Posters

Background: Personalised dosing of anti-CD20 disease modifying therapies for neurological conditions, including multiple sclerosis (MS), offers benefits but requires frequent and costly CD19 B-cell monitoring, using in-person whole-blood sampling. Remote dried blood spot (DBS) sampling could widen access to personalised dosing.

Aims: To evaluate the acceptability, feasibility, and costs of remote DBS sampling, compared to conventional blood sampling. Methods: 1) Capitainer-B50 (DBS) and venous blood samples were collected from people with MS to compare CD19 estimates. 2) Remote DBS packages were sent to MS patients. Returned samples were analysed for suitability, and questionnaires assessed for satisfaction.

Results: 3.4% of conventional lymphocyte-subset tests were unprocessed due to laboratory errors or delayed arrivals. In clinic, 72% of samples were successfully collected using one device. 11% required multiple devices. 17% were unsuccessful. Of 16 remote samples returned, 12 were sufficient for analysis. 53.8% of respondents preferred remote DBS sampling compared to in-person blood sampling. Of 34 samples collected in clinic, CD19 counts strongly correlated with FACS (R2=0.91), with 100% sensitivity and 81.3% specificity at <10 B-cell/uL. The mean turnaround time from posting remote samples to receiving results was 21.7 days, demonstrating postal feasibility. Costs of DBS sampling to support personalised dosing of ocrelizumab, based on n=50 and decreased annual infusion rate from 2 to 1.5, would result in £7,200 saving per patient.

Conclusions: Remote DBS sampling is accepted by patients, feasible, and provides comparable results to venous sampling. Its wider implementation has potential to improve patient convenience and reduce healthcare costs.

Spasticity affects a high proportion of people with multiple sclerosis (pwMS), drives pain and mobility loss, and commonly requires symptomatic prescribing. However, long term comparative effectiveness evidence is limited due to the costly and difficult nature of maintaining a traditional randomised controlled trial (RCT). We used the United Kingdom MS Register (UKMSR), a UK wide longitudinal registry with ~50,000 pwMS and >1 million patient-reported outcome (PRO) measures collected over 14 years, to emulate a pseudo-trial comparing Baclofen and Sativex. We derived a PRO spasticity scale (MSIS-Spasticity) from MSIS-29 items on stiffness and limb spasms, demonstrating good internal consistency and test-retest reliability. In participants reporting spasticity, the index date was first initiation of treatment. Baseline was the closest PRO within 180 days of initiation, with ≥3 follow-up assessments over 3.5 years. Treated participants were propensity matched 1:2 to comparators on baseline PROs, age, sex, disease duration and phenotype. Events were the first minimally clinically important change from baseline (direction-specific), analysed with time-dependent Cox models; linear mixed-effects models assessed longitudinal trajectories. In the emulated trial, Sativex was associated with reduced risk of clinically meaningful spasticity worsening (HR 0.41, 95%CI 0.18-0.92;p=0.032) and, over time, improvement in MSIS-Spasticity(β=-0.11;p<0.001) and MSWS-12 walking (β=-0.77;p=0.003). Baclofen showed no evidence of spasticity benefit and did not improve walking. By capturing routine UKMSR participation with broad eligibility, the pseudo-trial improved generalisability and provides robust comparative effectiveness evidence for clinical practice. Registry based pseudo-trials provide scalable, externally valid evidence for neurological treatments, complementing RCTs by informing and guiding clinical care.