To the editor,
We are glad to address the critics of van der Meer and co-authors as response to our recent paper in PLoS One [1], where we presented a small randomized, double-blind, placebo-controlled study of the CD20-directed antibody Rituximab as intervention in chronic fatigue syndrome (CFS/ME) patients fulfilling the Fukuda criteria.
At one point we totally agree with the critics: Rituximab should not be used to treat CFS/ME outside of clinical studies at present. We also agree that more research is needed.
van der Meer and coauthors have problems believing our results in part based on what they see as atypical response kinetics for Rituximab and a lack of theoretical understanding of why the drug should show efficacy in CFS/ME. We do not support their statement that a response within the first days to weeks necessarily is the usual pattern in other diseases where Rituximab intervention is used. Autoimmune cytopenias and some autoimmune diseases such as acute demyelinating polyneuropathy [2] may respond early, while in other diseases the responses usually start to occur after months. In accordance, the endpoints for clinical studies, or clinical benefit demonstrated in observational studies, are often defined at 6 - 12 months after treatment, with the clinical effects subsiding over the following months or year, such as in rheumatoid arthritis [3], lupus nephritis [4], and primary antiphospholipid antibody syndrome [5]. As an example, in a case study of pulmonary alveolar proteinosis, in which autoantibodies to GM-CSF have been demonstrated, the levels of autoantibodies declined the first three months after Rituximab treatment, while effect on alveolar-arterial gradient was seen at six months, and improvements in pulmonary function tests and CT scans were evident at nine months [6].
The full consequences of the action of Rituximab in other autoimmune diseases are only partly known, and also which of the known modes of action that are of greatest relative importance in individual patients [7]. CFS/ME is not a disease with evident widespread inflammatory lesions as in some established autoimmune conditions and the mode of action could thus be different from that seen in for example rheumatoid arthritis or lupus. Reduction of autoantibody levels (wash-out by reduced production following B-cell depletion) to a critical level before clinical responses become evident could therefore be one plausible mechanism explaining what we observed.
van der Meer and colleagues raise methodological concerns regarding the measurements of fatigue in our study. We used a visual analogue scale for self-reported symptoms every second week during 12 months follow-up. A similar scale was used in a previous study of intervention in CFS/ME [8]. We have not validated the symptom scales used further. However, as it was a randomized study, all patients (Rituximab and Placebo) had the same follow-up. The Fatigue-score was used as the main criterion for response assessment, and the GLM analysis for repeated measures showed a significant interaction time by intervention group, which we believe is the most important result in the study. The data in this analysis are a direct reflection of the self-reported symptom scores registered every second week, where the patients wrote down a number representative of the whole preceding two-week period, thus taking into account the day to day variations in symptoms typical for many CFS/ME patients. The critics question if a physician-rated fatigue score in a subjective complaint is relevant. The reason for including this in the protocol was to see if there was a good agreement between what the patients told us at the consultations and the figures they reported in their files. In general, the patient and doctor values were in good accordance (Figure 2, panels B and D).
van der Meer and colleagues argue that the clinical size of effect seems small. Figure 2 (panel A) shows the self-reported Fatigue-score for each patient, demonstrating the distribution of responders and non-responders at the different time intervals. In panel B, the mean values for Fatigue score in each group are shown and include the counts of the non-responders, also reflecting that the time frames for clinical responses vary among the patients, thus tending to reduce the numerical value at a given time. We do not agree that the clinical effects are small or clinically irrelevant, and as stated in the manuscript the patients assessed these to be important for their quality of life, generally affecting all CFS/ME related symptoms.
Some limitations of our study pointed at by the critics have already been addressed in the paper. The choice of the primary endpoint at three months (based on responses of the first two pilot patients given Rituximab) [9] and the lack of predetermined exact description of responses with respect to duration and extent of improvement, are caused by limited experience with the kinetics of responses after B-cell depletion in CFS/ME. After seeing the first two pilot patients with response on all CFS/ME symptoms, later followed by relapses, and seemingly as the result of the intervention, we made a decision to do a limited randomized study. A phase II study without a control group in a disease with subjective endpoints would probably have problems convincing both study readers and ourselves. Our study size was a balance between having a fair chance of detecting a difference between the intervention groups and what could be handled within a clinical university department, without any industrial or other external economical support for the clinical part of the study.
Concerning the blinding, randomization was done by the hospital pharmacy and the infusion bags packed in a manner that did not disclose their content for either the patients or the nurses administering the infusions. Any complaints during infusions were to be dealt with by the doctors on call and not the researchers. The observed side effects (Table 5) did not clearly indicate to patients or physicians which treatment had been given. We did not (as suggested by the critics) ask the patients which drug they thought they had been given. We believe the blinding for both patient and researchers was good.
van der Meer et al. comment on the high rate of other autoimmune diseases in the patients and their relatives in this series. This is not an original observation and has been noted also by others. One of the senior authors of our manuscript, prof. H. Nyland, is the single physician in Norway who has seen and systematically registered most CFS/ME patients and he confirms this finding his patients. We can agree that including a single case of carpal tunnel syndrome (CTS) as an autoimmune entity is questionable. Nevertheless, the CTS is greatly overrepresented in patients with known autoimmune disease, like diabetes type I, lupus, and rheumatoid arthritis, and is quite often a heralding condition before the systemic autoimmune disease is evident. Although multifactorial aetiology of CTS is a favoured view, a genetic and autoimmune component is probable in many patients. Concerning the question of a relation between a family history of autoimmune diseases and response, the material is of course too small to answer that. Table 1 shows that previous autoimmune disease in patients or first degree relatives is not higher in the Rituximab than in the placebo group.
