Aries et al. demonstrate in their study, the lack of response of
granulomatous disease in Wegener's granulomatosis to rituximab. From my
reading of the paper, this agent was given every fourth week. This is not
how this agent is usually prescribed, ie usually given as weekly pulses
for four weeks. Despite this, Bcell numbers were depleted in the
peripheral blood, though I would doubt at this low dose...
Aries et al. demonstrate in their study, the lack of response of
granulomatous disease in Wegener's granulomatosis to rituximab. From my
reading of the paper, this agent was given every fourth week. This is not
how this agent is usually prescribed, ie usually given as weekly pulses
for four weeks. Despite this, Bcell numbers were depleted in the
peripheral blood, though I would doubt at this low dose that lymph node
and or granulomatous B cells would have been depleted. This may well be
signficant reason why this study reached a negative conclusion.
It would also have been useful for the authors to have measured human
anti-chimeric antibodies. These can be induced by infliximab, and may well
have affected the efficacy of subsequent rituximab therapy.
In their article: Haematopoietic malignancies in rheumatoid
arthritis: lymphoma risk and characteristics after exposure to tumour
necrosis factor antagonists (and likewise in their subsequent article:
Risks of solid cancers in patients with rheumatoid arthritis and after
treatment with tumour necrosis factor antagonists) Askling et al. compare
occurrence rates for malignancies between different cohorts...
In their article: Haematopoietic malignancies in rheumatoid
arthritis: lymphoma risk and characteristics after exposure to tumour
necrosis factor antagonists (and likewise in their subsequent article:
Risks of solid cancers in patients with rheumatoid arthritis and after
treatment with tumour necrosis factor antagonists) Askling et al. compare
occurrence rates for malignancies between different cohorts of patients
with rheumatoid arthritis.
The authors conclude that patients treated with TNF antagonists in routine
care were not at any additional increased lymphoma risk compared with
patients with RA not treated with TNF antagonists (RR 1.1; 95%CI 0.6-2.1).
They urge us to view the, at most, marginally increased lymphoma risk
associated with TNF antagonists, in the context of higher disease activity
among patients who are offered anti-TNF treatment.
What the authors fail to note however, is that in routine care in
accordance with the labels of the different drugs, a pre-treatment
screening takes place before anti-TNF is prescribed. Rheumatologists have
long been aware of a possible increased risk for malignant lymphoma’s and
other malignancies during anti-TNF use. Product inlays and treatment
protocols also warn doctors and patients for this possible association.
Subsequently, in routine care, patients with current malignancies, a
(recent) history of a malignant disease, or a high suspicion for occult
malignancies are excluded from anti-TNF use. In the other two cohorts
studied, the Inpatient Register cohort and Early Arthritis cohort, such
exclusion is not likely to occur. Therefore the baseline risk for
malignancies might actually have been lower in the anti-TNF cohort than in
the other cohorts, in which case the detected increased lymphoma risk
might not be so marginal. We urge the authors to be careful in
interpreting their results.
Dr Ferrari has suggested that Whiplash Associated Disorders (WAD), is a
systemic illness in the September Annals of Rheumatic Disesases[1],
rather than an anatomically defineable lesion. They correctly quote that
in 90% of patients with WAD no pathology can be found. Dr Ferrari
reaches his conclusion from the acute subjective symptoms that describes
involvement of multiple systems. It is not too surpr...
Dr Ferrari has suggested that Whiplash Associated Disorders (WAD), is a
systemic illness in the September Annals of Rheumatic Disesases[1],
rather than an anatomically defineable lesion. They correctly quote that
in 90% of patients with WAD no pathology can be found. Dr Ferrari
reaches his conclusion from the acute subjective symptoms that describes
involvement of multiple systems. It is not too surprising that multiple
symptoms occur in “multiple trauma”, especially in the first 23 days.
What is very surprising and an extremely important omission, is that the
literature does not routinely report the precise location of multiple
symptoms, and even in this study, neck and shoulder pain are grouped
together.
We along with many others have found that one of the most common
locations of “neck pain” is not in the anatomical neck per se, nor even in
the shoulder, but nearby at the superior medial scapula. Indeed, there
are many reports that Whiplash Associated Disorders localized here are
diagnosed as Trapezius spasm, trigger points, etc. The problem is that in
addition to the absence of pathology, these conditions are generally
intractable to treatment.
We have considered that presentation of pain at this location is
related to the rotator cuff muscle origin, which is found here, and have
described a new cause of “neck pain” originating in the shoulder, which is
eminently treatable.[2] We now believe that many patients with “neck
pain” actually suffer from shoulder trauma, and that an anatomically
defined lesion does frequently exist, in the shoulder. Serendipitously
published at the exact time as our article, Turner et al.[3] showed that
many patients with neck pain had concurrent shoulder symptoms, and
consider this validation of the Referred Shoulder Impingement Syndrome.
Jerrold Gorski MD
Winthrop University Hospital
Mineola, New York
1. A re-examination of the whiplash associated disorders (WAD) as a
systemic illness R Ferrari, A S Russell, L J Carroll, J D Cassidy, Ann
Rhem Dis 2005; 64:1337-1442.
2. Shoulder Impingement Presenting as Neck Pain J M Gorski, L H
Schwartz, J Bone J Surg 85-American 2003;4: 635-638.
3. Impingement syndrome associated with whiplash injury S K Chauhan, T
Peckham, R Turner, J Bone J Surg 85-British 2003;3:408-410.
We very much welcome the thoughts of Drs Geborek and Saxne on our recent article,[1] and agree that a scientific approach should be taken to the synthesis of evidence in decision models. Our analysis is not the last word but rather the first decision analysis of adalimumab. The contention stems from the limitations in the evidence base. In this regard, to wait for all the evidence to be collected (possibly m...
