Conference Coverage

Plasma exchange in natalizumab-related PML shows no benefit, possible harm



The largest-ever study of natalizumab-associated progressive multifocal leukoencephalopathy (PML) in patients with multiple sclerosis (MS) demonstrated that plasma exchange is without beneficial effect on clinical outcomes and may well be harmful, Christopher McGuigan, MD, and colleagues reported at the annual congress of the European Committee for Treatment and Research in Multiple Sclerosis.

Christopher McGuigan, MD, consultant neurologist at St. Vincent’s University Hospital and clinical professor at University College Dublin. Bruce Jancin/MDedge News

Dr. Christopher McGuigan

“This is a very important issue for us as neurologists. I think this data suggests that plasma exchange does not seem to confer an advantage to our patients when it comes to mortality and – albeit with caveats – it actually seems to be associated with worsening of the disability level in survivors,” according to Dr. McGuigan, consultant neurologist at St. Vincent’s University Hospital in Dublin and clinical professor at University College Dublin.

Natalizumab is an effective treatment for patients with highly active MS, but its use is limited by the feared complication of PML. Plasma exchange to clear the drug from the patient’s system is a popular and biologically plausible therapy, but its impact on clinical outcomes had not been formally studied until now.

Examining a database of confirmed PML cases

Dr. McGuigan presented a retrospective study of 723 patients with confirmed PML included in the natalizumab pharmacovigilance database run by Biogen, which markets the drug. A total of 85% (616 patients) underwent plasma exchange. The study’s primary outcome was the survival rate 2 years after PML diagnosis in patients who received plasma exchange, compared with those who did not.

Since the viral load of John Cunningham (JC) virus is known to influence outcome in PML, Dr. McGuigan and coinvestigators stratified the primary outcome based on tertiles of baseline viral copy number (VCN). The key study finding was that, across the range of viral loads, plasma exchange was consistently associated with a numerically worse 2-year survival rate, although the difference did not reach statistical significance. Among patients with less than a log5 baseline JC virus VCN, 2-year survival was 88.2% in the plasma exchange group, compared with 89.3% in the patients who did not undergo plasma exchange. For patients with a JC virus VCN greater than log5 but not more than log7, the survival rate was 89.3% without plasma exchange versus 73.8% with the intervention. In the group with greater than a log7 VCN, the 2-year survival was 78.9% without plasma exchange and 68.2% with it.

Statistically significant covariates

In a multivariate Cox proportional hazards analysis, plasma exchange was associated with a statistically nonsignificant 44% increased risk of mortality at 2 years post PML diagnosis. However, several other covariates emerged as statistically significant independent predictors of 2-year mortality. These included age greater than 50 years at diagnosis of PML, with an associated hazard ratio of 1.56, compared with younger patients; male gender (HR, 1.48); a JC virus VCN greater than log 7, compared with log 5 or less (HR, 2.86); a VCN greater than log5 but not more than log7, compared with a VCN of log5 or less (HR, 2.11); and widespread as opposed to localized MRI brain lesions of PML (HR, 1.61). In contrast, an asymptomatic presentation of PML was protective, with an HR of 0.38, compared with patients with a symptomatic presentation.


Next Article: