 23.5 (18.75–32) | 0.02 (n = 26) |
| p‐value b | 0.02 (n = 25) | 0.7 (n = 11) | 0.008 (n = 8) | 0.03 (n = 6) |
Note: S1–S9 = from baseline to ninth injection.
Note: Median (interquartile range), mg/dL.
SMA subgroups (types) comparison by Kruskal‐Wallis test.
Two‐tailed Wilcoxon test (S1 vs. S9).
The median WBC count at baseline (i.e., before treatment initiation) and during treatment was 0 cells/mm2. Significant pleocytosis of more than 5 cells/mm2, was present in only three samples of different patients, all of whom had glucose and protein levels within the normal range and sterile cultures. The median protein concentration at baseline and during treatment was within the normal range, but increased during treatment, rising significantly compared to baseline by the ninth injection (Table 2, Figure 1, and Supplementary Figure 1). The Friedman test for repeated measures also demonstrated a significant increase in the CSF protein content (p = 0.0015).
FIGURE 1.
Changes in median CSF protein concentration at baseline and during nusinersen treatment. Nine time points are shown. The graph illustrates the trend of protein increase among patients with SMA types 2 and 3. At the same time, the protein concentration among patients with SMA type 1 remained stable. CSF protein concentration normal reference range is15–45 mg/dL
Kruskal‐Wallis test at baseline (S1) excluded any significant differences of protein concentrations between the SMA types before treatment initiation (p = 0.17). With time, statistical differences between the subtypes appeared starting from S7 (Table 2). The number of SMN2 copies correlated significantly with the increase in protein (Spearman rank: r = 0.51, p = 0.03, CI 0.06–0.79, n = 19). There were no significant correlations between the degree of protein increment and the improvement in the functional motor test scores, that is, CHOP/HINE for SMA type 1 and HMFCS and RULM for SMA types 2 and 3. Finally, age as a potential modifier was not correlated with protein level at baseline in any of the SMA groups.
4. DISCUSSION
This study shows that over the first 22‐mo treatment period with nusinersen, a gradual increase in CSF protein concentration can be expected for SMA types 2 and 3 patients, but not for SMA type 1.
Analyses of repeated measures revealed a significant increase in the CSF protein concentration. This finding confirms and expands the results of Wurster et al., 15 who indicated an increase of total protein in the CSF after the fourth administration of nusinersen in patients with SMA types 2 and 3. Paired Wilcoxon test of protein concentration at each time point versus baseline for the three SMA types revealed a rise in CSF protein concentration for SMA types 2 and 3, starting from the fourth injection, which was enhanced with each repeated intrathecal injection. Comparable result for the SMA type 1 group was not demonstrated. Given an identical treatment protocol for the three SMA subtypes, this finding makes it highly unlikely that the repeated intrathecal injections were the cause of the protein increase.
Comparison of protein concentration among type 1, type 2, and type 3 groups at baseline and at subsequent time points demonstrated that the protein concentration was not significantly different at baseline (S1) or at S2 to S6. This observation excludes that disease subtypes are associated with distinctive protein concentration. However, at the time of the seventh injection (S7) and after that, the protein concentration was significantly different among SMA groups which strengthens the hypothesis that nusinersen injections induce the increase in protein concentration and that it has a cumulative effect. In that context, it was not surprising to find a significant correlation between SMN2 copy number and the observed changes in the CSF protein concentration.
The observation that nusinersen affects the protein concentration is consistent with the medication mechanism for altering the splicing of the SMN2 gene mRNA, thus increasing the concentration of the functional SMN protein.2, 4 It is also consistent with the fact that the SMN protein plays a vital role in multiple fundamental cellular pathways, including mRNA trafficking and local translation.6, 20 We sought a correlation between clinical improvement, as reflected in specific functional SMA motor tests, with the degree of protein increase; no significant correlations were found, which substantially weakens the clinical importance of the main study observation. The relatively wide time range (S5 to S7) over which the specific motor tests were performed might explain the non‐significant results. Individual factors, such as body weight, sex, age, and spinal hardware that might influence the CSF protein concentration 21 were eliminated by applying paired analytic statistics. Advancing age is associated with decreased CSF protein concentration, leading to accepted age‐specific related normal references. In the current study, despite “aging” of the patients, an increase in the CSF protein concentration was observed, which strengths the magnitude of this observation.