We very much welcome the thoughts of Drs Geborek and Saxne on our recent article,[1] and agree that a scientific approach should be taken to the synthesis of evidence in decision models. Our analysis is not the last word but rather the first decision analysis of adalimumab. The contention stems from the limitations in the evidence base. In this regard, to wait for all the evidence to be collected (possibly many years) before making the policy decision is irrational, but to make the best possible decision with the limited evidence is the objective. To aid this decision, we reflected as far as we found possible the weaknesses of some elements of the evidence base in the uncertainty analysis.
On the specific points raised, our replies are as follows:
Probability of response: The main use of the Swedish observational study[2] in our model was to quantify the response rate to traditional DMARDs in a Swedish population, and the duration patients spend on each treatment. We are well aware of the patient mix issue and where possible attempted to adjust parameters to account for this. We searched the literature for evidence on the rates of ACR and EULAR response for traditional DMARDs in patients who have already failed a number of therapies. Such evidence is limited, as is apparent in the lack of suggested alternative sources from Geborek and Saxne or from the peer reviewers prior to publication. The alternative would be to use the short term ACR response results from the comparator arms of biologic clinical trials. This remains an option, yet we feel sure that the criticism would have been stronger had we used this and ignored the Swedish observational data.
Two further points are worth making. First, the probability of response (adjusted for disease duration) in the traditional DMARDs is assumed equal in all arms of the decision model and therefore to some extent cancels out. Second, the use of the Leflunomide data may over-estimate the probability of response to some of the other traditional DMARDs. If this is true, it would marginally improve the cost-effectiveness of TNF treatments ceteris paribus.
Decision rule for continuation of TNF therapy: Swedish guidelines suggest a consensus between clinician and patient before treatment decisions. To quantify the value of TNFs the policy decision maker needs to estimate the probability of an improvement to a level with which this clinician-patient pair might feel happy to continue therapy. We have used two alternatives i.e. moderate or good DAS28 as a proxy for this (in line with EULAR guidance). Again we would welcome an alternative, ideally evidence based, assumption.
We are clear in the paper that linking ACR20/50 to a moderate/good DAS response is an assumption. We did indeed take into account that all ACR20 responders are included in an ACR50 response. Both the etanercept and infliximab trials did not publish DAS results. We are not advocating the uncritical use of ACR or DAS response in treatment decisions, but what we are saying is that for our policy analysis model we need to make as strong a link as possible between the clinical trial and observational efficacy evidence and the likely use in practice. “Drug adherence is a better measure of continuation in practice” There may be a misunderstanding here because we have used adherence rates from Geborek[2] to quantify the probability of discontinuation each 6 months for patients who have initially responded.
Measuring Utility: Since health utilities were not measured in all the clinical trials, we were forced to use a surrogate outcome. The alternative was to wait for new trials which measured health utilities! The sensitivity analyses actually show that it is unlikely this assumption is going to have a major impact on the results.
HAQ alone has been shown to be a good determinant of utility.[3] Whether DAS, as suggested, contributes above and beyond the HAQ in explaining utility is far from being determined. Kobelt et al.[4], shows HAQ coefficient to remain stable at both time-points of the analysis whilst the DAS coefficient changed substantially. This would indicate some co-linearity between variables which would need further (probably a principal components) analysis to disentangle.
Long-term progression: Annual HAQ progression in post TNF patients is an important variable in the model. However, rather than use the crude figures as you suggest, we reflected the uncertainty in this parameter by forming a probability distribution using the maximum and minimum plausible values. These were the range 0.032 to 0.132 with a modal value of 0.07. Again because the value is assumed equal for patients not responding in both arms of the decision model there is an element of cancelling out and the effect on incremental cost-effectiveness is relatively small.
Sequential Use of TNFs: The fact that in Sweden 44% of patients initiating a biologic drug had previously had another biologic drug re-enforces the need for evidence of sequential biologic use. We recently reviewed the available evidence for the effectiveness of a 2nd biologic and found quality evidence to be sparse. We have undertaken some limited unpublished modelling which confirms the importance of such evidence for decision.
The pharmaceutical companies may well be unlikely to fund large scale clinical trials of sequential biologic use. The need for regulatory or governmental authorities or more local institutions to use both trial and observational registry based research designs (like the SSATG) remains high in order to further inform cost-effective practice with these drugs. This is the first work on adalimumab so it would seem sensible to analyse its use alone before moving onto a sequential strategy.
Indirect Comparison: Geborek and Saxne find indirect comparisons troublesome, as do health economists, medical statisticians and decision makers. That they are ‘almost universal’ is not our statement but reflects a long-term reality, particularly where there is little incentive for pharmaceutical companies to fund comparative studies. Decision science and evidence synthesis are evolving more sophisticated methodologies for such analyses and our model of adalimumab is part of that process, by representing the first probabilistic uncertainty analysis in the health economics of biologics.
It is extremely heartening for us, that our somewhat technical work has been for the most part well understood by eminent clinical colleagues and key critical points made concerning the evidence base. Our aim of transparency in the description of the modelling (every parameter and assumption we used is in the paper) seems to have been achieved. Consistent with economic guidelines,[5] we would like this to become a standard for rheumatology journals so that ‘black box’ models and hidden assumptions become an item of the past.
We would be happy for the authors to contact us with new evidence or indeed expert assumptions on input parameters to quantify the impact on results. The challenges for the clinical and economic research community now are to design data collection exercises to improve the evidence base in those uncertain parameters which could make a difference in policy decision making and to further improve evidence synthesis methodologies so that the full range observational and clinical trial evidence can be integrated to support policy decision making.