Importantly, we have not demonstrated that CFS/ME is an autoimmune disease with disease-specific autoantibodies, but we have stated as a hypothesis that the observed clinical pattern of responses and relapses could be compatible with such a mechanism. Alternatively, influence on other aspects of B-cell function could be the key to understand the effects of B-cell depletion in CFS/ME, such as antigen-presentation to T-cells, influence on other immune cells such as dendritic cells, or changes in Th1 and Th2 balance.
Finally, van der Meer and colleagues raise major concern on the fact that we are performing an open-label phase II study, exploring Rituximab induction followed by Rituximab maintenance, instead of a new high-quality randomized study. They also suggest that we overestimate the effects and downplay the possible serious side effects. The critics may have fallen into the same ditch of prejudice as they indicate we have.
We admit the strong and repeated responses to new Rituximab infusion seen in our three pilot patients may have influenced us. We have later treated another lymphoma patient with preexisting CFS/ME who also experienced a clear response on all CFS/ME symptoms from four months after start of treatment (8 cytotoxic chemotherapy courses (CHOP) with addition of Rituximab), and she still has response of CFS/ME symptoms 2 ½ years later. Other physicians treating patients having both CFS/ME and lymphoma elsewhere in the world have informed us of similar responses to Rituximab. The total experience with Rituximab in our opinion warrants a study to optimize the use of the drug.
Based on the current study, there is now a Norwegian national initiative to do a randomized, blinded, multicenter Phase III study to verify or reject the conclusions of the PLoS One study. In the planning of this national study, results of our exploratory phase II study with Rituximab induction and maintenance will be important for the scheduling of the drug administration.
CFS/ME according to Fukuda or Canadian criteria is in many patients a very serious and debilitating disease. Current treatment is for many patients highly unsatisfactory and we believe it is justified to find new interventions and learn more of the pathophysiology, of which we at the time being know little. The side effects of Rituximab in this particular patient group of patients is of course presently largely unknown, although there is vast experience with the drug in B-cell lymphomas and established autoimmune diseases. We believe the severity of disease in many CFS/ME patients balances the known risks. That was also the opinion of most patients who were given the option to participate in the current study at a time when only three pilot patients had been treated with the drug.
1. Fluge O, Bruland O, Risa K, Storstein A, Kristoffersen EK, et al. (2011) Benefit from B-cell depletion using the monoclonal anti-CD20 antibody Rituximab in chronic fatigue syndrome. A double-blind and placebo-controlled study. PLoS One 6:10 e 261238
2. Motoyama R, Yamakawa K, Suzuki S, Kusunoki S, Tanaka M (2011) Rapid improvement by rituximab treatment in a case of demyelinating polyneuropathy with anti-myelin-associated glycoprotein antibody. Rinsho Shinkeigaku 51: 761-764.
3. Edwards JC, Szczepanski L, Szechinski J, Filipowicz-Sosnowska A, Emery P, et al. (2004) Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 350: 2572-2581.
4. Diaz-Lagares C, Croca S, Sangle S, Vital EM, Catapano F, et al. (2011) Efficacy of rituximab in 164 patients with biopsy-proven lupus nephritis: Pooled data from European cohorts. Autoimmun Rev. epub ahead of print
5. Elazary AS, Klahr PP, Hershko AY, Dranitzki Z, Rubinow A, et al. (2011) Rituximab induces resolution of recurrent diffuse alveolar haemorrhage in a patient with primary antiphospholipid antibody syndrome. Lupus. epub ahead of print
6. Borie R, Debray MP, Laine C, Aubier M, Crestani B (2009) Rituximab therapy in autoimmune pulmonary alveolar proteinosis. Eur Respir J 33: 1503-1506.
7. Kessel A, Rosner I, Toubi E (2008) Rituximab: beyond simple B cell depletion. Clin Rev Allergy Immunol 34: 74-79.
8. Blacker CV, Greenwood DT, Wesnes KA, Wilson R, Woodward C, et al. (2004) Effect of galantamine hydrobromide in chronic fatigue syndrome: a randomized controlled trial. JAMA 292: 1195-1204.
9. Fluge O, Mella O (2009) Clinical impact of B-cell depletion with the anti-CD20 antibody rituximab in chronic fatigue syndrome: a preliminary case series. BMC Neurol 9: 28.
Øystein Fluge and Olav Mella
Department of Oncology and Medical Physics
Haukeland University Hospital
Competing interests declared: Haukeland University Hospital has patents and pending patent applications on the issue of B-cell depletion therapy for chronic fatigue syndrome. Family members of WO2009083602 A1 are pending, as well as granted US 12/348024. The two authors ØF and OM are named as inventors in these applications.
