We recruited 22 patients who had decided to start treatment with ocrelizumab (MS-O) and 14 age- and gender- matched, currently untreated MS patients (MS-nO). Demographic and clinical data are presented in Table 1, the study design is shown in Fig. 1. Despite similar disease duration, untreated patients had significantly better scores for EDSS and MSFC at baseline (Table 1). None of the patients treated with B-cell depletion (MS-O) had new T2-lesions at week 52, increased EDSS score or worsened MSFC-scores (Supplemental Fig. 1). Only one of the non-treated MS patients (MS-nO) had new T2-lesions at week 52 and only one patient from the MS-O and MS-nO group, respectively changed their diet habits during the study.
Differences in oral and stool microbiota composition of healthy controls and MS patients
First, baseline stool microbiota of both MS patient groups included in this study were compared to microbiota profiles from healthy controls (HCs) (Supplemental Table 1). Total abundance of bacterial phyla identified in stool revealed significant higher abundance of Bacteroidetes in MS patients compared to HCs (p = 0.0017) as well as significant lower abundance for Firmicutes (p = 0.0037) (Fig. 2A). These differences were mainly driven by the clinically more severely affected MS-O patient group [Figure 2A, Supplemental Table 2, Bacteroidetes: HC vs. MS-O (p = 0.0017), HC vs. MS-nO (p = 0.29), Firmicutes: HC vs. MS-O (p = 0.044), HC vs. MS-nO (p = 0.112)]. Next, we analysed the diversity and evenness of microbial composition (alpha diversity) in MS patients versus HCs We observed a generally lower alpha diversity in stool of MS patients that was again found to be more pronounced in MS-O patients (Fig. 2B, pShannon HC vs. MS-O = 0.017, pShannon HC vs. MS-nO = 0.109). Regarding overall bacterial composition and diversity, the MS patient groups showed a general shift indicating compositional differences between MS patients and HCs (Fig. 2C)[pAdonis = 0.001 (pAnosim HC vs. MS-O = 0.036, pAdonis2 HC vs. MS-nO = = 0.002)). Significant differentially abundant taxa between MS and HCs are shown in Fig. 2D (p < 0.05). As already observed at phyla level, in MS patients, and in particular in the MS-O group, a general higher abundance of several Bacteroidetes members could be observed (esp. the genera Prevotella, Butyricimonas, Coprobacter, Parabacteroides and Alistipes), while on the other hand members of the phylum Firmicutes were frequently less abundant (e.g., members of the Clostridiales as well Romboutsia and Turicibacter). Compared to HCs, both patient groups revealed a significant higher abundance of Parasutterella (Fig. 2D).
Secondly, the oral microbiota profile of MS patients was compared to that of HCs at baseline. Total abundance of bacterial phyla in the oral cavity is shown in Fig. 2E and revealed significant higher levels of Proteobacteria in MS patients (p = 0.008), whereas Actinobacteria were found to be decreased (p = 0.045). In contrast to the findings from stool, these results were mainly driven by the less affected MS-nO patient group [Figure 2E, Supplemental Table 2, Proteobacteria: HC vs. MS-O (p = 0.122), HC vs. MS-nO (p = 0.002), Actinobacteria: HC vs. MS-O (p = 0.255), HC vs. MS-nO (p = 0.127)]. Alpha-diversity was slightly, but not significantly decreased in MS patients versus HCs (Fig. 2F). With respect to overall bacterial composition, significant differences between MS patients and HCs were observed(pAdonis = 0.002) (Fig. 2G), which could also be verified in each of the MS patient groups and their respective HCs (pAnosim HC vs. MS-O = 0.001, pAdonis HC vs. MS-nO = 0.02). Significant different abundant taxa between MS and HCs are shown in Fig. 2H (p < 0.05). MS patients generally had significant higher abundances of the genera Campylobacter, Haemophilus and Neisseria – all belonging to the Proteobacteria, though this was only partly evident when splitting up the MS patients in groups (Fig. 2H).
Oral and stool microbial changes after B-cell depletion
When comparing microbial changes in MS-O patients during treatment, results from fitted models displayed a general slight increase in alpha diversity between HCs and MS patients before (TP1) and during treatment (TP2, TP3, TP4) in stool (Fig. 3A) and swab samples (Fig. 3C), though none of these changes was found to be significant (Supplemental Table 3). However, species richness determined by Chao1 index was significantly increased in swab samples at TP4 (pChao1 = 0.025) (Supplemental Table 3). Regarding beta diversity, a non-significant shift between MS patients at baseline compared to MS patients 6 months (TP3) and 12 months (TP4) after B-cell depletion in both stool and swab samples was observed (Fig. 3B,3D and 3E; Supplemental Table 3). Comparing the compositional differences in the oral cavity to HCs at TP1, the initially observed highly significant difference (pAnosim = 0.001) became less at TP4 (pAnosim = 0.023).
Next, we determined fine-scale abundance differences over time in treated patients (MS-O). We concentrated on changes persisting over an extended 6-months period (TP3-TP4, months 6–12) following B-cell depletion and found a significant decrease of members of the phylum Proteobacteria (particularly members of the genus Escherichia/Shigella) as well as varying genera from the Bacteroidetes (p < 0.05, e.g., Bacteroides, Parabacteroides and Prevotella; Fig. 3C) in stool samples. In contrast, numerous members of the Firmicutes were altered differentially (e.g., Faecalibacterium, Dialister and uncl. Ruminococcaceae), however were still less abundant compared to HCs. Within the oral cavity, several bacterial genera were consistently altered at 6- and 12-months following B-cell depletion (Fig. 3F). We found a significant decrease of members of the genera Haemophilus and Lautropia out of the Proteobacteria, Porphyromonas out of the Bacteroidetes, Veillonella and Granulicatella out of the Firmicutes as well as unclassified members from the phylum Candidatus Saccharibacteria. Different ASVs of the genera Streptococcus and Actinomyces were consistently increased following treatment.