References
[1] Bansback NJ, Brennan A, Ghatnekar O. Cost effectiveness of adalimumab in the treatment of patients with moderate to severe rheumatoid arthritis in Sweden. Ann Rheum Dis. 2005 64(7):995-1002
[2] Geborek P, Crnkic M, Petersson IF, Saxne T; South Swedish Arthritis Treatment Group. Etanercept, infliximab, and leflunomide in established rheumatoid arthritis: clinical experience using a structured follow up programme in southern Sweden. Ann Rheum Dis. 2002;61(9):793-8.
[3] Marra CA, Woolcott JC, Kopec JA, Shojania K, Offer R, Brazier JE, Esdaile JM, Anis AH. A comparison of generic, indirect utility measures (the HUI2, HUI3, SF-6D, and the EQ-5D) and disease-specific instruments (the RAQoL and the HAQ) in rheumatoid arthritis. Soc Sci Med. 2005;60(7):1571-82.
[4] Kobelt G, Eberhardt K, Geborek P. TNF inhibitors in the treatment of rheumatoid arthritis in clinical practice: costs and outcomes in a follow up study of patients with RA treated with etanercept or infliximab in southern Sweden. Ann Rheum Dis. 2004;63(1):4-10.
[5] Weinstein MC, O'Brien B, Hornberger J, Jackson J, Johannesson M, McCabe C, Luce BR; ISPOR Task Force on Good Research Practices--Modeling Studies. Principles of good practice for decision analytic modeling in health-care evaluation: report of the ISPOR Task Force on Good Research Practices--Modeling Studies. Value Health. 2003;6(1):9-17.
We read with interest the paper of Eriksson et al.[1], concluding
that infliximab, but not etanercept, induces anti dsDNA antibodies in
patients with rheumatoid arthritis (RA). Since conflicting data[2-6] are
being published about the autoantibody profile during anti TNF-alpha
treatment, we report our experience in a small, but closely followed-up,
cohort of patients with RA, which is not fully conco...
We read with interest the paper of Eriksson et al.[1], concluding
that infliximab, but not etanercept, induces anti dsDNA antibodies in
patients with rheumatoid arthritis (RA). Since conflicting data[2-6] are
being published about the autoantibody profile during anti TNF-alpha
treatment, we report our experience in a small, but closely followed-up,
cohort of patients with RA, which is not fully concordant with the results
obtained by Eriksson et al.[1]
One hundred and three serum samples were obtained from 11 RA patients
(F/M=9/2), of whom 3 received infliximab plus methotrexate (median follow
up 76 weeks - range 34-104, median number of samples per patient 14 -
range 6-14), 4 were given etanercept, alone or in association with
methotrexate (median follow up 38 weeks - range 4-84, median number of
samples per patient 5 - range 2-11), and 4 were switched from one to the
other drug (2 from infliximab to etanercept and 2 viceversa), due to
clinical reasons (unresponsiveness, second course of anti TNF-alpha
blockers). In the last cases, the median follow-up for infliximab
treatment was 52 weeks (range: 14-104) and for etanercept 24 weeks (range
8-52); the median number of samples per patient was 9 (range 3-14) and 4
(range 2-8), respectively. Considering the type of TNF-alpha blocker, we
can split our cases into 2 groups: 7 patients received an infliximab
course and 8 etanercept. As controls, we tested 21 serum samples obtained
from 4 patients with Crohn’s associated spondyloarthropathies (SpA), all
receiving infliximab (median follow up 29 weeks – range 14-46, median
number of sample per patient 4. – range 3-7).
All the sera were tested for: i. anti nuclear antibodies (ANA, by
indirect immunofluorescence – IFL - on HEp-2 cells, with serum diluted
1:40 and a fluorescent polyvalent anti human immunoglobulins as secondary
antibody); ii. anti double stranded DNA (anti dsDNA, by IFL on Chritidia
Luciliae, with serum diluted 1:20 and a fluorescent polyvalent anti human
immunoglobulins as secondary antibody); iii. anti extractable nuclear
antigens (anti ENA by counterimmunoelectrophoresis with an aqueous extract
of acetone powder of rabbit thymus – Peel Freeze, Ak, USA). Most sera were
tested for IgG, IgA and IgM anti dsDNA, anti single stranded DNA (anti
ssDNA), anti histone (AHA) and anti nucleosome antibodies (ANSA) by
different ELISA (Diamedix, Fl, USA). Most sera were also tested for anti
neutrophil cytoplasmic antibodies (ANCA, by IFL on alcohol-fixed human
neutrophils, with a serum dilution 1:20 and a fluoresecent polyvalent anti
human immunoglobulins as secondary antibody). Finally, most RA sera were
also tested for anti cyclic citrullinated peptide (anti CCP), by ELISA
(Axis Shield, UK). Sera exhibiting anti CCP positivity >100 AU were not
titrated.
As to ANA, all but 1 patients (treated with etanercept) either
increased ANA titre or newly developed these antibodies, irrespective of
the drug used and the underlying disease.
Anti dsDNA (by IFL) were transiently induced in 6 out of 7 (86%) RA
patients by infliximab, 4 out of 8 (50%) RA patients by etanercept and 1
out of 4 (25%) SpA patients by infliximab. Two out of the 4 RA patients
who experienced both anti TNF-alpha blockers transiently developed anti
dsDNA with both drugs. Of the remaining 2, one patient developed anti
dsDNA only with infliximab and the other did not produce anti dsDNA
either with infliximab or with etanercept. No patient developed lupus-
like syndrome. Unexpectedly, no IFL anti dsDNA reactivity was confirmed
positive by ELISA, unlike what happens in our experience with systemic
lupus erithematosus or type 1 autoimmune hepatitis sera. On the contrary,
transient ELISA anti dsDNA positivities, mainly of IgM class, were
detected in 3 IFL negative samples. It seems that these antibodies, though
reactive with Crithidia luciliae dsDNA as the lupus-associated antibodies,
are somehow different.
