A novel estrogen receptor 1: sphingomyelin phosphodiesterase acid-like 3B pathway mediates rituximab response in myositis patients, 2022, Parkes et al

[SNT Gatchaman]

A novel estrogen receptor 1: sphingomyelin phosphodiesterase acid-like 3B pathway mediates rituximab response in myositis patients
Parkes, Joanna E; Boehler, Jessica F; Li, Ning; Kendra, Ryan M; O’Hanlon, Terrance P; Hoffman, Eric P; Peterson, Jennifer M; Miller, Frederick W; Rider, Lisa G; Nagaraju, Kanneboyina

Objectives: The B-cell depleting biologic, rituximab, is used to treat refractory autoimmune myositis. However, the beneficial effects of rituximab appear to outweigh the known contribution of B cells in myositis. We aimed to elucidate how myositis patients respond differently to rituximab and possible alternative mechanisms of action.

Methods: Here we have: (i) comprehensively investigated concurrent mRNA and microRNA expression in muscle biopsies taken at baseline and 16 weeks post treatment in 10 patients who were part of the rituximab in myositis (RIM) trial; and (ii) investigated the beneficial effect of rituximab on myositis muscle cells.

Results: Our analyses identified an increased number of changes in gene expression in biopsies from patients who had a clinical response to rituximab (n=5) compared with non-responders (n=5). The two groups had completely different changes in microRNA and mRNA expression following rituximab therapy, with the exception of one mRNA, BHMT2. Networks of mRNA and microRNA with opposite direction of expression changes highlighted ESR1 as upregulated in responders. We confirmed ESR1 upregulation upon rituximab treatment of immortalized myotubes and primary human dermatomyositis muscle cells in vitro, demonstrating a direct effect of rituximab on muscle cells. Notably, despite showing a response to rituximab, human dermatomyositis primary muscle cells did not express the rituximab target, CD20. However, these cells expressed a possible alternative target of rituximab, sphingomyelinase-like phosphodiesterase 3 b (SMPDL3B).

Conclusion: In addition to B-cell depletion, rituximab may be beneficial in myositis due to increased ESR1 signalling mediated by rituximab binding to SMPDL3B on skeletal muscle cells.

Link | PDF (Rheumatology)

 

[SNT Gatchaman]

Possibly relates to SMPDL3B findings presented in IIMEC15 - Alain Moreau.

Also does this raise the question of whether the observed (though not borne-out) rituximab findings in ME could have been off-target effects in a sub-group (rather than B cell depletion)?

 

[SNT Gatchaman]

the beneficial effects of rituximab appear to outweigh the relative contribution of B cells to myositis. In the RIM trial, beneficial effects of the drug were delayed beyond time to B-cell depletion with B-cell depletion by week 4 but average time to reach definition of improvement at 20.1weeks. We hypothesized that an additional mechanism in skeletal muscle may be influencing response to rituximab in myositis patients.

At baseline, responders had higher expression of myeloid type I IFN signature genes, whereas non-responders had higher expression of classic non-myeloid IFN signature genes. Following rituximab treatment, responders had greater downregulation of both myeloid and non-myeloid IFN signature genes compared with non-responders.

Muscle biopsies were taken at baseline and at 16 weeks after treatment

In responders, the majority of significantly differentially expressed mRNAs were downregulated after treatment with rituximab […]. Whereas in nonresponders there was a more even split between up- and downregulated mRNAs after rituximab treatment. Furthermore […] an almost completely different set of mRNAs was found to be differentially expressed in responders compared with non-responders after rituximab treatment, with the exception of SLC6A2 upregulated in responders, downregulated in non-responders

Together, these data show that the response to rituximab treatment at a transcript level is very different in responders and non-responders.

findings suggest a different mechanism of action for rituximab in skeletal muscle cells compared with B cells.

CD20 […] could not be detected in healthy or dermatomyositis primary myoblasts or myotubes

A literature search identified SMPDL3B as an alternative target of rituximab in other cell types. We confirmed expression of SMPDL3B transcripts and protein in healthy and dermatomyositis primary human myoblasts by RTqPCR and western blotting. Immunofluorescence staining showed some co-localization of SMPDL3B and rituximab binding in primary human myoblasts

In our gene expression data, there was no significant change in SMPDL3B expression in responders, but a dramatic loss of SMPDL3B in non-responders, although this only reached marginal significance (–154.4 fold change, P = 0.0128). Notably, non-responders had lower SMPDL3B expression at baseline, although this did not reach significance, possibly due to the small sample sizes (–35.6 fold change, P = 0.055741)

 

[Jonathan Edwards]

These people seem to be completely ignorant of the immunological rationale for using rituximab. It doesn't surprise me but it is a bit depressing that it is still that bad, or worse.

It has nothing to do with a role for B cells themselves in muscle. It has to do with reducing antibodies!

 

[SNT Gatchaman]

And their reported average time to reach definition of improvement (20 weeks) sounds close to the 6 month peak with Rituximab you described in the saline thread.

The initial phase II study showed no significant effect at the primary endpoint. The reason for going to phase III was that the six month endpoint data would have been significant if chosen as primary endpoint. I had pointed out in PLOS One comments that six months should have been chosen for primary endpoint because that is when the effect of rituximab is maximal in autoimmune diseases. That meant that it was worth checking that the six month data are not a fluke. As it turned out they must have been a fluke. My confidence in getting a positive result went down a lot when looking at the detailed kinetics because there was no interpretable pattern in the blinded data. For diseases where rituximab works the kinetics are elegantly consistent, even if with some variation in parameters.

 

[Amw66]

I don't know if it would be a factor worth considering but did gender make a difference to responder v non responder ?