The study is limited by the relatively small cohort for carrying out subgroup analyses. Unfortunately, concurrent blood samples or additional CSF samples to investigate the CSF protein by a mass spectrometric analysis or a targeted approach for biomarker candidates in SMA such as neurofilaments, 12 gilial fibrillary acidic protein, 22 or chitotriosidase 1 23 were not available.
To conclude, a delayed increase in CSF protein levels can be expected under nusinersen treatment for SMA types 2 and 3, but not for SMA type 1. This study suggests that this is associated with the medication's effect and is a biochemical marker of the therapeutic intervention.
FUNDING INFORMATION
This research did not receive any specific grant from funding agencies in the public, commercial, or not‐for‐profit sectors.
List of Abbreviations
- CHOP INTEND
Children's Hospital of Philadelphia INfant TEst of Neuromuscular Disorders
- CSF
cerebrospinal fluid
- HFMSE
Hammersmith Functional Rating Scale Expanded
- HINE
Hammersmith Infant Neurological Examination
- RULM
Revised Upper Limb Module
- SMA
spinal muscular atrophy
- SMN
survival motor neuron
CONFLICTS OF INTEREST
None of the authors has any conflict of interest to disclose.
ETHIS STATEMENT
We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
Supporting information
Supplementary Figure 1 Longitudinal plots of individual patients over time, each patient is depicted in different color.
ACKNOWLEDGMENTS
The authors are grateful to all the participating patients and their families and the multidisciplinary team who cares for SMA patients treated in the Multidisciplinary SMA clinic at Tel Aviv Medical Center. The Authors thank Biogen for their support of the SMA registry. The authors thank Esther Eshkol for editorial assistance.
Orbach R, Sagi L, Sadot E, et al. Cerebrospinal fluid characteristics of patients treated with intrathecal nusinersen for spinal muscular atrophy. Muscle & Nerve. 2022;66(6):762‐766. doi: 10.1002/mus.27731
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1. Finkel RS, Chiriboga CA, Vajsar J, et al. Treatment of infantile‐onset spinal muscular atrophy with nusinersen: a phase 2, open‐label, dose‐escalation study. Lancet (London, England). 2016;388(10063):3017‐3026. doi: 10.1016/S0140-6736(16)31408-8 [DOI] [PubMed] [Google Scholar]
- 2. Mercuri E, Darras BT, Chiriboga CA, et al. Nusinersen versus sham control in later‐onset spinal muscular atrophy. N Engl J Med. 2018;378(7):625‐635. doi: 10.1056/NEJMoa1710504 [DOI] [PubMed] [Google Scholar]
- 3. Chiriboga CA. Nusinersen for the treatment of spinal muscular atrophy. Expert Rev Neurother. 2017;17(10):955‐962. doi: 10.1080/14737175.2017.1364159 [DOI] [PubMed] [Google Scholar]
- 4. Finkel RS, Mercuri E, Darras BT, et al. Nusinersen versus sham control in infantile‐onset spinal muscular atrophy. N Engl J Med. 2017;377(18):1723‐1732. doi: 10.1056/NEJMoa1702752 [DOI] [PubMed] [Google Scholar]
- 5. Chiriboga CA, Swoboda KJ, Darras BT, et al. Results from a phase 1 study of nusinersen (ISIS‐SMN(Rx)) in children with spinal muscular atrophy. Neurology. 2016;86(10):890‐897. doi: 10.1212/WNL.0000000000002445 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Li DK, Tisdale S, Lotti F, Pellizzoni L. SMN control of RNP assembly: from post‐transcriptional gene regulation to motor neuron disease. Semin Cell Dev Biol. 2014;32:22‐29. doi: 10.1016/j.semcdb.2014.04.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Liu Q, Dreyfuss G. A novel nuclear structure containing the survival of motor neurons protein. EMBO J. 1996;15(14):3555‐3565. [PMC free article] [PubMed] [Google Scholar]