Anti ENA were never detected. Anti ssDNA, AHA, ANSA and both p-ANCA
or c-ANCA were very rarely (<5%) positive in our samples, without any
apparent correlation with the other autoantibodies.
Finally, as to anti CCP, no relevant changes were observed during
therapy with both drugs: negative patients did not become positive, high
titres (>100 IU) patients did not show variation of their titres, while
patients with middle-low titres reactivities fluctuated with small
increases or decreases.
Although obtained in a small group of patients, but with a schedule
(every 8 weeks) specifically tailored to monitor transient phenomena, our
data suggest:
1. TNF-alpha blockade enhances the production of ANA, both in RA and
SpA, with no differences between etanercept and infliximab.
2. Both infliximab and etanercept induce anti dsDNA antibodies in a
high proportion of RA patients (86% and 50%, respectively), without the
emergence of any ANA-related clinical syndrome. Infliximab treated SpA
patients seem to develop these antibodies with a lower (25%) frequency.
3. Anti CCP seropositivity is not significantly influenced by TNF-
alpha inhibition.
They are in line with those of Eriksson et al.[1], with the exception
of the similar behaviour of etanercept and infliximab on autoantibody
induction. The two RA patients, who transiently redeveloped anti dsDNA
after switching from one biologic to the other, support this conclusion.
Other authors,[2] however, have recently reported that etanercept
treatment does not induce any autoantibody production. These differences
may be linked to the higher sensitivity of our close monitoring.
References:
1. Eriksson C, Engstrand S, Sundqvist K-G, Rantapaa-Dahlqvist S.
Autoantibody formation in patients with rheumatoid arthritis treated with
anti-TNF-alpha therapy. Ann Rheum Dis 2005;64:403-7.
2. De Rycke L, Baeten D, Kruithof E, Van den Bosch F, Veys EM, De
Keyser F. Infliximab, but not etanercept, induces IgM anti-double-stranded
DNA autoantibodies as main antinuclear reactivity. Arthritis Rheum
2005;52:2192-201.
3. Charles PJ, Smeenk RJ, De Jong J, Feldmann M, Maini RN. Assessment
of antibodies to double stranded DNA induced in rheumatoid arthritis
patients following treatment with infliximab, a monoclonal antibody to
tumor necrosis factor alpha: findings in open-label and randomized placebo
-controlled trials. Arthritis Rheum 2000;43:2383-90.
4. Bingham SJ, Buch MH, Kerr MA, Emery P, Valadao Barcelos AT.
Induction of antinuclear antibodies in patients with rheumatoid arthritis
treated with infliximab and leflunomide. Arthritis Rheum 2004;50:4072-3.
5. De Rycke L, Verhelst X, Kruithof E, Van den Bosch F, Hoffman IEA
et al. Rheumatoid factor, but not anti citrullinated protein antibodies,
is modulated by infliximab treatment in rheumatoid arthritis. Ann Rheum
Dis 2005;64:299-302.
6. Bobbio-Pallavicini F, Alpini C, Caporali R, Avalle S, Bugatti S,
Montecucco C. Autoantibody profile in rheumatoid arthritis during long-
term infliximab treatment. Arthritis Res Ther 2004;6:264-72.
An association has been made between systemic lupus
erythematosus (SLE) and NHL. A common genetic aetiology is suggested by the
observation of a marked over representation of lymphoma cases in families
with SLE. Likewise, families have been described with both lymphoma and
SLE. A tumour suppressor gene PTEN may link the two disorders via a
defective apoptosis pathway to eliminate hyperactive B and T cells...
An association has been made between systemic lupus
erythematosus (SLE) and NHL. A common genetic aetiology is suggested by the
observation of a marked over representation of lymphoma cases in families
with SLE. Likewise, families have been described with both lymphoma and
SLE. A tumour suppressor gene PTEN may link the two disorders via a
defective apoptosis pathway to eliminate hyperactive B and T cells in SLE. The accumulation of clonally expanded hyperactive B-cells that recognize
self-antigens in the lymph nodes of SLE may predispose these B-cells to
DNA breaks, facilitating tumourigenesis.
Methodological flaws of numerous positive industry-sponsored placebo-controlled clinical trials along with possible publication bias[1]
raised the question of whether orally administered glucosamine therapy
does work in osteoarthritis (OA)[2]. Scepticism increased after more
recent independent investigations failed to demonstrate efficacy[3-6]. In
light of the study by Biggee and colleagues[7] which sho...
Methodological flaws of numerous positive industry-sponsored placebo-controlled clinical trials along with possible publication bias[1]
raised the question of whether orally administered glucosamine therapy
does work in osteoarthritis (OA)[2]. Scepticism increased after more
recent independent investigations failed to demonstrate efficacy[3-6]. In
light of the study by Biggee and colleagues[7] which showed that
insignificant, trace amounts of glucosamine entered human serum after
ingestion of a standard oral dose of glucosamine sulphate (1500 mg), I
feel the actual issue is presently: “Can oral glucosamine really work in
OA?”
