Reader response: Prospective validation of the PML risk biomarker l-selectin and influence of natalizumab extended intervals
| Autor: | Bastian Tugemann |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Oncology
medicine.medical_specialty viruses JC virus JCV Antibody medicine.disease_cause 03 medical and health sciences 0302 clinical medicine Natalizumab Internal medicine Medicine Humans Immunologic Factors 030212 general & internal medicine Prospective Studies L-Selectin biology business.industry Progressive multifocal leukoencephalopathy Leukoencephalopathy Progressive Multifocal virus diseases medicine.disease Risk stratification Cohort biology.protein Biomarker (medicine) L-selectin Neurology (clinical) business 030217 neurology & neurosurgery Biomarkers medicine.drug |
| Zdroj: | Neurology. 95(11) |
| ISSN: | 1526-632X |
| Popis: | With great interest I read the article by Schwab et al.1 on the risk of natalizumab-associated progressive multifocal leukoencephalopathy (PML) as a function of CD62L-expression. However, for the statistical analyses, unlike in earlier studies,2,3 the authors also took blood samples drawn at diagnosis into account. Given that CD62L often increases during the acute stage of PML,2 this may have adversely affected the marker's estimated sensitivity. Another bias was presumably introduced through the inclusion of patients who were JC virus (JCV)-seronegative, disregarding that the latter have a substantially lower risk,4 which is relevant because expression of CD62L in fact negatively correlates with JCV antibody titers.3 Therefore, it would have been more clinically meaningful to restrict the assessments to users of natalizumab who were JCV-seropositive, and, for those developing PML, to samples obtained at least 6 months before the outbreak—cf. Biogen's risk stratification by JCV-index.4,5 In addition, it might be informative to have a chart showing the temporal evolution of CD62L for all patients with PML in the cohort (how long before diagnosis each available value was measured—cf. again work by the drugmaker5). The 2 outliers—with minimum expression in the range of 50%–60% and thus well above the median1—seem to impair the predictive power of the test. Finally, Schwab et al. remarked that the methodology “currently” limited widespread usage of their assay,1 suggesting that this may change in the future. It would be helpful to have further detail on this. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|