- 8. Pino MG, Rich KA, Kolb SJ. Update on biomarkers in spinal muscular atrophy. Biomark Insights. 2021;16:11772719211035644. doi: 10.1177/11772719211035643 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Krause K, Wulf M, Sommer P, et al. CSF diagnostics: a potentially valuable tool in neurodegenerative and inflammatory disorders involving motor neurons: a review. Diagnostics (Basel, Switzerland). 2021;11(9):1522. doi: 10.3390/diagnostics11091522 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kariyawasam DST, D'Silva A, Lin C, Ryan MM, Farrar MA. Biomarkers and the development of a personalized medicine approach in spinal muscular atrophy. Front Neurol. 2019;10:898. doi: 10.3389/fneur.2019.00898 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Faravelli I, Meneri M, Saccomanno D, et al. Nusinersen treatment and cerebrospinal fluid neurofilaments: an explorative study on spinal muscular atrophy type 3 patients. J Cell Mol Med. 2020;24(5):3034‐3039. doi: 10.1111/jcmm.14939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Olsson B, Alberg L, Cullen NC, et al. NFL is a marker of treatment response in children with SMA treated with nusinersen. J Neurol. 2019;266(9):2129‐2136. doi: 10.1007/s00415-019-09389-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Rich KA, Fox A, Yalvac M, et al. Neurofilament levels in CSF and serum in an adult SMA cohort treated with nusinersen. J Neuromuscul Dis. 2022;9(1):111‐119. doi: 10.3233/JND-210735 [DOI] [PubMed] [Google Scholar]
- 14. Tiziano FD, Lomastro R, Abiusi E, et al. Longitudinal evaluation of SMN levels as biomarker for spinal muscular atrophy: results of a phase IIb double‐blind study of salbutamol. J Med Genet. 2019;56(5):293‐300. doi: 10.1136/jmedgenet-2018-105482 [DOI] [PubMed] [Google Scholar]
- 15. Wurster CD, Gunther R, Steinacker P, et al. Neurochemical markers in CSF of adolescent and adult SMA patients undergoing nusinersen treatment. Ther Adv Neurol Disord. 2019;12:1756286419846058. doi: 10.1177/1756286419846058 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Welby E, Rehborg RJ, Harmelink M, Ebert AD. Assessment of cerebral spinal fluid biomarkers and microRNA‐mediated disease mechanisms in spinal muscular atrophy patient samples. Hum Mol Genet. 2021;31:1830‐1843. doi: 10.1093/hmg/ddab365 [DOI] [PubMed] [Google Scholar]
- 17. Swaiman KF, Ashwal S, Ferriero DM, et al. Swaiman's Pediatric Neurology. Elsevier; 2018. [Google Scholar]
- 18. Wurster CD, Koch JC, Cordts I, et al. Routine cerebrospinal fluid (CSF) parameters in patients with spinal muscular atrophy (SMA) treated with nusinersen. Front Neurol. 2019;10:1179. doi: 10.3389/fneur.2019.01179 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Biogen . Spinraza (nusinersen injection) [package insert]. U.S. Food and Drug Administration website. https://www.accessdata.fda.gov/drugsatfda_docs/label/2020/209531s010lbl.pdf. Revised June 2020. Accessed October 2022.
- 20. Chaytow H, Huang Y‐T, Gillingwater TH, Faller KME. The role of survival motor neuron protein (SMN) in protein homeostasis. Cell Mol Life Sci. 2018;75(21):3877‐3894. doi: 10.1007/s00018-018-2849-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Seyfert S, Kunzmann V, Schwertfeger N, Koch HC, Faulstich A. Determinants of lumbar CSF protein concentration. J Neurol. 2002;249(8):1021‐1026. doi: 10.1007/s00415-002-0777-2 [DOI] [PubMed] [Google Scholar]
- 22. Freigang M, Steinacker P, Wurster CD, et al. Glial fibrillary acidic protein in cerebrospinal fluid of patients with spinal muscular atrophy. Ann Clin Transl Neurol. 2022;9(9):1437‐1448. doi: 10.1002/acn3.51645 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Freigang M, Steinacker P, Wurster CD, et al. Increased chitotriosidase 1 concentration following nusinersen treatment in spinal muscular atrophy. Orphanet J Rare Dis. 2021;16(1):330. doi: 10.1186/s13023-021-01961-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1 Longitudinal plots of individual patients over time, each patient is depicted in different color.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.