Of note, similar concerns may be raised about chondroitin sulphate
(CS). Volpi assessed the bioavailability of CS extracted from bovine[8]
and shark[9] cartilage, the major sources of CS preparations used for the
treatment of OA. Twenty healthy volunteers were given a single oral dose
of 4g CS, and percentage increase and significance with respect to basal
(endogeneous CS) value were calculated in both studies. As such, oral
intake of bovine origin CS resulted in a significant increase in CS plasma
concentrations from the 2nd to the 6th hour. Although CS plasma levels
increased by more than 200% at 2 and 4 hours after CS administration, the
maximum measured plasma concentration (Cmax) was low, ranging from 5.2 to
23.0 microg/ml (mean ± SD: 12.7 ± 4.7 microg/ml).[8] After oral
administration of ichthyic CS, the increase in CS plasma levels was
significant between the 4th and the 16th hour. However, the mean Cmax was
as low as 4.87 ± 2.05 microg/ml (range: 1.5 to 9.7 microg/ml).[9] It
should be stressed that the dose of CS administered to these subjects was
far above the standard doses of CS. In France, for example, CS of bovine
origin has been approved as a prescription drug for OA and its
recommended dosage is 400 mg three times daily. Accordingly, the CS
plasma concentrations obtained in vivo are likely much lower than those
used in in vitro studies reporting that CS could exert beneficial effects
on cartilage metabolism.[10]
Finally, well designed, rigorously controlled clinical trials are
needed to elucidate the actual efficacy of both glucosamine and CS. In
this respect, the results of the National Institutes of Health
Glucosamine/Chondroitin Arthritis Intervention Trial (GAIT) are
impatiently awaited. This large placebo-controlled clinical trial tests
whether glucosamine and chondroitin used separately or in combination are
effective in reducing pain and improving functional ability in patients
with knee osteoarthritis. Furthermore, GAIT includes a sub-study that will
assess whether glucosamine and chondroitin could alter the progression of
knee osteoarthritis.
B Bannwarth
Department of Rheumatology Pellegrin Hospital, and Division of
Therapeutics
Victor Segalen University
33076 Bordeaux
France
bernard.bannwarth@u-bordeaux2.fr
References:
1. McAlindon TE, LaValley MP, Gulin JP, Felson DT. Glucosamine and
chondroitin for treatment of osteoarthritis. A systematic quality
assessment and meta-analysis. JAMA 2000;283:1469-75.
2. Chard J, Dieppe P. Glucosamine for osteoarthritis: magic, hype, or
confusion? BMJ 2001;322:1439-40.
3. Rindone JP, Hiller D, Collacott E, Nordhaugen N, Arriola G.
Randomized, controlled trial of glucosamine for treating osteoarthritis of
the knee. West J Med 2000;172:91-4.
4. Cibere J, Esdaile JM, Thorne A, Singer J, Kopec JA, Canvin J, et
al. Multicenter randomized double-blind placebo-controlled glucosamine
discontinuation trial in osteoarthritis. Arthritis Rheum
2002;46(suppl):S576.
5. Hughes R, Carr A. A randomized, double-blind, placebo-controlled
trial of glucosamine sulphate as an analgesic in osteoarthritis of the
knee. Rheumatology (Oxford) 2002;41:279-84.
6. McAlindon T, Formica M, LaValley M, Lehmer M, Kabbara K.
Effectiveness of glucosamine for symptoms of knee osteoarthritis: results
from an internet-based randomized double-blind controlled trial. Am J Med
2004;117:643-9.
7. Biggee BA, Blin CM, McAlindon TE, Nuite M, Silbert JE. Low levels
of human serum glucosamine after ingestion of glucosamine sulphate
relative to capability of peripheral effectiveness. Ann Rheum Dis 2005 Aug
3 ; [Epub ahead of print].
8. Volpi N. Oral bioavailability of chondroitin sulfate (Condrosulf®)
and its constituents in healthy male volunteers. Osteoarthritis Cartilage
2002;10:768-77.
9. Volpi N. Oral absorption and bioavailability of ichthyic origin
chondroitin sulfate in healthy male volunteers. Osteoarthritis Cartilage
2003;11:433-41.
10. Chan PS, Caron JP, Rosa GJM. Glucosamine and chondroitin sulfate
regulate gene expression and synthesis of nitric oxide and prostaglandin
E2 in articular cartilage explants. Osteoarthritis Cartilage 2005;13:387-
94.
We thank Dr Klauser and colleagues for their comments to our paper
"Ultrasound findings in healthy joints using two different contrast
media".
We would like to take the opportunity to respond to their comments.
Please bear in mind the findings that made us perform this study: In
the literature, the injection of contrast lead to more vessels being
visualised in inflamed joints while no...
We thank Dr Klauser and colleagues for their comments to our paper
"Ultrasound findings in healthy joints using two different contrast
media".
We would like to take the opportunity to respond to their comments.
Please bear in mind the findings that made us perform this study: In
the literature, the injection of contrast lead to more vessels being
visualised in inflamed joints while non-inflamed joints remained without
Doppler activity.[1-3] Opposed to this, two studies showed Doppler
activity in normal joints without the use of contrast agents.[4,5] One
explanation for this discrepancy could be that the latter used
machinery/settings with a higher Doppler sensitivity and our hypothesis
was that we would find Doppler activity in some normal joints before
contrast injection and in more normal joints after contrast injection. We
did not believe that contrast injection was so ideal that it would only
enhance and bring to visualisation flow in RA-joints. Our study supported
our hypothesis.
The appearance of blooming artefact is well know to us and is also
mentioned in the paper page 825 top section. We accept the presence of
blooming artefacts – to avoid blooming artefacts we would need to adjust
Doppler gain to an unacceptably low level, which would lead to non-visualisation of true flow. Another alternative to minimising blooming
artefact is to increase the Doppler frequency. Again an unsatisfactory
approach since it leads to reduced sensitivity. The blooming artefact is
of course present, since we see flow in vessels that we cannot see with
grey-scale imaging! All in all we do not regard blooming artefact as a
problem and accept its presence as a systematic error since it is flow-generated.
When Dr Klauser addresses the issue of blooming artefact it may be to
challenge our findings that were: 1) More joints were colour Doppler
positive after contrast injection, 2) A higher mean colour fraction was
found after contrast injection. Perhaps dr. Klauser implies that the extra
blooming caused by contrast explains our findings. The blooming artefact
cannot explain why a joint without Doppler activity before contrast
injection becomes Doppler-positive after. The elevation in colour fraction
after contrast injection is of course partly due to increased blooming but
as seen in our illustration extra vessels became visible.
