, data indicate.
In a multicenter study that emulated pairwise trials, AHSCT was associated with a higher likelihood of disability improvement, compared with fingolimod (hazard ratio, 2.70).
Once the immune system is reconstituted, it can contribute to the healing process. AHSCT thus provides the possibility of improvement in MS. Disease-modifying therapies that require continued use, however, could inhibit that process.
“If you can stop the inflammation that’s driving this disease fairly early on, [patients] do have the capacity to repair,” said study author Mark Freedman, MD, a professor of neurology at the University of Ottawa. “And we start to see that especially in a treatment [like AHSCT] where you don’t maintain the hit on the immune system.”
The study was published online in JAMA Neurology.
Pairwise analyses
Single-arm studies and an open-label, randomized trial have suggested that AHSCT has efficacy. The regimen is associated with a 0.3%-2.0% risk for mortality, but this risk has declined with better patient selection and clinical experience, according to the researchers. Comparative studies of AHSCT and DMTs are needed, but they are difficult to carry out, which is why the current team chose a propensity-score matched case–control design, according to lead author Tomas Kalincik, MD, PhD, head of the University of Melbourne’s Clinical Outcomes Research Unit.
The researchers examined data from 4,915 patients with relapsing-remitting MS. Of this population, 3.4% received AHSCT, 52.0% received fingolimod, 30.3% received natalizumab, and 14.2% received ocrelizumab. The proportion of women in the treatment groups ranged from 65% to 70%.
The researchers used pairwise matching to simulate the randomized, controlled trials comparing AHSCT with fingolimod, natalizumab, and ocrelizumab. Patients were matched on the basis of sex, age, disability, relapse events at 12 and 24 months before baseline, time from first symptoms to baseline, the most effective previous DMT, and geographical region.
Compared with fingolimod, AHSCT was associated with fewer relapses (annualized relapse rate, 0.09 vs. 0.20; P < .001). This finding was confirmed by a reduced cumulative hazard of relapse (HR, 0.26).
Compared with natalizumab, AHSCT had only a modestly greater effect on annualized relapse rates (ARR, 0.08 vs. 0.10; P = .03), with a cumulative HR for relapses of 0.51.
There was no significant difference in the risk for relapse between treatment with AHSCT and with ocrelizumab (ARR, 0.09 vs. 0.06; P = .86), nor was the risk of cumulative relapses significantly different between these treatments.
Among patients included in the pairwise analyses who received AHSCT, 23.3% developed febrile neutropenia during mobilization, 11.3% developed serum sickness, and 8.8% were admitted to an intensive care unit. There were 82 serious adverse events among 58 patients after they were discharged post AHSCT treatment, including infections (59.8% of adverse events) that were primarily due to viral sources (41.5% of adverse events). There was one death (0.6%) due to veno-occlusive disease of the liver following busulfan exposure.
Some AHSCT protocols are stronger than others, and milder immune ablative measures are less likely to produce lasting effects. “We completely remove the old immune system and put in a brand new one,” said Dr. Freedman, referring to the practice at his center. “That’s a fairly horrendous procedure, and, not surprisingly, we’ve had a slight increase in the types of side effects, compared with other groups who don’t use that heavy-duty conditioning regimen, but we’ve had absolutely zero return of inflammatory events over 23 years that we’ve been doing this. Nobody’s had another attack, no one has even developed a single new MRI lesion.”
The so-called medium and light conditioning regimens are associated with a return of disease activity in about 25% of patients within 3 years, Dr. Freedman added. “You start to see new MRI lesions form and relapses occurring. It’s still better than any of the higher efficacy therapies, but it’s not stopped the disease.”
A key limitation of the study is that its efficacy analysis did not distinguish between different intensities of AHSCT regimens, according to the authors.