I have started-up another email exchange with a colleague about the ORATORIO (ocrelizumab in PPMS) study. I have asked him/her if we could have this debate in the open on the blog, but he/she has asked to remain anonymous.
‘You were a co-author on the ocrelizumab in PPMS NEJM manuscript, which didn’t really address the issue of age/inflammatory activity as markers for treatment response. Both the rituximab data in OLYMPUS and the Gadolinium data in the supplement of the NEJM article indicate that the population that benefits is the young patient with inflammatory activity. B cell depletion is not benign, especially in the older population, and the ocrelizumab manuscript infers that all PPMS patients should be treated with ocrelizumab. Are you really treating all of your PPMS patients with B cell depletion, and, if not, why didn’t you push for more discussion about this issue in the NEJM paper?’
We tried to restrict the trial population as much as possible to mirror that of the responder subgroup in OLYMPUS (rituximab in PPMS) study. Please note the following specific inclusion criteria for the trial:
- Age cut-off of 55 years of age
- Disease duration of symptoms of less than 15 years in patients with an EDSS score of more than 5.0 at screening or less than 10 years in patients with an EDSS score of 5.0 or less at screening
- Documented history or the presence at screening of an elevated IgG index or at least one IgG oligoclonal band detected in the cerebrospinal fluid
The presence of absence of Gd-enhancing lesions was not part of the inclusion criteria and hence should not be used to select patients for ocrelizumab treatment in the real-life situation. The detection of Gd-enhancing lesions also depends on the frequency of imaging. If you are going to select patients for treatment I suggest you use the the three inclusion criteria above as your guide. I predict that these will probably exclude 70% of a typical PPMS clinic population.
I personally don’t buy into MS being 2 or 3 diseases. MS is one disease and PPMS is simply more advanced MS; if patients are active they should be offered treatment regardless of their presenting clinical phenotype.
Regarding my own practice. We can’t use rituximab in the UK; the NHS won’t pay for it and are unlikely to pay for ocrelizumab either. NICE will assess the cost-effectiveness of ocrelizumab based on its price for relapsing-forms of MS and in PPMS ocrelizumab will be compared to best-supportive care. The latter means ocrelizumab is unlikely to pass the NICE cost-effectiveness threshold. I am hoping that Roche, who will be marketing ocrelizumab in Europe, approach NICE and the NHS to discuss differential pricing and to offer ocrelizumab at a cheaper price for patients with PPMS. Differential pricing is a ‘hot potatoe’ and I am not sure the NHS is ready for it; but I live in hope for my patients.
I am not sure we have enough data on B-cell depletion in PPMS to make a call on whether it is benign or non-benign. Hopefully, good quality safety data will emerge from post-marketing surveillance studies. I suspect as with all immunosuppressive therapies it won’t be benign. I would recommend telling all your patients about infusion reactions, the herpetic infection risk, the possible malignancy risk and the likelihood that in time they may develop hypogammaglobulinaemia and need Ig-replacement therapy. Please note we have a large amount of clinical experience in children with agammaglobulineamia and these kids do well long-term, as long as we keep the Ig levels normal.
I am not sure you are correct in suggesting that we are recommending ocrelizumab for all-comers. The trial population defines the group we are advocating its use in. I am surprised the FDA did not include the CSF findings in the label. If I was a regulator I would state that CSF-ve PPMS should not be treated with ocrelizumab.
Regarding discussion in the NEJM paper; the editors of the NEJM essentially cull all speculative discussion and limit the discussion to issues in the trial. NEJM editors are renowned for rewriting submitted papers to keep them consistent for style. The NEJM article is not the forum for the kind of discussion you want to have. I suggest we have this discussion on the blog.
‘Thanks for your prompt and thoughtful response. It was very helpful. When do you think we’ll be able to see the age cutoffs for the study? Given that the average age for the study was 45 (and for OLYMPUS it was much higher—about 50), there should be enough patients who entered the study between 50 and 55, to know whether the relatively small treatment effect in the whole group was made up primarily of those under 50. I would certainly like for my older PPMS patients to avoid the potential adverse effects of this drug if there is no clear benefit for this group.
My second response:
There are numerous post-hoc subgroup analyses that are ongoing. I will ask the trial team to include your question about age on the list.
Regarding SPMS; yes, I would expect B cell depletion to work for SPMS as well. I wouldn’t limit it to young, or early, SPMS either. Based on our length-dependent axonopathy hypothesis I would even expect patients in wheelchairs to benefit, however, the benefit will be limited to arm and bulbar function. The latter is why we are lobbying Roche to do a trial of ocrelizumab in more advanced MS, with the primary outcome being upper limb function.