Dr Klauser states that bolus injection causes blooming, that blooming
causes bubble destruction before entering synovial vessels and that this
could explain our statement concerning vessel size! Vessel size? Nowhere
in our paper does the word size appear. Most of the synovial vessels we
see with colour Doppler are below the resolution of grey-scale US and
cannot be measured. With Doppler, measurements cannot be trusted due to
blooming artefacts.
Dr Klauser advocates the use of continuous contrast infusion because
it reduces significantly blooming artefacts, is cost-effective, and
because “Using this protocol, increased vascularity detection in early
rheumatoid arthritis joints was found, but no vascularity in healthy
controls”. We find the latter as an indication of low Doppler sensitivity.
With contrast injection they were not able to disclose flow in normal
joints – flow that we see without the use of contrast. We fail to
understand what it is Dr. Klauser questions in our equipment and settings
since our results cannot be called surprising: We saw flow before contrast
injection and more after.
We chose to use a bolus injection for both contrast agents as for
these two to be comparable. To our knowledge, Dr. Klauser and colleagues
are the only group to have used continuous contrast injection in the
evaluation of inflammatory flow.[3] All other studies have used bolus
injection or a slow injection.[1;2;6-8] That is another reason for
choosing bolus injection: the study would be comparable to most of those
already existing.
Dr Klauser shows some concern as to the time spent examining the
joints: First of all we think it is important to remember that this study
is carried out on healthy persons – this means that the positive Doppler
findings are much less compared to those seen in inflammatory conditions.
Bearing this in mind the time used for investigation of each joint is less
than for joints with pathology. Dr. Klauser and colleagues also mentions
the use of 3 RI measurements which also we recommend to be used for a mean
RI. When possible, as we state in methods, we obtained a mean RI. However,
only seldom did we obtain three measurements for lack of vessels. We did
not have difficulties with the time. The investigator has 23 years
experience with full time ultrasound. Our scanning protocol was first to
evaluate the joints in a predefined order for colour Doppler activity (our
highest priority) and then return to the colour Doppler positive joints
for the RI measurements (that were performed after the critical 60 seconds
mentioned by dr. Klauser). The reason for measuring the RI was twofold: 1)
Obtaining the Doppler spectrum proved that we were seeing true flow and
not some artefact, and 2) The vessels we were seeing had a high (normal)
RI.
Dr. Klauser worries about microbubble destruction in our study. We fail to
see that there is any indication of this in our findings. We do find more
vascularity after contrast injection and with our RI-measurements we know
that we are not seeing artefacts.
Dr. Klauser argues that our hypothesis (as to the behaviour of low-
end and high-end equipment) is only a hypothesis and not based on data. We
can only agree. We have not investigated the behaviour of low-end
equipment. That is why it is a hypothesis as also stated in the paper!
We find that the one of our core statements was not understood.
Namely, that in rheumatology, resources may probably be better spent on
improving the Doppler sensitivity (acquiring high-end equipment) than on
injecting contrast agents (expensive and invasive) both serving the
purpose of seeing more vascularity. The methodological limitations
mentioned by dr. Klauser should (if present at all) have resulted in poor
sensitivity after contrast injection, which was not the case in our study!
Respectfully
Lene Terslev
Søren Torp-Pedersen
Michael Bachmann Nielsen
References:
1. Szkudlarek M, Court-Payen M, Strandberg C, Klarlund M, Klausen
T, Ostergaard M. Contrast-enhanced power Doppler ultrasonography of the
metacarpophalangeal joints in rheumatoid arthritis. Eur Radiol 2003;
13:163-168.
2. Magarelli N, Guglielmi G, Di Matteo L, Tartaro A, Mattei PA,
Bonomo L. Diagnostic utility of an echo-contrast agent in patients with
synovitis using power Doppler ultrasound: a preliminary study with
comparison to contrast-enhanced MRI. Eur Radiol 2001; 11(6):1039-1046.
3. Klauser A, Frauscher F, Schirmer M, Halpern E, Pallwein L,
Herold M et al. The value of contrast-enhanced color Doppler ultrasound in
the detection of vascularization of finger joints in patients with
rheumatoid arthritis. Arthritis Rheum 2002; 46(3):647-653.
4. Hau M, Schultz H, Tony HP, Keberle M, Jahns R, Haerten R et al.
Evaluation of pannus and vascularization of the metacarpophalangeal and
proximal interphalangeal joints in rheumatoid arthritis by high-resolution
ultrasound (multidimensional linear array). Arthritis Rheum 1999;
42(11):2303-2308.
5. Terslev L, Torp-Pedersen S, Qvistgaard E, Von der Recke P,
Bliddal H. Doppler Ultrasound Findings In Healthy Wrists And Finger
Joints. Ann Rheum Dis 2004; 63:644-648.
6. Qvistgaard E, Rogind H, Torp-Pedersen S, Terslev L, Danneskiold-
Samsoe B, Bliddal H. Quantitative ultrasonography in rheumatoid arthritis:
evaluation of inflammation by Doppler technique. Ann Rheum Dis 2001;
60(7):690-693.
7. Carotti M, Salaffi F, Manganelli P, Salera D, Simonetti B,
Grassi W. Power Doppler sonography in the assessment of synovial tissue of
the knee joint in rheumatoid arthritis: a preliminary experience. Ann
Rheum Dis 2002; 61(10):877-882.
8. Salaffi F, Carotti M, Manganelli P, Filippucci E, Giuseppetti
GM, Grassi W. Contrast-enhanced power Doppler sonography of knee synovitis
in rheumatoid arthritis: assessment of therapeutic response. Clin
Rheumatol 2004; 23(4):285-290.
Methodological problems in the detection of IgG antibodies to glucose
-6-phosphate isomerase in serum and synovial fluid from patients with
inflammatory arthritis.
