CD62L Test Expands Natalizumab PML Risk Pool in RRMS

Daniel M. Keller, PhD

May 05, 2014

PHILADELPHIA, Pennsylvania — A new study shows that combining 2 markers for risk for progressive multifocal leukoencephalopathy (PML) during long-term therapy for relapsing-remitting multiple sclerosis (RRMS) with natalizumab (Tysabri, Biogen Idec) may expand the population of patients at risk compared with using the traditional single parameter of positivity for JC virus (JCV) antibody alone.

"Patients that had high antibody titers had lower expression of CD62L. And actually this was only true for patients that had not been immune suppressed beforehand," Johanna Breuer, MSc, from the Department of Neurology at the University of Münster, Germany, told Medscape Medical News. Previous studies showed that JCV antibodies could be used as a predictive marker for PML only in patients who had not been immunosuppressed.

Currently, patients are stratified for PML risk according to previous immune suppression, the duration of natalizumab treatment, and the presence of serum anti-JCV antibodies at a certain level.

PML is an opportunistic viral disease causing inflammation of the white matter of the brain and is usually fatal. The immune system normally keeps JCV in check, but natalizumab, by blocking the migration of inflammatory immune cells across the blood-brain barrier, can allow JCV-induced PML.

In a poster presentation here at the American Academy of Neurology (AAN) 66th Annual Meeting, the researchers proposed a new approach to individual risk stratification based on 2 parameters — anti-JCV antibody titers expressed as an index value and CD62L (L-selectin) as a possible biomarker of PML risk during natalizumab treatment.

CD62L is a cell adhesion molecule on lymphocytes, including some T cells.

Interaction of CD62L and Anti-JCV Antibodies

The researchers assessed the correlation of the CD62L and anti-JCV parameters in a large German cohort of patients with RRMS receiving long-term treatment with natalizumab, which provided 949 peripheral blood mononuclear cell (PBMC) samples and 1921 serum samples.

CD62L was assessed on CD3+, CD4+, and CD8– T cells among the PBMCs. Anti-JCV index was measured by optical density and is a corollary to antibody titers.

Of the entire cohort, 41.23% had an anti-JCV index of 0.9 or greater, a cut-point previously proposed to avoid 98.04% of presumed PML cases. Anti-JCV antibodies did not correlate with the duration of natalizumab treatment in an analyzed patient cohort and were stable over time among JCV-positive patients.

"We could show for L-selectin or CD62L is also a risk and that patients that develop PML had very low levels of CD62L expression before developing the disease," Breuer said.

CD62L expression decreased with increasing cycles of natalizumab, and the slope of the decrease was twice as high for immunosuppressed patients as for nonimmunosuppressed ones, confirming the higher PML risk with a history of immune suppression.

CD62L expression also decreased with increasing anti-JCV index values, but only for nonimmunosuppressed patients. However, there was large scatter in the data for all numbers of cycles of natalizumab and all anti-JCV index values.

Two patients developed PML. One had previous immunosuppression and showed a rise in the number of CD62L-positive cells at the time of development of acute PML. The anti-JCV antibody index remained below the 0.9 threshold throughout, illustrating that this parameter has no utility in patients with a history of immune suppression.

The second patient was not immunosuppressed and had normal CD62L (L-selectin) cell numbers for some time, but these decreased 2.5 years before PML developed. In parallel with the drop in CD62L cells, the anti-JCV antibody index rose from about 2.75 (well above the 0.9 threshold) to 3.25.

Given the long lag time between the drop in CD62L cells and the emergence of PML, Breuer concluded that "L-selectin can only be used as a marker for predicting PML and not actually to diagnose it."

The researchers developed an algorithm of successive stratification incorporating both markers to predict PML risk among patients with RRMS who received 18 or more cycles of natalizumab:

All eligible patients → CD62L confirmed low on 2 tests (3.74% of all) → JCV+ (2.57% of all) → nonimmunosuppressed (2.10% of all) → JCV index ≥ 0.9 (1.71% of all).

To the 1.71%, they added the population of immunosuppressed patients (0.47%), for whom the JCV index would be meaningless and therefore not included in the previous group. So the total group at risk for PML is 1.71% + 0.47%, or 2.18% of all patients.

In summary, JCV index alone determined 43.37% of the study cohort to be at risk for PML. CD62L set the proportional risk at 3.74% of the cohort. Combining the 2 parameters reduced the at-risk percentage to 2.18%, as above.

Better Prediction

Giancarlo Comi, MD, professor in the Department of Neurology at the Vita-Salute San Raffaele University in Milan, Italy, commented to Medscape Medical News that the study provides an interesting message "because if we combine the predictivity of this new parameter with the titer of the anti-JC virus antibody, you may increase by about 6 times the predictivity of the risk to develop PML."

The combination of the 2 tests results in the possibility of detecting patients that are at about 2% risk if they never received immunosuppressive agents.

"This population usually has a risk of about 1 out of 300. So [prediction] is much better now," he said. "Of course, 98% of these patients will never develop a PML."

If we combine the predictivity of this new parameter with the titer of the anti-JC virus antibody, you may increase by about 6 times the predictivity of the risk to develop PML. Dr. Giancarlo Comi

Professor Comi said there is still a need for better predictive biomarkers for PML. As it is, "in some cases we are likely to stop natalizumab treatment in those patients, for example, that didn't respond to the other therapies." Stopping natalizumab therapy risks very strong reactivation, disability, and severity and progression of the disease, putting the 98% who would never develop PML at undue risk.

He said even the tests proposed by this research will not solve this conundrum because its predictive power for PML — especially who is not at risk — is inadequate.

Breuer said the CD62L+ T cell test is already being used on a research basis to monitor patients receiving natalizumab therapy on an ongoing basis because the values can change during therapy.

The study had no commercial support. Ms. Breuer has disclosed no relevant financial relationships. Professor Comi reported commercial relationships with Teva, Novartis, Genzyme, Merck Serono, Biogen, Bayer, Actelion, Almirall, and Serono Symposia International Foundation.

American Academy of Neurology (AAN) 66th Annual Meeting. Abstract P4.348. Presented April 30, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: