Background: People with multiple sclerosis (pwMS) treated with certain disease-modifying therapies (DMTs) have attenuated IgG response following COVID-19 vaccination; however, the clinical consequences remain unclear. Objective: To report COVID-19 rates in pwMS according to vaccine serology. Methods: PwMS with available (1) serology 2–12 weeks following COVID-19 vaccine 2 and/or vaccine 3 and (2) clinical data on COVID-19 infection/hospitalisation were included. Logistic regression was performed to examine whether seroconversion following vaccination predicted risk of subsequent COVID-19 infection after adjusting for potential confounders. Rates of severe COVID-19 (requiring hospitalisation) were also calculated. Results: A total of 647 pwMS were included (mean age 48 years, 500 (77%) female, median Expanded Disability Status Scale (EDSS) 3.5% and 524 (81%) exposed to DMT at the time of vaccine 1). Overall, 472 out of 588 (73%) were seropositive after vaccines 1 and 2 and 222 out of 305 (73%) after vaccine 3. Seronegative status after vaccine 2 was associated with significantly higher odds of subsequent COVID-19 infection (odds ratio (OR): 2.35, 95% confidence interval (CI): 1.34–4.12, p = 0.0029), whereas seronegative status after vaccine 3 was not (OR: 1.05, 95% CI: 0.57–1.91). Five people (0.8%) experienced severe COVID-19, all of whom were seronegative after most recent vaccination. Conclusion: Attenuated humoral response to initial COVID-19 vaccination predicts increased risk of COVID-19 in pwMS, but overall low rates of severe COVID-19 were seen.