Examination of the methods used by M Schaller et al in their article
on the detection of IgG antibodies to glucose-6-phosphate isomerase (GPI)
in patients with inflammatory arthritis (1) revealed serious deficiencies.
Methodological problems in the detection of IgG antibodies to glucose
-6-phosphate isomerase in serum and synovial fluid from patients with
inflammatory arthritis.
Examination of the methods used by M Schaller et al in their article
on the detection of IgG antibodies to glucose-6-phosphate isomerase (GPI)
in patients with inflammatory arthritis (1) revealed serious deficiencies.
The authors expressed the results in micrograms/ml of IgG antibodies
to GPI in the studied serums and synovial fluids. To calculate the
micrograms/ml of antibodies to GPI from the measured optical densities,
the authors used the linear range of a standard curve on the same ELISA
plate. The standard curve was constructed by coating the wells with
serial dilutions of a human serum with known concentrations of total IgG,
ranging from 0.02 to 5 micrograms/ml in the dilutions. The optical
densities measured for the serial dilutions of the standard serum were
plotted against the amount of IgG in the dilution for each data point.
Unfortunately, the authors used the whole serum with many other proteins
which competed for binding to the wells of the ELISA plate. Thus, the
amount of IgG that actually bound to the wells was not known. Using the
calculated amount of IgG in the used dilutions of serum for the
construction of the standard curve is wrong and renders the reported data
useless. As an example, if only a tenth of the IgG present in the diluted
serum bound to the wells, the calculated amount of the antibodies to GPI
would be one tenth of the reported values. The same problem renders the
calculations of IgG subclass antibodies to GPI useless.
Another concern is that using serum and synovial fluid specimens the
authors did not use a blank for each specimen. The use of a blank is
essential for testing the presence of any antibodies in the serum or
synovial fluid of patients with RA or with other autoimmune diseases owing
to the possible nonspecific binding of IgG or of immune complexes
containing IgG to the wells or to the blocking agent used. This point is
well illustrated by the authors in Figure 1A. The binding of IgG from one
patient with RA to rabbit or human GPI is only a little higher than IgG
binding to bovine serum albumin, which was used as a blocking agent in all
ELISA wells.
In conclusion, the reported quantities of IgG antibodies to GPI in
the tested serum and synovial fluid specimens are wrong.
Reference
1. Schaller M, Stohl W, Tan S M, Benoit V M, Hilbert D M, Ditzel H J.
Raised levels of anti-glucose-6-phosphate isomerase IgG in serum and
synovial fluid from patients with inflammatory arthritis. Ann Rheum Dis;
64:743-749.
The article "Cost effectiveness of adalimumab in the treatment of
patients with moderate to severe rheumatoid arthritis in Sweden" by
Bansbeck, Brennan, and Ghatnekar [1] deals with an important subject, i.e.
is it justifiable from a cost-effectiveness point of view to use tumour
necrosis factor (TNF) inhibitors in rheumatoid arthritis (RA).
We acknowledge that the authors have used data from...
The article "Cost effectiveness of adalimumab in the treatment of
patients with moderate to severe rheumatoid arthritis in Sweden" by
Bansbeck, Brennan, and Ghatnekar [1] deals with an important subject, i.e.
is it justifiable from a cost-effectiveness point of view to use tumour
necrosis factor (TNF) inhibitors in rheumatoid arthritis (RA).
We acknowledge that the authors have used data from our publications
(references 14,15 in the article). We are concerned when these data are
used in a context that they were not intended for. The authors draw wide
ranging conclusions on adalimumab cost-effectiveness from results in our
observational study describing early experience of etanercept, infliximab,
and leflunomide but not of adalimumab. We were very clear in pointing out
that the patients treated with these remedies in southern Sweden at that
time were not randomly selected, but were the most severely affected and
difficult to treat patients that had tried almost all treatment modalities
in the pre-TNF blocking years. On the other hand we agree with the
authors, that observational data may mirror clinical reality more closely,
and is complimentary to randomised control trials, which represent a
different patient mix. However, the uncritical modelling of data from
different data sets without accounting for differences in the patient mix
raises our concern.
From our perspective as rheumatologists we would like to raise
several points of concern regarding the study:
Although it is reasonably correct to state that an ineffective
treatment in RA patients is replaced by a new treatment remedy, this has
not been formally investigated. In contrast, it is not true that treatment
decisions in Sweden rely on changes in DAS28 and fulfilment of a specific
response criteria set. This is not stipulated in the guidelines issued
from Swedish Rheumatology Society. Instead treatment cessation relies on
clinician’s and patient’s consensus (be it due to adverse events or
inadequate response).
It is questionable to equate ACR20=EULAR moderate responders and
ACR50=EULAR good responders. The authors may possibly not take into
account that all ACR50 responders are included in the ACR20 responders,
whereas EULAR moderate responders are separate from EULAR good responders.
We have investigated how different response criteria set perform in our
setting and warned against their uncritical use in treatment decisions in
clinical practice [2]. We believe that drug adherence is a better measure
of drug continuation in clinical practice.
The systematic translation of a certain ACR response into a
specific HAQ improvement needs validation, as well as the translation of a
DAS score into a specific HAQ score (table 2 in the article). These
instruments have been developed for quite different purposes and measures
different aspects of the disease. We actually found that HAQ alone was not
sufficiently exact as utility and economic outcome measure in our
particular patient setting [3].
Annual HAQ progression of 0.132 in post anti-TNF treated as opposed
to 0.03 in other RA patients seems little substantiated but the almost 4-
fold increased HAQ progression rate has huge impact on long term estimated
functional and economic outcome in patients stopping anti-TNF drugs in the
model and therefore favours such drug continuation.
It is not likely that three disease modifying drugs (DMARDs) will
be used after failure of one biologic drug and no further biologic agent
will be tried. The use and performance of post-biologic DMARD treatment is
not known with any reasonable certainty but real life experience in
southern Sweden shows that 44 % of initiation of a biologic drug in 2003
occured in patients previously being treated with another such drug [4].
We seriously doubt that our relatively restricted leflunomide
adherence and efficacy data can be used as a surrogate for all post-TNF
blocking DMARD outcomes.
Finally we find the statement "indirect comparisons are almost
universal in health economic analyses" troublesome. From a clinician´s
perspective we think that health economic analyses in RA should always be
done with intimate knowledge of the patient cohort on which the
calculations are based. Any projection and modelling in other settings
must be carried out with great care in close collaboration between
scientists trained in health economics, which the authors are, and
rheumatologists. We thoroughly agree with Symmons that this type of
investigation is a legitimate scientific endeavour and should therefore
always be submitted to critical scientific appraisal [5].
References
1. Bansback NJ, Brennan A, Ghatnekar O. Cost effectiveness of
adalimumab in the treatment of patients with moderate to severe rheumatoid
arthritis in Sweden
Ann Rheum Dis 2005;64:995-1002.
2. Gülfe A, Saxne T, Geborek P. Response criteria for rheumatoid
arthritis in clinical practice: how useful are they? Ann Rheum Dis
2005;64:1186-1189
3. Kobelt G, Eberhardt K, Geborek P. TNF-Inhibitors in the Treatment
of Rheumatoid Arthritis in Clinical Practice. Costs and outcomes in 1 year
follow-up of patients with rheumatoid arthritis treated with etanercept or
infliximab in Southern Sweden. Ann Rheum Dis 2004;63:4-10.
4. Geborek P, Leden I, Nitelius E, Noltorp S, Petri H, Jacobsson L,
Larsson L, Saxne T. Regional population based studies of biological anti-
rheumatic drug use in southern Sweden validated against pharmaceutical
sales. Ann Rheum Dis in press
5. Symmons DPH. Anti-tumour necrosis factor therapy: can we afford
it? Ann Rheum Dis 2005; 64: 969-970.
Dear Editor
Aries et al. demonstrate in their study, the lack of response of granulomatous disease in Wegener's granulomatosis to rituximab. From my reading of the paper, this agent was given every fourth week. This is not how this agent is usually prescribed, ie usually given as weekly pulses for four weeks. Despite this, Bcell numbers were depleted in the peripheral blood, though I would doubt at this low dose...
Dear Editor
In their article: Haematopoietic malignancies in rheumatoid arthritis: lymphoma risk and characteristics after exposure to tumour necrosis factor antagonists (and likewise in their subsequent article: Risks of solid cancers in patients with rheumatoid arthritis and after treatment with tumour necrosis factor antagonists) Askling et al. compare occurrence rates for malignancies between different cohorts...
Dear Editor,
Dr Ferrari has suggested that Whiplash Associated Disorders (WAD), is a systemic illness in the September Annals of Rheumatic Disesases[1], rather than an anatomically defineable lesion. They correctly quote that in 90% of patients with WAD no pathology can be found. Dr Ferrari reaches his conclusion from the acute subjective symptoms that describes involvement of multiple systems. It is not too surpr...
Dear Editor
We very much welcome the thoughts of Drs Geborek and Saxne on our recent article,[1] and agree that a scientific approach should be taken to the synthesis of evidence in decision models. Our analysis is not the last word but rather the first decision analysis of adalimumab. The contention stems from the limitations in the evidence base. In this regard, to wait for all the evidence to be collected (possibly m...
Dear Editor,
We read with interest the paper of Eriksson et al.[1], concluding that infliximab, but not etanercept, induces anti dsDNA antibodies in patients with rheumatoid arthritis (RA). Since conflicting data[2-6] are being published about the autoantibody profile during anti TNF-alpha treatment, we report our experience in a small, but closely followed-up, cohort of patients with RA, which is not fully conco...
Dear Editor,
An association has been made between systemic lupus erythematosus (SLE) and NHL. A common genetic aetiology is suggested by the observation of a marked over representation of lymphoma cases in families with SLE. Likewise, families have been described with both lymphoma and SLE. A tumour suppressor gene PTEN may link the two disorders via a defective apoptosis pathway to eliminate hyperactive B and T cells...
Dear Editor,
Methodological flaws of numerous positive industry-sponsored placebo-controlled clinical trials along with possible publication bias[1] raised the question of whether orally administered glucosamine therapy does work in osteoarthritis (OA)[2]. Scepticism increased after more recent independent investigations failed to demonstrate efficacy[3-6]. In light of the study by Biggee and colleagues[7] which sho...
Dear Editor,
We thank Dr Klauser and colleagues for their comments to our paper "Ultrasound findings in healthy joints using two different contrast media".
We would like to take the opportunity to respond to their comments.
Please bear in mind the findings that made us perform this study: In the literature, the injection of contrast lead to more vessels being visualised in inflamed joints while no...
Dear Editor
Methodological problems in the detection of IgG antibodies to glucose -6-phosphate isomerase in serum and synovial fluid from patients with inflammatory arthritis.
Examination of the methods used by M Schaller et al in their article on the detection of IgG antibodies to glucose-6-phosphate isomerase (GPI) in patients with inflammatory arthritis (1) revealed serious deficiencies.
The a...
Dear Editor
The article "Cost effectiveness of adalimumab in the treatment of patients with moderate to severe rheumatoid arthritis in Sweden" by Bansbeck, Brennan, and Ghatnekar [1] deals with an important subject, i.e. is it justifiable from a cost-effectiveness point of view to use tumour necrosis factor (TNF) inhibitors in rheumatoid arthritis (RA).
We acknowledge that the authors have used data from...
Pages